MONOGRAPHS JOURNAL OF THE NATIONAL CANCER INSTITUTE Number 23 ISSN 0027-8874 ISBN 0-19-922371-8 EDITORIAL BOARD Barnett S. Kramer Editor-in-Chief J. Gordon McVie European Editor Eric J. Seifter Book Reviews Editor J. Paul Van Nevel News Editor Frederic J. Kaye Douglas L. Weed Reviews Editors Martin L. Brown Economics Editor ASSOCIATE EDITORS Susan G. Arbuck Frank M. Balis William J. Blot Peter M. Blumberg John D. Boice, Jr. Louise A. Brinton Bruce A. Chabner Ross C. Donehower Susan S. Ellenberg Suzanne W. Fletcher Michael A. Friedman Patricia A. Ganz John K. Gohagan Frank J. Gonzalez Michael M. Gottesman Peter Greenwald Donald E. Henson Susan M. Hubbard Colin R. Jefcoate Frederic J. Kaye Hynda K. Kleinman Theodore S. Lawrence W. Marston Linehan Marc E. Lippman Scott M. Lippman Darrell T. Liu Dan L. Longo Reuben Lotan Douglas R. Lowy Susan G. Nayfield David L. Nelson Kenneth Olden David G. Poplack Ross L. Prentice Alan S. Rabson Edward A. Sausville Robert H. Shoemaker David Sidransky Richard M. Simon Michael B. Sporn Maryalice Stetler-Stevenson J. Paul Van Nevel Douglas L. Weed Noel S. Weiss Bernard Levin EDITORIAL ADVISORY BOARD Hans-Olov Adami Harry B. Burke Pelayo Correa Ethan Dmitrovsky Merrill J. Egorin Soldano Ferrone Isaiah J. Fidler Edward L. Giovannucci Jean LL. Grem Curtis C. Harris Richard D. Klausner Ilan R. Kirsch Kenneth H. Kraemer Alexandra M. Levine Martha S. Linet Henry T. Lynch Frank L. Meyskens Anthony B. Miller Malcolm S. Mitchell David L. Page Kenneth J. Pienta Emily L. Harris Kathy J. Helzlsouer Maureen M. Henderson Gloria H. Heppner Allan Hildesheim Waun Ki Hong William J. Hoskins Peter J. Houghton David H. Johnson V. Craig Jordan Director, National Cancer Institute Susan Molloy Hubbard Director, International Cancer Information Center Julianne Chappell Chief, Scientific Publications Branch; Executive Editor 1998 STATISTICAL EDITORS Janet W. Andersen Jacques Benichou Donald A. Berry Barry W. Brown Bernard F. Cole Susan S. Ellenberg Scott S. Emerson Eric Feuer Edmund A. Gehan Barry I. Graubard Sylvan B. Green Susan G. Groshen Richard A. Olshen Philip C. Prorok Timothy R. Rebbeck Philip S. Rosenberg Harland N. Sather Daniel J. Schaid Richard M. Simon Donald M. Stablein Robert E. Tarone Sholom Wacholder David Schottenfeld Richard K. Severson Jerry W. Shay Sandra M. Swain G. Marie Swanson Mario Sznol Raymond Taetle Ian Tannock Peter R. Twentyman EDITORIAL STAFF Scientific Editor: Ellen Feigal, M.D. Monograph Coordinator: Elizabeth Horowitz Manuscript Editors: Elaine Price Beck Joan O’Brien Rodriguez Elizabeth Horowitz Editorial Assistant: Jamia V. Capehart MARKETING STAFF Marketing Director: Jean Griffin Baum Press Contact: Ellen Scott EDITORIAL POLICY: Manuscripts from key conferences dealing with cancer and closely related research fields, or a related group of papers on specific subjects of importance to cancer research, are considered for publication, with the understanding that they have not been published previously and are submitted exclusively to the Journal of the National Cancer Institute Monographs. All material submitted for consideration will be subject to review, when appropriate, by at least one outside reviewer and one member of the Editorial Board of the Journal of the National Cancer Institute. Opinions expressed by the authors are not necessarily those of the publisher or the editors. Proposals for monographs should be submitted to the Editor-in-Chief, Journal of the National Cancer Institute, National Cancer Institute, P.O. Box 31110, Bethesda, MD 20824-1110. Journal of the National Cancer Institute Monographs are available on request to members of the National Cancer Institute Information Associates Program (one copy of each monograph per member). Monographs are also available from Oxford University Press, Journals Subscription Department, 2001 Evans Road, Cary, NC 27513 USA. Telephone toll-free in the U.S. and Canada, 1-800-852-7323, or 919-677-0977. Fax: 919-677-1714. E-mail: jnlorders @oup-usa.org. © Oxford University Press First National AIDS Malignancy Conference Proceedings of a Conference Held at the National Institutes of Health Bethesda, Maryland April 28-30, 1997 Conference Sponsor National Cancer Institute Richard Klausner, M.D. Director First National AIDS 1998 Malignancy Conference Number 23 Contents First National AIDS Malignancy Conference Monograph: Introduction Ellen Feigal Overview of the Epidemiology of Immunodeficiency-Associated Cancers Valerie Beral, Robert Newton Current Perspectives on the Molecular Pathogenesis of Virus-Induced Cancers in Human Immunodeficiency Virus Infection and Acquired Immunodeficiency Syndrome Elliott Kieff Human Papillomavirus Infection and Anogenital Neoplasia in Human Immunodeficiency Virus-Positive Men and Women Joel M. Palefsky Acquired Immunodeficiency Syndrome-Related Cancers: the Community Perspective Michael Marco Association of Non-Acquired Immunodeficiency Syndrome-Defining Cancers With Human Immunodeficiency Virus Infection Charles S. Rabkin Papillomaviruses and Cervical Cancer: Pathogenesis and Vaccine Development Douglas R. Lowy, John T. Schiller Cancers in Human Immunodeficiency Virus-Infected Children Brigitta U. Mueller Hodgkin’s Disease in the Setting of Human Immunodeficiency Virus Infection Alexandra M. Levine Management of Cervical Neoplasia in Human Immunodeficiency Virus-Infected Women Mitchell Maiman Human Herpesvirus Type 8 and Kaposi’s Sarcoma Robin A. Weiss, Denise Whitby, Simon Talbot, Paul Kellam, Chris Boshoff Some Aspects of the Pathogenesis of HIV-1-Associated Kaposi’s Sarcoma Robert C. Gallo Clinical Overview: Issues in Kaposi’s Sarcoma Therapeutics Susan E. Krown Kaposi’s Sarcoma-Associated Herpesvirus-Encoded Oncogenes and Oncogenesis Patrick S. Moore, Yuan Chang Human Herpesvirus 8—the First Human Rhadinovirus Frank Neipel, Jens-Christian Albrecht, Bernhard Fleckenstein Novel Organizational Features, Captured Cellular Genes, and Strain Variability Within the Genome of KSHV/HHVS John Nicholas, Jian-Chao Zong, Donald J. Alcendor, Dolores M. Ciufo, Lynn J. Poole, Robert T. Sarisky, Chuang-Jiun Chiou, Xiaoqun Zhang, Xiaoyu Wan, Hong-Guang Guo, Marvin S. Reitz, Gary S. Hayward Immunotherapy for Epstein-Barr Virus-Associated Cancers Cliona M. Rooney, Marie A. Roskrow, Colton A. Smith, Malcolm K. Brenner, Helen E. Heslop Genetic Basis of Acquired Immunodeficiency Syndrome-Related Lymphomagenesis Gianluca Gaidano, Antonino Carbone, Riccardo Dalla-Favera Clinical Management of Human Immunodeficiency Virus-Associated Non-Hodgkin’s Lymphoma Lawrence D. Kaplan vii 31 37 43 51 55 59 65 73 79 89 95 101 First National AIDS Malignancy Conference Monograph: Introduction Ellen Feigal* Scientists, physicians, health care workers, and community and patient advocates from around the world gathered at the first national forum focused on acquired immunodeficiency syn- drome (AIDS)-associated cancers in April 1997 on the grounds of the National Institutes of Health campus in Bethesda, MD. The first National AIDS Malignancy Conference, sponsored by the National Cancer Institute, featured multidisciplinary presen- tations on the epidemiology, biology, virology, immunology, and treatment of cancers in the setting of human immunodefi- ciency virus (HIV) infection. Cancers have been associated with HIV since the beginning of the epidemic in the late 1970s. With almost 1 million individuals infected with HIV across the United States and 10 million infected worldwide, AIDS-associated can- cers offer a unique perspective on the interplay of viruses, im- mune dysregulation, and cancer pathogenesis. This monograph is a comprehensive compilation of papers from 18 plenary speakers. Overview of Cancers, Viruses, and Immunodeficiency Dr. Valerie Beral opened the conference with a presentation of the epidemiologic evidence on the effect of immunosuppres- sion on cancer risk. Immunodeficiency, whether congenital, therapeutic, or infectious in origin, increases the risk of certain, but not all, types of cancer. A common feature of these cancers is that specific infectious agents appear to be important in their etiology. The study of cancer in immunodeficient populations offers a unique opportunity to investigate the role of the immune system in controlling the development, growth, and dissemina- tion of tumors. Such studies have already contributed substan- tially to knowledge about the role of infectious agents in human cancer. Given the strategic role that infectious agents may have in the etiology of these cancers, Dr. Elliott Kieff discussed the mo- lecular mechanisms by which human papillomavirus (HPV), Ep- stein-Barr virus (EBV), human herpesvirus type 8 (HHV-8), and HIV persist and effect changes in cell growth that result in malignancy. Pharmacologic strategies to inhibit the mechanisms by which these viruses cause persistent infection and cell growth transformation may be useful in preventing and treating these cancers. Those approaches may also be useful in considering prevention and treatment of the underlying HIV and other op- portunistic infections. The focus of the overview talks next centered on a discussion of the epidemiologic role of HPV in the development of ano- genital neoplasia in both men and women, the higher incidence of HPV in the HIV-infected population, and clinical features of the disease in the HIV-infected as contrasted to the immuno- competent individual. Dr. Joel M. Palefsky postulated that the Journal of the National Cancer Institute Monographs No. 23, 1998 higher incidence of HPV may reflect loss of systemic immune response to HPV antigens or local HPV-HIV interactions at the tissue or cellular level. Activist Community The AIDS activist community is a source of expertise on how science and clinical medicine interact with politics and policy. Activists have had an enormous impact on how the research community shapes its directions, and they have opened the doors for other community and patient advocates to be involved in the decision-making process that ultimately has an impact on pa- tients and individuals at risk for the disease. In his plenary talk, Mr. Michael Marco challenged the National Cancer Institute, the Food and Drug Administration, and the biotechnology and phar- maceutical industries to become more engaged in the basic and clinical scientific communities. A primary emphasis of his pre- sentation was asking how the National Cancer Institute can as- sure that patients with AIDS and cancer will receive the best medical care for their cancer as well as for their underlying HIV infection. Non-AIDS-Defining, HPV-Associated, and Pediatric Cancers A frequently asked question is whether new cancers are emerging in the HIV-infected population. Dr. Charles S. Rabkin examined evidence on whether there are age-related as well as geographic variations in their prevalence. Varying levels of evi- dence link several additional neoplasms to HIV infection, in- cluding Hodgkin's disease and anogenital intraepithelial neopla- sia, although invasive disease is still uncertain for both cervical and anal cancers. Leiomyosarcoma and benign leiomyomas are increased in HIV-infected children, but they are unusual in HIV- infected adults. Conjunctival cancer is seen in HIV-infected in- dividuals from sub-Saharan Africa, but it is uncommon in West- ern countries. The strong epidemiologic data linking particular subsets of HPV to anogenital disease, combined with knowledge gained from molecular biology and immunology, have spurred recent efforts that have focused on the development of vaccines against HPV. Dr. Douglas R. Lowy raised the possibility that virus-like particles, resembling native HPV capsids, might be effective immunogens for a prophylactic HPV vaccine. Dr. Mitchell Maiman provided his insight on the diagnosis and management *Affiliation of author: Division of Cancer Treatment and Diagnosis, National Cancer Institute, Bethesda, MD. Correspondence to: Ellen Feigal, M.D., National Institutes of Health, Bldg. 31, Rm. 3A44, Bethesda, MD 20892. vii of cervical intraepithelial neoplasia and cervical cancer in HIV- infected women. Children with HIV infection have a lower incidence of cancer than adults with HIV infection, but they appear to have unusual and extremely rare tumors, such as leiomyosarcomas and leio- myomas, and a high prevalence of lymphoproliferative disor- ders. Dr. Brigitta U. Mueller noted that there are approximately 130 cases of cancer per million non-HIV-infected children (0.013%) per year. A conservative estimate is that children with HIV infection appear to have at least a 100-fold higher incidence of cancers. The need for oncologic and infectious disease ex- pertise in treating these types of uncommon tumors was stressed. Hodgkin's disease represents another cancer that is not AIDS- defining, although its clinical and pathologic characteristics dif- fer from those in the immunocompetent setting. Dr. Alexandra M. Levine compared and contrasted the characteristics of Hodg- kins disease in the immunocompetent and immunodeficient set- tings. Kaposi’s Sarcoma (KS): KSHV/HHV-8, Inflammatory Cytokines, and Treatment In the next series of overviews, Dr. Robin A. Weiss set the scene with a presentation on the epidemiology, current serologic tests, and various KS-associated herpesvirus (KSHV)/HHV-8- encoded proteins that may play a role in the promotion of cel- lular growth. Later in the conference, molecular biologists ana- lyzed the viral genome and described how it replicates and functions. Challenging prevailing views on KSHV/HHV-8 and its etio- logic relationship to disease, Dr. Robert C. Gallo suggested that the role of HIV-1 is more active in the KS disease process than simply promoting immunodeficiency. He suggested that HIV-1 acts directly by promoting an increase in inflammatory cyto- kines, which, through sustained release, influences early stage KS by inciting local micro-inflammatory responses and affects growth of the inflammatory cells through the Tat protein. The clinical aspects of KS were presented by Dr. Susan E. Krown. Four questions critical to the development of improved therapeutic and prophylactic strategies included the following: 1) Can we identify risk factors and predict who will develop KS? 2) Can we translate hypotheses about pathogenesis into im- proved therapeutic or prophylactic strategies, e.g., target new blood vessel development or inhibit inflammatory cytokines? 3) How does improved anti-HIV therapy have an impact on KS treatment? 4) How can we best evaluate the benefit of therapy beyond conventional measures of tumor response? KSHV/HHV-8 Symposium The KSHV/HHV-8 symposium was led by Dr. Patrick S. Moore, co-discoverer with Dr. Yuan Chang and colleagues of KSHV/HHV-8. Molecular biologic studies of KSHV have iden- tified a number of potential oncogenes that may contribute to neoplasia. With the publication of the full-length genomic se- quence, new opportunities to investigate its molecular biology and virology are becoming available. This presentation provided viii a road map for subsequent papers in this symposium on the major features of the KSHV genome and its potential oncogenes. The virus is already providing unique insights into tumorigen- esis and may serve as an important model for virus-induced oncogenesis. The first known human member of the genus Rhadinovirus is HHV-8/KSHV. Dr. Bernhard Fleckenstein noted that acquisition of genes from the host cell genome is a common feature of most herpesviruses and of rhadinoviruses in particular. Although dif- ferent rhadinoviruses acquire different host-cell genes, their pu- tative functions apparently converge to achieve three common goals: 1) enhance DNA replication independent from the cell cycle, 2) expand the pool of infectable cells, and 3) counteract the host’s responses to infection. Dr. Gary S. Hayward delved into the HHV-8 genome struc- ture and strain differences. These differences would provide the opportunity to address the origins, geographical preferences, transmissibility, and disease association. Lymphomas: EBV, Molecular Pathogenesis, and Treatment EBV-associated lymphoproliferative disease is a frequently fatal complication of organ transplantation and HIV infection. Dr. Cliona M. Rooney presented her study on the safety and efficacy of adoptively transferred, gene-marked. virus-specific cytotoxic T lymphocytes as prophylaxis and treatment of EBV- associated lymphoproliferative disease and its implications for HIV-related cancers. Molecular pathogenesis of AIDS-associated non-Hodgkin's lymphoma is a complex process involving both host factors and the accumulation of genetic lesions within the tumor clone. Dr. Riccardo Dalla-Favera reviewed the pattern of molecular lesions in these tumors and several distinct pathogenic pathways in AIDS-related lymphomagenesis. These pathways selectively as- sociate with the different clinicopathologic variants of AIDS- associated non-Hodgkin’s lymphoma. On the basis of the dif- ferences in molecular characteristics and presumed mechanisms of pathogenesis among the AIDS-associated non-Hodgkin's lymphomas and given the relatively modest survival of patients treated with conventional cytotoxic chemotherapy, Dr. Lawrence D. Kaplan suggested that development of future thera- peutic approaches should take advantage of some of these unique molecular characteristics. This monograph is just a taste of the energizing atmosphere of the first National AIDS Malignancy Conference. Multidisci- plinary, interactive discussions took place during the abstract presentations and poster sessions. The Second National AIDS Malignancy Conference occurred on April 6-8, 1998. Next-day summaries for those who were unable to attend the conference are available on http://www healthcg.com. I would like to acknowledge and thank Dr. Richard D. Klaus- ner and Dr. Robert Wittes for their guidance and support and Dr. James Goedert and the other members of the 16-person program steering committee for their time and effort in the organization of this conference. Journal of the National Cancer Institute Monographs No. 23, 1998 Overview of the Epidemiology of Immunodeficiency-Associated Cancers Valerie Beral, Robert Newton* Immunodeficiency, be it congenital, therapeutic, or infec- tious in origin, increases the risk of certain, but not all, types of cancer. A common feature of these cancers is that specific infectious agents appear to be important in their etiology, not only in immunodeficient subjects but also in the general population. People with acquired immunodeficiency syn- drome (AIDS) are at an increased risk of Kaposi’s sarcoma, non-Hodgkin’s lymphoma, Hodgkin’s disease, squamous cell carcinoma of the conjunctiva, and childhood leiomyosarco- ma. It is striking that most of these cancers have been asso- ciated with specific human herpesvirus (HHV) infections: HHYV-8 with Kaposi’s sarcoma and the closely related Ep- stein-Barr virus with non-Hodgkin’s lymphoma, Hodgkin’s disease, and possibly also with childhood leiomyosarcoma. Moreover, similar associations between these viruses and cancer have been found, albeit inconsistently, in people who are not immunosuppressed. Further research is needed to establish whether the risk of other cancers is also increased in people with AIDS, although, if so, the cancers are likely to be rare or to have comparatively small associated relative risks. Existing evidence suggests that there may be no marked increase in the risk of two common cancers that are known to be caused by infectious agents—hepatocellular carcinoma and invasive carcinoma of the uterine cervix. The apparent lack of an increase in invasive cervical cancer is unexpected and needs further investigation, especially since the prevalence of cervical infection with human papilloma- viruses and of low-grade preneoplastic changes in the cervi- cal epithelium is increased in women with AIDS. With the prospect of improved survival in people with AIDS, the ef- fect of immunosuppression on cancer is likely to become an increasingly important issue. [Monogr Natl Cancer Inst 1998;23:1-6] The study of cancer in immunodeficient populations offers a unique opportunity to investigate the role of the immune system in controlling the development, growth, and dissemination of tumors. Such studies have already contributed substantially to knowledge about the role of infectious agents in human cancer. This article reviews the epidemiologic evidence about the effect of immunosuppression on cancer risk. Immunodeficiency and Cancer: The Evidence Before the Acquired Immunodeficiency Syndrome (AIDS) Epidemic It has long been thought that the immune system plays a vital role in the etiology of cancer. In 1965, Nobel laureate Sir Mc- Farlane Burnet argued that immunosurveillance was a central Journal of the National Cancer Institute Monographs No. 23, 1998 mechanism by which tumor development was kept in check and predicted that individuals who were immunosuppressed would be at an increased risk of cancer (/). It was already known that children with rare congenital defects of their immune system, such as X-linked gammaglobulinemia or ataxia telangiectasia, were at increased risk of lymphoma, but the number of children with such defects was exceedingly small and so it was not pos- sible to tell whether they were also at an increased risk of other types of cancer (2). Since the 1970s, with the increasing use of immunosuppressive drugs in relation to tissue transplantation, it has been possible to investigate Burnet’s hypothesis in detail. Studies of individuals on long-term immunosuppressive drug therapy have shown that such people were at an increased risk of certain, but not all, types of cancer. The most marked increases were for non-Hodgkin's lymphoma (the most commonly re- ported tumor in most studies of immunosuppressed transplant recipients), Kaposi's sarcoma, hepatocellular carcinoma, and squamous cell carcinoma of the skin, including lip and vulval cancers (3,4). The magnitude of the increase in the relative risk of these tumors was very large indeed. For example, there was about a 100-fold increase in the risk of Kaposi's sarcoma and a 10-fold or greater increase in the risk of other cancers. Some studies (4) reported an increase in Hodgkin's disease, and others (3) reported an increase in cervical cancer, but it is not clear whether this was for premalignant epithelial changes in the cer- vix and/or for invasive cervical cancer (3). The findings in transplant recipients suggested that immuno- suppression led to the selective development of cancers that were caused by infections. The first specific infectious agent implicated as causing cancer in transplant recipients was the Epstein-Barr virus (EBV), which has been consistently identi- fied in transplant-related lymphomas (5). Human herpesvirus (HHV)-8 has been associated with Kaposi's sarcoma, and hepa- titis viruses B and C (HBV and HCV, respectively) have been associated with most hepatocellular carcinomas (6). In addition, human papillomavirus (HPV) types 16/18 have been associated with cervical cancer, and HPV types 5/8 appear to be responsible for some skin cancers in the immunosuppressed (7). The clinical course of the cancers that occur in immunosup- pressed transplant recipients tends to be more aggressive than in the general population. It has also been noted that the cessation of immunosuppressive therapy can halt or even reverse tumor *Affiliation of authors: Imperial Cancer Research Fund, Cancer Epidemiology Unit, Gibson Building, Radcliffe Infirmary, Oxford, U.K. Correspondence to: Valerie Beral, M.D., ICRF Cancer Epidemiology Unit, Gibson Bldg., Radcliffe Infirmary, Oxford OX2 6HE, U.K. See “Note” following **References.”’ © Oxford University Press growth (8). Furthermore, similar risk factors seemed to deter- mine who developed the cancer, irrespective of immune status. For example, the transplant recipients who developed skin can- cer tended to be fair skinned and have excessive lifetime expo- sures to the sun (9). By the late 1970s, before the AIDS epidemic, it was widely believed that immunosuppression led to the selective develop- ment of cancers that were caused by infectious agents. However, because the immunodeficient patients tended to be ill for other reasons, it was not always clear whether the cancers that oc- curred among them were due to the immunodeficiency itself or to the underlying medical conditions, related exposures, or even to the immunosuppressive drugs themselves. For example, the increased risk of hepatocellular carcinoma seen in transplant recipients may have been due to the high prevalence of HBV or HCV infection in this group, and the high rates of cervical can- cer may have been due to more frequent screening for that cancer in transplant recipients than in the general population. Cancers Associated With AIDS The AIDS epidemic has provided an unprecedented opportu- nity to study the effects of immunosuppression on cancer. The fact that large numbers of people throughout the world are in- fected with the human immunodeficiency virus (HIV) permits the study of the effects of immunosuppression on cancer risk on a scale and in populations that had not been studied before. Furthermore, the reason for the immunodeficiency in people with AIDS was not the same as that for the populations studied previously. As will be discussed below, many of the increases in cancer risk found in people with AIDS are similar to the findings in immunodeficient children and in transplant recipients, sug- gesting that it is the impairment of immune function, rather than other factors, that is leading to the appearance of these tumors. Cancers Definitely Increased in People With AIDS Although many cancers have been reported to be increased in people with AIDS, for only five cancers is the evidence suffi- ciently strong and consistent that it is possible to conclude that there is a definite increase in risk. These cancers are listed in Table 1. The evidence for three of those five cancers—Kaposi’s sarcoma, non-Hodgkin's lymphoma, and squamous cell carci- noma of the conjunctiva—is well established and has been re- viewed in depth elsewhere (6,70). Recent evidence for the other two cancers—Hodgkin’s disease and childhood leiomyosarco- Table 1. Cancers that are definitely increased among human immunodeficiency virus (HIV)-infected people Approximate relative risk in HIV-seropositive Cancer individuals Infectious agent* Kaposi's sarcoma 10 000 HHV-8 Non-Hodgkin's lymphoma 50 EBV Hodgkin's disease 10 EBV Squamous cell carcinoma 10 Possibly HPV 16/18 of the conjunctiva Childhood leiomyosarcoma Unclear Possibly EBV *HHV = human herpesvirus; EBV = Epstein-Barr virus; HPV = human papillomavirus. ma—also suggests a definite increase in risk associated with HIV infection (//-13). The relative risks for HIV-seropositive compared with HIV-seronegative people for the five cancers listed in Table 1 tend to be very large—generally 10-fold or greater. Furthermore, each of those cancers is believed to be caused by a specific infectious agent: HHV-8 for Kaposi's sar- coma, HPV 16/18 for conjunctival cancers (although the evi- dence for this is not consistent), and EBV for non-Hodgkin's lymphoma and Hodgkin's disease and possibly, also, for leio- myosarcoma in children. Some, but not all, of the cancers listed in Table | have been associated with other forms of immunodeficiency. Kaposi's sar- coma and non-Hodgkin’s lymphoma have been strongly associ- ated with both immunosuppressive drug therapy and with AIDS. Hodgkin's disease has also been associated with both types of immunosuppression, although the relative risks are not as large as those for the other two cancers (4). Some of the associations found in people with AIDS that have not been reported in trans- plant recipients can probably be explained by the fact that HIV infection affects a much broader range of people living in dif- ferent parts of the world. For example, squamous cell carcinoma of the conjunctiva is common in equatorial Africa, where HIV infection is common, but where tissue transplantation is rare. Also, leiomyosarcoma appears to occur exclusively as a rare complication of HIV infection in children, and few children have been given long-term immunosuppressive drug therapy. Other Cancers in People With AIDS The available evidence suggests that people with AIDS are not experiencing large increases in the risk of most types of cancer. Cancer trends for men living in San Francisco, where HIV prevalence is relatively high, show marked increases over time in the incidence of Kaposi's sarcoma and non-Hodgkin’s lymphoma only (7/4). Furthermore, in a case—control study of about 1000 people with all types of cancer in South Africa, the only cancers that were significantly increased in HIV- seropositive people compared with HIV-seronegative people were Kaposi's sarcoma and non-Hodgkin’s lymphoma (75). Re- sults broadly similar to those from South Africa were found in a case—control study of 250 people in Rwanda (/6). Likewise, record linkage of data from AIDS registries and cancer registries in the United States and in Australia has tended to find increased risks mainly for the cancers listed in Table 1 (717,17). In the Australian study linking information from an AIDS registry to cancer registry data, Grulich et al. (//) reported a relative risk of 5.8 (95% confidence interval [CI] = 1.2-17) for multiple myeloma. There have been a number of case reports of plasmacytomas occurring in people with AIDS (6). It seems likely, therefore, that some form of plasmacytoma may be a rare consequence of HIV-associated immunosuppression, although, at this stage the evidence is not as firm as it is for the five cancers listed in Table 1. Some researchers have linked multiple my- eloma to infection with EBV, and others have linked it to in- fection with HHV-8 (6,18). Perhaps the most fascinating results are the consistent reports that people with AIDS do not appear to be at an increased risk for two common cancers that are known to be caused by infec- tious agents: invasive cervical cancer and hepatocellular cancer (0). Journal of the National Cancer Institute Monographs No. 23, 1998 Cervical Cancer Cervical cancer is the most common cancer among women in Africa, and although HIV infection is highly prevalent in central and southern Africa, no epidemic of cervical cancer has been observed. By contrast, there are marked epidemics of AIDS- related Kaposi's sarcoma in these African countries (/9,20). Fig. I summarizes the relative risk of invasive cervical cancer in HIV-seropositive subjects compared with HIV-seronegative subjects from two published studies (75,16) and preliminary results from another study (2/7) conducted in three African coun- tries, which together total 363 women with the tumor. The rela- tive risk for invasive cervical cancer is not elevated in any study. The overall relative risk is 0.8 (95% CI = 0.5-1.4), thus arguing against the possibility of a large increase in risk. It should be noted that the relative risk estimates are adjusted by age, but not by sexual practices, and the failure to adjust by sexual practices may even lead to a slight overestimate of the relative risk. Although some of the HIV-seropositive subjects may not be severely immunosuppressed, such a bias would act to reduce the relative risk, but not to eliminate an association al- together. There are very few data about invasive cervical cancer in North America or Europe, chiefly because of the low prevalence of HIV infection in women in those countries, but also because women in the West tend to have regular Pap smears and so preclinical cervical neoplasia is generally treated early. How- ever, the few studies in the West that have looked at this ques- tion have also failed to find an increase in the risk of invasive cervical cancer for HIV-infected women (6,22). In contrast to the apparent lack of an increase in invasive cervical cancer, HIV infection is associated with a definite in- crease in the expression of HPV by cervical cells and in the prevalence of apparently premalignant cervical lesions. Numer- ous studies have shown markedly higher prevalences of cervical HPV infection, by either single or multiple types, among HIV- infected women. There is also a corresponding increase in the prevalence of low-grade cervical lesions; however, few women with high-grade cervical lesions or with in situ cervical cancer have been found in these studies (6). At this stage, there is no obvious explanation as to why HIV infection appears to increase the risk of low-grade cervical le- sions, but not the risk of invasive cervical cancer. It has been Country Number of cases Relative Risk" (% HIV positive) (95% CI) Rwanda (16) 23 (0%) 0.0 (0.05.4) = SAfrica(15) 180 (4%) 0602-19) —Ji} Uganda (21)* 160 (21%) 1.0 (0.6-1.8) ~m, SUMMARY 363 (11%) 0.8 (0.5-1.4) 00 1.0 20 3.0 suggested that HIV-infected women in Africa may die of other causes before they have time to develop invasive cervical can- cer. However, cervical cancer is so common in Africa that there should be many women who already had premalignant cervical lesions and even in situ cancer prior to HIV infection. If immu- nosuppression associated with HIV infection hastened the clini- cal course of disease in such women, it is surprising that an epidemic of cervical cancer has not been seen in Africa and that an increase in the risk of invasive cervical cancer has not been found in association with HIV infection. The possibility that there is little or no increase in the risk of invasive cervical cancer among HIV-infected women is worth considering, and, if so, this observation offers an important clue to the pathogenesis of cervical cancer. In women who are chronically infected with HPV, the immune system may play an important role in determining what type of cervical lesion ulti- mately develops. Low-grade cervical lesions mainly reflect HPV infection, and only a small proportion of women with these lesions go on to develop invasive disease. The HPV infections that occur in association with HIV infection may have little immunologic control, and this possibility may favor the persis- tence of low-grade cervical lesions that do not progress. By contrast, the typical, chronic HPV infections in women who are not immunosuppressed and have strong immunologic control may favor progression to invasive disease. Hepatocellular Carcinoma It is curious that there appears to be no increase in the risk of hepatocellular carcinoma in people with AIDS. Fig. 2 summa- rizes the results from two case—control studies in Africa (15,16), preliminary results from a case—control study in Uganda (27), and from one cohort study of patients with hemophilia in the U.K. (23). In no study is there an increase in the relative risk of hepatocellular carcinoma in HIV-seropositive subjects com- pared with HIV-seronegative subjects. The overall relative risk is 0.8 (95% CI = 0.5-1.5). Hepatocellular carcinoma is some- times difficult to diagnose in the absence of modern techniques, and it might be argued that this absence contributes to the ab- sence of an association with HIV infection in Africa. Neverthe- less, the study of U.K. hemophiliacs found a 17-fold overall increase in the risk of liver cancer (because of the high preva- lence of HBV and HCV in hemophiliacs), but no difference in Country Number of cases Relative Risk* (% HIV positive) (95% CI) Rwanda (16) 35 (3%) 0.9 (0.16.1) S Africa (15) 64 (6%) 0.9 (0.3-2.9) Uganda (21)* 62 (13%) 0.8 (0.4-1.8) UK (23) 3 (33%) 0.8 (0.0-15.1) SUMMARY 164 (9%) 0.8 (0.51.5) vCyclin TippNF-KB Journal of the National Cancer Institute Monographs No. 23, 1998 lytic infection-associated protein such as the HHV8 IL-6 homo- log could be important for KS endothelial cell growth if it is expressed in large amounts from lytically infected cells and such cells are sufficiently prevalent to sustain high cytokine levels. Persistent lytic poxvirus infections, for example, cause hyper- trophic cutaneous lesions through secretion of epidermal cell growth factor-related and immune modulating factors. Poxvi- ruses that cause hypertrophic lesions appear to be unique in that infection is highly localized, all infected cells are persistently lytically infected, and there is almost a complete blockade of an effective immune response. Human Immunodeficiency Virus The frequency with which HIV itself causes cancer is uncer- tain and only limited data are available. The positive data are from analyses of non-B-cell lymphomas of all types that were HIV p24 positive (59). Of 22 specimens analyzed in two studies, 18 were positive for HIV proviral DNA by inverse polymerase chain reaction. Of the 18 cases in which proviral DNA was detected, 10 were integrations of HIV upstream of the c-fes gene. Proviral integration upstream of c-fes was found in one T-cell lymphoma. However, almost all of the other integrations were in macrophage-like cells and not in the tumor cells. These findings have led to the hypothesis that HIV integration near c-fes might result in up-regulation of a cytokine that might con- tribute to lymphomagenesis. Given the frequent finding of de- fective proviral integrations in HIV infection overall and the substantial activity of the HIV promoter, particularly in cells with high basal NF-kB activation, HIV may be a more signifi- cant factor in HIV/AIDS-associated cancer than is currently ap- preciated. References (1) Poeschla E, Wong-Staal F. Etiology of cancer: RNA viruses. In: DeVita VT Jr, Hellman S, Rosenberg SA, editors. Cancer: principles and practice of oncology. Philadelphia: Lippincott-Raven, 1997:153-67. Howley PM, Ganem D, Kieff ED. Etiology of cancer: DNA viruses. In: DeVita VT Ir, Hellman S, Rosenberg SA, editors. Cancer: principles and practice of oncology. Philadelphia: Lippincott-Raven, 1997:168-84. Chisari FV, Ferrari C. Hepatitis B virus immunopathogenesis. Ann Rev Immunol 1995:13:29-60. Gootenberg JE, Ruscetti FW, Mier JW, Gazdar A, Gallo RC. Human cutaneous T cell lymphoma and leukemia cell lines produce and respond to T cell growth factor. J Exp Med 1981:154:1403-18. zur Hausen H. Papillomavirus infections: a major cause of human cancer. BBA Reviews on Cancer 1966;1288(2):F55-78. Ostrow RS, Bender M, Niimura M, Seki T, Kawashima M, Pass F, et al. Human papillomavirus DNA in cutaneous primary and metastasized squa- mous cell carcinomas from patients with epidermodysplasia verruciformis. Proc Natl Acad Sci U S A 1982;79:1634-8. Durst M, Gissmann L, Ikenburg H, zur Hausen H. A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different geographic regions. Proc Natl Acad Sci U S A 1983:80:3812-5. (8) Iftner T, Bierfelder S, Csapo Z, Pfister H. Involvement of human papillo- mavirus type 8 genes E6 and E7 in transformation and replication. J Virol 1988:62:3655-61. (9) Palefsky JM, Barrasso R. HPV infection and disease in men. Obstet Gy- necol Clin North Am 1996;23:895-916. (10) Thierry F, Yaniv M. The BPVI-E2 trans-acting protein can be either an activator or a repressor of the HPV 18 regulatory region. EMBO J 1987:6: 3391-17. Durst M, Croce CM, Gissmann L, Schwarz E, Huebner K. Papillomavirus sequences integrate near cellular oncogenes in some cervical carcinomas. Proc Natl Acad Sci U S A 1987:84:1070-4. (12) Schwarz E, Freese UK, Gissmann L, Mayer W, Roggenbuck B, Stremlau A, et al. Structure and transcription of human papillomavirus sequences in cervical carcinoma cells. Nature 1985;314:111-4. ro ~— (3 (4 (5 (6 — Z (11 Journal of the National Cancer Institute Monographs No. 23, 1998 (13) Romanczuk H, Howley PM. Disruption of either the El or E2 regulatory region of human papillomavirus type 16 increases viral immortalization capacity. Proc Natl Acad Sci U S A 1992:89:3159-63. (14) Munger K, Phelps W, Bubb V, Howley PM, Schlegel R. The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes. J Virol 1989; 63:4417-21. (15) Munger K, Werness BA. Dyson N, Phelps WC, Harlow E, Howley PM. Complex formation of the human papillomavirus E7 proteins with the retinoblastoma tumor suppressor gene product. EMBO J 1989:8:4099-105. (16) Chellappan S, Kraus VB, Kroger B, Munger K, Howley PM, Phelps WC, et al. Adenovirus EIA, simian virus 40 tumor antigen, and human papil- lomavirus E7 protein share the capacity to disrupt the interaction between the transcription factor E2F and the retinoblastoma gene product. Proc Natl Acad Sci U S A 1992:89:4549-53. (17) Johnson DG, Schwarz JK. Cress WD, Nevins JR. Expression of the tran- scription factor E2F1 induces quiescent cells to enter S phase. Nature 1993:365:349-52. (18) Scheffner M, Huibregtse JM, Vierstra RD, Howley PM. The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiqui- tination of p53. Cell 1993:;75:495-505. (19) Yokota J, Tsukada Y, Nakajima T, Gotoh M, Shimosato Y, Mori N, et al. Loss of heterozygosity on the short arm of chromosome 3 in carcinoma of the uterine cervix. Cancer Res 1989:49:3598-601. (20) Ohta M, Inoue H, Cotticelli MG, Kastury K, Baffa R, Palazzo I. et al. The FHIT gene, spanning the chromosome 3p14.2 fragile site and renal carci- noma-associated (3:8) breakpoint, is abnormal in digestive tract cancers. Cell 1996;84:587-97. (21) Kieff ED. Epstein-Barr virus and its replication. In: Fields B, Knipe D, Howley PM, editors. Fields virology. Philadelphia: Lippincott-Raven, 1996:2343-96. (22) Rickinson A, Kieff ED. Epstein-Barr virus. In: Fields B, Knipe D, Howley PM, editors. Fields virology. Philadelphia: Lippincott-Raven, 1996:2397. (23) Shope T, Dechairo D, Miller G. Malignant lymphoma in cottontop mar- mosets after inoculation with Epstein-Barr virus. Proc Natl Acad Sci U S A 1973:70:2487-91. (24) Tomkinson BE, Maziarz R, Sullivan JL. Characterization of the T cell- mediated cellular cytotoxicity during acute infectious mononucleosis. J Immunol 1989:143:660-70. (25) Khanna R, Burrows SR. Kurilla MG, Jacob CA, Misko IS, Sculley TB. et al. Localization of Epstein-Barr virus cytotoxic T cell epitopes using re- combinant vaccinia: implications for vaccine development. J Exp Med 1992:176:169-76. (26) Murray RJ. Kurilla MG, Brooks JM, Thomas WA, Rowe M, Kieff ED, et al. Identification of target antigens for the human cytotoxic T cell response to Epstein-Barr virus (EBV): implication for the immune control of EBV- positive malignancies. J Exp Med 1992;176:157-68. (27) Levitskaya J, Coram M, Levitsky V, Imreh S, Steigerwald-Mullen P, Klein G, etal. Inhibition of antigen processing by the internal repeat region of the Epstein-Barr virus nuclear antigen-1. Nature 1995:375:685-8. (28) Izumi KM, Kaye KM, Kieff ED. The Epstein-Barr virus LMP1 amino acid sequence that engages tumor necrosis factor receptor associated factors is critical for primary B lymphocyte growth transformation. Proc Natl Acad Sci US A 1997:94:1447-52. (29) Yates J, Warren N, Reisman D, Sugden B. A cis-acting element from the Epstein-Barr viral genome that permits stable replication of recombinant plasmids in latently infected cells. Proc Natl Acad Sci U S A 1984:81: 3806-10. Johannsen E, Miller CL, Grossman SR, Kieff ED. EBNA-2 and EBNA-3C extensively and mutually exclusively associate with RBPJkappa in Epstein- Barr virus-transformed B lymphocytes. J Virol 1996;70:4179-83. (31) Dash AB, Orrico FC, Ness SA. The EVES motif mediates both intermo- lecular and intramolecular regulation of c-Myb. Genes Dev 1996:10: 1858-69. (32) Harada S. Kieff ED. Epstein-Barr virus nuclear protein LP stimulates EBNA-2 acidic domain-mediated transcriptional activation. J Virol 1997; 71:6611-8. (33) Pear WS, Aster JC, Scott ML, Hasserjian RP, Soffer B, Sklar J, et al. Exclusive development of T cell neoplasms in mice transplanted with bone marrow expressing activated Notch alleles. J Exp Med 1996:183:2283-91. (34) Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C, Kieff ED. The Epstein-Barr virus transforming protein LMPI engages signaling proteins for the tumor necrosis factor receptor family. Cell 1995:80: 389-99. (35) Izumi KM, Kieff ED. The Epstein-Barr virus oncogene product latent membrane protein | engages the tumor necrosis factor receptor-associated death domain protein to mediate B lymphocyte growth transformation and activate NF-kappaB. Proc Natl Acad Sci U S A 1997,94:12592-7. (30 (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) (40) (47) (48) Devergne O, Hatzivassilliou E, Izumi KM, Kaye KM, Kleijnen MF, Kieff ED, et al. Association of TRAFI, TRAF2, and TRAF3 with an Epstein- Barr virus LMP1 domain important for B-lymphocyte transformation: role in NF-kappaB activation. Mol Cell Biol 1996:16:7098-108. Heslop HE, Ng CY, Li C, Smith CA, Loftin SK, Krance RA, et al. Long- term restoration of immunity against Epstein-Barr virus infection by adop- tive transfer of gene-modified virus-specific T lymphocytes. Nat Med 1996;2:551-5. Lacerda JF, Ladanyi M, Louie DC, Fernandez JM, Papadopoulos EB, O'Reilly RJ. Human Epstein-Barr virus (EBV)-specific cytotoxic T lym- phocytes home preferentially to and induce selective regressions of autolo- gous EBV-induced B cell lymphoproliferations in xenografted C.B-17 scid/ scid mice. J Exp Med 1996:183:1215-28. Gaidano G, Dalla-Favera R. Molecular pathogenesis of AIDS-related lym- phomas. Adv Cancer Res 1995:67:113-53. Dolcetti R, Frisan T, Sjoberg J, De Campos-Lima PO, Pisa P. De Re V, et al. Identification and characterization of an Epstein-Barr virus specific T-cell response in the pathologic tissue of a patient with Hodgkin's disease. Cancer Res 1995:55:3675-81. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Ka- posi’s sarcoma. Science 1994;266:1865-9. Ensoli B, Gendelman R, Markham P, Fiorelli V, Colombini S, Raffeld M, et al. Synergy between basic fibroblast growth factor and HIV-1 Tat protein in induction of Kaposi's sarcoma. Nature 1994:371:674-80. Beral V, Peterman TA, Berkelman RL, Jaffe HW. Kaposi's sarcoma among persons with AIDS: a sexually transmitted infection? Lancet 1990:335: 123-8. Moore PS. Chang Y. Detection of herpesvirus-like DNA sequences in Kaposi's sarcoma patients with and without HIV infection. N Engl J Med 1995;332:1181-5. Staskus KA, Zhong W, Gebhard K, Herndier B, Wang H, Renne R. et al. Kaposi's sarcoma-associated herpesvirus gene expression in endothelial (spindle) tumor cells. J Virol 1997;71:715-9. Kedes DH, Operskatski E, Busch M, Kohn R, Flood J, Ganem D. The seroepidemiology of human herpesvirus 8 (Kaposi's sarcoma-associated herpesvirus): distribution of infection in KS risk groups and evidence for sexual transmission. Nat Med 1996:2:918-24. Gao SJ, Kingsley L, Li M, Zheng W, Parravicini C, Zeigler J, et al. KSHV antibodies among Americans, Italians, and Ugandans with and without Kaposi's sarcoma. Nat Med 1996:2:925-8. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. Kaposi's sar- coma-associated herpesvirus-like DNA sequences in AIDS-related body- cavity-based lymphomas. N Engl J Med 1995:332:1186-91. (49) Rainbow L, Platt GM, Simpson GR, Sarid R, Gao SJ, Stobier H, et al. The 222- to 234-kilodalton latent nuclear protein (LNA) of Kaposi's sarcoma- associated herpesvirus (human herpesvirus 8) is encoded by orf 73 and is a component of the latency-associated nuclear antigen. J Virol 1997:71: 5915-21. (50) Renne R, Zhong W, Herndier B, McGrath M, Abbey N., Kedes D. et al. Lytic growth of Kaposi's sarcoma-associated herpesvirus (human herpes- virus 8) in culture. Nat Med 1996:2:342-6. (51) Russo JJ, Bohenzky RA, Chien MC, Chen J, Yan M, Maddalena D, et al. Nucleotide sequence of the Kaposi sarcoma-associated herpesvirus (HHVS). Proc Natl Acad Sci U S A 1996;93:14862-7. Neipel F, Albrecht JC, Fleckenstein B. Cell-homologous genes in the Ka- posi’s sarcoma-associated rhadinovirus human herpesvirus 8: determinants of its pathogenicity? J Virol 1997;71:4187-92. (53) Jung JU, Desrosiers RC. Association of the viral oncoprotein STP-C488 with cellular ras. Mol Cell Biol 1995;15:6506—12. (54) Guo J, Duboise M, Lee H, Li M, Choi JK, Rosenzweig M, et al. Enhanced down-regulation of Lck-mediated signal transduction by a Y 114 mutation of herpesvirus Saimiri tip. J Virol 1997;71:7092-6. (55) Miles SA, Rezai AR, Salazar-Gonzalez JF, Vander Mayden M, Stevens RH, Logan DM, et al. AIDS Kaposi sarcoma-derived cells produce and respond to interleukin 6. Proc Natl Acad Sci U S A 1990:87:4068-72. (56) Moore PS, Boshoff C, Weiss RA, Chang Y. Molecular mimicry of human cytokine and cytokine response pathway genes by KSHV. Science 1996; 274:1739-44. (57) Thome M, Schneider P, Hofmann K, Fickenscher H, Meinl E, Neipel F, et al. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 1997;386:517-21. (58) Sarid R, Sato T, Bohenzky RA, Russo JJ, Chang Y. Kaposi's sarcoma- associated herpesvirus encodes a functional bel-2 homologue. Nat Med 1997:3:293-8. (59) McGrath MS, Herndier B, Shiramizu B. Evidence for direct HIV involve- ment in a broad spectrum of AIDS associated lymphomas. Blood 1994:84: 520a. Note (52 -— Supported by Public Health Service grant (CA47006) from the National Cancer Institute, National Institutes of Health, Department of Health and Human Ser- vices. Journal of the National Cancer Institute Monographs No. 23, 1998 Human Papillomavirus Infection and Anogenital Neoplasia in Human Immunodeficiency Virus-Positive Men and Women Joel M. Palefsky* Human immunodeficiency virus (HIV)-positive women have a higher prevalence of human papillomavirus (HPV) infec- tion in the cervix and anus, as well as squamous intraepi- thelial lesions (SILs) at these sites, than do HIV-negative women matched for age and HIV risk factors. Similarly, HIV-positive homosexual or bisexual men have a higher prevalence of anal HPV infection and anal SIL than do HIV- negative homosexual or bisexual men. In HIV-positive indi- viduals, the prevalence of HPV infection, the proportion in- fected with multiple HPV types, and the prevalence of anogenital SILs increase with decreasing CD4 count. This situation may reflect loss of systemic immune response to HPV antigens or local HPV-HIV interactions at the tissue or cellular level. Despite the high levels of anogenital SILs, to date, there has not been a significant increase in reported cases of invasive anogenital cancer in HIV-positive individu- als. However, several years may be required for SIL to pro- gress to invasive cancer, and the advent of newer therapies for HIV that are expected to prolong survival may paradoxi- cally increase the risk of progression to cancer in individuals with SILs if these lesions do not regress spontaneously and remain untreated. [Monogr Natl Cancer Inst 1998;23:15- 20] The association between human papillomavirus (HPV) and invasive cervical cancer has been recognized for many years, initially through the recognition that cervical cancer had the characteristics of a sexually transmitted disease, i.e., association with number of sexual partners and the age at first intercourse. In the last 10 years, with the advent of molecular probes for HPV, a clear association between the presence of HPV DNA in cervical cancer cells has been shown. Furthermore, increasing understanding of the function of HPV proteins known to be expressed in cervical cancer tissues, such as E6 and E7, has established strong biologic plausibility for HPV as a key factor in the pathogenesis of this disease. Most of current understanding of the epidemiology of HPV infection and anogenital neoplasia was derived from human im- munodeficiency virus (HIV)-negative study populations. In re- cent years, it has become apparent that the prevalence of ano- genital HPV infection is higher among HIV-positive men and women than in their HIV-negative counterparts. Likewise, po- tentially precancerous lesions of the anogenital region are more common among HIV-positive men and women. While this in- creased prevalence has not yet led to a substantial increase in the incidence of anogenital cancer in these populations, the inci- Journal of the National Cancer Institute Monographs No. 23, 1998 dence of cancer may well increase as HIV-positive patients live longer as a result of improvements in medical therapy for HIV infection. The consequences of HPV infection may be a para- digm for a possible shift in complications of HIV disease from acute opportunistic infections to malignancies associated with chronic viral infection. An understanding of HPV infection and anogenital neoplasia in HIV-positive men and women is there- fore important, particularly since many, if not all, of the HPV- associated anogenital cancers that occur in these patients may be preventable. The goals of this article are to summarize current understand- ing of the role of HPV in the pathogenesis of anogenital neo- plasia, to describe current knowledge of the epidemiology of HPV infection and anogenital neoplasia in HIV-positive men and women, to describe HIV-HPV interactions that may play a role in anogenital disease pathogenesis in HIV-positive indi- viduals, and to speculate on the effect of improved therapy for HIV infection on the natural history of anogenital neoplasia. Role of HPV in Pathogenesis of Anogenital Neoplasia There are many different anogenital HPV types, and these are generally divided into oncogenic and nononcogenic types by virtue of the frequency of their association with invasive cervical cancer. Of the oncogenic types, HPV type 16 is the most im- portant by virtue of its being the most common in cervical can- cer, followed by HPV types 18, 31, and 45 (/). Many HPV types are also found, although with less frequency, in some cases of cervical cancer; these include HPV types 33, 35, 39, 52, 56, 58, 59, and 68. Lastly, there is a large group of nononcogenic types rarely if ever found by itself in cervical cancer, and the ones that are most prominent in the genital region are types 6, 11,42, 43, and 44. The mechanisms of oncogenicity of HPV are increasingly, but not completely, understood. The HPV E6 protein binds and de- grades its cellular p53 protein target through a ubiquitin- mediated pathway, whereas the HPV E7 protein inactivates the cellular RB protein, as reviewed previously (2). Among their many functions, p53 and RB shut down the cell cycle and nega- tively regulate cell growth. The p53 protein also has been shown to stimulate DNA repair enzymes after DNA damage, presum- *Correspondence to: Joel M. Palefsky, M.D., Department of Laboratory Medicine and Stomatology, Box 0100, Rm. C634, University of California, San Francisco, San Francisco, CA 94143. See ‘Notes’ following * ‘References.’ © Oxford University Press ably to limit the amount of chromosomal damage that occurs in a cell (3,4). Consequently, through the effects of E6 and E7, HPV infection may lead to genomic instability in cells that con- tinue to cycle despite the presence of chromosomal damage (3,5). Consistent with this hypothesis, chromosomal mutations as shown by studies of loss of heterozygosity may be found in cervical cancer (6-8). The stage at which these changes occur is not yet known, but one or more genetic changes may be required for progression from high-grade squamous intraepithelial lesion (HSIL) to cancer. Cervical HPV Infection and Cervical Squamous Intraepithelial Lesions (SILs) HPV on the cervix typically infects in the transformation zone, where the squamous epithelium of the exocervix meets the columnar epithelium of the endocervix. This squamocolumnar junction is a relatively thin, highly metabolically active area of epithelium. Most HPV infections occur here, and most HPV- related lesions, including invasive cancer, arise from this area. A spectrum of histopathologic abnormalities resulting from HPV infection has been described. At the more benign end of the disease spectrum are changes described as low-grade SILs (LSILs). The primary features of LSILs include limited prolif- eration of basal or parabasal cells with a high nuclear-to- cytoplasmic ratio and koilocytosis (characterized by cells in the more differentiated cell layers with an enlarged, irregular nucleus surrounded by a clear area or halo). At the other end of the disease spectrum are changes described as HSILs. The car- dinal feature of HSILs is marked proliferation of immature basal cells, which may replace most or all of the normal epithelium, and mitoses in the more superficial cell layers. The clinical importance of this grading scheme is that almost all the invasive cancers arise from HSILs and few, if any, arise from LSILs. Consistent with this observation, almost all HSILs contain high- or medium-risk oncogenic HPV types, whereas LSILs may con- tain a wider range of types (9). Epidemiologic studies of cervical HPV infection suggest that the age-related prevalence of HPV infection as determined by polymerase chain reaction (PCR) is highest among women in their late teens and early twenties (/0). These data suggest that most women acquire HPV infection relatively early after initia- tion of sexual activity. The age-related prevalence of cervical HPV infection declines thereafter, suggesting that immune mechanisms may either be clearing the HPV infection or reduc- ing it to levels that are undetectable with the use of current technology. In addition, some of this age-related decline may represent a cohort effect, given changes in sexual behaviors that have occurred in the last few decades. The age-related prevalence of cervical LSILs parallels that of infection, but only a small proportion of women who acquire HPV infection develop clinically detectable LSILs (70). An even smaller proportion of these women develop HSILs or invasive cervical cancer. Currently, the rates in the United States are approximately 8 per 100000, and many of these cancers are occurring in women who never have Pap smear screening (717). In developing countries, where there is no routine cervical cy- tology screening, the rates are much higher, and cervical cancer constitutes a major cause of mortality among young women. 16 Anal HPV Infection and Anal SILs In the general population, anal cancer is more common among women than among men. In the pre-HIV era, anal cancer was reported in women approximately four times more often than in men, with an incidence of 13 per 1 000 000 per year in the United States (7/2). In the last 15 years, the incidence of anal cancer has increased over 35% in women (Hauser A: personal communi- cation). In Denmark, rates of anal cancer are lower than those in the United States; however, between 1957 and 1987, the inci- dence of anal cancer among Danish women more than tripled to 7.4 per 1000000 (73). Rates of anal cancer also rose among Danish men during that time, increasing by 1.5-fold to 3.8 per 1 000 000. The precise incidence of anal cancer among men with a his- tory of receptive anal intercourse has been difficult to determine because cancer registries do not collect information on sexual orientation or behavior. Men with a history of receptive anal intercourse, however, may be at especially high risk of anal cancer. Daling et al. (/4) estimated that the incidence of anal cancer among homosexual men was approximately 35 per 100000, rendering the incidence of anal cancer in this group several times higher than current rates of cervical cancer in women in the United States (//) and similar to rates of cervical cancer prior to the introduction of routine cervical cytology screening. Anal cancer shares many biologic properties with cervical cancer. These cancers are similar histopathologically, and both are strongly associated with HPV infection (75,16). The anal canal has a transformation zone, where the columnar epithelium of the rectum joins the squamous epithelium of the anus, which closely resembles that of the cervix. Like the cervical transfor- mation zone, the anorectal junction is a common site of anal HPV infection and HPV-associated SIL. The histology of anal SIL is similar to that of cervical SIL, and anal HSIL is associated with the same HPV types as cervical HSIL (75). Although stud- ies have never been done to determine if untreated anal HSIL is the precursor to invasive anal cancer, this is likely to be the case, based on knowledge of the natural history of cervical HSIL. From a clinical standpoint, this is important because the anorec- tal junction is about 2 cm inside the anal canal and thus would not be visible with routine perianal inspection. Anogenital HPV Infection and SILs in HIV-Positive Men and Women The data collected on the epidemiology and natural history of HPV infection and SIL described above were derived from HIV- negative individuals. There are several reasons to suspect that these data may be different in HIV-positive men and women. First, both HIV and HPV are sexually transmissible. Behaviors that increase the risk of acquiring HIV infection may also in- crease the risk of acquiring HPV infection and vice versa. Sec- ond, since the immune response to HPV may play an important role in control of SIL, it is possible that, as HIV-related immu- nosuppression increases, immunity specific to HPV declines. Consequently, HIV-positive individuals may be at higher risk of both acquiring HPV and developing SIL related to HPV once HPV is acquired. Clear patterns are emerging from studies of large numbers of Journal of the National Cancer Institute Monographs No. 23, 1998 HIV-positive men and women. These patterns confirm the in- creased risk of HPV infection and SIL in these groups. One of these studies is a prospective cohort study known as the WIHS Study, the Women’s Interagency HIV Study. This ongoing study enrolled 2015 HIV-positive women and 577 HIV-negative women who were matched for age and HIV risk factors at six different cities around the United States. HPV testing of a cer- vicovaginal lavage specimen was performed on the women at baseline by use of PCR. The procedure used was a modification of that employed by Ting et al. (/7) and used consensus primers from the LI region of the HPV genome, followed by specific typing for 39 different HPV types. Among these high-risk, HIV- negative women, the prevalence of HPV was approximately 26% (Palefsky JM: unpublished data), relatively high for their age when compared with the data of Schiffman (/0), which were collected in a population of women with Kaiser Permanente medical insurance. However, the HIV-positive women had even higher rates of prevalent HPV infection, and the rates were high- est among those with the lowest baseline CD4 count. Among women with CD4 levels less than 200/mm? at baseline, approxi- mately 70% had detectable HPV infection (Palefsky JM: unpub- lished data). Independent risk factors for cervical HPV infection at baseline included HIV status, current smoker, younger age, and ethnicity (with African-American women having the highest rates of HPV infection). Similar data have been obtained from a cohort study of 346 HIV-positive and 262 HIV-negative homosexual or bisexual men in San Francisco. By use of a PCR HPV testing method similar to that used for cervicovaginal lavage specimens in the WIHS study, anal HPV infection was characterized in this study population at baseline. Approximately 60% of the HIV-negative men had anal HPV infection. Similar to our findings in the cervix, the proportion of men with anal HPV infection was even higher among the HIV-positive group and was highest among those with the lowest CD4 levels. Among the HIV-positive men with CD4 counts less than 500/mm®, HPV infection was nearly universal (Palefsky JM: unpublished data). These data were similar to those reported previously among homosexual men in Seattle (18). Another interesting feature of HPV infection among both HIV-positive women and men was the multiplicity of HPV types. Of HIV-negative women in the WIHS study who were HPV-positive, 16% had more than one HPV type, compared with 42% of HIV-positive women (Palefsky JM: unpublished data). Among HIV-negative men, 23% had more than one type of HPV in the anal canal, compared with 73% of HIV-positive men (Palefsky JM: unpublished data). With respect to cervical cytologic changes in the WIHS co- hort at baseline, a relatively high proportion of HIV-negative women had abnormal cervical cytology (16%). Similar to the HPV data, the proportion of HIV-positive women with abnormal cervical cytology at baseline increased inversely with the base- line CD4 levels, and 53% of women with CD4 levels less than 200/mm* had abnormal cytology (Fruchter R: personal commu- nication). Finally, among HIV-negative men in the San Francis- co cohort study, 21% had anal cytologic abnormalities at base- line, while 72% of HIV-positive men with CD4 counts less than 200/mm® had abnormal anal cytology (Palefsky JM: unpub- lished data). Taken together, these findings indicate that HIV- Journal of the National Cancer Institute Monographs No. 23, 1998 positive women and men have a higher prevalence of anogenital HPV infection, a higher number of HPV types, and a higher rate of prevalent anogenital lesions—each of these clearly associated with lower CD4 levels. Relatively few data are currently available on the natural his- tory of HPV infection and SIL in these large study groups, and these studies are in progress. However, some prospective data on the projected incidence of anal HSILs are available from the San Francisco men’s cohort study. Men who entered the study with- out HSIL were stratified according to HIV status and baseline CD4 level. Among HIV-positive men who had CD4 levels less than 200/mm”* at baseline or between 200/mm? and 500/mm”, the 4-year projected incidence of anal HSIL was greater than 50% (Palefsky JM: unpublished data). Even among those who entered the study with a CD4 level greater than 500/mm?, the projected incidence of HSILs was high (approximately 30%). Finally, HIV-negative men in the study had a projected 4-year incidence of 17%, putting them potentially at the highest risk, since many of these men will have a completely normal life span. These data indicate that a high proportion of HIV-positive men, as well as a substantial proportion of HIV-negative men, will develop HSILs if they are followed for a sufficient period of time. Invasive Anogenital Cancer in HIV-Positive Women and Men The above data indicate a high prevalence and incidence of anogenital HPV infection and putative precancerous anogenital lesions. An important question that has emerged in the last few years is whether these findings will translate into higher rates of invasive anogenital cancer. Thus far, there has been no signifi- cant increase in the rate of invasive cervical cancer among HIV- positive women in the United States or in the developing world (19). Data on anal cancer are not as clear, since studies using different methods lead to different conclusions. Rabkin and Yellin (79) did not find a significant increase in anal cancer in single, never-married men in the San Francisco Bay Area in the early 1990s when compared with the pre-HIV years. In contrast, Melbye et al. (20) reported increased relative risk of anal cancer compared with the general population with increasing proximity to an AIDS diagnosis, implying a role for increasing immuno- suppression. However, these data should be interpreted cau- tiously, since the results may be confounded by the length of time that an individual may have been infected with HPV. Over- all, it seems likely that there has been some increase (but not very large) in anal cancer, and the exact magnitude is still un- known. Based on knowledge of the natural history of cervical disease, it seems likely that progression of HSIL to invasive cancer may require several years. Until recently, most HIV-positive indi- viduals would have died of other HIV-related complications before HSIL would have had the chance to progress, which may explain why the high prevalence of anogenital HSIL in HIV- positive women and men has not led thus far to a dramatically large increase in cervical or anal cancer. Despite the absence of a clear increase in cervical cancer in HIV-positive women, it is important to recognize that, once it develops, cervical cancer may be very aggressive in HIV- positive women, and extreme vigilance on the part of the clini- 17 cian is very important (2/,22). Conversely, cervical cancer may be an indicator of HIV infection, particularly in those regions with the highest HIV prevalence, and women with cervical SIL or cervical cancer should be strongly considered for HIV testing. In the case of cervical disease, the clinical implications of these findings are that aggressive screening, treatment, and fol- low-up for cervical SIL are necessary. In the case of anal dis- ease, the data indicating the need for an anal-screening program modeled on that used for cervical disease are compelling. Stud- ies have been performed recently that validate the use of anal colposcopy as a diagnostic tool (23) and anal cytology as a screening tool (24). Barriers to implementation of such a screen- ing program include the paucity of studies showing that anal HSIL progresses to invasive cancer and that a screening program would have an impact on the rate of anal cancer. However, such studies are very unlikely to be performed, given the ethical is- sues of following an individual without treatment once HSIL is diagnosed. Other barriers include the absence of widespread expertise in the performance of anal diagnostic techniques and the morbidity associated with most treatments for anal lesions. Were such a screening program to be implemented in the future, individuals most likely to benefit would include HIV-positive and HIV-negative men with a history of receptive anal inter- course. Because of data showing that women with cervical HSIL or vulvar cancer have a high prevalence of anal SIL and anal cancer (25,26), HIV-positive and HIV-negative women with high-grade cervical or vulvar lesions should also be considered for screening. Data on anal HPV infection and anal SIL in HIV- positive women without cervical disease are still emerging, but several earlier studies (27,28) have shown that, similar to cer- vical HPV infection and cervical SIL, the prevalence of anal lesions is highest among HIV-positive women with the lowest CD4 levels. Mechanisms of HIV-HPV Interaction Fig. 1. Schematic representation of possible interactions between human immu- nodeficiency virus (HIV) and human papillomavirus (HPV) at the tissue and cellular levels in the pathogenesis of anogenital squamous intraepithelial lesions and invasive anogenital cancer. HPV infection is restricted to the epithelium, as shown in the shaded cells. HIV infection, as shown in the striped cells, is located primarily in stromal cells (1) such as circulating T cells and possibly fibroblasts or other stromal elements or Langerhans’ cells within the epithelium (2). HIV and HPV co-infection in the same epithelial cells (3) (dotted cells) is also a theoretical possibility. Possible interactions include secretion of HIV-1 tat by Langerhans’ or stromal cells which may up-regulate HPV gene expression or other factors such as cytokines. Cells infected with both viruses may also trans- activate each other. pathogenesis of anogenital disease and may explain in part the higher levels of HPV infection and SIL among those HIV- positive individuals with the lowest CD4 levels. Local interactions at the tissue and cellular levels between HIV and HPV may also play a role in potentiation of anogenital neoplasia in HIV-positive individuals (Fig. 1). Although the vi- Several mechanisms of interaction between HIV and HPV may play a role in the higher prevalence and incidence of anogenital SIL in HIV-positive women and men. One such mechanism is the systemic im- mune response to HPV. It is known that women who have iatrogenic immunosuppression secondary to or- gan transplantation are at increased risk of developing cervical and vulvar cancers when compared with age- matched women with normal immunity (29). Re- cently, one study (30) showed that a smaller propor- tion of women with HPV type 16 infection and cervical SIL have cytotoxic T-cell responses to the HPV type 16 E6 and E7 proteins than women who were HPV type 16 positive without cervical disease. HPV Infection HIV infection Anogenital Disease 1 1 Cancer - Other studies of T-cell proliferative responses (317,32) also confirm that HPV-positive women without dis- ease have higher response rates than women with le- sions. These data suggest that, once a woman acquires HPV infection, having a cell-mediated immunity (CMI) response is associated with protection against lesion development. Conversely, if HIV infection leads to global impairment of CMI responses, includ- ing those to HPV antigens, then loss of that CMI response to HPV antigens may be important in the 18 Fig. 2. Model for the role of systemic immune response in the pathogenesis of anogenital squamous intraepithelial lesions (SILs) and invasive anogenital cancer. Human papillomavi- rus (HPV) infection is acquired early after initiation of sexual activity, followed thereafter by human immunodeficiency virus (HIV) acquisition. Additional HPV types, represented by different lines, may be acquired with new exposures. Early in the natural history of HIV infection, systemic immune responses remain relatively intact. HPV replication levels remain low, and there are no anogenital lesions. With advancing HIV disease, systemic immunity is attenuated, and loss of control of HIV replication is seen as depicted by increased HPV levels. This is accompanied by development of low-grade SIL (LSIL), which may progress over time to high-grade SIL (HSIL) if the lesion remains untreated and the HIV disease continues to progress. In most individuals, progression of HSIL to invasive cancer does not occur because they die of other HIV-related complications first. N = normal. Journal of the National Cancer Institute Monographs No. 23, 1998 ruses are likely to be found in different cell types (33) (HPV in the epithelium and HIV in local Langerhans’ cells and stromal cells), one possibility for interaction between these two viruses locally is through production of HIV-1 tat, which has been shown to be secreted by HIV-infected cells and may be able to up-regulate expression of the HPV type 16 E6 and E7 genes (34). Aberrant cytokine expression by HIV-infected cells may also play a role in potentiating HPV infection. Perhaps most controversial is the possibility of co-infection with HPV and HIV in the same epithelial cell and the possibility of mutual transactivation between the viruses. Taken together, these disparate observations may be used to construct a model for the interplay between HPV and HIV in the pathogenesis of anogenital SIL and cancer (Fig. 2). In this model, individuals acquire HPV infection relatively early on after initiation of sexual activity and, in many cases, before HIV infection. HIV is acquired at one or more moments in time, and these individuals may continue to acquire different strains of HPV as well. As long as the systemic and local immune re- sponses remain intact, HPV replication and gene expression are controlled and there are no anogenital lesions. If HIV infection is allowed to persist and the immune response deteriorates, higher levels of viral replication and expression of HPV- transforming genes may lead to development of a low-grade lesion. With time, which may vary considerably from one indi- vidual to another, the disease may progress to HSIL. As de- scribed earlier, the high prevalence of HSIL does not appear to be accompanied by increases in the rate of anogenital cancer because of HIV-related mortality before progression of HSIL to cancer occurs. Conclusions: What Does the Future Hold? The natural history of anogenital HPV infection and SIL in the era of highly active antiretroviral therapy (HAART) may represent a paradigm for HIV-related complications of the fu- ture. HAART has been shown to dramatically reduce systemic HIV viral load and to reduce the incidence of some opportunistic infections. Because HAART has not been in use for an extended period of time, it is not yet clear how much it will prolong the survival of individuals with HIV infection. Two scenarios are possible (Fig. 3). If individuals go on HAART after developing HSIL and if partial or complete restoration of immune response to HPV occurs, then one might expect HPV levels to decrease and HSIL to regress. If that is the case, then a decreased inci- dence of anogenital cancer from current levels would be ex- pected. Conversely, if HAART permits individuals to live longer but does not have a significant impact on the immune response to HPV, then the disease in individuals with HSIL might now have the time to progress to invasive cancer, resulting in an increased incidence of anogenital cancer. It is not yet clear which of these two scenarios is correct. However, the most that HAART therapy could be expected to achieve would be to re- store the immune system of an HIV-positive individual to one as functional as that of an HIV-negative individual. Since most HSILs do not regress spontaneously, even in HIV-negative in- dividuals, the second scenario seems more likely at this time. Clearly, further natural history studies of men and women on HAART are needed, as are studies of the interactions between HIV and HPV in the pathogenesis of anogenital cancer. Journal of the National Cancer Institute Monographs No. 23, 1998 HPV Infection HIV infection H 0 os SEE Lim nt re . HPV oe" — od bees ; Anogenital Disease HPV Infection HIV infection HPV HPV HPV LSIL Anogenital Disease Invasive cancer Fig. 3. Model for the role of systemic immune response in the pathogenesis of anogenital squamous intraepithelial lesions (SILs) and invasive anogenital can- cer in the era of highly active antiretroviral therapy (HAART). HSIL = high- grade SIL: LSIL = low-grade SIL. Upper panel: If HAART therapy results in both restoration of systemic immune responses, including those to human pap- illomavirus (HPV) antigens, and prolongation of life, then individuals with HSIL may have decreased levels of HPV and regression of HSIL. The net result would be a rate of invasive anogenital cancer that parallels that of human immunode- ficiency virus-negative individuals. Lower panel: If HAART leads to prolon- gation of life but does not fully restore immunity to HPV, then no regression of HSIL will be seen. In this case, the prolongation of life may increase the risk of progression to invasive cancer, since there may now be sufficient time for progression before the individual dies of other causes. N = normal. References (1) Bosch FX, Manos MM, Munoz N. Sherman M, Jansen AM, Peto J, et al. Prevalence of human papillomavirus in cervical cancer: a worldwide per- spective. International Biological Study on Cervical Cancer (IBSCC) Study Group. J Natl Cancer Inst 1995:87:796-802. (2) Palefsky JM, Holly EA. Molecular virology and epidemiology of human papillomavirus and cervical cancer. Cancer Epidemiol Biomarkers Prev 1995:4:415-28. Livingstone LR, White A, Sprouse J, Livanos E, Jacks T, TIsty TD. Altered cell cycle arrest and gene amplification potential accompany loss of wild- type p53. Cell 1992:70:923-35. Vousden KH, Vojtesek B, Fisher C, Lane D. HPV-16 E7 or adenovirus EIA can overcome the growth arrest of cells immortalized with a tempera- ture-sensitive p53. Oncogene 1993;8:1697-702. Carder P, Wyllie AH, Purdie CA, Morris RG, White S, Piris J, et al. Stabilised p53 facilitates aneuploid clonal divergence in colorectal cancer. Oncogene 1993:8:1397-401. (6) Magnusson PK, Wilander E, Gyllensten U. Analysis of loss of heterozy- gosity in microdissected tumor cells from cervical carcinoma using fluo- rescent dUTP labeling of PCR products. Biotechniques 1996:21:844-7. Rader JS, Kamarasova T, Huettner PC, Li L, Li Y, Gerhard DS. Allelo- typing of all chromosomal arms in invasive cervical cancer. Oncogene 1996:13:2737-41. (8) Douc-Rasy S. Barrois M, Fogel S, Ahomadegbe JC, Stehelin D, Coll J, et al. High incidence of loss of heterozygosity and abnormal imprinting of H19 and IGF2 genes in invasive cervical carcinomas. Uncoupling of HI19 and IGF2 expression and biallelic hypomethylation of HI19. Oncogene 1996:12:423-30. (3 — (4 (5 -— (7 (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21 ~ (22) (23) Lorincz AT, Reid R, Jenson AB, Greenberg MD, Lancaster W, Kurman RJ. Human papillomavirus infection of the cervix: relative risk associations of 15 common anogenital types. Obstet Gynecol 1992:79:328-37. Schiffman MH. Recent progress in defining the epidemiology of human papillomavirus infection and cervical neoplasia. J Natl Cancer Inst 1992; 84:394-8. Qualters JR, Lee NC, Smith RA, Aubert RE. Breast and cervical cancer surveillance, United States, 1973-1987. MMWR Morb Mortal Wkly Rep 1992:41:1-15. Young J Jr, Percy CL, Asire Al, Berg JW, Cusano MM, Gloeckler LA, et al. Cancer incidence and mortality in the United States, 1973-77. Monogr Natl Cancer Inst 1981:57:1-187. Frisch M, Melbye M, Moller H. Trends in incidence of anal cancer in Denmark. BMJ 1993:306:419-22. Daling JR, Weiss NS, Hislop TG, Maden C, Coates RJ, Sherman KJ, et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med 1987:317:973-7. Palefsky JM, Holly EA, Gonzales J, Berline J, Ahn DK, Greenspan JS. Detection of human papillomavirus DNA in anal intraepithelial neoplasia and anal cancer. Cancer Res 1991;51:1014-9. Zaki SR, Judd R, Coffield LM, Greer P, Rolston F, Evatt BL. Human papillomavirus infection and anal carcinoma. Retrospective analysis by in situ hybridization and the polymerase chain reaction. Am J Pathol 1992; 140:1345-55. Ting Y, Manos MM. Detection and typing of genital human papillomavi- ruses. In: Innis MA, Gelfand DH, Sninsky JJ. White TJ, editors. PCR protocols and applications. San Diego (CA): Academic Press, 1990: 356-67. Kiviat NB, Critchlow CW, Holmes KK, Kuypers J. Sayer J, Dunphy C, et al. Association of anal dysplasia and human papillomavirus with immuno- suppression and HIV infection among homosexual men. AIDS 1993.7: 43-9. Rabkin CS, Yellin F. Cancer incidence in a population with a high preva- lence of infection with human immunodeficiency virus type 1. J Natl Can- cer Inst 1994:86:1711-6. Melbye M, Cote TR, Kessler L, Gail M, Biggar RJ. High incidence of anal cancer among AIDS patients. The AIDS/Cancer Working Group. Lancet 1994:343:636-9. Maiman M, Fruchter RG, Serur E, Remy JC, Feuer G, Boyce J. Human immunodeficiency virus infection and cervical neoplasia. Gynecol Oncol 1990;38:377-82. Rellihan MA, Dooley DP, Burke TW, Berkland ME, Longfield RN. Rap- idly progressing cervical cancer in a patient with human immunodeficiency virus infection. Gynecol Oncol 1990;36:435-8. Jay N, Holly EA, Berry M, Hogeboom CJ, Darragh JM, Palefsky JM. Colposcopic correlates of anal squamous intraepithelial lesions. Dis Col Rectum 1997:40:919-28. (24) Palefsky JM, Holly EA, Hogeboom CJ, Jay N, Berry M, Darragh TM. Anal cytology as a screening tool for anal squamous intraepithelial lesions. J Acquir Immun Defic Syndr 1997;14:415-22. (25) Scholefield JH, Hickson WG, Smith JH, Rogers K, Sharp F. Anal intra- epithelial neoplasia: part of a multifocal disease process. Lancet 1992;340: 1271-3. (26) Ogunbiyi OA, Scholefield JH, Robertson G, Smith JH, Sharp F, Rogers K. Anal human papillomavirus infection and squamous neoplasia in patients with invasive vulvar cancer. Obstet Gynecol 1994:83:212-6. (27) Williams AB, Darragh TM, Vranizan K, Ochia C, Moss AR, Palefsky JM. Anal and cervical human papillomavirus infection and risk of anal and cervical epithelial abnormalities in human immunodeficiency virus- infected women. Obstet Gynecol 1994:83:205-11. (28) Melbye M, Smith E, Wohlfahrt Nielson J, Osterlind A, Orholm M, Berg- mann OJ, et al. Anal and cervical abnormality in women—prediction by HPV tests. Int J Cancer 1996:68:559-64. (29) Penn I. Cancer in the immunosuppressed organ recipient. Transplant Proc 1991;23:1771-2. (30) Nakagawa M, Stites DP, Farhat S, Sisler JR, Moss B, Kong F, et al. Cytotoxic T lymphocyte responses to E6 and E7 proteins of human papil- lomavirus type 16: relationship to cervical intraepithelial neoplasia. J Infect Dis 1997;175:927-31. (31) Nakagawa M, Stites D, Farhat S, Judd A, Moscicki AB, Canchola AJ, et al. T cell response to human papillomavirus type 16: relationship to cervical intraepithelial neoplasia. Clin Diagnost Lab Immunol 1996:3:205-10. (32) Tsukui T, Hildesheim A, Schiffman MH, Lucci JR, Contois D, Lawler P, et al. Interleukin 2 production in vitro by peripheral lymphocytes in re- sponse to human papillomavirus-derived peptides: correlation with cervical pathology. Cancer Res 1996:56:3967-74. (33) Nuovo GJ, Forde A, MacConnell P, Fahrenwald R. In situ detection of PCR-amplified HIV-1 nucleic acids and tumor necrosis factor. Am J Pathol 1993:143:40-8. (34) Vernon SD, Hart CE, Reeves WC, Icenogle JP. The HIV-1 tat protein enhances E2-dependent human papillomavirus 16. Virus Res 1993;27: 133-45. Notes Supported by Public Health Service (PHS) grants CA54053 and CA63933 from the National Cancer Institute, National Institutes of Health (NIH), Depart- ment of Health and Human Services, as well as by PHS grant AI-DE34989 from the National Institute of Allergy and Infectious Diseases and the National Insti- tute of Dental Research, NIH. We acknowledge the subjects and investigators of the University of Califor- nia, San Francisco, Anal Cancer Study and the Women’s Interagency HIV Study. Journal of the National Cancer Institute Monographs No. 23, 1998 Acquired Immunodeficiency Syndrome-Related Cancers: the Community Perspective Michael Marco* Those individuals who work for acquired immunodeficiency syndrome (AIDS) community-based organizations (treatment education, dissemination, advocacy, and care) and the people with AIDS whom they represent are commonly referred to as the “AIDS community.” While the primary interest of the AIDS community is safe and effective antiretroviral therapy, many of us are concerned with the opportunistic neoplasms that affect approximately 25% of people with AIDS. The community’s per- spective on AIDS-related cancers can be summarized as follows: 1) It’s about time the National Cancer Institute (NCI) got its act together. 2) How and when are diseases such as AIDS-related Kaposi's sarcoma and non-Hodgkin’s lymphoma going to be cured and prevented? 3) Until then, how can we be sure that we will receive the best medical care for these cancers and the underlying human immunodeficiency virus (HIV) infection? With regard to NCI’s finally getting its act together, my per- ception is that the AIDS community is pleased that there has been a change in leadership at the NCI. The previous era gave us zidovudine (AZT) and didanosine, but unfortunately little was done when it came to AIDS-related cancers. Many of us have wondered why the NCI—with its 250 million dollars in AIDS research funds—is doing so little extramurally to address these cancers. There are some in the AIDS community who trace this failure to the division of labor among the various institutes of the National Institutes of Health (NIH) regarding AIDS. More recently, the NCI should be praised for spearheading new research initiatives on AIDS-related cancers. The NCI re- cently formed the AIDS Malignancy Consortium (AMC), a clinical trials network that concentrates solely on AIDS-related cancers. In its first year of operation, the AMC already has five studies open to treat either Kaposi's sarcoma or non-Hodgkin’s lymphoma; two of these studies use conventional chemotherapy together with protease inhibitors, and three studies are employ- ing cytokine inhibitors or immune modulators. The vast majority of the AMC members came directly from the AIDS Clinical Trials Group’s (ACTG) Oncology Commit- tee, whose clinical trials in the late 1980s and early 1990s helped develop a clinical standard of care for Kaposi's sarcoma and non-Hodgkin's lymphoma. During that period, we knew very little about the pathogenesis of Kaposi's sarcoma and non- Hodgkin's lymphoma, and it was up to these researchers to determine the proper therapeutic doses of the limited chemo- therapeutic agents that we knew had activity. Thus, these clinical trials taught us how to care properly for people with AIDS who were diagnosed as having Kaposi's sarcoma and non-Hodgkin's lymphoma with a scant arsenal of drugs. The ACTG Oncology Committee was, however, working within a clinical trials network filled with infectious disease Journal of the National Cancer Institute Monographs No. 23, 1998 specialists who had limited expertise in or commitment to AIDS-related cancer clinical research. The oncologists, who used less than 10% of the ACTG’s financial resources, were relegated to a lesser role. Now that the AMC is in its second year, we should ensure its success by providing it with the resources that it needs. The first resource is, of course, money. Increased funding will help labo- ratory intensive studies that will measure cytokine levels as well as patient's HIV RNA. The second resource is patients for its studies. For large phase IT and phase III studies that are generated from the AMC or the Eastern Cooperative Oncology Group’s AIDS Committee, we will need a majority of the cooperative oncology groups to spon- sor these protocols. If this does not work, we might have to return to ACTG for patients. With regard to advisory panels, the NCI needs to use the members of its AIDS Malignancy Working Group (AMWG) more actively. Many from the AMWG helped organize the ex- cellent (and badly needed) National AIDS Malignancy Confer- ence at the NIH, but putting on a conference is not enough. The NCI should form a working group to discuss and formulate AIDS-related cancer RFAs (i.e., requests for application). Yes, there are some legal matters to consider, since some of these experts in the working group would themselves apply, but we should try to work around this situation. The NCI cannot suc- cessfully develop such programs without the full involvement of AMWG. The AMWG should be more than window dressing. The community’s second concern can be summarized in the following question: How and when are cancers, such as Kaposi’s sarcoma and non-Hodgkin’s lymphoma, going to be cured and prevented? This is a difficult question, especially coming from the AIDS community who, understandably, has never been known for its patience. The cure for these cancers will eventually come from the work completed by many of the researchers who presented papers at the National AIDS Malignancy Conference. We are just beginning to understand the etiology and pathogenesis of Kaposi's sarcoma and non-Hodgkin's lymphoma. Basic scien- tists and clinicians must work together in the old-fashioned ‘‘bench-to-clinic’” mode. When trying to turn their discoveries into active agents, basic scientists must use clinicians as con- sultants. For example, there is still much to learn about Kaposi's sar- coma herpesvirus (KSHV). We must first answer three integral *Correspondence to: Michael Marco, Treatment Action Group, 105 West 73rd St., Apartment 1D, New York, NY 10023. © Oxford University Press 21 questions: 1) Is KSHV a true cause of Kaposi's sarcoma or just a helper virus? 2) What are the roles of cytokines and growth factors? 3) Which assay is most sensitive, and how soon can we start using these assays for screening patients? In our quest for these cures, industry—especially the biotech companies—will need to take a more active role in AIDS-related cancer clinical research. They must provide their agents to knowledgeable AIDS oncologists for pilot studies or to the NCI for use in the AMC or cooperative oncology group studies. The U.S. Food and Drug Administration (FDA) must also get more involved. Its Oncologic Drug Division needs to fully un- derstand all aspects of AIDS-related Kaposi's sarcoma and AIDS-related non-Hodgkin’s lymphoma. While it should ensure that industry carries out safe and well-designed clinical trials, the Oncologic Drug Division must realize that AIDS-related Kapo- si’s sarcoma not only causes morbidity and mortality but also emotionally damages patients with AIDS. The Oncologic Drug Division should ensure that the Oncologic Drug Advisory Com- mittee has at least three reviewers who are knowledgeable in all aspects of AIDS-related cancers when AIDS-related Kaposi's sarcoma and AIDS-related non-Hodgkin’s lymphoma new drug applications (NDAs) are reviewed. Likewise, the Oncologic Drug Division should attempt to work closely with the Antiviral Drug Division when it is reviewing AIDS-related Kaposi's sar- coma and AIDS-related non-Hodgkin's lymphoma NDAs. The FDA and the NCI should be applauded for working to- gether on developing clinical end points for Kaposi's sarcoma that will be used alongside classic tumor response. The use of these newly developed clinical end points, such as pain, skin lesion color, and edema, will help us evaluate the true effective- ness of experimental agents. To test new agents and possibly use these new end points, we will need patients for these studies. Accrual for studies on AIDS- related cancer, like that for most cancer studies, is abysmal. One solution to this problem might be to form close ties with primary care physicians and AIDS clinicians at hospitals and clinic sites so that patients with Kaposi's sarcoma and non-Hodgkin's lym- phoma are routinely referred to AIDS oncologists and are in- formed of studies that have opened. The final concern of the AIDS community is a frequently asked question: How can we be sure that we will receive the best medical care for these cancers and the underlying HIV infection? To ensure that patients receive the best medical care for their cancers, we must attempt to adopt standard (and evolving) prac- tices and principles of the therapy. As stated earlier, clinical trials that were completed in the last 10 years helped us to understand what drugs (and at what doses) were best for patients with disease at various stages. We know how to use radiation, interferon, ABV (i.e., doxorubicin—bleomycin—vinblastine), li- 22 posomal doxorubicin, liposomal daunorubicin, and paclitaxel to treat Kaposi's sarcoma, and we know that modified doses of doxorubicin hydrochloride and cyclophosphamide are warranted for the severely immunocompromised patients with AIDS- related non-Hodgkin's lymphoma. Although convening a government panel to draft practices and principles of therapy for AIDS-related Kaposi's sarcoma was suggested to the NCI at its AMWG meeting in September 1996, no action has yet been taken on this recommendation. My sense is that many in the AIDS community believe that this lack of action represents an abdication of the government's public health responsibility. The U.S. Public Health Service convened two panels that recently published guidelines for anti-HIV therapy and treatment for opportunistic infections. Even though the field is still evolv- ing in these areas, the guidelines will ensure that patients receive the best up-to-date AIDS clinical care. Nevertheless, a Kaposi's sarcoma practices and principles of therapy panel was convened. However, it was sponsored by a small biotech company that took the idea and ran with it. Granted, this company has a financial interest in the outcome of these recommendations, but it was able to assemble an excellent panel of experienced AIDS oncologists, an AIDS clinician, a dermatologist, and an radiation oncologist to discuss methods of treatment of patients with various clinical presentations of Ka- posi’s sarcoma. The recommendations that will come from this panel will not hold as much weight as ones that should come from the NCI. It is necessary to give certain guidelines to all those who treat patients with Kaposi’s sarcoma. To instruct a physician about when it is time to use systemic therapy with either interferon or cytotoxic chemotherapy and when it is no longer appropriate to use cryotherapy or intralesional vinblastine will undoubtedly help patients. Until the NCI decides that this a worthwhile initiative, they should see to it that the new Public Health Service guidelines for HIV therapy and opportunistic infection treatment and prophy- laxis get into the hands of as many oncologists as possible. For the treatment of patients with Kaposi's sarcoma, non-Hodgkin’s lymphoma, or anal or cervical dysplasia, the underlying HIV infection must always be taken into account and treated prop- erly. The AIDS community is anxious to see new developments in AIDS-related cancer research. Now that people with AIDS are living longer with the advent of protease inhibitors in combina- tion with nucleoside analogues, it is certain that the rates of AIDS-related Kaposi's sarcoma and non-Hodgkin’s lymphoma will increase. Thus, clinical and basic AIDS-related cancer re- search needs to be of concern for all those working in AIDS. Journal of the National Cancer Institute Monographs No. 23, 1998 Association of Non-Acquired Immunodeficiency Syndrome-Defining Cancers With Human Immunodeficiency Virus Infection Charles S. Rabkin* Kaposi’s sarcoma and non-Hodgkin’s lymphoma were among the earliest recognized manifestations of the acquired immunodeficiency syndrome (AIDS) epidemic. Excluding these two tumors, the overall risk of all other cancers in human immunodeficiency virus (HIV)-infected individuals is similar to that of the general population. However, varying levels of evidence link several additional neoplasms to HIV infection. The evidence is strongest for an association with Hodgkin’s disease, with lower relative and absolute risks than for non-Hodgkin’s lymphoma. Anogenital intraepithe- lial neoplasia also appears to be HIV associated, but in- creases of invasive disease are still uncertain for both cervi- cal and anal cancers. Various studies have suggested associations with testicular seminoma, multiple myeloma, oral cancer, and melanoma, but the data are inconsistent. Leiomyosarcoma and benign leiomyomas have increased in incidence in HIV-infected children but are unusual in HIV- infected adults. Conjunctival carcinoma is seen in HIV- infected individuals in sub-Saharan Africa but it is uncom- mon in Western countries. Most other cancers do not seem to have increased incidences in HIV infection. The etiologic mechanisms of HIV-related cancer likely differ among these diverse cancers and do not globally increase cancer risk. [Monogr Natl Cancer Inst 1998;23:23-25] Kaposi’s sarcoma and non-Hodgkin's lymphoma are rela- tively frequent outcomes of human immunodeficiency virus (HIV) infection. Several additional tumors appear to be associ- ated with HIV infection, albeit with smaller relative and absolute risks. The evidence is strongest for associations of HIV with anogenital neoplasia, Hodgkin's disease, testicular seminoma, pediatric leiomyosarcoma, and conjunctival cancer. Most other cancers, however, including the carcinomas most common in the general population, do not appear to be increased in HIV infec- tion. Excluding Kaposi's sarcoma and non-Hodgkin's lymphoma, the overall risk of all other cancers in HIV-infected individuals is similar to that of the general population. The Viral Epidemi- ology Branch follows two cohorts of HIV-infected hemophilia patients. The cancer experience (through March 1991) of the 1261 HIV-infected subjects in the Multicenter Hemophilia Co- hort Study has been previously reported (7). The National Can- cer Institute Registry of HIV-Infected Hemophilia Patients fol- lows an additional 1639 subjects recruited in 1991 and 1992. Through June 1996, there were a total of 7675 person-years of observation for the two cohorts combined. With the exclusion of Journal of the National Cancer Institute Monographs No. 23, 1998 Kaposi’s sarcoma and non-Hodgkin’s lymphoma, a total of 20 cases of other cancers were observed in comparison with the 13.2 expected, for a relative risk (RR) of 1.5 (95% confidence interval [CI] = 0.9-2.3). Unlike non-Hodgkin’s lymphoma in- cidence, the incidence of other cancers did not significantly in- crease with the duration of HIV infection (Fig. 1). HIV and concomitant immunosuppression has been associ- ated with cervical carcinoma in situ in studies by Vermund et al. (2), Williams et al. (3), Klein et al. (4), Ho et al. (5), and others. Although cervical cancer was added to the 1993 surveillance definition for AIDS by the U.S. Centers for Disease Control and Prevention, an effect of HIV on invasive cancer is uncertain. Between 1976 and 1988, invasive cervical cancer incidence de- creased approximately 40% in New York City black women, a group with a high prevalence of HIV infection (6). Furthermore, New York City AIDS cases from 1992 through 1993 had only a fourfold increase of cervical cancer relative to the general popu- lation (7), despite likely confounding by shared behavioral risk factors for HIV and human papillomavirus (HPV) infection. Similarly, there was only one case of invasive cervical cancer in 3612 woman-years of observation after AIDS in the AIDS- Cancer Match Registry (Goedert JJ: personal communication), which matches population-based cancer and AIDS registries in Puerto Rico and seven regions of the United States (8). While the impact of Pap screening is difficult to assess, a large excess risk of invasive cervical cancer seems unlikely by these data. Anal intraepithelial neoplasia has also been associated with HIV and/or HIV-related immunosuppression in studies by Palef- sky et al. (9), Kiviat et al. (/0), and others. As with cervical cancer, an etiologic association of HIV with invasive anal cancer is uncertain. Incidence in single San Francisco men aged 25-54 years (a population approximately 50% homosexual) was 2.0 per 100000 from 1973 through 1979, which was 9.9 (95% CI = 4.5-18.7) times that of the general population in the same pe- riod. Thus, even before the HIV epidemic, these men were at very high risk of invasive anal cancer. While the rate increased to 3.9 per 100000 in 1988 through 1990, incidence in the general population increased in parallel, so the RR was virtually un- changed at 10.1 (95% CI = 5.0-18.0) times expected (1/7). In more recent data through 1994, the number of anal cancer cases *Affiliation of author: Viral Epidemiology Branch, Division of Cancer Epi- demiology and Genetics, National Cancer Institute, Bethesda, MD. Correspondence to: Charles S. Rabkin, M.D., M.Sc., National Institutes of Health, Executive Plaza North, Rm. 434, Bethesda, MD 20892. See “Note” following ‘‘References.”’ 23 jects, 17780 HIV-positive person-years), the New South Wales AIDS-Cancer Match (1/6) (3616 subjects), and the National Cancer In- stitute hemophilia studies. The RR of Hodg- kin’s disease was 2.5 in the hepatitis B cohorts and ranged from 5.6 to 8.5 in the other studies; the 95% CI excluded 1.0 in all four studies (Table 1). The hepatitis cohort found the low- est risk of total non-AIDS cancers (Table 1), perhaps because cases were ascertained by matching with cancer registrations rather than by active follow-up. The four studies are less consistent with re- spect to HIV-associated increases in other can- cers (Table 1). An increase in testicular semi- noma was previously reported in the Years After HIV-Seroconversion 2.07 — ct ————— rere: — — 1.5 11 == Non-Hodgkin's Lymphoma Xs =—e— Other Cancers (excluding Kaposi's Sarcoma) a : / =~ 1.0 0 0 I) © oO 0.5 AN 0.0 - > 0 5 10 . Multicenter AIDS Cohort Study by Lyter et al. 15 (15) but was not observed in the other studies. Statistically significant increases (i.e., 95% CI Fig. 1. Incidence of non-Hodgkin's lymphoma and of other cancers (excluding Kaposi's sarcoma) in the Multicenter Hemophilia Cohort Study by duration of HIV infection. in San Francisco men has continued to increase. Moreover, about half of the cases since 1985 have occurred in association with AIDS, diagnosed either before or after anal cancer (Fig. 2). While these data may reflect some excess risk with advanced HIV infection, the relative excess appears to be small in com- parison with the 10-fold RR predating AIDS. Early evidence that Hodgkin's disease is increased by HIV infection came from the San Francisco City Clinic Cohort (12), which has been updated in a combined analysis (total 15565 subjects) with the New York City hepatitis B natural history and vaccine trial cohorts (13). Additional evidence came from a San Francisco AIDS and cancer registry-linkage study (/4). These findings have been corroborated in several additional studies, including the Multicenter AIDS Cohort Study (/5) (2683 sub- excluding 1.0) of multiple myeloma and oral cancer were observed in the Multicenter AIDS Cohort Study and the New South Wales AIDS— Cancer Match. Malignant melanoma was significantly increased in the Multicenter AIDS Cohort Study yet not in New South Wales where the disease is more common. These inconsistent associations warrant further investigation be- fore being accepted or discarded. Notably absent from this list are increases in leiomyosarcoma and squamous cell conjunctival cancer. HIV-infected children are at greatly increased risk of leiomyosarcoma and benign leio- myomas, as first noted by Chadwick et al. (/7). These tumors have only rarely been noted in HIV-infected adults (/8), despite the much larger number potentially at risk. The HIV-associated cases uniformly contain Epstein-Barr virus (EBV), leading to speculation that prior EBV infection protects HIV-infected adults from this disorder. Similarly, conjunctival squamous cell cancer is found relatively frequently in AIDS cases in sub- Saharan Africa (19), yet is rare in the United States. Differences in solar ultra- 10 AIDS-Cancer Match 8 [J Other Cases 2 6 n © oO 4 2 0 1973 1976 1979 1982 1985 1988 Year violet radiation exposure or in prevalence of conjunctival HPV infection may ex- plain the variation. Despite their limited variety, HIV- associated tumors are likely to be diverse in their pathology and etiology. Viral co- factors, such as EBV, HPV, and human herpesvirus type 8 may play a role in some of these disorders, but not all viral- associated tumors are increased in AIDS. Notable exceptions are hepatocellular carcinoma (despite high prevalence of both hepatitis B and hepatitis C in HIV- infected hemophilia patients) and naso- pharyngeal carcinoma (despite near universal prevalence of EBV). Immuno- 1991 1994 Fig. 2. Anal cancer cases and AIDS comorbidity in San Francisco County men aged 25-54 years, SEER’ Program and AIDS—Cancer Match Registry, from 1973 through 1994. Shaded boxes = cases with AIDS at or after cancer diagnosis; unshaded boxes = difference in a given year between the total cases detected by cancer surveillance and the number detected by the AIDS—Cancer Match Registry. 24 dysregulation or cytokine imbalance may underlie the various tumor associations. More importantly, even advanced immu nodeficiency does not appear to lead to in- Journal of the National Cancer Institute Monographs No. 23, 1998 Table 1. Relative risks of selected cancers in cohort- and registry-matching studies of HIV and cancer* Cancer type or site Cohort No. of cases Relative risk 95% Cl Total non-AIDS NCI Hemophilia Cohort and Registry 20 ) 1.5 09-2.3 Multicenter AIDS Cohort Study 51 2.6 1.9-3.4 NYC and SF Hepatitis B Studies 168 0.7 0.6-0.8 New South Wales AIDS—Cancer Match 70 _— _— Hodgkin's disease NCI Hemophilia Cohort and Registry 3 5.6 1.1-16 Multicenter AIDS Cohort Study 5 6.7 2.2-16 NYC and SF Hepatitis B Studies 18 23 1.5-3.9 New South Wales AIDS—Cancer Match 10 8.5 4.1-16 Testicular seminoma NCI Hemophilia Cohort and Registry 2 2.5 0.3-8.8 Multicenter AIDS Cohort Study 7 3.9 1.6-8 NYC and SF Hepatitis B Studies 9 0.5 0.2-0.9 New South Wales AIDS—Cancer Match 4 1.2 0.3-2.9 Multiple myeloma NCI Hemophilia Cohort and Registry 0 0 — Multicenter AIDS Cohort Study 3 14.2 2.941 NYC and SF Hepatitis B Studies 1 0.5 0-2.8 New South Wales AIDS—Cancer Match 4 5.8 1.2-17 Melanoma NCI Hemophilia Cohort and Registry 3 37 0.8-11 Multicenter AIDS Cohort Study 6 2.9 1.1-6.3 NYC and SF Hepatitis B Studies 11 0.6 0.31.1 New South Wales AIDS-Cancer Match 15 1.3 0.8-2.2 Oral NCI Hemophilia Cohort and Registry 2 33 0.4-12 Multicenter AIDS Cohort Study 5 3.9 1.3-9.1 NYC and SF Hepatitis B Studies 8 0.6 0.3-1.2 New South Wales AIDS—Cancer Match 6 4.1 1.6-9.5 *¥HIV = human immunodeficiency virus; NCI = National Cancer Institute; NYC = New York City: SF = San Francisco; AIDS = acquired immunodeficiency syndrome; and CI = confidence interval. creases in most types of cancers. The spectrum of HIV- associated cancers may further develop as HIV-infected persons survive longer with highly active antiretroviral therapies. Deter- mining the mechanisms of the specific cancers increased with HIV infection will likely advance the general understanding of cancer etiology. References (1) Rabkin CS, Hilgartner MW, Hedberg KW, Aledort LM, Hatzakis A, Eichinger S, et al. Incidence of lymphomas and other cancers in HIV- infected and HIV-uninfected patients with hemophilia. JAMA 1992;267: 1090-4. Vermund SH, Kelley KF, Klein RS, Feingold AR, Schreiber K, Munk G, et al. High risk of human papillomavirus infection and cervical squamous intraepithelial lesions among women with symptomatic human immuno- deficiency virus infection. Am J Obstet Gynecol 1991;165:392-400. Williams AB, Darragh TM, Vranizan K, Ochia C, Moss AR, Palefsky JM. Anal and cervical human papillomavirus infection and risk of anal and cervical epithelial abnormalities in human immunodeficiency virus- infected women. Obstet Gynecol 1994;83:205-11. Klein RS, Ho GY, Vermund SH, Fleming I, Burk RD. Risk factors for squamous intraepithelial lesions on Pap smear in women at risk for human immunodeficiency virus infection. J Infect Dis 1994;170:1404-9. (5) Ho GY. Burk RD, Fleming I, Klein RS. Risk of genital human papilloma- virus infection in women with human immunodeficiency virus-induced immunosuppression. Int J Cancer 1994;56:788-92. (6) Rabkin CS, Biggar RJ, Baptiste MS, Abe T, Kohler BA, Nasca PC. Cancer incidence trends in women at high risk of human immunodeficiency virus (HIV) infection. Int J Cancer 1993:55:208-12. (7) Chiasson MA, Kelley KF, Williams R, Mikl J, Forlenza S, Smith PF. Invasive cervical cancer (ICC) in HIV+ women in New York City (NYC). 3rd Conf Retro and Opportun Infect 130, 1996 [abstract]. (8) Cote TR, O’Brien TR, Ward JW, Wilson SE, Blattner WA. AIDS and cancer registry linkage: measurement and enhancement of registry com- pleteness. The National AIDS/Cancer Match Study Group. Prev Med 1995; 24:375-7. (9) Palefsky JM, Gonzales J, Greenblatt RM, Ahn DK, Hollander H. Anal (2 ~ (3 ~— (4 = Journal of the National Cancer Institute Monographs No. 23, 1998 intraepithelial neoplasia and anal papillomavirus infection among homo- sexual males with group IV HIV disease. JAMA 1990:;263:2911-6. (10) Kiviat NB, Critchlow CW, Holmes KK, Kuypers J, Sayer J. Dunphy C, et al. Association of anal dysplasia and human papillomavirus with immuno- suppression and HIV infection among homosexual men. AIDS 1993;7: 43-9. (11) Rabkin CS, Yellin F. Cancer incidence in a population with a high preva- lence of infection with human immunodeficiency virus type 1. J Natl Can- cer Inst 1994:86:1711-6. (12) Hessol NA, Katz MH, Liu JY, Buchbinder SP, Rubino CJ, Holmberg SD. Increased incidence of Hodgkin disease in homosexual men with HIV infection. Ann Intern Med 1992;117:309-11. (13) Koblin BA, Hessol NA, Zauber AG, Taylor PE, Buchbinder SP, Katz MH, et al. Increased incidence of cancer among homosexual men, New York City and San Francisco, 1978-1990. Am J Epidemiol 1996;144:916-23. (14) Reynolds P, Saunders LD. Layetsky ME, Lemp GF. The spectrum of acquired immunodeficiency syndrome (AIDS)-associated malignancies in San Francisco, 1980-1987. Am J Epidemiol 1993;137:19-30. (15) Lyter DW, Kingsley LA, Rinaldo CR, Bryant J. Malignancies in the mul- ticenter AIDS cohort study (MACS), 1984-1994 (Meeting abstract). Proc ASCO 1996;15:A852. (16) Grulich A, Wan X, Law M, Coates M, Kaldor J. Rates of non-AIDS defining cancers in people with AIDS. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A18 [abstract]. (17) Chadwick EG, Connor EJ, Hanson IC, Joshi VV, Abu-Farsakh H, Yoger R, et al. Tumors of smooth-muscle origin in HIV-infected children. JAMA 1990:263:3182-4. (18) Steel TR, Pell MF, Turner JJ, Lim GH. Spinal epidural leiomyoma occur- ring in an HIV-infected man. Case report. J Neurosurg 1993;79:442-5. (19) Ateenyi-Agaba C. Conjunctival squamous-cell carcinoma associated with HIV infection in Kampala, Uganda. Lancet 1995;345:695-6. Note "Editor's note: SEER is a set of geographically defined, population-based central tumor registries in the United States, operated by local nonprofit orga- nizations under contract to the National Cancer Institute (NCI). Each registry annually submits its cases to the NCI on a computer tape. These computer tapes are then edited by the NCI and made available for analysis. Papillomaviruses and Cervical Cancer: Pathogenesis and Vaccine Development Douglas R. Lowy, John T. Schiller* A subset of human papillomaviruses (HPVs) has been impli- cated as the principal etiologic agents of cervical cancer. Cervical cancers consistently retain and express two of the viral genes, E6 and E7. Although infection with HPV seems to be necessary, other factors, such as cellular immune func- tion, play an important role in determining whether cervical infection will regress, persist, or progress to cancer. The close relationship between viral infection and cancer makes HPV an attractive target for prophylactic and therapeutic vaccines. Candidate vaccines have been shown to have effi- cacy in animal models, and human clinical trials are planned or in progress. [Monogr Natl Cancer Inst 1998;23:27-30] In addition to inducing benign papillomas of the skin and mucous membranes, some human papillomaviruses (HPVs) are clearly associated with the development of malignant epithelial tumors (/—3)." These cancers include anogenital cancers, espe- cially cancer of the cervix, which is the second most common cancer among women worldwide. A wealth of epidemiologic and molecular biologic data now points to an etiologic link between HPV infection and most cervical cancers. Recognition of the clinical importance of papillomaviruses has stimulated efforts to develop vaccines that may treat or prevent benign and malignant diseases associated with papillomavirus infection. This article considers the relationship between papillomavirus infection and cervical cancer and describes recent approaches to papillomavirus vaccines. Subset of Genital/Mucosal HPV Types Found in Cervical Cancer The HPV replicative cycle is limited to stratified squamous epithelia, with no viremic phase (4). Most of the virus gene expression and replication take place in suprabasal cells that are undergoing terminal differentiation, a strategy that seems de- signed to evade immune surveillance. More than 70 different HPV types have been identified on the basis of the sequence divergence between their DNA genomes (2). Their genomes, which are a closed, circular, double-stranded DNA approxi- mately 8 kilobases in length, share a similar structural and ge- netic organization (5). Only a subset of the HPV types appears to regularly infect the genital epithelia. Some of these so-called genital/mucosal types, such as HPV6 and HPV11, are almost never found in cervical cancer and have thus been designated “low-risk” viruses. Others, such as HPV16 and HPV 18, are found regularly in cervical cancer and have therefore been des- ignated ‘‘high-risk’” HPV types (2,3,6). Many different HPV types have been found in cervical cancer. However, a recent analysis of almost one thousand cervical can- Journal of the National Cancer Institute Monographs No. 23, 1998 cers from different regions of the world (6) showed that HPV 16 was consistently the most common type, being present in about one half of the cancers from any region. HPV16, HPVIS, HPV31, or HPV45 was detected in about 80% of the cancers in every region. HPV infection of the cervix precedes the onset of cancer by many years. Reliable epidemiologic evidence has shown that HPV infection is by far the most important risk factor (10-fold to 200-fold, compared with controls) for the development of cervical dysplasias, from which almost all cervical cancers arise (3,7-9). The peak in cervical cancer incidence is more than 20 years after the peak in incidence of high-risk genital HPV in- fection, suggesting that infection per se is insufficient to cause cancer. In most women, cervical infection even with a high-risk HPV is self-limited. Low-risk viruses do not seem to possess the intrinsic capacity to induce cervical cancer. Inactivation by High-Risk HPVs of Proteins Controlling Cell Proliferation Considerable progress has been made in identifying poten- tially important differences between high-risk and low-risk geni- tal/mucosal HPVs (10). Normal human keratinocytes in culture have a finite life span, which is not increased when low-risk HPVs are introduced into them. However, high-risk types, in contrast to low-risk types, can reproducibly induce the immor- talization of the keratinocytes. This ability to be passaged in- definitely is a characteristic of partial cellular transformation. Further analysis of high-risk HPV indicated that keratinocyte immortalization requires two of the viral genes, E6 and E7, which are the same two viral genes that are preferentially re- tained and expressed in cervical tumors (//,12). Biochemical analysis of the proteins encoded by E6 and E7 has shown that they form complexes with and inactivate specific cellular pro- teins that probably contribute to the limited life span of cultured keratinocytes and inhibit cell growth (10,13, 14). These activities are much lower or are lacking in the E6 and E7 of low-risk viruses. The biological significance of such complexes with high-risk E6 and E7 has been shown most clearly for E6 and the p53 protein and for E7 and the pRB protein, which is encoded by the retinoblastoma tumor susceptibility gene RB. Since p53 and pRB normally control cell proliferation, abrogating these func- tions places the cells at much greater risk of malignant progres- sion. *Affiliation of authors: Laboratory of Cellular Oncology, Division of Basic Sciences, National Cancer Institute, Bethesda, MD. Correspondence to: Douglas R. Lowy, M.D., National Institutes of Health, Bldg. 36, Rm. 1D32, Bethesda, MD 20892. See “Note” following ‘ ‘References.’ 27 As in the clinical situation, cultured keratinocytes that have been recently immortalized by high-risk HPV are often resistant to differentiation signals but are not fully transformed. Contin- ued passage or the addition of an activated ras oncogene can, in some instances, render these cells tumorigenic for nude mice (15). Additional Changes Required for Tumorigenic Progression The long interval between HPV infection and the develop- ment of cervical cancer suggests that factors other than viral infection are required for progression to high-grade dysplasia and tumor development. Both virus-specific factors and immune reactivity appear to be important. While the viral DNA in benign lesions remains extrachromo- somal, it is integrated into the host genome in most cervical cancers. During progression, viral expression is usually limited to E6 and E7. Other changes associated with progression likely involve E6- and E7-induced chromosome instability that results from deregulation of cellular growth-control genes and telom- erase activity (16). Persistence of viral infection is associated with progression (17). The likelihood of progression to invasive cancer appears to depend in part on the relative oncogenic potential of the HPV type (18). Some evidence also suggests that genotypic variants within a high-risk type may affect the potential for progression to high-grade dysplasia (/9). It remains to be determined wheth- er these apparent differences in likelihood of progression mainly represent biological differences between the viruses or differ- ences in host response. Cellular immunity to the viral infection seems to represent a critical determinant of whether dysplastic lesions will develop, regress, persist, or progress. HPV 16-infected patients with more severe cervical cytology are less likely to show positive cellular immune responses to E6 and E7 antigens than are patients with less severe cytology or with a history of previous HPV 16 infec- tion (20,21). There is some evidence that loss of major histo- compatibility complex (MHC) class 1 expression may allow some lesions to evade immune surveillance and progress more rapidly (22). Progression is more common in long-term renal transplant patients on immunosuppressive therapy or in women who are human immunodeficiency virus (HIV) positive, indi- cating the important role of cellular immune function in host defense (3). Vaccination Against HPV Infection The recognition that HPV infection plays the central etiologic role in cervical cancer has fostered efforts to develop vaccines against HPV. Both prophylactic and therapeutic forms of vac- cines are under development (23-28). They seek, respectively, to prevent infection or to induce regression of established infec- tion via immune recognition of specific HPV-encoded proteins or peptides. Such vaccines can be delivered either directly as protein, as DNA that encodes and expresses the requisite viral protein(s), or by heterologous viral vectors (29). General Considerations A major barrier to developing a practical vaccine is that most critical HPV-immune determinants are likely to be type specific 28 because the proteins of different HPV types are quite divergent at the amino acid level. This limitation implies that protection induced by protein from a given HPV type is likely to be type specific. It will therefore be necessary either to have a specific vaccine for each HPV type or to incorporate viral protein from an appropriate spectrum of HPV types in a polyvalent vaccine. Further difficulties are that HPV does not cause disease in animals and is not infectious for them. This means that animal studies must be carried out with animal papillomaviruses, with grafted human material in immunologically suppressed hosts, or with model systems that incorporate specific HPV genes or pro- teins. It is not always clear whether experimental results ob- tained with these animal models will apply directly to clinical HPV infection in humans. Prophylactic Vaccines Encouraging results have come from animal studies of vac- cines to prevent papillomavirus infection. Consistent protection (90%—-100%) has been obtained by immunizing animals with virus-like particles (VLPs) composed of the major structural viral protein L1 (30-32). The papillomavirus capsid is primarily composed of 360 mol- ecules of L1 protein, and L1, when expressed in the absence of other papillomavirus genes, can self-assemble into VLPs that are morphologically and immunologically similar to infectious pap- illomavirus (33,34). Since the VLPs are produced by genetically engineered cells that do not contain the nonstructural viral genes, such as E6 or E7, the VLPs do not contain the papillomavirus DNA genome, are not infectious, and cannot cause neoplastic changes in cells. Immunization with VLPs (30) or with the L1 gene (35) has produced substantial protection in rabbits against experimental challenge with the Shope cottontail rabbit papillomavirus (CRPV), which induces cutaneous papillomas that can progress to malignant squamous cell carcinomas. VLP immunization can also prevent experimental oral mucosal infection in dogs by canine oral papillomavirus (37) and in cows by bovine papillo- mavirus type 4 (BPV4) (32). In these studies, the protection, which can be passively trans- ferred by antibodies from immune animals to nonimmune ani- mals, is mediated by antibodies directed against conformational epitopes that are present on the VLP as well as on infectious papillomaviruses. Since the conformational epitopes are type specific (36-38), protection is type specific (30). The efficacy of these protocols seems to be limited to prevention; BPV4 VLP immunization did not induce regression of established BPV4 papillomas (32). These promising animal studies are leading to the testing of a candidate prophylactic HPV vaccine in humans. In addition to deciding which HPV types to include in such a vaccine, it will be desirable to optimize the adjuvant, dosage, and route of ad- ministration. Efforts to improve the mucosal immunity induced by VLPs may increase their efficacy against mucosal genital infection. However, there are reasons to believe that preventing genital mucosal HPV infection may not require the induction of muco- sal immunoglobulin (Ig) A. In addition to the animal studies that show protection against experimental oral infection, transuda- tion of IgG into vaginal secretions can be induced by systemic Journal of the National Cancer Institute Monographs No. 23, 1998 immunization with purified protein (39). Furthermore, it is be- lieved that initiation of genital HPV infection occurs only when there is sufficient trauma to allow the virions to come in contact with the proliferating epithelial cells, which are located in the basal layer of the epithelium. This hypothesis is supported by the tentative identification of a, integrin, which is not expressed in suprabasal cells, as a candidate receptor for papillomaviruses (40). It is reasonable to expect that trauma which was sufficient to abrade the epithelium would usually be associated with exu- dation of systemic IgG into the abraded area, where it might neutralize the HPV virions. Therapeutic Vaccines Therapeutic vaccines might be used in various settings, in- cluding the treatment of invasive cancers, as adjunct therapy to prevent recurrence or metastasis, against dysplasias, or in benign disease. A major theoretical obstacle to developing such vac- cines is that the immunologic determinants for viral persistence or regression remain poorly defined (41,42), although it is clear that patients with impaired cellular immunity are at increased risk of persistent HPV infection and carcinogenic progression. Most efforts have been directed toward using the E6 and E7 proteins, or peptides derived from them, largely because these are the viral proteins that are retained and expressed in cervical tumors (43). However, vaccines directed against benign lesions would not need to be limited to these two proteins. For example, El and E2, which are required for the viral DNA to be main- tained as an extrachromosomal element that is unintegrated in the host DNA, represent potentially interesting targets for benign lesions. Since these papillomavirus proteins are not expressed on the cell surface, there is little potential for antibody-dependent cy- totoxicity to mediate regression. Instead, potentially effective cytotoxic responses will probably require a vaccine that induces the presentation of small virally encoded peptides to antigen- presenting cells. In cells that possess class I molecules, the nor- mal process of partial intracellular degradation of cytoplasmic or nuclear viral proteins can, following the binding of small viral peptides to the class I molecules, lead to the induction of anti- gen-specific reactivity of CD8-positive cytotoxic T lymphocytes (CTLs). Introduction of the relevant viral gene or protein into target cells is normally required to develop virus-specific CTLs. Pap- illomavirus genes can be introduced into cells as naked DNA, which is a relatively inefficient process, or as part of a viral vector such as vaccinia virus, which is usually more efficient. Rodents immunized with HPV 16 E6 or HPV 16 E7, delivered either in vaccinia virus vectors containing one of the two viral genes or in killed tumor cells that expressed one of the genes and the gene for immune co-stimulatory protein B7, were protected against subsequent challenge with tumor cells expressing the corresponding viral protein (44-46). A recombinant vaccinia human safety trial of an HPV 16 and HPV 18 E6 and E7 virus has been completed (47), and others are under way or planned (28). A CTL response can also be induced by the viral protein itself if it is taken up by cells in a manner that leads to its partial degradation and presentation with class I antigen. Although soluble protein by itself does not usually generate such a re- sponse, injection of rabbits with bacterially derived CRPV El or Journal of the National Cancer Institute Monographs No. 23, 1998 E2 protein was shown to increase the rejection rate of CRPV- induced papillomas (48). The precise immunologic mechanism underlying this effect remains to be established. However, class I-dependent CD8 CTLs can be reproducibly induced if viral protein is presented in particulate form (49), as part of a VLP, in combination with some adjuvants, or encapsidated into lipo- somes. Another alternative is to immunize with small viral peptides, which, following their binding to empty class I molecules on cell surfaces, can induce cytotoxic CD8 T cells (50). A difficulty associated with this approach is that the immunogenicity of a given peptide is genetically determined by the class I alleles present in a given individual, which means that only some in- dividuals will respond to even an ‘‘immunogenic’’ peptide. Also, small peptides are often quite unstable in vivo (51). In some tumor models, immunologic rejection is mediated by antigen-specific CD4 T cells rather than by CDS T cells (52,53). While the class I CD8 pathway is characteristic of cytoplasmic and nuclear proteins, CD4 T cells can be activated by the pro- cessing of endogenously expressed membrane-associated pro- teins for processing and presentation by MHC class II molecules to CD4 cells. To induce E7-specific CD4 cells, a study (54) used genetic engineering to target the ordinarily nuclear HPV16 E7 protein to the lysosomal compartment by adding a lysosomal membrane targeting signal to E7. Such a lysosomally targeted E7 protein, which was introduced into mice via a vaccinia virus vector carrying the engineered E7 gene, was highly effective in protecting mice against tumors derived from a mouse epithelial cell line transformed by HPV16 E6 and E7 genes and a ras oncogene (55). In this tumor model, protection by the lysosomal E7 protein was found to depend on both CD4 and CDS cells. This observation may explain why authentic E7, which presum- ably induced only active CD8 cells, was much less effective against the tumors than was the lysosomally targeted protein. Thus, various approaches are being taken to develop HPV vaccines. Information from human clinical trials with candidate vaccines may be expected in the next few years. References (1) Lowy DR, Kirnbauer R. Schiller JT. Genital human papillomavirus infec- tion. Proc Natl Acad Sci U S A 1994;91:2436-40. (2) zur Hausen H. Molecular pathogenesis of cancer of the cervix and its causation by specific human papillomavirus types. Curr Top Microbiol Immunol 1994;186:131-56. (3) Human papillomaviruses. IARC Monogr Eval Carcinog Risks Hum, vol 64. 1995. (4) Taichman LB, LaPorta RF. The expression of papillomaviruses in human epithelial cells. In: Salzman NP, Howley PM, editors. The papovaviridae: vol 2. The papillomaviruses. New York: Plenum Press, 1987:109-39. (5) Turek L. The structure, function and regulation of papillomaviral genes in infection and cervical cancer. Adv Virus Res 1994;44:305-56. (6) Bosch FX, Manos MM, Munoz N, et al. Prevalence of human papilloma- virus in cervical cancer: a worldwide prospective. J Natl Cancer Inst 1995; 87:796-802. (7) Schiffman M, Bauer HM, Hoover RN, et al. Epidemiologic evidence show- ing that human papillomavirus infection causes most cervical intraepithelial neoplasia. J Natl Cancer Inst 1993;85:958-64. (8) Schiffman MH. Epidemiology of cervical human papillomaviruses. In: zur Hausen H, editor. Human pathogenic papillomaviruses. Heidelberg: Springer-Verlag, 1994:55-81. (9) Munoz N, Bosch FX, de Sanjose S, et al. The role of HPV in the etiology of cervical cancer. Mutat Res 1994:305:293-301. (10) Werness BA, Munger K, Howley PM. Role of the human papillomavirus oncoproteins in transformation and carcinogenic progression. In: DeVita 29 VT Jr, Hellman S, Rosenberg SA, editors. Important advances in oncology 1991. Philadelphia: Lippincott, 1991:3-18. (11) Vousden K. Interactions of human papillomavirus transforming proteins with the products of tumor suppressor genes. FASEB J 1993;7:872-9. (12) Mansur CP, Androphy AJ. Cellular transformation by papillomavirus on- coproteins. Biochim Biophys Acta 1993;1155:323-45. (13) Scheffner M, Huibregtse JM, Vierstra RD, et al. The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53. Cell 1993;75:495-505. (14) Chen JJ, Reid CE, Band V, et al. Interaction of papillomavirus E6 onco- proteins with a putative calcium-binding protein. Science 1995:269: 529-31. DiPaolo JA, Woodworth CD, Popescu NC, et al. Induction of human cervical squamous cell carcinoma by sequential transfection with human papillomavirus 16 DNA and viral Harvey ras. Oncogene 1989;4:395-9. (16) Klingelhutz AJ, Foster SA, McDougall JK. Telomerase activation by the E6 gene product of human papillomavirus type 16. Nature 1996;380:79-82. (17) Gloria YF, Ho F, Burk RD, et al. Persistent genital human papillomavirus infection as a risk factor for persistent cervical dysplasia. J Natl Cancer Inst 1995:87:1365-71. (18) Lorincz AT, Reid R, Jenson AB, et al. Human papillomavirus infection of the cervix: relative risk associations of 15 common anogenital types. Obstet Gynecol 1992:79:328-37. (19) Xi LF, Koutsky LA, Galloway DA, et al. Genomic variation of human papillomavirus type 16 and risk for high grade cervical intraepithelial neo- plasia. J Natl Cancer Inst 1997;89:796-802. (20) Tsukui T, Hildesheim A, Schiffman M, et al. Interleukin 2 production in vitro by peripheral lymphocytes in response to human papillomavirus- derived peptides: correlation with cervical pathology. Cancer Res 1996:56: 3967-74. (21) Nakagawa M, Stites D, Farhat S, et al. Cytotoxic T lymphocyte responses to E6 and E7 proteins of human papillomavirus type 16: relationship to cervical intraepithelial neoplasia. J Infect Dis 1997;175:927-31. (22) Connor ME, Stern PL. Loss of MHC class-1 expression in cervical carci- nomas. Int J Cancer 1990;46:1029-34. (23) Galloway DA. Human papillomavirus vaccines: a warty problem. Infect Agents Dis 1994:3:187-93. (24) Hines JF, Ghim S, Schlegel R, et al. Prospects for a vaccine against human papillomavirus. Obstet Gynecol 1995;86:860—6. (25) Munoz N, Crawford L, Coursaget P. HPV vaccines and their potential use in the prevention and treatment of cervical neoplasia. Papillomavirus Rep 1995:6:54-5. (26) Schiller JT, Okun M. Papillomavirus vaccines: current status and future prospects. Adv Dermatol 1996;11:355-80. (27) Frazer 1. Strategies for immunoprophylaxis and immunotherapy of papil- lomaviruses. Clin Dermatol 1997;15:285-97. (28) Hanissian J. Emerging HPV vaccines. Infect Med 1997;14:266,273-275, 330. (29) Rabinovich NR, McInnes P, Klein DL, et al. Vaccine technologies: view to the future. Science 1994;265:1401-4. (30) Breitburd F, Kirnbauer R, Hubbert NL, et al. Immunization with virus-like particles from cottontail rabbit papillomavirus (CRPV) can protect against experimental CRPV infection. J Virol 1995;69:3959-63. (31) Suzich JA, Ghim S, Palmer-Hill FJ, et al. Systemic immunization with papillomavirus LI protein completely prevents the development of viral mucosal papillomas. Proc Natl Acad Sci U S A 1995;92:11553-7. (32) Kirnbauer R, Chandrachud L, O’Neil B, et al. Virus-like particles of bovine papillomavirus type 4 in prophylactic and therapeutic immunization. Vi- rology 1996:;219:37-44. (33) Kirnbauer R, Booy F, Cheng N, et al. Papillomavirus L1 major capsid protein self-assembles into virus-like particles that are highly immuno- genic. Proc Natl Acad Sci U S A 1992;89:12180-4. (34) Hagensee ME, Yaegashi N, Galloway DA. Self-assembly of human pap- illomavirus type 1 capsids by expression of the L1 protein alone or by coexpression of the L1 and L2 capsid proteins. J Virol 1993;67:315-22. (35) Donnelly JJ, Martinez D, Jansen KU, et al. Protection against papilloma- virus with a polynucleotide vaccine. J Infect Dis 1996;173:314-20. (15 =! (36) Christensen ND, Kirnbauer R, Schiller JT, et al. Human papillomavirus types 6 and 11 have antigenically distinct strongly immunogenic confor- mationally dependent neutralizing epitopes. Virology 1994:205: 329-35. (37) Roden RB, Hubbert NL, Kirnbauer R, et al. Assessment of the serological relatedness of genital human papillomaviruses by hemagglutination inhi- bition. J Virol 1996;70:3298-301. (38) Roden RB, Greenstone HL, Kirnbauer R, et al. In vitro generation and type-specific neutralization of a human papillomavirus type 16 virion pseu- dotype. J Virol 1996;70:5875-83. (39) Bouvet JP, Belec L, Pires R, et al. Immunoglobulin G antibodies in human vaginal secretions after parenteral vaccination. Infect Immun 199462: 3957-61. (40) Evander M, Frazer IH, Payne E, et al. Identification of the a integrin as a candidate receptor for papillomaviruses. J Virol 1997:;71:2449-56. (41) Tindle RW, Frazer IH. Immune response to human papillomaviruses and the prospects for human papillomavirus-specific immunization. Curr Top Microbiol Immunol 1994;186:217-53. (42) Dillner J. Serology of human papillomavirus. Cancer J 1995:8:264-9. (43) Steller MA, Schiller JT. Human papillomavirus immunology and vaccine prospects. Monogr Natl Cancer Inst 1996:21:145-8. (44) Meneguzzi G, Cerni C, Kieny MP, et al. Immunization against human papillomavirus type 16 tumor cells with recombinant vaccinia virus ex- pressing E6 and E7. Virology 1991;181:62-9. (45) Chen L, Thomas EK, Hu SL, et al. Human papillomavirus type 16 nucleo- protein E7 is a tumor rejection antigen. Proc Natl Acad Sci U S A 1991; 88:110-4. (46) Chen L, Mizuno MT, Singhal MC, et al. Induction of cytotoxic T lympho- cytes specific for a syngeneic tumor expressing the E6 oncoprotein of human papillomavirus type 16. J Immunol 1992;148:2617-21. (47) Borysiewicz LK, Fiander A, Nimako M, et al. A recombinant vaccinia virus encoding human papillomavirus types 16 and 18, E6 and E7 proteins as immunotherapy for cervical cancer. Lancet 1996;1:1523-7. (48) Selvakumar R, Borenstein LA, Lin YL, et al. Immunization with nonstruc- tural proteins El and E2 of cottontail rabbit papillomavirus stimulates regression of virus-induced papillomas. J Virol 1995:69:602-5. (49) Tindle RW, Herd K, Londono P. Chimeric hepatitis B core antigen par- ticles containing B- and Th- epitopes of human papillomavirus type 16 E7 proteins induce specific antibody and T-helper responses in immunized mice. Virology 1994;200:547-57. (50) Feltkamp MC, Smits HL, Vierboom MP, et al. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur J Immunol 1993; 23:2242-9. (51) Falo LDJ, Colarusso LJ, Benacerraf B, et al. Serum proteases alter the antigenicity of peptides presented by class I major histocompatibility com- plex molecules. Proc Natl Acad Sci U S A 1992;89:8347-50. (52) Golumbek PT, Lazenby AJ, Levitsky HI, et al. Treatment of established cancer by tumor cells engineered to secrete interleukin-4. Science 1991; 254:713-6. (53) Topalian SL, Rivoltini L, Mancini M, et al. Human CD4+ T lymphocytes specifically recognize a shared melanoma-associated antigen encoded by the tyrosine gene. Proc Natl Acad Sci U S A 1994;92:9481-5. (54) Wu TC, Guarnieri FG, Staveley O, et al. Engineering an intracellular pathway for major histocompatibility complex class 11 presentation of an- tigens. Proc Natl Acad Sci U S A 1995;92:11671-5. (55) Lin KY, Guarnieri FG, Staveley-O’Carroll KF, et al. Treatment of estab- lished tumors with a novel vaccine that enhances major histocompatibility class II presentation of tumor antigen. Cancer Res 1996:;56:21-6. Note "Portions of this article are adapted from Lowy DR, Schiller JT. Oncogenesis and vaccine prospects for the papillomaviruses. Curr Opin Dermatol 1997:4: 256-61. Journal of the National Cancer Institute Monographs No. 23, 1998 Cancers in Human Immunodeficiency Virus-Infected Children Brigitta U. Mueller* Although the exact incidence of cancers in human immuno- deficiency virus (HIV)-infected children is not clear, an ex- cess of non-Hodgkin’s lymphomas and soft tissue tumors as well as a multitude of otherwise rare tumors in childhood, such as cervical, thyroid, or pulmonary carcinoma, has been reported. In contrast to the findings in HIV-infected adults, Kaposi’s sarcoma is rare in children in industrialized coun- tries but not in children living in the sub-Saharan area. Treatment of the neoplastic disease is often complicated by multiple HIV-associated organ dysfunctions as well as drug interactions and infectious complications secondary to se- vere immunosuppression. Nonetheless, preliminary results with dose-intensive, but brief, chemotherapeutic regimens have been encouraging, and HIV-infected children who de- velop cancer are likely to benefit from aggressive treatment combined with adequate supportive care. Furthermore, in- sights gained from the study and treatment of this very chal- lenging group of patients may benefit other immunocompro- mised hosts as well as increase our understanding of oncogenesis in general. [Monogr Natl Cancer Inst 1998;23: 31-35] Background and Epidemiology Worldwide, the Joint United Nations Programme on AIDS [acquired immunodeficiency disease syndrome] (UNAIDS) es- timates that 400 000 new human immunodeficiency virus (HIV) infections occurred in children in 1996 and that currently 830000 children are living with symptoms of HIV/AIDS (UNAIDS Web site). Of the 2.6 million children who were infected since the epidemic began, an estimated 1.4 million have already died. The number of children with HIV infection who develop a cancer is poorly defined. Among the 7629 children diagnosed with AIDS by the end of December 1996, 156 (2.04%) had a cancer as their AIDS-defining diagnosis (7). However, the AIDS definition revised by the Centers for Disease Control and Prevention (CDC) in 1994 lists only a limited num- ber of neoplasms: Kaposi's sarcoma (KS) lesions, primary non- Hodgkin's lymphomas (NHLs) of the central nervous system (CNS), small non-cleaved cell lymphomas, and immunoblastic large-cell NHLs of B-cell or unknown phenotype (2). Leiomy- osarcomas are listed under category B of the CDC classification, representing a ‘ ‘moderate’ clinical symptom. Furthermore, chil- dren who meet the diagnosis of AIDS with another symptom before they develop a tumor will not be registered again. It is therefore clear that the CDC registry will provide only limited information about the true incidence of neoplasms in HIV- infected children. Journal of the National Cancer Institute Monographs No. 23, 1998 Among the 156 tumors captured in the CDC AIDS definition in children, there were 50 Burkitt's lymphomas, 48 immuno- blastic lymphomas, 30 NHLs of the CNS, and 28 KS lesions. The number of KS cases is probably an overestimation because of initial problems with the pathologic diagnosis (Lindegren ML [CDC]: personal communication). If one takes into account only these numbers, a conservative estimate is that children with HIV infection appear to have at least a 100-fold higher incidence of cancers. In 1996 alone, 10 of 678 children who were newly diagnosed with AIDS received this diagnosis because of an NHL or a KS. This represents an incidence of neoplasms in children with AIDS of 1.47%, compared with healthy, non-HIV-infected children, who develop cancer in approximately 130 cases per million children (0.013%) per year (1,3). Several efforts have been made to better define the number and range of neoplastic manifestations in HIV-infected children. In a combined, retrospective survey performed by the HIV and AIDS Malignancy Branch of the National Cancer Institute (NCI), Bethesda, MD, and the Children’s Cancer Group, 59 tumors were identified in 57 children (Granovsky M [NCI]: personal communication). It is interesting that 42% of these tumors would not have been captured by the CDC classification. These data are very comparable to a prospective survey per- formed by the Pediatric Oncology Group (4). The tumors ob- served are often unusual and extremely rare in non-HIV-infected children; they include carcinoma of the thyroid, oat cell tumor of the tracheo—esophageal region, carcinomas of the vagina and cervix, hepatoblastoma, and soft tissue tumors (e.g., leiomyo- sarcomas), among others (5-8). Several reports from Africa have noted an increased incidence of retinoblastomas, nasopha- ryngeal carcinomas, and rhabdomyosarcomas, tumors not com- monly associated with immune deficiency states (9,70). There are clear differences between HIV-infected children and HIV-infected adults. For example, KS is a rare tumor in children in industrialized countries (United States and Europe), perhaps with the exception of Haiti and Romania (7/7). It is interesting that KS is the AIDS-defining illness in only 3% of adolescents between 13 and 19 years of age, but this incidence increases to 9% in young adults between 20 and 24 years of age and to 13% in adults older than 25 years (7,12). KS is more common in homosexual HIV-infected men than in heterosexual HIV-infected men or women. However, there is an increasing incidence of KS among children living in Africa (9,10,13-15). A *Correspondence to: Brigitta U. Mueller, M.D., Assistant Professor of Pedi- atrics, Harvard Medical School, Children’s Hospital, HU-215, 300 Longwood Ave., Boston, MA 02115. E-mail: mueller b@al.tch.harvard.edu © Oxford University Press 31 further difference between adults and children with HIV infec- tion is the increased incidence in children of smooth muscle tumors, leiomyomas, and leiomyosarcomas (8,/6-20) as well as the high prevalence in children of lymphoproliferative disorders (Table 1) (21-24). It is currently not clear whether there is an increased tendency of these lymphoproliferative disorders to evolve into a “‘true,”” monoclonal cancer with a rapid and inva- sive growth pattern. Non-Hodgkin’s Lymphoma As indicated above, the CDC classification includes small, non-cleaved cell lymphomas (mainly B-cell phenotype), large- cell lymphomas (immunoblastic and anaplastic tumors), and pri- mary CNS lymphomas. In non-HIV-infected children, there are two forms of small, non-cleaved cell NHLs (Burkitt's and Burkitt’s-like lympho- mas). The endemic form, more than 95% of which is associated with Epstein-Barr virus infection, presents typically with jaw tumors and has an especially high prevalence in equatorial Af- rica. The sporadic form is observed in the United States, Europe, and Asia and is 20- to 100-fold less common than the endemic form (25). Burkitt’s lymphoma constitutes about 50% of the childhood lymphomas. It has an annual incidence in children of 76.2 per million in Nigeria compared with 0.3 (U.S. blacks) and 2.0 (U.S. whites) per million in the United States. However, either form is rare compared with an extrapolated annual inci- dence of more than 4400 cases per million children with AIDS (based on the CDC statistics of three of 678 children diagnosed with AIDS because of a Burkitt's NHL in 1996) (1,12). In non-HIV-infected children, only about 20% of all NHLs are of the large-cell type, and most of them are of the B-cell phenotype (26). The incidence of large-cell tumors appears to be higher in HIV-infected children, and anaplastic T-cell NHL can occur. Among the 13 HIV-infected children with NHL followed at the NCI, five (38%) had large-cell tumors. Two were immu- noblastic B-cell tumors, two were anaplastic T-cell tumors, and one was an anaplastic tumor of non-B, non-T phenotype. Lym- phoblastic NHLs, which represent about one third of all NHLs in children in the United States, are predominantly of T-cell origin and do not appear to be more common in the HIV-infected population. They are not included in the CDC AIDS definition. Table 1. Lymphoproliferative disorders in human immunodeficiency virus-infected children* Disorder Comments Polyclonal hypergammaglobulinemia Common Reactive lymphadenopathy Common Lymphocytic interstitial pneumonitis Common, can lead to oxygen and steroid dependency Lungs, gastric mucosa, and parotid or lacrimal glands Polyclonal, often indolent Parotid or lacrimal glands Mucosa-associated lymphoid tumors Cystic mediastinal tumors Diffuse infiltrative lymphocyte syndrome Lymphomatoid papulosis Monoclonal, indolent, T-cell process with waxing and waning course Castleman’s disease Rare in children *The distinction among the different manifestations is not always well de- fined. Most of these processes are polyclonal or oligoclonal, but they might evolve into a monoclonal process. 32 Although secondary involvement of the CNS is observed, especially in children with bone marrow involvement, it is ex- tremely rare to see a primary CNS lymphoma in the non-HIV- infected child (26). However, it is the AIDS-defining diagnosis in 0.4% of children, constituting 23% of all the NHLs registered for AIDS diagnosis. However, in 1996, only one (10%) of 10 children had a CNS NHL as an AIDS-defining tumor, compared with 230 (22%) of 1030 adults (/). This finding possibly reflects a change in the disease course in children, since CNS NHLs are most prevalent in patients with very low CD4 counts and have often been diagnosed at autopsy. As in adults. the differential diagnosis, especially in the adolescent, has to include cerebral toxoplasmosis. However, we also have observed at least one case of an intracranial lymphoproliferative lesion similar to what has been described in the post-transplant situation. Furthermore, since HIV-infected children are now surviving longer, it has to be kept in mind that brain tumors are the most common solid tumors in otherwise healthy children (3). An empirical treatment for toxoplasmosis may not be warranted in the younger child, since cerebral toxoplasmosis with mass lesion(s) is extremely rare in preadolescent children, and an early biopsy should be considered. Unusual, extranodal sites of presentation of NHLs are rela- tively common in the HIV-infected child (Table 2) (27-31). Intriguing is the high incidence of mucosa-associated lymphoid tumors, observed in pulmonary or gastric mucosa as well as in the parotid, salivary, or lacrimal glands (29,32,33). These oli- goclonal or monoclonal processes are usually curable by surgi- cal excision alone and may represent an interface between lym- phoproliferative disorder and cancer. Lymphocytic interstitial pneumonitis is a common manifestation of pediatric HIV dis- ease, and it is possible that pulmonary mucosa-associated lym- phoid tumors and lymphocytic interstitial pneumonitis are re- lated (34-37). Kaposi’s Sarcoma KS is rarely diagnosed in HIV-infected children from the United States, with the exception of vertically infected children from Haiti or older adolescents (38-42). However, in African countries such as Zambia or Uganda, KS now constitutes almost 20% of all childhood cancers, compared with 6% before 1986, and the male-to-female ratio has changed from 3:1 before the AIDS epidemic to about 1.7:1 at the present time (9,70,15). It is interesting that this more than 40-fold increase in the occurrence Table 2. Unusual sites of non-Hodgkin's lymphomas in the human immunodeficiency virus-infected child Site Comment Lungs Presenting as cavitary lesion or as mucosa-associated lymphoid tumor Reminiscent of lymphomatoid papulosis Heart Rare Liver Differential diagnosis with lymphoproliferative disorder In gastric mucosa presenting as mucosa-associated lymphoid tumor Bony lesions without overt bone marrow involvement As single or multiple lesions Gastrointestinal tract Bone Central nervous system Journal of the National Cancer Institute Monographs No. 23, 1998 of childhood KS, although mainly occurring in HIV-infected children, has also been observed in noninfected pediatric popu- lations. The lymphadenopathic form is most prevalent, and com- mon sites include head and neck (82%), extremities (7%), in- guinal (5%), and abdomen (4%). In adults, human herpesvirus type 8 (HHV-8) has been shown to be present in more than 90% of all KS lesions, as well as in AIDS-related body cavity NHLs, lymphomatous effusions without tumor mass, and Castleman’s disease (43-46). In the general population of the United States, about 25% of adults have HHV-8-related antibodies, compared with 18% of adolescents and less than 4% of children under 13 years of age (47-51). However, in African countries such as Nigeria or Zaire, the seroprevalence in the population is much higher (56% and 82%, respectively). In a study from Uganda, nine of 18 children with KS were also positive for HHV-8, compared with seven of 22 children with other tumors (9,15). Smooth Muscle Tumors Leiomyomas and leiomyosarcomas are extremely rare in non- HIV-infected children, but at least 18 cases have now been observed in HIV-infected children. While leiomyomas can be incidental findings and can remain clinically indolent, leiomyo- sarcomas often present as widespread disease. As described for lymphomas, smooth muscle tumors can also occur in unusual locations (e.g., in the lungs, adrenal glands, or skin) (8,/6— 20,52,53). Leiomyosarcomas from children with HIV infection as well as from patients after receiving a transplant but not from other patients have been shown to be positive for Epstein-Barr virus by in situ hybridization (20,54). Quantitative polymerase chain reaction showed relatively high levels of Epstein-Barr vi- rus in tumor tissue (as many as 4.3 genome copies per cell), and discrete episomal Epstein-Barr virus clones were found in le- sions taken from different sites in the same patient, indicating simultaneous emergence of separate clones (20). It is currently unclear whether HIV infection and the associated disturbance in immune surveillance lead to an increased expression of the Ep- stein-Barr virus receptor or whether Epstein-Barr virus infection of the smooth muscle cells results in increased expression. Miscellaneous Tumors Anal (55,56) and cervical (57,58) carcinomas might be seen in sexually active adolescents, since infection with human papillo- mavirus, which is known to be associated with the occurrence of these tumors, is frequently acquired by adolescents (59). Hodg- kin’s disease is more common in adolescents than in younger children and, as in adults, it is possible that a slight increase in the number of this relatively common tumor will be observed in the future (60). Hepatic tumors might become more prevalent, since many HIV-infected children, especially those who are he- mophiliacs, are also infected with hepatitis B or C (61-63). HIV-infected children (like HIV-infected adults) are surviv- ing longer and are in better health, thanks to improved and expanded therapeutic options. It is therefore possible that more children in the future will develop the ‘normal’ pediatric can- cers, especially leukemia and brain tumors. While these tumors may not be more common in the HIV-infected child, their di- agnosis and therapy will certainly pose unique problems. Journal of the National Cancer Institute Monographs No. 23, 1998 Children are being exposed to multiple drugs for the treatment of HIV disease and its complications not only during childhood but also prenatally. To decrease the transmission rate of HIV disease, it is currently recommended that all pregnant HIV- positive women receive zidovudine during pregnancy and birth and that their infants be treated with zidovudine for the first 6 weeks of life (64,65). Recently, studies of the offsprings of mice that had been treated with zidovudine during the last trimester (66) revealed an increased risk of developing liver and lung tumors as well as tumors of the reproductive organs. Olivero et al. have demonstrated that zidovudine is incorporated into the DNA of newborn mice and monkeys, as well as into the nuclear DNA of cord blood samples drawn from children whose mothers had been treated with the drug. In January 1997, the National Institutes of Health (Bethesda, MD) convened a panel to review these studies and, although acknowledging the validity of the findings, also recognized that the benefit of preventing transmis- sion of HIV disease in the vast majority of children currently outweighs the potential concerns of carcinogenicity. However, the panel also strongly emphasized the need for careful, long- term follow-up of all children exposed in utero to antiretroviral therapy, including those children who are not HIV infected. Present and Future Challenges There are several dilemmas that the physician who is treating HIV-infected children is facing. First, as mentioned, the scope of the problem, including the exact incidence and spectrum of tu- mors, is not well defined. It may be necessary in the future to adapt the CDC definition, and it clearly remains important to use the AIDS malignancy registries as well as to publicize unusual cases in order to capture the (potentially very broad and unex- pected) spectrum of pediatric AIDS-related tumors. Second, the treating physician is faced with diagnostic dilem- mas. Atypical symptoms and courses can occur, unusual cancers may be encountered, and the differential diagnosis is often con- fusing because of similar presentations of opportunistic infec- tions (e.g., CNS lesion and toxoplasmosis or fever and night sweats in disseminated Mycobacterium avium complex infec- tion). Furthermore, even if a child is successfully treated for one tumor, he or she remains at risk for another cancer as long as the immunosuppression continues. Thirdly, there are therapeutic dilemmas. Children are often followed by an infectious disease specialist who has no oncol- ogy experience or at a cancer center that has no HIV experience, and an inter-disciplinary approach appears certainly more ad- vantageous. Furthermore, there are currently no standardized treatment guidelines or protocols available for HIV-infected (or otherwise immune compromised) children who develop a can- cer. Many of these children have unique problems, such as pre- existing organ damage, especially diminished bone marrow re- serves, and concurrent opportunistic infections, and they are also receiving multiple drugs (antiretroviral and other) that might lead to potentially dangerous drug interactions. However, there is increasing evidence that children with HIV infection, despite the multiple problems involved, are able to tolerate even inten- sive chemotherapy (at least for limited periods of time) and can be cured of their cancer without acceleration of their underlying disease (31,67-69). One such example is the dose-intensive pro- tocol currently used by the NCI for the treatment of NHL in Ww Ww immunocompromised children (3/7). A combination of cytoxan (day 1) and methotrexate (day 10) is given for three cycles as soon as the absolute neutrophil count reaches 500 cells/mm". Seven of nine patients achieved a complete remission, and no patient died during or as a result of chemotherapy. In summary, there are many unsolved problems to be ad- dressed regarding the epidemiology, etiology, diagnosis, and therapy of cancers occurring in HIV-infected children. Knowl- edge gained in studying these questions not only will benefit the affected children but also will further our understanding of on- cogenesis and of the interactions between viral infections and the immune system, and shorter but still effective therapies might benefit the nonimmune compromised cancer patient as well. References (1 - —~ ro Nad (3 = (4 (5 (0) (7 = (8 = (9 (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) Centers for Disease Control and Prevention. U.S. HIV and AIDS cases reported through December 1996. HIV/AIDS Surveillance Report 1996:8: 1-39. Centers for Disease Control and Prevention. Revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morb Mortal Wkly Rep 1994:43:1-12. Robison LL. General principles of the epidemiology of childhood cancer. In: Pizzo PA, Poplack DG, editors. Principles and practice of pediatric oncology. Philadelphia: Lippincott-Raven Publishers, 1997:1-10. Pollock BH, Jenson HB. McClain KL. et al. Risk factors for HIV-related malignancies in children. J Acquir Immune Defic Syndr Hum Retrovirol 1997:14:A17. Diamond FB, Price LJ, Nelson RP Jr. Papillary carcinoma of the thyroid in a seven-year-old HIV-positive child. Pediatr AIDS and HIV Infect: From Fetus to Adolescent 1994:5:232-5. Ninane J, Moulin D, Latinne D, et al. AIDS in two African children—one with fibrosarcoma of the liver. Eur J Pediatr 1985:144:385-90. Scully RE, Mark RE, McNeely BU. Case records of the Massachusetts General Hospital. N Engl J Med 1986:314:629-40. Chadwick EG, Connor EJ. Guerra Hanson IC, et al. Tumors of smooth muscle origin in HIV-infected children. JAMA 1990;263:3182-4. Ziegler JL, Katongole-Mbidde E. Kaposi's sarcoma in childhood: an analy- sis of 100 cases from Uganda and relationship to HIV infection. Int J Cancer 1996:65:200-3. Athale UH, Patil PS, Chintu C, et al. Influence of HIV epidemic on the incidence of Kaposi's sarcoma in Zambian children. J Acquir Immune Defic Syndr Hum Retrovirol 1995:8:96-100. Orlow SJ, Cooper D, Petrea S, et al. AIDS-associated Kaposi's sarcoma in Romanian children. J] Am Acad Dermatol 1993:28:449-53. Centers for Disease Control and Prevention. AIDS among children— United States, 1996. MMWR Morb Mortal Wkly Rep 1996:45:1005-10. Davachi F, Ngoie K, Ndoko K, et al. Pediatric AIDS and associated cancers in Kinshasa, Zaire. Antibiot Chemother 1991:43:140-9. Chintu C, Athale UH, Patil PS. Childhood cancers in Zambia before and after the HIV epidemic. Arch Dis Child 1995:73:100-5. Mbulaiteye SM, Ziegler JL, Katongole-Mbidde E, et al. Risk factors for childhood Kaposi's sarcoma in Uganda: a case control study. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A18. McLoughlin LC, Nord KS, Joshi VV, et al. Disseminated leiomyosarcoma in a child with acquired immune deficiency syndrome. Cancer 1991:67: 2618-21. Sabatino D, Martinez S. Young R, et al. Simultaneous pulmonary leiomyo- sarcoma and leiomyoma in pediatric HIV infection. Pediatr Hematol Oncol 1991:8:355-9. Orlow SJ, Kamino H, Lawrence RL. Multiple subcutaneous leiomyosar- comas in an adolescent with AIDS. Am J Pediatr Hematol Oncol 1992:14: 265-8. Ross JS, Del Rosario A, Bui HX, et al. Primary hepatic leiomyosarcoma in a child with the acquired immunodeficiency syndrome. Hum Pathol 1992: 23:69-72. McClain KL, Leach CT, Jenson HB, et al. Association of Epstein-Barr virus with leiomyosarcomas in young people with AIDS. N Engl J Med 1995;332:12-8. Joshi VV. Systemic lymphoproliferative lesions in children with AIDS. Pediatr AIDS and HIV Infect: From Fetus to Adolescent 1990; 1:44-8. Soberman N, Leonidas JC, Berdon WE, et al. Parotid enlargement in chil- dren seropositive for human immunodeficiency virus: imaging findings. Am J Radiol 1991;157:553-6. (23) lItescu S. Diffuse infiltrative lymphocytosis syndrome in children and adults infected with HIV-1: a model of rheumatic illness caused by acquired viral infection. Am J Reprod Immunol 1992:28:247-50. (24) Kontny HU, Sleasman JW, Kingma DW, et al. Multilocular thymic cysts in children with human immunodeficiency virus infection. Clinical and patho- logical aspects. J Pediatr 1997:131:264-70. (25) Magrath I, Shad A, Epelman S, et al. Pediatric oncology in countries with limited resources. In: Pizzo PA, Poplack DG, editors. Principles and prac- tice of pediatric oncology. Philadelphia: Lippincott-Raven Publishers, 1997:1395-1420. (26) Shad A, Magrath IT. Malignant non-Hodgkin's lymphomas in children. In: Pizzo PA, Poplack DG, editors. Principles and practice of pediatric oncol- ogy. Philadelphia: Lippincott-Raven Publishers, 1997:545-88. Guarner J, Brynes RK, Chan WC, et al. Primary non-Hodgkin’s lymphoma of the heart in two patients with the acquired immunodeficiency syndrome. Arch Pathol Lab Med 1987:111:254-6. (28) Siskin GP, Haller JO, Miller S, et al. AIDS-related lymphoma: radiologic features in pediatric patients. Radiology 1995:196:63-6. (29) Teruya-Feldstein J, Temeck BK, Sloas MM, et al. Pulmonary malignant lymphoma of mucosa-associated lymphoid tissue (MALT) arising in a pediatric HIV-positive patient. Am J Surg Pathol 1995:19:357-63. (30) Mueller BU, Shad AT, Magrath IT, et al. Malignancies in children with HIV infection. In: Pizzo PA, Wilfert CM, editors. Pediatric AIDS. The challenge of HIV infection in infants, children, and adolescents. Baltimore: Williams & Wilkins, 1994:603-22. (31) Shad A, Mueller BU, Adde M, et al. Results of a treatment protocol for children with HIV and non-Hodgkin's lymphomas (NHL). J Acquir Im- mune Defic Syndr Hum Retrovirol 1997;14:A54. (32) Isaacson PG. The MALT lymphoma concept updated. Ann Oncol 1995:6: 319-20. (33) Slater DN. MALT and SALT: the clue to cutaneous B-cell lymphoprolif- erative disease. Br J Dermatol 1994:131:557-61. (34) Connor EM, Marquis J, Oleske JM. Lymphoid interstitial pneumonitis. In: Pizzo PA, Wilfert CM, editors. Pediatric AIDS. The challenge of HIV infection in infants, children, and adolescents. Baltimore: Williams & Wilkins, 1991:343-54. (35) Pitt J. Lymphocytic interstitial pneumonia. In: Edelson PJ, editor. Child- hood AIDS. Philadelphia: W. B. Saunders Co.. 1991:89-96. (36) Scott GB, Hutto C, Makuch RW, et al. Survival in children with perinatally acquired human immunodeficiency virus type | infection. N Engl J Med 1989;321:1791-6. (37) Kingma DW, Weiss A, Sorbora L, et al. Is lymphocytic interstitial pneu- monitis an early stage of MALT lymphoma in HIV-infected pediatric pa- tients? J Acquir Immune Defic Syndr Hum Retrovirol 1997:14:A54. Buck BE, Scott GB, Valdes-Dapena M, et al. Kaposi sarcoma in two infants with acquired immune deficiency syndrome. J Pediatr 1993:103: 911-3. (39) Baum LG, Vinters HV. Lymphadenopathic Kaposi's sarcoma in a pediatric patient with acquired immune deficiency syndrome. Pediatr Pathol 1989; 9:459-65. (40) Bouquety JC, Siopathis MR, Ravisse PR, et al. Lympho-cutaneous Kapo- si’s sarcoma in an African pediatric AIDS case. Am J Trop Med Hyg 1989:40:323-5. (41) Gutierrez-Ortega P, Hierro-Orozco S, Sanchez-Cisneros R, et al. Kaposi's sarcoma in a 6-day-old infant with human immunodeficiency virus. Arch Dermatol 1989:;125:432-3. (42) Connor E. Boccon-Gibod L, Joshi V, et al. Cutaneous acquired immuno- deficiency syndrome-associated Kaposi's sarcoma in pediatric patients. Arch Dermatol 1990:126:791-3. (43) Boshoff C, Schulz TF, Kennedy MM, et al. Kaposi's sarcoma-associated herpesvirus infects endothelial and spindle cells. Nature Med 1995:1: 1274-8. (44) Gaidano G, Pastore C, Gloghini A, et al. Distribution of human herpesvi- rus-8 sequences throughout the spectrum of AIDS-related neoplasia. AIDS 1996:10:941-9. (45) Otsuki T, Kumar S, Ensoli B, et al. Detection of HHV-8/KSHV DNA sequences in AIDS-associated extranodal lymphoid malignancies. Leuke- mia 1996:10:1358-62. (46) Strauchen JA, Hauser AD, Burstein D. et al. Body cavity-based malignant lymphoma containing Kaposi sarcoma-associated herpesvirus in an HIV- negative man with previous Kaposi sarcoma. Ann Intern Med 1996:125: 822-5. (47) Schalling M, Ekman M, Kaaya EE, et al. A role for a new herpes virus (KSHV) in different forms of Kaposi's sarcoma. Nature Med 1995:1: 707-8. (48) Gao S-J, Kingsley L, Li M, et al. KSHV antibodies among Americans, (27 ~ (38 Journal of the National Cancer Institute Monographs No. 23, 1998 (49) (50) (51) (52 ~ (53) (54) (55) (56) (57) (58) (59) (60) Italians and Ugandans with and without Kaposi's sarcoma. Nature Med 1996:2:925-8. Simpson GR, Schulz TF, Whitby D, et al. Prevalence of Kaposi's sarcoma associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet 996:349: 1133-8. Monini P, De Lellis L, Fabris M, et al. Kaposi's sarcoma-associated her- pesvirus DNA sequences in prostate tissue and human semen. N Engl J Med 1996;334:1168-72. Lennette ET, Blackbourn DJ, Levy JA. Antibodies to human herpesvirus type 8 in the general population and in Kaposi's sarcoma patients. Lancet 1996:348:858-61. Balsam D, Segal S. Two smooth muscle tumors in the airway of an HIV- infected child. Pediatr Radiol 1992;22:552-3. Parola P, Petit N, Azzedine A, et al. Symptomatic leiomyoma of the adrenal gland in a woman with AIDS. AIDS 1996:10:340-1. Lee ES, Locker J, Nalesnik M, et al. The association of Epstein-Barr virus with smooth muscle tumors occurring after organ transplantation. N Engl J Med 1995;332:19-25. Melbye M, Cote TR, Kessler L, et al. High incidence of anal cancer among AIDS patients. Lancet 1994;343:636-9. Palefsky JM. Anal human papillomavirus infection and anal cancer in HIV-positive individuals: an emerging problem. AIDS 1994:8:283-95. Maiman M. Fruchter RG, Guy L, et al. Human immunodeficiency virus infection and invasive cervical carcinoma. Cancer 1993;71:402-6. Tweddel G, Heller P, Cunnane M. et al. The correlation between HIV seropositivity, cervical dysplasia, and HPV subtypes 6/11, 16/18, 31/33/35. Gynecol Oncol 1994:52:161-4. Cripe TP. Human papillomavirus: pediatric perspectives on a family of multifaceted tumorigenic pathogens. Pediatr Infect Dis J 1990:9:836-44. Tirelli U, Errante D, Dolcetti R, et al. Hodgkin's disease and human im- Journal of the National Cancer Institute Monographs No. 23, 1998 (61) (62) (63) (64) (65) (60) (67) (68) (69) munodeficiency virus infection: clinicopathologic and virologic features of 114 patients from the Italian Cooperative Group on AIDS and Tumors. J Clin Oncol 1995;13:1758-67. Mazzaro C., Zagonel V, Monfardini S, et al. Hepatitis C and non-Hodgkin's lymphoma. Br J Haematol 1996:94:544-50). White YS, Johnson PJ, Davison F, et al. Frequency of hepatic HBV-DNA in patients with cirrhosis and hepatocellular carcinoma: relation to serum HBV markers. Br J Cancer 1990;61:909-12. Pontisso P, Morsica G, Ruvoletto MG, et al. Latent hepatitis B virus in- fection in childhood hepatocellular carcinoma. Analysis by polymerase chain reaction. Cancer 1992;69:2731-5. Connor EM, Mofenson LK. Zidovudine for the reduction of perinatal hu- man immunodeficiency virus transmission: Pediatric AIDS Clinical Trials Group protocol 076-results and treatment recommendations. Pediatr Infect Dis J 1995:14:536-41. Centers for Disease Control and Prevention. Recommendations of the U.S. Public Health Service Task Force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR Morb Mortal Wkly Rep 1994:43:1-20. Olivero OA, Anderson LM, Diwan BA, et al. AZT is a genotoxic trans- placental carcinogen in animal models. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A29. Neumann Y, Toren A, Mandel M, et al. Favorable response of pediatric AIDS-related Burkitt's lymphoma treated by aggressive chemotherapy. Med Pediatr Oncol 1993:21:661-4. Nadal D, Caduff R, Frey E, et al. Non-Hodgkin's lymphoma in four chil- dren infected with the human immunodeficiency virus. Cancer 1994:73: 224-30. Desai N, Miller ST, Rao SP. et al. Malignancies associated with HIV-1 infection in children: one center's experience over 6 years. J Acquir Im- mune Defic Syndr Hum Retrovirol 1997;14:A49. 35 Hodgkin’s Disease in the Setting of Human Immunodeficiency Virus Infection Alexandra M. Levine* Although Hodgkin’s disease (HD) is not usually associated with congenital or acquired immunodeficiency disorders, re- cent evidence would suggest a statistically significant in- crease in HD among individuals infected with human immu- nodeficiency virus (HIV). In the setting of underlying HIV infection, clinical and pathologic characteristics of HD may differ from usual expectations. Thus, 70%-100% of HIV- infected patients with HD present with systemic ‘‘B’’ symp- toms. Likewise, disseminated, stage III or IV disease is re- ported in approximately 75%-90%. Bone marrow is a common site of extranodal HD, occurring in 40%-50%. Complete response rates after multiagent chemotherapy range from approximately 45% to 70%, although median survival has been only in the range of approximately 18 months. Hematologic toxicity from multiagent chemo- therapy may be substantial, even with the use of hematopoi- etic growth factor support. It is apparent that new strategies of therapeutic intervention must be explored. [Monogr Natl Cancer Inst 1998;23:37-42] Background: De Novo Hodgkin’s Disease Relationship Between Hodgkin’s Disease and Underlying Immunodeficiency Non-Hodgkin's lymphoma is known to occur with increased frequency in various settings of congenital, acquired, or iatro- genic immunosuppression (/—4). It was thus not surprising to observe significantly increased rates of lymphoma in patients infected with human immunodeficiency virus (HIV). Similarly, the other current acquired immunodeficiency syndrome (AIDS)- defining cancers, Kaposi's sarcoma and cervical cancer, are also associated with diverse types of abnormal immunity, including immunosuppression related to organ transplantation or to therapy for autoimmune or malignant disorders (2,3,5,6). Within the context of this background, Hodgkin's disease (HD) repre- sents a somewhat different situation. Thus, while the risk of HD is increased in children with ataxia telangiectasia (7), the mag- nitude of this increase is quite modest, and HD is not seen with increased frequency in other types of congenital immune defi- ciency disease. Furthermore, the incidence of HD is not in- creased in the setting of organ transplantation or in the setting of prior cancer chemotherapy or immunosuppressive therapy for autoimmune disease (2,3). The expectation that HIV-infected patients would be at risk for HD was thus not initially considered a prominent concern. Epidemiology of De Novo HD In patients who are not infected by HIV, the epidemiology of HD has been well described. Correa, MacMahon, and others Journal of the National Cancer Institute Monographs No. 23, 1998 (8-10) have demonstrated a clear relationship between geo- graphic and socioeconomic factors and the ensuing clinicopath- ologic characteristics of HD. Thus, the “‘type 1 pattern’ of HD is seen in underdeveloped areas of the world, associated with lower socioeconomic conditions. In this setting, two age peaks of disease are observed, with the first in childhood (primarily affecting boys) and the second in the fifth or sixth decade. The pathologic spectrum of HD in the type 1 setting includes a predominance of mixed-cellularity or lymphocyte-depletion dis- ease, with proclivity toward advanced stage and presence of systemic “‘B’’ symptoms. The overall prognosis of such patients is relatively poor, when compared with that of patients in other settings and with other patterns of disease. The “‘type 3 pattern’ of HD is seen in industrialized nations and is also characterized by two modal peaks of disease (8-10). However, in these areas of higher socioeconomic background, the first modal peak occurs in adolescent or young-adult fe- males, who tend to present with nodular sclerosis HD, often presenting with rather limited stage disease localized to the me- diastinal lymph nodes. The second peak occurs in the fifth and sixth decades, is seen primarily in males, and may be associated with a broader spectrum of pathologic subtypes. Various char- acteristics associated with higher socioeconomic status have been described with increased frequency in young patients who develop HD within type 3 areas. In the ‘“‘type 2 pattern,” all three age peaks of disease are encountered, with a broad spectrum of pathologic subtypes and clinical outcome. This pattern has been described in developing areas of the world, such as Mexico. Of interest, areas of lower socioeconomic status within the United States have been described in which the clinicopatholog- ic spectrum of HD resembles that seen in type 2 or developing nations (//). Because of these divergent epidemiologic, clinical, and pathologic characteristics, any differences between HIV- associated and de novo HD must be considered within the con- text of the usual expectations for HD in that particular region under study. Immune Function in De Novo HD Defects in cell-mediated immunity have been well described in patients with de novo HD, occurring even in patients with early stage disease and in those who have been free of HD for many years (/2). Thus, patients with HD are often anergic to *Correspondence to: Alexandra M. Levine, M.D., Division of Hematology, University of Southern California (USC), USC/Norris Cancer Hospital, 1441 Eastlake Ave., MS 34, Los Angeles, CA 90033. See ‘Note’ following ‘‘References.”’ © Oxford University Press 37 delayed cutaneous hypersensitivity testing, as described by Dorothy Reed as early as 1902 [see (13)]. While more common in patients with advanced, symptomatic HD, dinitrochloroben- zene skin testing may also be abnormal in patients with good prognosis, early stage HD (1/2). In vitro testing of immune func- tion is also abnormal in patients with de novo HD. Thus, de- creased mitogenic responses to phytohemagglutinin and to con- canavalin A have been described, even prior to the initiation of chemotherapy (7/4). While significant reductions in T cells have been described in peripheral blood and spleen of approximately 30% of patients with untreated, advanced HD (75), no alteration in the ratio of helper or cytotoxic/suppressor T cells has been found; moreover, there is no evidence of cellular activation (75). Furthermore, most patients with HD do not have lymphocyto- penia, indicating that the major cell-mediated defect is a func- tional impairment, as opposed to an absolute depletion of T cells in blood or tissues. These defects of cell-mediated immunity may also be sur- mised by the fact that patients with HD may be at increased risk for certain infections known to be associated with depressed T-cell function. Of interest, increased susceptibility to tubercu- losis among patients with HD was well described as early as 1928, while the earliest description of HD was complicated by the fact that some of Thomas Hodgkins initial HD patients actually had tuberculosis (76). Furthermore, in a large series of 300 adult patients with HD, Notter et al. (/7) noted the devel- opment of opportunistic infections in 4%, consisting of Pneu- mocystis carinii pneumonia, disseminated vaccinia, herpes zos- ter, and progressive multifocal leukoencephalopathy, among others. When considering the immune function of patients with HIV- related HD, it is important to realize that any immune defect etiologically related to HIV will, by definition, be augmented by the known defects in cell-mediated immunity that are inherent to HD itself. Epidemiology of HD in the Setting of HIV Infection Incidence of HD Among HIV-Infected Persons Of interest, at the outset of the AIDS epidemic, the incidence of de novo HD appeared to have decreased among HIV-negative persons in New York and Los Angeles (78,19). In the evaluation of young, single men in New York City (presumably including HIV-infected individuals or homosexual men at risk for HIV), stationary rates were documented, while an actual decrease in HD among married men in New York State was observed at the same time (/8). In 1992, Hessol et al. (20) noted an increase in the standard- ized morbidity ratio for HD among HIV-infected homosexual men from San Francisco. By use of data from 6704 homosexual men enrolled in the San Francisco City Clinic HIV Cohort Study, cases of HD were ascertained through the computer- matched identification of participants in the population-based Northern California Cancer Center registry. Comparisons were made with rates determined in the general population by use of the Surveillance, Epidemiology, and End Results (SEER)' can- cer registry. A total of 90 cases of non-Hodgkin's lymphoma were identified, along with eight cases of HD. The age-adjusted 38 standardized morbidity ratio for HD among the HIV-infected men was 5.0 (95% confidence interval [CI] = 2.0-10.3), indi- cating a statistically significant increase. Furthermore, when these data were compared with those from the SEER registry, the excess risk of HD among HIV-infected homosexual men was 19.3 cases per 100000 person-years, again indicating a signifi- cant increase. While pathologic review was not performed, and while diagnostic misclassification is known to occur, especially in lymphocyte-depleted HD (27,22), the data did indicate that the incidence of HD may have increased in HIV-infected ho- mosexual men (20). More recently, Lyter et al. (23) described an increase in HD among homosexual and bisexual men, who were followed as part of the national Multicenter AIDS Cohort Study (MACS). Thus, within the Pittsburgh MACS cohort, who were followed between 1984 and 1993, two cases of HD were seen among 430 HIV-infected men (with 2344 person-years of follow-up) versus no cases among 769 HIV-negative control subjects (followed for 5708 person-years). While consisting of only two cases, the standardized incidence ratio for seropositives was increased 19.8-fold (95% C1 = 2.4-71.5) when compared with that for seronegatives. Furthermore, when these data were compared with those from the population-based SEER registry, the rate of HD among HIV-positive men was 85 per 100000 person-years versus 4.3 per 100000 person-years in the general population. In an attempt to ascertain the full spectrum of malignant dis- ease occurring in association with HIV infection, Goedert and colleagues (24) performed a linkage analysis in which all cases of cancer reported in over 98 000 cases of AIDS reported to the Centers for Disease Control and Prevention were linked, case by case, to the population-based cancer registries in the United States. AIDS-related cancers were defined as those that had increased statistically after an initial AIDS diagnosis, with in- creasing prevalence from 5 years before the diagnosis of AIDS, through 2 years after the initial AIDS diagnosis. In this analysis, the risk of HD was increased 7.6-fold, with the relative risk (RR) statistically increasing from the period prior to AIDS to that occurring after the initial diagnosis of AIDS. Significant increases in HD have also recently been docu- mented in the National Cancer Institute Hemophilia Cohort (RR = 5.6) (25); in injection drug users with AIDS in New Jersey (odds ratio [OR] = 4.2: 95% CI = 1.4-14.7, P = .04) (26); in HIV-infected individuals from Denver, CO (27); and in homo- sexual men from New South Wales, Australia (28). In the latter report, the RR of HD was 18.3 (95% CI = 8.4-35), with the increase in risk confined to the 2 years immediately before an AIDS diagnosis and the period after AIDS was diagnosed. These data would be consistent with the concept that HD is a late manifestation of HIV disease, which may become even more apparent as HIV-infected individuals live long enough to de- velop the disease. HIV-Infected Populations at Risk for HD Early studies from Europe indicated that HD did occur in HIV-infected individuals, although no formal population-based comparisons were published at that time. However, Monfardini et al. (29) from the Italian Cooperative Group for AIDS-related Tumors sent 1962 questionnaires to various professional groups of oncologists, infectious disease specialists, and others, in an Journal of the National Cancer Institute Monographs No. 23, 1998 attempt to ascertain potential cases of HD among HIV-infected individuals. A total of 35 cases of HD and 95 cases of lymphoma were identified. Of interest, 89% of the patients with HD had a history of injection drug use, while an additional 9% were ho- mosexual with a history of injection drug use. In contrast, a history of injection drug use was present in 68% of the lym- phoma patients, while 5% had a history of both injection drug use and homosexuality. While retrospective and biased by the method of case ascertainment, this study suggested the possibil- ity that injection drug use might be a specific risk factor for HIV-associated HD. Subsequent work from France (30) con- firmed this suggestion, with 38% of 45 HIV-infected patients with HD providing a history of injection drug use, versus only 12.5% of 168 patients with non-Hodgkin’s lymphoma. Roith- mann et al. (3/7) confirmed the increased likelihood of HD among injection drug users versus homosexual men from France, while investigators from Spain (32) also documented a similar relationship, with 85% of 46 cases of HD occurring in injection drug users and 9% occurring in homosexual men. Fur- thermore, Ahmed et al. (33) also noted that the incidence of HD was three to 10 times higher among prisoners in the United States who were injection drug users, versus those who were not, although the overall incidence of HD had not increased in these prison inmates. While these earlier studies seemed to indicate an increased proclivity for HD among HIV-infected injection drug users, sub- sequent studies have not really confirmed this fact. Thus, statis- tically significant increases in HD have now been described in HIV-infected hemophiliacs (25) as well as in HIV-infected ho- mosexual men (23,28). More time will be required to elucidate the true increase in risk of HD among injection drug users, when compared to other HIV-infected populations. Degree of Immune Deficiency Among HIV-Infected Persons With HD Relatively little is currently known about the precise level of immune dysfunction in patients with HIV-related HD. Ames et al. (34) evaluated 23 such patients, whose median CD4 cell count was 201/mm’, ranging from 7 to 882/mm”. In another report published in 1991 from Italy (35), the median CD4 cell count in 38 patients was 254/mm® (range, 27-1316/mm?*). A study from France published in 1993 (30) described 45 patients with HIV-related HD in whom the median CD4 cell count was 306/mm?. More recently, CD4 cells were determined at baseline in a group of 21 patients with HIV-related HD (36) who were treated prospectively as part of the AIDS Clinical Trials Group (ACTG study #149). The median CD4 cell count was 128/mm°, ranging from 2 to 972/mm>. The apparently lower median CD4 cell count seen in this latter series may reflect simple chance alone, or it may be reflective of the fact that this series was accrued later in the epidemic, at a time when patients are living longer, despite falling CD4 cell counts. Most patients with HIV-related HD have not had an AIDS- defining diagnosis prior to the onset of HD. Thus, in the group of 21 patients studied as part of ACTG study #149 (36), 20% had a history of AIDS prior to HD, consisting of Mycobacterium avium-intracellulare in two, Pneumocystis carinii pneumonia in two, and Kaposi's sarcoma in one. An additional 50% had symp- tomatic HIV-related disease prior to the diagnosis of HD, con- Journal of the National Cancer Institute Monographs No. 23, 1998 sisting of oral candidiasis, herpes zoster, and/or immune throm- bocytopenic purpura. Of interest, although the majority of patients have not had full-blown AIDS prior to HD, many have had a history of reactive lymphadenopathy (persistent, general- ized lymphadenopathy), which was reported in 36%-83% of individuals (30,33,35). Pathologic Aspects of HIV-Associated HD In de novo HD in the United States and Europe, the majority of patients are diagnosed with nodular sclerosis disease, de- scribed in 52%—-62% of large series (37). Lymphocyte- predominant disease accounts for 8%-21%, while mixed cellu- larity HD is seen in 24% and lymphocyte depletion subtype in 3%—6% of patients (37). In contrast, the most common type of HD in underdeveloped areas of the world (type | pattern) is mixed cellularity, while lymphocyte depletion is the next most frequently diagnosed pathologic subtype of disease (8-10). Similar to the pathologic spectrum in type 1 areas, patients with HIV-related HD tend to present with either mixed cellu- larity or lymphocyte depletion HD, diagnosed in 41%—-100% of patients (29,32,34,38—41). Serrano et al. (40) compared the pathologic characteristics in 22 patients with HIV-related HD with those in 125 uninfected control subjects with HD diagnosed in the same time period, in the same institution in Spain. Mixed cellularity or lymphocyte depletion HD was diagnosed in 68% of the HIV-infected group, versus 35% of the uninfected control subjects. Of interest, when nodular sclerosis was diagnosed in the setting of HIV, it did not occur within the mediastinum, in contrast to HIV-negative patients with nodular sclerosis, in whom mediastinal involvement was documented in 74%. Clinical Features of HIV-Related HD Systemic B Symptoms In general, fever, drenching night sweats, and/or weight loss occur more often in patients with HIV-related HD than in pa- tients with de novo disease. Thus, in a large series from the United States or Europe, the prevalence of systemic B symptoms ranged from 30% (42) to 62%, the latter described in U.S. pa- tients from minority racial/ethnic backgrounds and lower socio- economic status (7/7). In contrast, between 70% and 100% of patients with HIV-related HD have presented with B symptoms (29,30,32,34,38—41). In the series of HIV-positive versus HIV- negative cases of HD from Spain, Serrano et al. (40) noted B symptoms in 81% of 22 HIV-infected patients and in 57% of 125 HIV-negative patients. While thus apparent that fever, night sweats, and/or weight loss may be commonly encountered in the setting of HIV-related HD, it is important to recognize that these symptoms may also be seen in Mycobacterium avium-intracellulare, cytomegalovi- rus disease, and other opportunistic infections. Thus, the assig- nation of such symptoms to HD must be made only after a careful evaluation has excluded the presence of underlying oc- cult infection. Clinical Stage of HIV-Related HD at Presentation In de novo HD in the United States or Europe, stage III dis- ease is diagnosed in approximately 27%, while stage IV is docu- mented in 16%-26% (37). In contrast, widely disseminated HD is expected in the majority of patients with HIV-related HD. 39 Thus, in the study of HIV-positive versus HIV-negative HD in Spain, Serrano et al. (40) noted stage III or IV disease in 91% of HIV-infected patients, versus only 46% of those with de novo HD. In a similar comparative study from France, Andrieu et al. (30) reported stage III or IV disease in 75% of 45 HIV-infected patients and in 33% of 407 HIV-negative individuals. These data would indicate that HD associated with HIV is clearly distinct in terms of the expectation of widely disseminated disease at initial presentation. This is very similar to the expectation in AIDS- related lymphoma, which is distinct from de novo lymphoma- tous disease (/). The most common site of extranodal HD in HIV-infected patients is the bone marrow, which is involved in approximately 419%-50% of all reported patients (32,34,39,40,43). Of interest, bone marrow involvement may be the first presentation of HIV- related HD, with patients seeking medical attention because of systemic B symptoms and/or signs or symptoms of pancytope- nia. The primary diagnosis of HD within bone marrow may be difficult, revealing lymphohistiocytic aggregates and large atypi- cal mononuclear cells that are suggestive but not diagnostic of HD (44). The rather common occurrence of HD within the bone marrow of HIV-infected individuals is in sharp contrast to de novo HD, in which Colby et al. (45) described such involvement in only 3.5% of 659 patients. While bone marrow involvement is distinctly unusual in de novo HD in the United States, when one considers only those patients with mixed cellularity or lym- phocyte depletion HD, the prevalence of bone marrow involve- ment is much higher, at 20%-30%, as reported by O’Carrol et al. (46). Even when one considers only mixed cellularity and lym- phocyte depletion disease, however, bone marrow involvement by HD is still more likely in the setting of HIV infection. Thus, Serrano et al. (40) described bone marrow involvement by HD in 50% of HIV-positive patients from Spain and in only 10% of 125 HIV-negative patients who were diagnosed and evaluated in the same institution. Aside from bone marrow, other sites of extranodal HD have included rectum (34,47), tongue (34), and skin (43,49). Involve- ment of the brain by HD is extremely unusual in de novo disease, but it has been described in the setting of HIV (49) in an injec- tion drug user who presented with a seizure and was found to have a 3-cm, partially enhancing lesion in the left parietal lobe. Spinal fluid was negative, and the diagnosis of mixed cellularity HD was made after stereotaxic brain biopsy. More usual sites of extranodal HD have been described in patients with HIV-related disease, but even in this circumstance, the disease is distinct from that expected. Thus, HD in the lung has been described in the absence of mediastinal lymphadenop- athy (43,50). Furthermore, hepatic HD has been diagnosed in the absence of splenic involvement (43). Results of Therapy for HIV-Related HD The majority of all patients with de novo HD experience long-term disease-free survival, with probable cure of disease expected in approximately 70%. When one considers results of radiotherapy for localized stage I or II disease, complete remis- sion is expected in 80%-95%, with 80% of responders in con- tinuous complete remission at 7 years (5/7). In patients with stage II or IV disease, ABVD chemotherapy (i.e., chemotherapy with doxorubicin, bleomycin, vinblastine, and dacarbazine) is asso- 40 ciated with complete remission in 70%-80% of patients, with relapse-free survival in approximately 60%—70% of responders (52). Results of Retrospective Therapeutic Trials for HIV-Related HD Prognosis in patients with HIV-related HD is distinctly poorer than that expected in patients with de novo disease. Thus, me- dian survival has been in the range of only 8-18 months (30,32,34,35,40,41,53). In the series of HIV-positive versus HIV-negative HD patients reported from Spain (40), 5-year sur- vival was achieved in 80% of 125 HIV-negative patients and in none of 22 HIV-infected patients. The potential causes of decreased survival in HIV-related HD patients could include early demise as a result of other AIDS- related disorders, decreased efficacy of standard therapy, and/or increased toxicity of treatment. Of interest, while no large pro- spective series of uniformly treated patients has yet been pub- lished, preliminary data would suggest that the results of mul- tiagent chemotherapy may be somewhat less optimal than expected in de novo disease. Thus, the comparative study by Serrano et al. (40) documented a complete remission in 54% of HIV-infected patients with HD, versus an 84% complete remis- sion rate in those with de novo disease. As depicted in Table I, complete remission rates have ranged from 44% to 100% in patients with HIV-related HD after being treated with a variety of regimens (32,35,38,40,54,55). Prospective Treatment Trials When one considers prospective therapeutic trials, Errante et al. (53) treated 17 patients with epirubicin, vinblastine, and bleo- mycin. Patients with poor prognostic disease (Eastern Coopera- tive Oncology Group performance status =3 or a history of opportunistic infection) received 50% of planned epirubicin and vinblastine dosing. In addition, these patients received zidovu- dine at the initiation of chemotherapy, while the others received full-dose chemotherapy with zidovudine initiated only after cycle 3. In patients with good prognosis disease, a complete remission rate of 67% was achieved, while complete remission was achieved in only one (20%) of five patients with poor prog- Table 1. Hodgkin's disease associated with human immunodeficiency virus: therapeutic results No. of Complete Reference Regimen(s)* patients remission, % No. Various 27 44 (32) Various 22 54 (40) Various 32 47 (35) Various 18 83 (38) MOPP or ABVD 8 100 (55) MOPP 22 46 (54) MOPP/ABVD 19 65 Epirubicin, bleomycin, and vinblastine 29 69 (56) Standard ABVD with G-CSF 21 56 (36) *MOPP = combination chemotherapy with mechlorethamine, vincristine, procarbazine, and prednisone; ABVD = combination chemotherapy with doxo- rubicin, bleomycin, vinblastine, and dacarbazine; G-CSF = granulocyte colony- stimulating factor. Journal of the National Cancer Institute Monographs No. 23, 1998 nosis HIV-related HD. Overall median survival in the 17 pa- tients was 11 months. Tirelli et al. (56) used a similar regimen in 29 patients with HIV-related HD; they employed epirubicin (70 mg/m’), bleo- mycin (10 mg/m?), and vinblastine (6 mg/m?) given intrave- nously on day 1, with prednisone (40 mg/m?) given orally on days 1 through 5. Cycles were given every 21 days. Zidovudine (500 mg/day) was given to all patients who had not taken this drug before, while didanosine was given to the other patients. Granulocyte colony-stimulating factor was given at 5 pg/kg per day subcutaneously from day 6 to day 20 in all cycles. The median CD4 cell count at study entry was 219/mm’ (range, 6-812/mm*). Stage III or IV disease was present in 80% of the patients, while 90% had systemic B symptoms. Pathologic evaluation revealed either mixed cellularity or lymphocyte depletion HD in 69% of the patients. The complete remission rate was 69%, with 58% of these patients (i.e., 40% of all pa- tients showing complete remission) remaining disease free at 2 years. Relapse of HD was documented in 30% of patients who had shown complete remission. Opportunistic infections oc- curred in 28% during therapy. Grade 3 or 4 leukopenia was documented in 34%. Median survival for the group was 14 months. HD was the cause of death in 50% of these patients; HD and opportunistic infection were the causes of death in an addi- tional 5%. Levine et al. (36) have recently reported results from the AIDS Clinical Trials Group, employing standard-dose doxoru- bicin, bleomycin, vinblastine, and dacarbazine (52) in patients with newly diagnosed HIV-related HD. Cycles were repeated every 28 days, and therapy was continued until two cycles be- yond complete remission, until six cycles had been given, or until toxicity or progressive disease. Granulocyte colony- stimulating factor was given subcutaneously at a dose of 5 pg/ kg per day from days 2 through 14 and days 16 through 28 of each cycle. Antiretroviral agents were withheld during the first two cycles and then initiated with subsequent therapy. Doses of doxorubicin and vinblastine were reduced 75% in the presence of initial bone marrow involvement. A total of 21 patients were accrued, of whom 16 are currently assessable for response. The median age of the patients was 34 years, and 75% were male. The majority (91%) had no history of injection drug use. Median CD4 cell count at study entry was 128/mm® (range, 2-972/ mm?), and 20% had a history of AIDS prior to the diagnosis of HD. Systemic B symptoms were present in 86%. Stage IV HD was detected in 62%, while 14% had stage III disease. Sites of extranodal disease included bone marrow in 25%, while kidney, liver, brain, spinal cord, lung, and muscle involvement was also documented. A median of six cycles of ABVD chemotherapy was given (/-8). Complete remission was attained in 56% of the patients, and an additional 25% attained a partial response. Tox- icity was significant, with grade 4 neutropenia in 48%, despite the use of granulocyte colony-stimulating factor. Hepatic toxic- ity was also observed; it was detected in 52% of the patients and consisted of grade 4 transaminasemia in four of 21. Other toxic effects were rather mild. The median survival for the group is 19.5 months, and eight patients have now died, as a result of opportunistic infection in three, liver failure due to reactivation of hepatitis B in one, sepsis in one, bacterial pneumonia in one, and HD in two. Journal of the National Cancer Institute Monographs No. 23, 1998 It is apparent from these studies that complete remission may be achieved in the majority of patients with HIV-related HD. However, toxicity from chemotherapy is substantial and is re- lated especially to bone marrow compromise, even when hema- topoietic growth factors are routinely employed. Furthermore, the median survival is significantly shorter than that described in patients with de novo HD, with the majority of deaths due to AIDS-related complications and other bacterial infections. It is clear that alternative therapeutic strategies must be explored in HIV-infected patients with HD. References (1) Levine AM. AIDS-associated malignant lymphoma [review]. Blood 1992; 80:8-20. Penn I. The occurrence of malignant tumors in immunosuppressed states. Prog Allergy 1986:37:259-300. Penn I. The incidence of malignancies in transplant recipients. Transplant Proc 1975:7:323-6. Swinnen LJ, Costanzo-Nordin MR, Fisher SG, O'Sullivan EJ, Johnson MR, Heroux AL, et al. Increased incidence of lymphoproliferative disorder after immunosuppression with the monoclonal antibody OKT3 in cardiac- transplant recipients. N Engl J Med 1990;323:1723-8. Penn I. Kaposi's sarcoma in organ transplant recipients: report of 20 cases. Transplantation 1979:27:8-11. (6) Penn I. Depressed immunity and the development of cancer. Clin Exp Immunol 1981:46:459-74. Spector BD. Perry GS 3d. Kersey JH. Genetically determined immunode- ficiency disease (GDID) and malignancy: report from the immunodeficien- cy—cancer registry. Clin Immunol Immunopathol 1978;11:12-29. Correa P, O’Conor GT, Berard CW, Axtell LM, Myers MH. International comparability and reproducibility in histologic subclassification of Hodg- kin’s disease. J Natl Cancer Inst 1973;50:1429-35. (9) MacMahon B. Epidemiology of Hodgkin's disease. Cancer Res 1966:26: 1189-200. (10) Grufferman S, Delzell E. Epidemiology of Hodgkin's disease. Epidemiol Rev 1984:6:76—-106. (11) Hu E, Hufford S, Lukes R, Bernstein-Singer M, Sobel G. Gill P, et al. Third-World Hodgkins disease at the Los Angeles County—University of Southern California Medical Center. J Clin Oncol 1988:6:1285-93. Eltringham JR, Kaplan HS. Impaired delayed-hypersensitivity responses in 154 patients with untreated Hodgkin's disease. Natl Cancer Inst Monogr 1973:36:107-15. (13) Kumar RK, Penny R. Cell-mediated immune deficiency in Hodgkin's dis- ease. Immunol Today 1982:3:269-73. (14) Levy R, Kaplan HS. Impaired lymphocyte function in untreated Hodgkin's disease. N Engl J Med 1974:290:181-6. (15) Posner MR, Reinherz EL, Breard J, Nadler LM, Rosenthal DS, Schlossman SE. Lymphoid subpopulations of peripheral blood and spleen in untreated Hodgkin's disease. Cancer 1981:48:1170-6. (16) Ewing J. Neoplastic diseases. Philadelphia: Saunders, 1928. (17) Notter DT, Grossman PL, Rosenberg SA, Remington JS. Infections in patients with Hodgkin's disease: a clinical study of 300 consecutive adult patients. Rev Infect Disease 1980:2:761-800. (18) Biggar RJ, Horm J, Goedert JJ, Melbye M. Cancer in a group at risk of acquired immunodeficiency syndrome (AIDS) through 1984. Am J Epide- miol 1987:126:578-86. (19) Bernstein L, Levin D, Menck H, Ross RK. AIDS-related secular trends in cancer in Los Angeles County men: a comparison by marital status. Cancer Res 1989:49:466-70. (20) Hessol NA, Katz MH, Liu JY, Buchbinder SP, Rubino CJ, Holmberg SD. Increased incidence of Hodgkin disease in homosexual men with HIV infection. Ann Intern Med 1992;117:309-11. (21) Glaser SL, Swartz WG. Time trends in Hodgkin's disease incidence. The role of diagnostic accuracy. Cancer 1990;66:2196-204. (22) Kant JA, Hubbard SM, Longo DL, Simon RM, DeVita VT Jr, Jaffe ES, et al. The pathologic and clinical heterogeneity of lymphocyte-depleted Hodgkin's disease. J Clin Oncol 1986:4:284-94. (23) Lyter DW, Bryant J, Thackeray R, Rinaldo CR, Kingsley LA. Incidence of human immunodeficiency virus-related and nonrelated malignancies in a large cohort of homosexual men. J Clin Oncol 1995:13:2540-6. (24) Cote TR, Biggar RJ, Rosenberg PS, DeVesa SS. Percy C, Yellin FJ, et al. Non-Hodgkin's lymphoma among people with AIDS: incidence presenta- tion and public health burden. Int J Cancer 1997;73:645-50. (25) Rabkin CS, Hilgartner MW, Hedberg KW, Aledort LM, Hatzakis A, Ei- (2 (3 (4 (5 J (8 ~ ro 41 (26) (27, (28 (29 — (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) chinger S, et al. Incidence of lymphomas and other cancers in HIV-infected and HIV-uninfected patients with hemophilia. JAMA 1992:267:1090-4. Weiss SH, Karcnik T, Patel S, Skurnick JH, Fromowitz F, Schwartz RA, et al. The ten-year cancer and HIV experience of the UMDNIJ-University Hospital cancer registry: excess proportions of Hodgkin's lymphoma and lung cancer. J AIDS Human Retrovirol 1997:14:A18. Myers AM, Toi M, Muldrow M. Rajan G. Kane M. Curiel T, et al. Non AIDS-defining malignancies in HIV positive individuals. J AIDS Human Retrovirol 1997;14:A31. Grulich A, Wan X, Law M, Coates M, Kaldor J. Rates of non-AIDS defining cancers in people with AIDS. J AIDS Human Retrovirol 1997; 14:A18. Monfardini S, Tirelli U, Vaccher E, Ambrosini A, Andriani A, Silvestroni IB, et al. Malignant lymphomas in patients with or at risk for AIDS: a report of 50 cases observed in Italy. Cancer Detect Prev 1988;12:237-41. Andrieu JM, Roithmann S, Tourani JM, Levy R, Desablens B, le Maignan C, et al. Hodgkin's disease during HIV infection: the French registry experience. French registry of HIV-associated tumors. Ann Oncol 1993:4: 635-41. Roithmann S, Tourani JM, Andrieu JM. Hodgkin's disease in HIV-infected intravenous drug abusers. N Engl J Med 1990;323:275-6. Rubio R. Hodgkin's disease associated with human immunodeficiency vi- rus infection. A clinical study of 46 cases. Cooperative Study Group of Malignancies Associated with HIV Infection of Madrid. Cancer 199473: 2400-7. Ahmed T, Wormser GP, Stahl RE, Mamtani R. Cimino J, Glasser M, et al. Malignant lymphomas in a population at risk for acquired immune defi- ciency syndrome. Cancer 1987:60:719-23. Ames ED, Conjalka MS, Goldberg AF, Hirschman R, Jain S, Distenfeld A, et al. Hodgkin's disease and AIDS. Twenty-three new cases and a review of the literature. Hematol Oncol Clin North Am 1991:5:343-56. Monfardini S, Tirelli U, Vaccher E, Foa R, Gavosto F. Hodgkin's disease in 63 intravenous drug users infected with human immunodeficiency virus. Gruppo Italiano Cooperativo AIDS & Tumori (GICAT). Ann Oncol 1991:2(suppl 2):201-5. Levine AM, Cheung T, Tulpule A, Huang J, Testa M. Preliminary results of AIDS Clinical Trials Group (ACTG) study #149: phase II trial of ABVD chemotherapy with G-CSF in HIV-infected patients with Hodgkin's dis- ease. J AIDS Human Retrovirol 1997;14:A12. Davis S, Dahlberg S, Myers MH, Chen A, Steinhorn SC. Hodgkin's disease in the United States: a comparison of patient characteristics and survival in the Centralized Cancer Patient Data System and the Surveillance, Epide- miology, and End Results Program. J Natl Cancer Inst 1987;78:471-8. Gold JE, Altarac D, Ree HJ, Khan A, Sordillo PP, Zalusky R. HIV- associated Hodgkin disease: a clinical study of 18 cases and review of the literature. Am J Hematol 1991:36:93-9. Ree HJ, Strauchen JA, Khan AA, Gold JE, Crowley JP, Kahn H, et al. Human immunodeficiency virus-associated Hodgkin's disease. Clinico- pathologic studies of 24 cases and preponderance of mixed cellularity type characterized by the occurrence of fibrohistiocytoid stromal cells. Cancer 1991:67:1614-21. Serrano M, Bellas C, Campo E, Ribera J, Martin C, Rubio R, et al. Hodg- kin’s disease in patients with antibodies to human immunodeficiency virus. A study of 22 patients. Cancer 1990;65:2248-54. Tirelli U, Errante D, Vaccher E, Repetto L, Rizzardini G, Spina M, et al. Hodgkin's disease in 92 patients with HIV infection: the Italian experience. (43) (44) (45) (46) (47) (48) (49) (50) (51) (52 (53) (54) (55) (56) 2 tr ~ GICAT (Italian Cooperative Group on AIDS & Tumors). Ann Oncol 1992:3(suppl 4):69-72. Kaplan HS. Hodgkin's disease. 2nd ed. Cambridge, MA: Harvard Univer- sity Press, 1980. Schoeppel SL, Hoppe RT, Dortman RF, Horning SJ, Chew TG, Weiss LM, et al. AIDS and Hodgkins disease. Front Radiat Ther Oncol 1985:19: 66-73. Karcher DS. Clinically unsuspected Hodgkin disease presenting initially in the bone marrow of patients infected with the human immunodeficiency virus. Cancer 1993:71:1235-8. Colby TV, Hoppe RT, Warnke RA. Hodgkin's disease at autopsy: 1972-1977. Cancer 1982:49:1848-58. O’Carrol DI, McKenna RW, Brunning RD. Bone marrow manifestations of Hodgkin's disease. Cancer 1976:38:1717-28. Picard O, de Gramont A, Krulik M, Gonzalez G, Debray I, Cady J. Rectal Hodgkin disease and the acquired immunodeficiency syndrome. Ann Intern Med 1987:106:775. Shaw MT, Jacobs SR. Cutaneous Hodgkin's disease in a patient with human immunodeficiency virus infection. Cancer 1989;64:2585-7. Hair LS, Rogers JD, Chadburn A, Sisti MB, Knowles DM, Powers JM. Intracerebral Hodgkin's disease in a human immunodeficiency virus- seropositive patient. Cancer 1991:67:2931-4. Schieb RG, Siegel RS. Atypical Hodgkin's disease and the acquired im- munodeficiency syndrome. Ann Intern Med 1985:102:554. Hoppe RT, Hanlon AL, Hanks GE, Owen JB. Progress in the treatment of Hodgkin's disease in the United States, 1973 versus 1983. The Patterns of Care Study. Cancer 1994:74:3198-203. Canellos GP, Anderson JR, Propert KJ, Nissen N, Cooper MR, Henderson ES, et al. Chemotherapy of advanced Hodgkin's disease with MOPP, ABVD, or MOPP alternating with ABVD. N Engl J Med 1992:327: 1478-84. Errante D, Tirelli U, Gastaldi R, Milo D, Nosari AM, Rossi G, et al. Combined antineoplastic and antiretroviral therapy for patients with Hodg- kin’s disease and human immunodeficiency virus infection. A prospective study of 17 patients. The Italian Cooperative Group on AIDS and Tumors (GICAT). Cancer 1994:73:437-44. Tirelli U, Vaccher E. Rezza G, Broccia G, Fassio PG, Gobbi M, et al. Hodgkin disease and infection with the human immunodeficiency virus in Italy. Ann Intern Med 1988;108:309-10. Newcom SR, Ward M, Napoli VM, Kutner M. Treatment of human im- munodeficiency virus-associated Hodgkin disease. Is there a clue regarding the cause of Hodgkin disease? Cancer 1993;71:3138-45. Tirelli U, Errante D, Santarossa S, Spina M, Gisselbrecht C, Marolleau P, et al. Epirubicin, bleomycin, vinblastine, and prednisone chemotherapy in combination with antiretroviral therapy and primary use of G-CSF for patients with Hodgkin's disease and HIV infection. J AIDS and Human Retrovirol 1997:14:A32. Note "Editor's note: SEER is a set of geographically defined, population-based central tumor registries in the United States. operated by local nonprofit orga- nizations under contract to the National Cancer Institute (NCI). Each registry annually submits its cases to the NCI on a computer tape. These computer tapes are then edited by the NCI and made available for analysis. Journal of the National Cancer Institute Monographs No. 23, 1998 Management of Cervical Neoplasia in Human Immunodeficiency Virus-Infected Women Mitchell Maiman* The existence of cervical neoplasia in women with human immunodeficiency virus (HIV) represents one of the most serious challenges in the oncologic care of immunosup- pressed patients. While the development of most cancers in the immunosuppressed patient can be attributed solely to immune deficiency, the relationship between squamous cell neoplasia of the cervix and HIV is quite unique because of common sexual behavioral risk factors. Screening strategies in HIV-positive women must take into account the high prevalence of cervical dysplasia in this subgroup as well as the limitations of cytologic screening. Cervical dysplasia in HIV-positive women may be of higher grade than in HIV- negative patients, with more extensive involvement of the lower genital tract with HPV-associated lesions. The pres- ence and severity of cervical neoplasia in HIV-positive women correlate with both quantitative and qualitative T- cell function. Standard therapies for preinvasive cervical dis- ease have yielded suboptimal results with high recurrence rates. While poor treatment results of standard ablative and excisional therapies warrant unique therapeutic strategies, one must recognize that close surveillance and repetitive treatment have been successful in preventing progressive neoplasia and invasive cervical carcinoma. The disease char- acteristics of invasive cervical carcinoma may take a more aggressive clinical course in HIV-infected women. HIV- positive women with cervical cancer have higher recurrence and death rates with shorter intervals to recurrence and death than do HIV-negative control subjects. CD4 status does influence subsequent outcome. In general, the same principles that guide the oncologic management of cervical cancer in immunocompetent patients should be applied. However, extremely close monitoring for both therapeutic efficacy and unusual toxicity must be instituted. [Monogr Natl Cancer Inst 1998;23:43-49] Human immunodeficiency virus (HIV) infection continues to be a national and international health problem of epidemic pro- portions. Despite the fact that new acquired immunodeficiency syndrome (AIDS) cases increased by less than 5% for the fifth year in a row, a number well below the rate of increase in the epidemic’s first decade, the incidence of AIDS cases among women continues to increase, particularly in minority women. Approximately 19% of new adult and adolescent cases of AIDS in the United States last year were in women, and women rep- resent the subgroup with the greatest rate of increase compared with any other defined population in North America. As with cervical neoplasia, HIV infection is largely a disease of women Journal of the National Cancer Institute Monographs No. 23, 1998 in their reproductive years, with the incidence of both diseases significantly higher in women of color. The existence of cervical neoplasia in women infected with HIV represents one of the most serious challenges in the onco- logic care of immunosuppressed patients. While the develop- ment of most illnesses and cancers in HIV-infected patients can largely be attributed to immune deficiency, the relationship be- tween cervical neoplasia and HIV infection is quite unique. Both cervical carcinoma and HIV infection are, in part, sexually trans- mitted diseases, with oncogenic types of human papillomavirus (HPV) infection the implicated viral carcinogen associated with cervical cancer. Therefore, an association between cervical can- cer and HIV can be anticipated not only on the basis of immu- nosuppression but also because of shared common sexual be- havioral risk factors. Thus, while immunosuppressed women, such as renal transplant patients receiving highly immunosup- pressive drugs, are known to be at high risk for lower genital tract neoplasia, immunodeficient HIV-infected women are per- haps the highest risk subgroup that we know of. While these factors likely play the major role in the pathogenesis of cervical neoplasia in HIV-infected women, direct interactions between HIV and HPV at the molecular level, the effects of HIV on the local mucosal immune response, enhancement of HPV regula- tory expression by the HIV-1 tat protein, and HIV-induced per- turbations of paracrine or autocrine factors that influence HPV ene expression must also be considered. us Screening Issues Because a well-defined precursor lesion for cervical cancer can be detected by screening, organized screening programs for cervical neoplasia can be expected to reduce both the incidence and the mortality rates. Since the prevalence of disease is far greater in HIV-positive patients than in the general population, optimal screening strategies take on an increased importance. HIV-positive women have as much as a 10-fold increased rate of abnormal cytology, including a wide range of cellular and in- flammatory changes, such as hyperkeratosis, parakeratosis, trichonomas, herpes, inflammatory atypia, HPV-related changes, and varying degrees of cervical neoplasia (/). Higher rates of abnormalities have also been demonstrated after sero- conversion than before seroconversion. Studies from several centers have demonstrated that women who are immunodefi- cient from HIV infection have cytologic abnormality rates of between 30% and 60% (2-4) and Pap smears consistent with *Correspondence to: Mitchell Maiman, M.D., Department of Gynecologic Oncology, State University of New York-Health Science Center at Brooklyn, 450 Clarkson Ave., Box 24, Brooklyn, NY 11203. © Oxford University Press cervical dysplasia from 15% to 40% (5). Most studies also con- sistently demonstrate that the prevalence of such abnormalities increases as immunodeficiency becomes more severe. Screening strategies in HIV-positive women must take into account the high prevalence of cervical dysplasia in this sub- group as well as the limitations of cytologic screening, the rela- tively high noncompliance rate, and the possibility of acceler- ated progression of disease. These factors make initial accurate diagnosis critical. Although the Centers for Disease Control and Prevention (CDC) continues to recommend Pap smears as the sole screening method, many authors have advocated the need for baseline colposcopy in HIV-infected women because of the inaccuracy of cytology in consistently predicting histology, low negative predictive values of the Pap smear, decreased sensitiv- ity, and discordances between cytologic smear and biopsy re- sults (6-10). Other authors have found that HIV-infected women had significantly more smears of limited adequacy due to ob- scuring blood or inflammation and higher rates of concomitant vaginal infections leading to under-read smears (//). Maiman et al. (12) in a unique study of 248 HIV-infected women, all of whom had cytology, colposcopy, and biopsy, found that 38% of all cervical intraepithelial neoplasia (CIN) in 13% of the total patients would have been missed if routine colposcopy and bi- opsy were not performed. Interestingly, similar high false- negative rates of cytology have been reported in immunosup- pressed women after renal transplantation (/3). The limitations of cytologic screening become more glaring as the prevalence of cervical dysplasia increases in a given population. Although more frequent cytologic screening (every 6 months) is to be advocated, a reasonable but not universally accepted approach that we use at our institution involves base- line colposcopy or cervicography in HIV-positive women once the diagnosis is made (Fig. 1). Patients with normal colposcopy could then undergo more aggressive cytologic screening than the normal population, while those diagnosed and treated for CIN undergo colposcopy with liberal biopsy every 4-6 months for 2 years and seminannual Pap tests thereafter. Alternatively, screening strategies may be based on baseline immune status, with more aggressive techniques reserved for patients with CD4 counts of less than 500/mm®. Of course, all strategies must take into account availability of colposcopic resources and individualized knowledge of risk in given patient popula- tions. In contrast to the inherent lack of sensitivity of normal cer- vical cytology, abnormal cytology is extremely accurate in predicting CIN on histology, and Pap smears indicating CIN must be taken very seriously (/4). Specificities of 84% or greater have been reported with even higher predictive values (10). Therefore, the vast majority of such pa- tients will indeed have cervical pathology, jus- tifying immediate treatment during initial col- - Baseline colposcopy or cervicography -Evaluation of entire lower genital tract -Pap smear -CD4 count CIN | | Normal poscopic evaluation of a significantly abnormal Pap smear. This ‘“‘see-and-treat’” approach us- ing excisional methods such as the loop elec- trosurgical excision procedure may be particu- larly appropriate for the HIV-infected woman. The prevalence of HIV infection in women — with invasive cervical carcinoma may be even higher than in those with preinvasive disease. Maiman et al. (/5) reported a 19% seropositiv- ity rate in women under the age of 50 years in a high-risk population in Brooklyn. Of course, Cervical cytology every 6 months Primary Treatment (Loop excision, Laser, Cone Biopsy) this high prevalence rate is, in fact, reflective of the high HIV infection rate in our community and would be expectedly lower in lower risk geographic areas. Most important, the majority of these women were asymptomatic with regard to HIV disease and die of cervical cancer, not AIDS; therefore, only HIV screening programs 6 months 6 months -Colposcopy with liberal biopsy every 4-6 months for 2 years, then every -Cervical cytology every would have detected their positive serostatus. At our institution, we currently recommend HIV counseling and testing in all younger (<50 years of age) patients with cervical cancer, since test results may have a significant impact on oncologic therapeutic strategies. The Gyneco- logic Oncology Group (GOG) is currently con- ducting a nationwide screening study involving HIV testing and follow-up of newly diagnosed patients with invasive cervical carcinoma. Invasive Cervical Carcinoma Fig. 1. Screening for cervical intraepithelial neoplasia in human immunodeficiency virus-positive women. 44 On January 1, 1993, the CDC expanded the surveillance case definition of AIDS to include Journal of the National Cancer Institute Monographs No. 23, 1998 HIV-positive women with invasive cervical cancer. As an AIDS-defining illness, it is required that physicians and hospi- tals report cases of cervical cancer in HIV-infected women to their local health departments. These changes have served to educate the health care community concerning this relationship and stimulate more aggressive testing programs. In the first year of the expanded definition, approximately 1.3% of women with AIDS-defining illnesses 13 years of age or older had cervical cancer (1/6). However, HIV testing in women with cervical can- cer was far from routine, and documentation of AIDS cases varied considerably based on the reporting practices of physi- cians and hospitals. Other factors may also contribute to the underdiagnosis of cervical cancer in HIV-infected women, in- cluding lack of HIV testing in older women, poor access to health care, and the coexistence of more acute and life- threatening AIDS-defining opportunistic infections that inhibit both the diagnosis and the reporting of cervical cancer. In an urban population at high risk for both diseases in Brooklyn, where a routine HIV testing program was instituted in all cer- vical cancer patients 50 years of age or younger (/7), cervical cancer was the sixth most common AIDS-defining illness in women representing 4% of the subjects as well as the most common AIDS-related cancer in women (55%), followed by lymphoma (29%) and Kaposi's sarcoma (16%). The signs and symptoms of cervical cancer may take on typi- cal or atypical presentations in HIV-infected women. Ideally, patients should be diagnosed by classic screening methods of cytology, colposcopy, and biopsy if disease is to be detected in the preinvasive or early invasive phase. When patients present with symptoms, more advanced disease is often found, and vagi- nal bleeding or postcoital bleeding is most commonly reported. Malodorous vaginal discharge is also quite common, as is pelvic pain, back pain, or lower abdominal pain. Leg pain, edema, weight loss, or obstructive uropathy is indicative of more ad- vanced disease. In HIV-positive patients, metastatic disease may occur in both common and uncommon sites (/8), and unusual extracervical metastases have been described in the psoas muscles, periclitoral area, and spinal cord as well as malignant ascites (19,20). Diagnosis can be even more difficult, because the classic signs of systemic cancer may mimic the subtle mani- festations of HIV disease. Low-grade fevers, unexplained weight loss, gastrointestinal disturbances, and fatigue may occur in both disease processes. Lymphadenopathy, either clinically detected in the left supraclavicular (scalene) or inguinal nodes, or retroperitoneal (pelvic or para-aortic) nodes, discovered at the time of surgery or radiologic imaging, must not be assumed to be metastatic cancer; HIV-infected patients will often have very large, suspicious nodes that may be secondary to follicular hy- perplasia, not to a tumor. This pathologic diagnosis, which in- cludes mononuclear cell proliferation, polykaryoctes, epithelioid histiocytes, and mantle-zone loss, while not pathognomonic for HIV, is highly suggestive. In addition, coexistent pelvic infection may present diagnostic and therapeutic dilemmas. HIV-infected women have high rates of concomitant pelvic inflammatory disease that is more often refractory to antibiotic therapy. Pelvic abscesses may mimic metastatic cervical cancer and pelvic infection may contribute to the development and spread of disease and increase the failure rate and morbidity of therapeutic interventions. Journal of the National Cancer Institute Monographs No. 23, 1998 Cervical cancer is the only gynecologic cancer that is clini- cally staged, which is assigned by the current staging system of the International Federation of Gynecology and Obstetrics (FIGO). Once a clinical stage is assigned and treatment has been initiated, the stage must not be changed because of subsequent findings by either extended clinical or surgical staging. Bi- manual rectovaginal pelvic examination, performed under anes- thesia if necessary, is the most important part of clinical staging, and only certain additional studies for extracervical disease are allowed by FIGO, including intravenous pyelogram, barium en- ema, chest and skeletal x ray, cystoscopy, and sigmoidoscopy. However, other more practical and useful studies are often em- ployed that provide more specific information, including com- puterized axial tomography, magnetic resonance imaging, and laparoscopic lymph node biopsies or extraperitoneal lymph node biopsies by laparotomy. Tumor markers, specifically squamous cell carcinoma antigen or CA 125 for the respective cell types of squamous or adenocarcinoma, are useful parameters to follow patients with during and after treatment. Cancers of the cervix most often spread by direct invasion into the cervical stroma and laterally into the parametrial tissues as well as into the vagina and corpus uteri. Lymphatic metastases are also quite common and are involved in a progressive fashion beginning with the pelvic (obturator, external and internal iliac) nodes and then involving the para-aortic, inguinal, and scalene nodes. The incidence of positive pelvic and para-aortic nodes ranges from 15% to 5%, respectively, for patients with stage IB disease to 45% and 30% for patients with stage III. However, HIV-positive patients may have higher grade tumors and higher rates of lymph node involvement than expected by stage alone. Cervical cancer may also spread by blood-borne metastases or intraperitoneal implantation or present or recur with painful bony metastases. Although much of the data, thus far, has been limited to case reports and single institution studies, it seems apparent that in some HIV-infected women, the disease characteristics may take a more aggressive clinical course (75,27). This parallels the examples observed in other AIDS-related cancers. HIV-positive women present with more advanced disease than do HIV- negative women with cervical cancer. In the largest study to date at the Health Science Center at Brooklyn comparing 16 sero- positive to 68 seronegative women, significantly more advanced disease was found in the HIV-positive women (75). Half of the seropositive patients presented with stage 111 or IV disease, com- pared with 19% in seronegative controls, and only one infected patient remained with early disease after careful surgical— pathologic staging. Almost 70% of the HIV-infected women had stage IIT or higher surgical-pathologic stage compared with 28% of seronegative controls. HIV-infected women with cervical cancer may be quite young, such as the case of a 16-year-old who presented with stage IIIB disease and died at age 17 years (18). Most HIV-positive patients can be expected to have squa- mous cell carcinomas as was found in the above study where 15 of 16 patients had squamous cell cancers, with the remaining patient having an adenosquamous tumor (/5). Most important, the majority of patients die of cervical cancer before they died of AIDS and are usually asymptomatic with regard to HIV infec- tion. Therefore, only HIV testing programs of such patients will 45 enable the physician to make the diagnosis and realize the full impact of the interaction of these two disease processes. The HIV-positive women with cervical cancer have higher recurrence and death rates with shorter intervals to recurrence and death than do HIV-negative control subjects (/5). Mean intervals to recurrence are extremely short, and many patients retain persistent disease after primary treatment. Median time to death was found to be 10 months in seropositive women com- pared with 23 months in seronegative patients. Several other case reports have described examples of rapidly progressive cer- vical cancer in such women. As with preinvasive disease, the relationship between immune function and disease status is ap- parent. The mean absolute CD4 count in such patients was 360/ mm” compared with 830/mm® in uninfected women. One can infer from these values that the typical HIV-infected woman with cervical cancer would not meet the immunologic definition for AIDS solely by CD4 count (<200/mm?), emphasizing the importance of the recent inclusion of seropositive women with cervical cancer as AIDS patients. Although stage of cancer may not predict CD4 levels, immune status does influence subse- quent outcome. Patients with counts greater than 500/mm?* have had more favorable disease courses; therefore, management de- cisions in HIV-infected women with cervical cancer should carefully consider pretreatment immune function, since positive serostatus alone may not necessarily and uniformly confer a dismal outcome (75). HIV-related immunodeficiency may con- tribute heavily to the natural history of cervical carcinoma, and HIV-positive patients need not demonstrate other signs of im- munosuppression such as opportunistic infections for their neo- plasia to be adversely affected by HIV. The characteristics of HIV disease in cervical cancer may be different in cervical cancer patients when compared with HIV- positive patients with other AIDS-related cancers. Women with invasive cervical cancer are less immunosuppressed than women with other AIDS opportunistic illnesses and may be ex- pected to have CD4 counts about twice as high as those with should be performed in most cases of stage IA2 and IB1 and in some cases of stage IB2 and IIA cervical cancer where cervical size is not too enlarged and vaginal involvement is minimal. Radical oncologic surgery in HIV-positive women can be per- formed for the usual indications, and surgical decisions should be based on oncologic issues not HIV status. As has been dem- onstrated in other types of operations in the HIV-positive pa- tients, women with relatively good immune function tolerate surgery well with no significant excess morbidity. Prophylactic antibiotics should be routinely used, and standard surgical pre- cautions should be taken to prevent surgical transmission. The transmission rate of HIV from patient to health care worker is extremely low, estimated to be about I in 320. Although radiation therapy can be used in all stages of cer- vical cancer and has identical cure rates when compared with radical surgery in stage IB, the issues of ovarian conservation and better vaginal sexual function make radical hysterectomy the preferred modality in younger patients. Ovarian transposition should be considered at the time of surgery if postoperative pelvic radiation is a strong possibility. One may expect a uro- logic fistula rate of 1%-2% and a chronic bladder atony rate of 3% after radical hysterectomy compared with an intestinal and urinary stricture and fistula rate of 3%-5% and rate of chronic radiation fibrosis of the bowel or bladder of 6%—-8% with pelvic radiation therapy. Another advantage of surgery over radiation is the availability of surgical-pathologic data such as lymph node status for which postoperative therapy can then be designed. Since most HIV-infected patients with cervical cancer present with more advanced disease, radiation therapy is usually the cornerstone of treatment and is the basis for therapy in stages [I-1V. Pelvic radiation therapy begins with external-beam therapy designed to shrink the primary tumor and create better geometry for the brachytherapy insertions to follow. Pelvic fields are usually 15 x 5 cm, extending to a 2-cm margin lateral Kaposi's sarcoma and non-Hodgkin's lymphoma (312-153/ mm’) (17). In the majority of women with cervical cancer, HIV infection was diagnosed at the time of cancer presen- tation, whereas in women with other cancers, HIV diagnosis preceded cancer diagnosis by a mean of 2.7 years. Although the interval from cancer diagnosis to death may be similar in all AIDS-related cancers, cancer was found to be the cause of death more often with cervical cancer patients (95%) compared with those with other cancers (60%). Although patients with cervical cancer as their AIDS-defining illness may be slightly younger than those women without cervical cancer, the distribution of mode of HIV transmission (het- erosexual versus injection drug abuse) and race (black ver- sus Hispanic versus white) was found to be remarkably simi- lar (16). The management of HIV-positive patients with cervical cancer is among the most challenging tasks faced by the oncologic team. In general, the same principles that guide the oncologic management of cervical cancer in immuno- competent patients should be applied (Fig. 2). However, extremely close monitoring for both therapeutic efficacy Stage Ial: Cold knife therapeutic cone biopsy if fertility desired; otherwise, simple hysterectomy Stage IA2 Stage IB1 Radical hysterectomy with pelvic lymphadenectomy Alternatively, radiation therapy in poor surgical candidates Stage IB2 Stage ITA Radiation therapy +/- simple hysterectomy; or Radical hysterectomy with pelvic lymphadenectomy; or Neoadjuvant chemotherapy + radical surgery Stage IIB-IVA: Radiation therapy +/- chemosensitization Stage IVB: Chemotherapy +/- radiation therapy Recurrent Disease: Pelvic Exenteration (central disease), otherwise Palliative Chemotherapy and unusual toxicity must be instituted. Radical hysterectomy and pelvic lymphadenectomy 46 Fig. 2. Treatment recommendations for cervical carcinoma in human immunodefi- ciency virus-infected women. Journal of the National Cancer Institute Monographs No. 23, 1998 to the bony pelvis and inferiorly to the border of the obturator foramen. The superior margin can be extended to 18 cm to cover the common iliac nodes or even higher if para-aortic lymph node metastases are evident. The usual dosages of pelvic radiotherapy delivered include 7000-8000 cGy to point A and 6000 cGy to point B. Some HIV-positive patients may respond poorly to radiation therapy alone and regimens that incorporate radiation sensitizers that are being used more frequently in general should be strongly considered. The two most common sensitizing regi- mens include cisplatin (50 mg/m’, weeks | and 6) combined with fluorouracil (1000 mg/m?>, continuous infusion x 4 days, weeks 1 and 6) or oral hydroxyurea (3 g/m’, twice per week x 5 weeks). One must also keep in mind that pelvic radiation is associated with transient lymphopenia and depressed T-cell function, which may lead to further immunologic compromise in the HIV-positive patient, and that anecdotal reports of poor tol- erance and increased morbidity from pelvic radiation therapy have been made, although adequate studies have not yet been performed to evaluate this issue. Chemotherapy should be used in cases of systemic disease (pulmonary or liver metastases) or recurrent disease after radia- tion failures in those patients not eligible for pelvic exenteration. However, recurrent cervical cancer in this setting is not consid- ered curable with chemotherapy, and treatment is palliative. Regimens that incorporate drugs that are both active in cervical carcinoma and relatively bone marrow sparing should be used, with close monitoring of all hematologic indices. Cisplatin (50— 75 mg/m?) is the drug of choice and may be combined with bleomycin (20 U/m?; maximum, 30 U) and vincristine (1 mg/ m?). Recently, neoadjuvant chemotherapy has been used in pa- tients with stage IB2 or stage II cervical cancer with similar agents followed by radical hysterectomy with good results; how- ever, the efficacy of this approach in HIV-positive patients is unknown. Other novel therapeutic approaches, such as the use of inter- ferons, retinoids, bone marrow support, or vaccine therapy, may represent future investigative treatment options. The Gynecol- ogy Oncology Group is presently evaluating the role of inter- feron alfa and isotretinoin with or without antiretroviral therapy in the neoadjuvant or advanced/recurrent setting in the treatment of HIV-infected women with cervical cancer. Innovative treat- ment regimens and research protocols such as these are needed to combat the interaction of these two potentially life- threatening disease processes. Last, the treatment of HIV and its sequelae with anti-HIV-1 agents and prophylactic regimens is exceedingly complex and always changing. Combination therapy with nucleoside ana- logues and protease inhibitors is now standard and initiated ear- lier in the course of HIV disease. Careful attention to overlap- ping side effects of these regimens when combined with oncologic therapy for cervical cancer as well as potential syn- ergistic antineoplastic effects will be an important area of future investigation. Preinvasive Cervical Neoplasia HIV-seropositive women represent perhaps the highest risk group encountered for the development of CIN. In the most recent classification system of HIV-related diseases, the CDC has identified moderate to severe cervical dysplasia and carci- Journal of the National Cancer Institute Monographs No. 23, 1998 noma in situ as category B conditions. Independent studies (2,4,6,10) have estimated the prevalence of CIN in this subgroup to be between 20% and 50%. Maiman et al. (12), in a study of HIV-positive women without AIDS-defining illness, found the prevalence of CIN on histology to be 32% in a cohort of 248 patients. Cervical dysplasia of HIV-positive women may be of higher grade than in seronegative women, with more extensive involvement of the lower genital tract with HPV-associated le- sions (/8). Extensive cervical involvement, endocervical in- volvement, and multisite (vagina, vulva, and perianal) disease are more common. Natural history studies examining the bio- logic behavior of cervical HPV and CIN strongly suggest that disease is more aggressive in HIV-positive patients. Conti (22) found fourfold higher progressions and threefold lower regres- sion rates of untreated HPV-related cervical lesions in infected women compared with HIV-negative controls, and Petry et al. (23) found only a 27% regression rate of CIN I lesions in im- munosuppressed HIV-positive and transplant patients compared with 62% in immunocompetent control subjects. Higher rates of more oncogenic HPV subtype infection, multiple-type HPV in- fection, and unspecified-type HPV infection have been reported by many investigators and may help explain the more aggressive cervical pathology that develops in HIV-infected women (12,24-26). Numerous studies (6,9,27) have demonstrated the relationship between HIV-associated immunosuppression and the develop- ment of CIN, and the presence and severity of cervical neoplasia are associated with quantitative T-cell function. In one study (6), HIV-positive patients with CIN had absolute CD4 counts and T4:T8 ratios roughly half those of HIV-positive patients without CIN, and patients with AIDS-defining illness are more likely to have cervical disease than are asymptomatic HIV-positive pa- tients. Schafer et al. (27) and Wright et al. (28) concluded that a CD4 lymphocyte count of 200/mm® was independently associ- ated with CIN. The concept of worsening immunodeficiency increasing the risk of cervical pathology may be used to indi- vidualize screening and surveillance strategies in HIV-positive women. The treatment of preinvasive cervical disease in HIV-infected women is among the most challenging and frustrating tasks faced by the practicing gynecologist. In general, standard thera- peutic strategies for immunocompetent women apply, but the increased risk for treatment failures and chronic nature of dis- ease in such women is well documented. Excisional methods, which include LEEP (loop electrosurgi- cal excision procedure), laser cone, and cold knife cone biopsy, are preferred over ablative methods, such as cryotherapy and laser vaporization. Excisional methods have the advantage of confirmation of histology and documentation of negative mar- gins, which is of particular importance in HIV-positive women in whom disease may be more extensive. Del Priore et al. (29) reported that colposcopically directed biopsies may be poor pre- dictors of histology on excisional cone specimens in HIV- seropositive women, since 47% with CIN II-III on cone biopsy had only CIN I or HPV on punch biopsy compared with only 9% in HIV-seronegative patients Additionally, with advances in the technology in electrosurgical generators and the development of large wire loops with insulated bases, LEEP has allowed exci- sion of the cervical transformation zone and distal canal with a 47 single pass of the loop with the patient under local anesthesia in an outpatient setting. Therefore, in most centers, LEEP excision has become the preferred treatment, although laser ablation is still quite adequate if the CIN lesion lies within the range of satisfactory colposcopic assessment. Although the use of cryo- therapy in HIV-positive women may seem particularly attractive because of the absence of bleeding and reduced theoretical risk of iatrogenic transmission of HIV, studies have indicated that the use of this modality in HIV-infected women may offer a specific treatment disadvantage. In one study (30), 48% of HIV-infected women with low-grade CIN developed recurrence compared with 1% of seronegative women. Diagnostic procedures, such as cold knife cone conization, diagnostic LEEP, or laser cone, must be performed for the usual indications. Diagnostic procedures may then be considered therapeutic if histologic results are fa- vorable. An algorithm for the management of CIN in HIV- positive patients is presented in Fig. 3. Recurrence rates for CIN in HIV-positive women with stan- dard therapies have been reported to be as high as 40% at one year and 60% with longer follow-up (30,31). Maiman et al. (30) and Del Priore and Lurain (9) reported recurrence rates of 39% and 40%, respectively, compared with 9% and 10% on seropos- itive controls. Fruchter et al. (32), at 36 months, reported a 62% failure rate compared with 18% in control subjects, with an 87% failure rate in patients with CD4 less than 200/mm?*. In addition, progression to higher grade dysplasia was more common in HIV-positive patients as well as multiple episodes of recurrent disease requiring many repeated procedures. It has been well documented that the frequency of recurrence is closely related to immune function. Patients with CD4 counts less than 500/mm” are at extremely high risk for recurrence, whereas women with counts greater than 500/mm’ may be expected to have a recurrence women and be particularly reserved for the multiparous patient with relatively good immune function who has undergone mul- tiple therapeutic procedures for recurrent high dysplasia in whom repeated evaluation is exceedingly difficult. In response to the high rates of treatment failures for prein- vasive cervical disease in HIV-positive women, the AIDS Clini- cal Trials Group (ACTG) is investigating novel therapeutic ap- proaches. For patients with high-grade dysplasia (CIN-II-III), ACTG examined the role of topical vaginal flourouracil cream maintenance therapy as prophylaxis against recurrent CIN. In this study, patients were randomized to receive standard ablative or excisional therapy alone versus standard therapy plus 2 g of vaginal fluorouracil every 2 weeks for 6 months. Fluorouracil has previously been used with considerable success in the care of immunocompromised women with lower genital tract neoplasia, especially after conventional therapy has resulted in repetitive recurrence. Krebs (34) found that prophylactic maintenance therapy with vaginal fluorouracil was effective in the treatment of HPV-associated lesions of the vulva and vagina, especially in immunosuppressed women with multiple organ involvement. Effective local adjunctive therapy with topical chemotherapeutic agents is particularly attractive in HIV-positive patients in light of their favorable therapeutic index, and results of such random- ized controlled studies are extremely important. ACTG 293 is a study involving the treatment of HIV-positive patients with CIN-I. While the standard of care today in immu- nocompetent women with CIN-I (mild dysplasia) is close fol- low-up without surgical therapy, it is unknown whether such conservative strategy is safe in seropositive patients in light of reports suggesting more aggressive disease. In this trial, patients are randomized between observation only without any surgical at only twice the rate of seronegative women (30). Therefore, diagnostic and therapeutic strategies may be stratified based on the degree of im- munosuppression. Unfortunately, complications after treatment for CIN in HIV-positive women may be in- creased since one study demonstrated higher rates of excessive bleeding and cervicovaginal infections (33). While poor treatment results of stan- dard ablative and excisional therapies in HIV-infected women with CIN cer- tainly warrant unique therapeutic strat- egies, one must recognize that close and meticulous post-therapy surveil- lance and repetitive aggressive re- treatment for persistent and recurrent disease have been successful in pre- venting progressive neoplasia and in- vasive cervical carcinoma and should therefore not be abandoned. Hysterec- tomy, which is rarely used today in the management of CIN in immunocom- CINI Close observation LEEP Small lesion CD4>200 Large lesion CD4 <200 Adequate colposcopy CIN IT IIX CIN II-IT1 Entire lesion not seen Positive ECC Possible microinvasion Cytology/Histology discordance Adenocarcinoma in situ x LEEP Cold knife conization LASER ablation LASER cone LEEP cone a RECURRENCE Retreatment +/- Maintenance 5-FU petent patients, may be considered in individual cases in HIV-seropositive 48 Fig. 3. Treatment options for cervical intraepithelial neoplasia in human immunodeficiency virus-positive women. 5-FU = fluorouracil. Journal of the National Cancer Institute Monographs No. 23, 1998 therapy versus oral isotretenoin at a dose of 0.5 mg/kg per day for 6 months, with close attention to follow-up for regression, persistence, or progression. The results of this trial are likely to have implications for women with CIN in general. Optimization of immune function and lowering of viral load with proven and newer anti-HIV drugs are also desirable, and many patients today are placed on multiple drug regimens. It is unknown whether any of such interventions has any impact on cervical disease, and the presence of multiple confounding vari- ables will make this issue exceedingly difficult to study. At present, the principles of aggressive initial evaluation, more fre- quent cytologic screening, and meticulous post-therapy surveil- lance with liberal repeat colposcopy and re-treatment for recur- rent dysplasia in managing preinvasive disease in HIV-infected women seem prudent as we investigate novel treatment strate- gies in clinical trials. References (1) Provencher D, Valme, S, Averette HE, Ganjei P, Donato D, Penalver M, et al. HIV status and positive Papanicolaou screening: identification of a high-risk population. Gynecol Oncol 1988:31:184-90. Schrager LK, Friedland GH, Maude D. Cervical and vaginal squamous cell abnormalities in women infected with the human immunodeficiency virus. J AIDS 1990:2:570-5. Marte C, Kelly P, Cohen M, Fruchter RG, Sedlis A, Gallo L, et al. Papa- nicolaou smear abnormalities in ambulatory care sites for women infected with the human immunodeficiency virus. Am J Obstet Gynecol 1992;166: 1232-7. Feingold AR, Vermund SH, Kelley KF, Schreiber LK, Munk G, Friedland GH, et al. Cervical cytological abnormalities and papillomavirus in women infected with HIV. J AIDS 1990:3:896-903. Smith JR, Kitchen VS, Bothcherby M, Hepburn M, Wells C, Gor D, et al. Is HIV infection associated with an increase in the prevalence of cervical neoplasia? Br J Obstet Gynaecol 1993:100:149-53. Maiman M, Tarricone N, Viera J. Suarez J. Serur E, Boyce JG. Colpo- scopic evaluation of human immunodeficiency virus-seropositive women. Obstet Gynecol 1991:78:84-8. Kell PD, Shah PM, Barton SE. Colposcopic screening of HIV seropositive women—help or hinderance? Int Conf AIDS 1992:8:B96. Fink MJ, Fruchter RG, Maiman M, Kelly P. Sedlis A, Weber CA, et al. The adequacy of cytology and colposcopy in diagnosing cervical neoplasia in HIV-seropositive women. Gynecol Oncol 1994:55:133-7. Del Priore G, Lurain J. The value of cervical cytology in HIV-positive women. Gynecol Oncol 1995:56:395-8. Korn A, Autry M, DeRemer P. Sensitivity of the Papanicolaou smear in human immunodeficiency virus-infected women. Obstet Gynecol 1994:83: 401-4. Fruchter RG, Maiman M, Silman FH, Camilien L, Webber CA, Kim DS. Characteristics of cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus. Am J Obstet Gynecol 1994;171: 531-7. Maiman M, Fruchter RG, Sedlis A, Feldman J, Chen P, Burk RD, et al. Prevalence, risk factors and accuracy of cytologic screening for cervical intraepithelial neoplasia in women with the human immunodeficiency vi- rus. Accepted for publication Am J Obstet Gynecol. Alloub MM, Barr BB, McLaren KM, Smith GE. Bunney MH, Smart GE. Human papillomavirus infection and cervical intraepithelial neoplasia in women with renal allografts. Br Med J 1989:298:153-6. Adachi A, Fleming I, Burk RD, Ho GY, Klein RS. Women with human immunodeficiency virus infection and abnormal Papanicolaou smears: a prospective study of colposcopy and clinical outcome. Obstet Gynecol 1993:81:372-7. (4 (5 (6) (7 — (8) (9 (10) (11) (13) (14) Journal of the National Cancer Institute Monographs No. 23, 1998 (15) (16) (17) (18) (19) (23) (24) (25) (26) (28) (29) (30) (31) (32) (33) (34) Maiman M. Fruchter RG, Guy L, Cuthill S, Levine P, Serur E. Human immunodeficiency virus infection and invasive carcinoma. Cancer 1993; 71:402-6. Klevens MR, Fleming PL, Mays MA, Frey R. Characteristics of women with AIDS and invasive cancer. Obstet Gynecol 1996:88:169-73. Maiman M, Fruchter MG, Clark M, Arrastia CD, Matthews R, Gates EJ. Cervical cancer as an AIDS-defining illness. Obstet Gynecol 1997:89: 76-80. Maiman M, Fruchter RG, Serur E, Remy JC, Feuer G, Boyce JG. Human immunodeficiency virus and cervical neoplasia. Gynecol Oncol 1990:38: 377-82. Singh GS, Aikins JK, Deger R, King S, Mikuta JJ. Case report on meta- static cervical cancer and pelvic inflammatory disease in the AIDS patient. Gynecol Oncol 1994:54:372-6. Schwartz LB, Carcangiu ML, Bradham L, Schwartz PE. Rapidly progres- sive squamous cell carcinoma of the cervix coexisting with human immu- nodeficiency virus infection: clinical opinion. Gynecol Oncol 1991:41: 255-8. Relliman MA, Dooley DP, Burke TW, Berkland ME, Longfield RN. Rap- idly progressing cervical cancer in a patient with human immunodeficiency virus infection. Gynecol Oncol 1990;36:435-8. Conti M. Prevalence and risk of progression of genital intraepithelial neo- plasia in women with human immunodeficiency virus infection. Adv Gy- necol Obstet Res 1991:3:283-7. Petry KU, Scheffel D, Bode U. Cellular immunodeficiency enhances the progression of human papillomavirus-associated cervical lesions. Int J Can- cer 1994:57:836-40. Sun XW, Ellerbrock TV, Lungu O, Chiasson MA, Bush TJ, Wright TC. Human papillomavirus infection in human immunodeficiency virus- seropositive women. Obstet Gynecol 1995:85:680-6. Laga M, Icenogle JP, Marsella R, Manoka AT,N Nzila, RW Ryder. Genital papillomavirus infection and cervical dysplasia-opportunistic complica- tions of HIV infection. Int J Cancer 1992:50:45-8. Johnson J, Burnett A, Willet G, Young MA, Doniger J. High frequency of latent and clinical human papillomavirus cervical infections in immuno- compromised human immunodeficiency virus-infected women. Obstet Gy- necol 1992:9:321-7. Schafer A, Friedmann W, Mielke M, Schwartlander B, Koch MA. The increased frequency of cervical dysplasia-neoplasia in women infected with the human immunodeficiency virus is related to the degree of immunosup- pression. Am J Obstet Gynecol 1991;164:593-9. Wright TC, Ellerbrock TV, Chiasson MA, Van Devanter N, Sun XW, and the New York Cervical Disease Study. Cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus: prevalence, risk factors and validity of Papanicolaou smears. Obstet Gynecol 1994:84: 591-7. Del Priore G, Gilmore PR, Maag T, Warshal DP, Cheon TH. Colposcopic biopsies versus loop electrosurgical excision procedure cone histology in human immunodeficiency virus-positive women. J Reprod Med 1996:41: 653-7. Maiman M, Fruchter RG, Serur E, Levine PL, Arrastia C. Sedlis A. Re- current cervical intraepithelial neoplasia in human immunodeficiency vi- rus-seropositive women. Obstet Gynecol 1993;82:170-4. Wright T, Koulos J, Schnoll F, Swanback J, Ellerbrock TV, Chiasson MA, et al. Cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus: outcome after loop electrosurgical excision. Gy- necol Oncol 1994;55:253-8. Fruchter RG, Maiman M, Sedlis A, Bartley L, Camilien L, Arrastia CD. Multiple recurrences of cervical intraepithelial neoplasia in women with human immunodeficiency virus. Obstet Gynecol 1996;87:338—44. Cuthill S, Maiman M, Fruchter RG, Lopatinsky I, Cheung CC. Complica- tions after treatment of cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus. Reprod Med 1995:40: 823-8. Krebs HB. Prophylactic topical 5-fluorouracil following treatment of hu- man papillomavirus-associated lesions of the vulva and vagina. Obstet Gynecol 1986:68:837. 49 Human Herpesvirus Type 8 and Kaposi's Sarcoma Robin A. Weiss, Denise Whitby, Simon Talbot, Paul Kellam, Chris Boshoff* Kaposi’s sarcoma-associated herpesvirus or human herpes- virus type 8 (HHV-8) is present in all forms of Kaposi’s sarcoma (KS) as well as in primary effusion lymphomas and some cases of Castleman’s disease. In KS tissues, HHV-8 is present in endothelial and spindle cells. Current serologic tests suggest that HHV-8 is predominantly found in those at risk of KS and is not as widespread as most other human herpesviruses. HHV-8 encodes various proteins that may play a role in promotion of cellular growth, including cyclin- and G-coupled protein receptor homologues, and anti- apoptotic proteins, including Bel-2, IL-6 (i.e., interleukin 6), and FLIP (i.e., FLICE inhibitory protein) homologues. In addition, HHV-8 encodes two macrophage inflammatory- like proteins with anti-human immunodeficiency virus and angiogenic potential. [Monogr Natl Cancer Inst 1998;23: 51-54] This conference is a timely opportunity to focus on the can- cers that occur more frequently in acquired immunodeficiency syndrome (AIDS). Kaposi's sarcoma (KS) with Pneumocystis carinii infection heralded the AIDS epidemic in its earliest days. KS is seen frequently in Africa and is linked both to human immunodeficiency virus (HIV) (the epidemic form) and HIV- negative KS (the endemic form). It is also endemic in northern Mediterranean and eastern European countries. We shall focus on human herpesvirus type 8 (HHV-8)/Kaposi’s sarcoma- associated herpesvirus (KSHV), but many more details are pre- sented later in the conference when molecular biologists analyze the viral genome and describe how it replicates and functions. The study of HHV-8/KSHV started less than 3 years ago when Yuan Chang and colleagues announced its discovery in December 1994 (7). Since then, about 200 publications on this new herpesvirus have appeared, which indicates the growing interest in it. Without this discovery through the representational difference analysis, this conference might not have been con- vened. Chang et al. (/) called the new virus ‘*Kaposi’s sarcoma- associated herpesvirus (KSHV)™ because they discovered it in KS tissue of patients with AIDS. Schulz and Weiss (2), in a commentary on their discovery, suggested that *“HHV-8"" was a slightly more neutral name regarding etiology. The name does not really matter and, throughout the conference, some partici- pants use “‘KSHV’’ and others say ‘*“HHV-8."" It has been ar- gued that, because this virus is associated not only with KS but also with other lesions, such as lymphoma, we should not call it “KSHV.” But on that account we should not call HTLV-I “human T-cell leukemia virus’ anymore because it also causes the neurologic disease tropical spastic paraparesis; we should Journal of the National Cancer Institute Monographs No. 23, 1998 Table 1. Detection of human herpesvirus type 8 in Kaposi's sarcoma (KS) tissue by polymerase chain reaction® Positive Tested / Acquired immunodeficiency 108 113 96 syndrome-associated KS Classic KS 46 49 94 Posttransplant KS 9 9 100 Human immunodeficiency 5 5 100 virus-negative gay men African endemic 41 47 87 *Data compiled from references (3,29-32). rename ‘ ‘hepatitis G virus’ because it does not appear to cause hepatitis. After Chang et al. (7) first identified HHV-8 in lesions of KS of AIDS patients, we examined the presence of HHV-8 among the different KS epidemiologic groups. Data on the polymerase chain reaction (PCR) detection of HHV-8 in paraffin-imbedded KS tissue are shown in Table 1. The great majority of KS samples were positive for the presence of this virus, whatever the epidemiologic group (3,4). If we exclude the ones that had poor-quality DNA, as judged by weak amplification of a cellular gene, then virtually 100% of KS samples were HHV-8 positive. Several other groups have accrued similar results. Thus, there is genuine consensus that this virus is universally present in KS lesions. When we looked for HHV-8 by in situ techniques, we ob- served that a majority of spindle cells showed the presence of HHV-8, although negative nuclei were also present (5). Again, several other groups have confirmed this result. For example, the presence of HHV-8 genes, antigens, and indeed viruses particles has been demonstrated by Orenstein et al. (6). A continuing question is whether HHV-8 plays a causal role or whether it is a passenger virus. Another question is whether HHV-8, like Epstein-Barr virus, is ubiquitous in the human population or is restricted to KS groups? We favor a causal role, and a consensus is emerging that HHV-8 contributes to the pathogenesis of KS. O *Affiliation of authors: Institute of Cancer Research, Chester Beatty Labora- tories, London, U.K. Correspondence to: Robin A. Weiss, Ph.D., Institute of Cancer Research, Chester Beatty Laboratories, 237 Fulham Rd., London SW3 6JB, U.K. See ““Notes™” following “‘References.”’ © Oxford University Press 51 Fig. 1. Nuclear stippling seen in human herpesvirus type 8 latently infected cell line (BCP-1) (72) with sera of patients with Kaposi's sarcoma. Using PCR techniques, some groups reported that the virus is more widespread and detected KSHV in non-KS skin tumors in immunosuppressed patients and in the semen of healthy donors. We could not detect HHV-8 in squamous cell carcinomas of the skin in immunosuppressed transplant recipients (7) or in the semen of healthy donors (8). However, one could argue that this virus is ubiquitous and that it might become activated by angiogenesis but that it is normally latent at very low load or not expressed at all until it is activated by angiogenic cytokines or chemokines. We could not detect HHV-8 in benign hemangiosarcomas or granulomas; as a result, among angiogenic lesions (3), its expression seems to be specific to KS, although the cytokine profiles in these lesions might differ. Nevertheless, HHV-8 is associated with other lymphoprolif- erative lesions. Multicentric Castleman’s disease was first re- ported by Soulier et al. (9) in France to contain HHV-8, with 14 of 14 HIV-positive patients with Castleman's disease and a rather smaller proportion of HIV-negative patients being HHV-8 positive; we have obtained similar results, as have Dupin et al. (10), among others. Multicentric Castleman’s disease is a lym- phoproliferative lesion that frequently includes angiogenic pro- liferation and sometimes presents together with KS itself. Better known are the primary effusion lymphomas (or body cavity-based lymphomas) first described as containing HHV-8 by Cesarman et al. (//). These lymphomas tend to grow as effusions in the pleural or pericardial cavity. They lack many of the cell surface adhesion molecules that one might expect to be Fig. 2. A) Incidence of Kaposi's sarcoma (KS) in various risk groups [data from Beral et al. (/7)]. B) Presence of human herpesvirus type 8 (HHV-8) by immunofluorescence assay (IFA) in different risk groups [data from Simpson et al. (/4)]. Numbers = number of patients’ sera tested. C) Presence of HHV-8 by lytic IFA [data obtained from Len- nette et al. (/3)]. B KS Greek 18 HIV+ gay men HIV- gay men 65 HIV+ IV drug users — 38 HIV+ hemophilia <4 26 Blood donors UK: HIV- Uganda HIV+ Ugana-} 1 0 25 50 75 100 USA gay men 1 USA heterosexual USA transfusion recipient USA hemophilia: % 0 10 20 30 40 50 60 % HIV+ gay men = HIV+ IV drug users HIV+ women — Haemophilia Blood donors USA — Uganda Dn ro Journal of the National Cancer Institute Monographs No. 23, 1998 present on lymphomas (7/2). While many primary effusion lymphomas are positive for EBV, practi- ne 120 130 140 kb cally all of them are positive for HHV-8. Primary effusion lymphoma cell lines have provided the basis for many studies presented at this conference on the induction, replication, and cloning of HHV-8 genes, as well as for HHV-8 serology. With the use of primary effusion lymphoma cell lines in culture with latent infection as an assay for immunofluorescence antibody detection, sera from patients with KS show a punctate nuclear staining (Fig. 1). This assay has been the basis of several studies for first-generation serologic surveys of HIV infection (73-16). 65 66 67 68 69 Lief go 4 Terminal repeat fo! K14 ) = v-cyclin aaNgle yg (a) These cell lines also can be induced by phorbol esters or sodium butyrate to enter a lytic viral rep- lication in which many more of the HHV-8 pro- teins are expressed, and hence more antigens are there to be recognized. This induction of lytic viral protein expression should produce greater sensitiv- ity in detection of antibodies, but the question is whether it reduces specificity. Recombinant antigen assays are not yet as sensitive as immunoflurescence assays (14). As discussed at the beginning of this conference, the inci- dence of KS is different in different parts of the world and in different categories of patients with AIDS (77). Our serology for HHYV-8 antibodies fits the same pattern (Fig. 2). Among all the different categories of AIDS patients from the United States and Northern Europe, we can see that HHV-8 infection occurs most frequently in gay men and least often in patients with hemo- philia. Heterosexual acquisition of HIV in Africa is also asso- ciated with KS. We showed in 1995 that HHV-8 genomes are detectable in the blood of a proportion of KS-negative, HIV-positive gay men and that their presence is strongly predictive of later development of KS (18). If we consider serology and genome detection together, there appears to be a strong link between HHV-8 infection and later development of KS. Geographically, too, the detection of HHV-8 antibodies is associated with high-incidence areas for KS (e.g., Greece, Italy, and Uganda) (7/9). In a collaboration with Freddy Sits in Johannesburg, South Africa, we have shown that mother-to-child transmission is an important route for acquiring HHV-8 in Africa (20). The antigen specifying the nuclear speckle staining used in immunofluorescence studies has recently been identified as the HHV-8 orf 73 product (21,22). This gene has a complex pattern of transcription, differing in latent and lytic infections. In latent infection, a large transcript including v-cyclin (orf 72) and K13 FLIP (i.e., FLICE inhibitory protein) (orf 71) is evident (Fig. 3) (22). With regard to v-cyclin, we together with Mittnacht, Chang, and Moore (23,24), have shown that the v-cyclin protein can phosphorylate retinoblastoma protein. This phosphorylation is mediated mainly through cdk6 (24,25). HHV-8 v-cyclin can also block the function of the Rb protein as a tumor suppressor in a transfection assay in osteosarcoma cells that lack Rb (23). This is an example of one HHV-8 gene that could be potentially oncogenic. Others include the genes for a G-coupled protein receptor, an interferon response element (26,27), and apoptosis Journal of the National Cancer Institute Monographs No. 23, 1998 Fig. 3. Human herpesvirus type 8 LNA-1 (latent nuclear antigen) is encoded by orf 73 and forms part of a larger messenger RNA (mRNA) transcript including orf 72 (cyclin) and K13 (v-FLIP). During latent infection, orf 73 (LNA-1) is transcribed with the viral cyclin and anti-apoptotic protein K13 (a). During lytic induction, orf 73 is spliced out and K13 and cyclin (b) are tran- scribed as a bicistronic mRNA [data from Kellam et al. (22)]. inhibitors such as the v-FLIP (K 13), v-bcl-2, and v-1L-6. HHV-8 also encodes for chemokine homologues (v-MIP-I and v-MIP- IT), which we have shown to have HIV-inhibitory activity (26) and angiogenic potential in an in vivo assay (28). Several of these genes are discussed in more detail in other papers delivered at this conference. A caveat is that the presence of transforming genes in a human virus does not necessarily mean that the virus is oncogenic in humans. The adenovirus genes EIA and EIB are classical oncogenes, but there is no evidence that adenoviruses cause tumors in humans. The same is true for BK papovavirus. Thus, we have to be cautious about interpreting our findings with genes when we clone them and express them in NIH3T3 cells or other experimental systems. What makes HHV-8 look increasingly guilty of causing KS and lymphoma is the combination of epidemiologic and experimen- tal data: The prevalence of the virus is generally high in those human populations with a high incidence of KS, and the virus carries the tools to stimulate cell proliferation and to induce neovascularization. References (1) Chang Y, Cesarman E, Pessin MS. et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 1994:266: 1865-9. (2) Schulz TF, Weiss RA. Kaposi's sarcoma. A finger on the culprit. 1995:373:17-8. (3) Boshoff C, Whitby D, Hatziioannou T, et al. Kaposi’s-sarcoma-associated herpesvirus in HIV-negative Kaposi's sarcoma. Lancet 1995:345:1043—4. (4) Chang Y, Ziegler J, Wabinga H, et al. Kaposi's sarcoma-associated her- pesvirus and Kaposi's sarcoma in Africa. Arch Intern Med 1996:156: 202-4. (5) Boshoff C, Schulz TF, Kennedy MM, et al. Kaposi's sarcoma-associated herpesvirus infects endothelial and spindle cells. Nat Med 1995;1:1274-8. (6) Orenstein JM, Alkan S, Blauve HA, et al. Visualization of human herpes- virus-8 in Kaposi's sarcoma by light and transmission electron microscopy. AIDS 1997:11:735-45. (7) Boshoff C, Talbot S, Kennedy M, et al. HHV8 and skin cancers in immu- nosuppressed patients. Lancet 1996;347:338-9. Howard MR, Whitby D, Bahadur G, et al. Detection of human herpesvirus 8 DNA in semen from HIV-infected individuals but not healthy semen donors. AIDS 1997;11:F15-F19. (9) Soulier J, Grollet L, Oksenhendler E, et al. Kaposi's sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood 1995;86:1276-80. Nature 53 (10) (11) (12 < (13) (14) (16) (17) (18) (19) (20) (21) Dupin N, Gorin I, Deleuze J, et al. Herpes-like DNA sequences, AIDS related tumors, and Castleman’s disease. N Engl J Med 1995:333:798-9. Cesarman E, Chang Y, Moore PS, et al. Kaposi's sarcoma-associated her- pesvirus-like DNA sequences in AIDS-related body-cavity-based lympho- mas. N Engl J Med 1995:332:1186-91. Boshoff C, Gao SJ, Healy LE, et al. Establishing a KSHV positive cell line (BCP-1) from peripheral blood and characterizing its growth in Nod/SCID mice. Blood. In press. Lennette ET, Blackbourn DJ, Levy JA. Antibodies to human herpesvirus type 8 in the general population and in Kaposi's sarcoma patients. Lancet 1996:348:858-61. Simpson GR, Schulz TF, Whitby D. et al. Prevalence of Kaposi's sarcoma associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet 1996:348: 1133-8. Kedes DH, Operskalski E, Busch M, et al. The seroepidemiology of human herpesvirus 8 (Kaposi's sarcoma-associated herpesvirus): distribution of infection in KS risk groups and evidence for sexual transmission. Nat Med 1996:2:918-24. Gao SJ, Kingsley L, Li M, et al. KSHV antibodies among Americans, Italians and Ugandans with and without Kaposi's sarcoma. Nat Med 1996; 2:925-8. Beral V, Peterman TA, Berkelman RL, et al. Kaposi's sarcoma among persons with AIDS: a sexually transmitted infection? Lancet 1990:335: 123-8. Whitby D, Howard MR, Tenant Flowers M, et al. Detection of Kaposi sarcoma associated herpesvirus in peripheral blood of HIV-infected indi- viduals and progression to Kaposi's sarcoma. Lancet 1995;346:799-802. Whitby D, Luppi M, Barozzi P, et al. Human herpesvirus 8 seroprevalence in blood donors and lymphoma patients from different regions of Italy. J Natl Cancer Inst 1998:90:395-397. Bourboulia D, Whitby D, Boshoff C, et al. Serological evidence for vertical transmission of KSHV/HHV-8 in healthy South African children. Manu- script submitted for publication. Rainbow L, Platt GM, Simpson GR, et al. The 222-234 kDa nuclear protein (LNA) of Kaposi's sarcoma-associated herpesvirus (human herpes- virus 8) is encoded by orf 73 and a component of the latency-associated nuclear antigen. J Virol 1997:71:5915-21. (22) Kellam P, Boshoff C, Whitby D, et al. Identification of a major latent (24) (2: (26) (27 (28) (29) (30) (3 3 9 © ~ nuclear antigen (LNA-1) in the human herpesvirus 8 (HHV-8) genome. J Hum Virol 1997:1:19-29. Chang Y, Moore PS. Talbot SJ, et al. Cyclin encoded by KS herpesvirus. Nature 1996:382:410. Godden-Kent D, Talbot SJ, Boshoft C. et al. The cyclin encoded by Ka- posi’s sarcoma-associated herpesvirus (KSHV) stimulates ¢dk6 to phos- phorylate the retinoblastoma protein and histone HI. J Virol 1997:71: 4193-8. Li M, Lee H, Yoon DW, et al. Kaposi's sarcoma-associated herpesvirus encodes a functional cyclin. J Virol 1997;71:1984-91. Moore PS, Boshoft C, Weiss RA, et al. Molecular mimicry of human cytokine and cytokine response pathway genes by KSHV. Science 1996; 274:1739-44. Gao SJ, Boshoff C, Jayachandra S, et al. KSHV ORF K9 (vIRF) is an oncogene which inhibits the interferon signaling pathway. Oncogene 1997; 15:1979-85. Boshoff C, Endo Y, Collins PD, et al. Angiogenic and HIV inhibitory functions of KSHV-encoded chemokines. Science 1997:278:290-4. Chang Y, Ziegler J, Wabinga H, et al. Kaposi's sarcoma-associated her- pesvirus and Kaposi's sarcoma in Africa. Uganda Kaposi's Sarcoma Study Group. Arch Intern Med 1996;156:202-4. Schalling M, Ekman M, Kaaya BE, et al. A role for a new herpes virus (KSHV) in different forms of Kaposi's sarcoma. Nat Med 1995:1:705-6. 1) Dupin N, Grandadam M, Calvez V. Herpes-like DNA sequences in patients with Mediterranean Kaposi's sarcoma. Lancet 1995:345:761-2. Moore PS, Chang Y. Detection of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma lesions from persons with and without HIV infection. N Engl J Med 1995:332:1181-5. = 5 = Notes in Supported by the Cancer Research Campaign and the Medical Research Council. The work presented here is part of a much larger consortium of collaborators epidemiology and molecular pathology as seen in the authorship of our pri- mary publications. Journal of the National Cancer Institute Monographs No. 23, 1998 Some Aspects of the Pathogenesis of HIV-1-Associated Kaposi's Sarcoma Robert C. Gallo* Kaposi’s sarcoma (KS) is a very rare tumor except after human immunodeficiency virus type 1 (HIV-1) infection when it becomes common. Most investigators assume that the role of HIV-1 is passive, i.e., via inducing immune defi- ciency, thereby enhancing cancer development, and specifi- cally, in the case of human herpesvirus 8 (KSHV) by enhanc- ing HHV-8 replication. We suggest that the role of HIV-1 is more active in the disease process by at least two events: 1) promoting an increase in inflammatory cytokines, which through sustained release influences early stage KS by incit- ing local microinflammatory responses, and 2) by the Tat protein that effects growth of the inflammatory cells. Cultures of all activated endothelial spindle cells, whether hyperplastic or neoplastic, are negative for HHV-8; trans- mission of HHV-8 does not induce cell growth or transfor- mation; monkeys immune suppressed by simian immunode- ficiency virus infection and infected also with HHV-8 do not develop KS; polymerase chain reaction analysis of blood cells shows HHV-8 sequences in monocytes and B cells (about 20% of normal donors in Maryland); M. Starzl showed that early KS has few cells (mostly macrophage) positive for HHV-8, increasing and present in endothelial cells only late in the disease; no increase in HHV-8 has been found in association with progressive immune deficiency; and recent studies in Gambia by others showed that HHV-8 is a very common infection, and though HHV-2 is known to be relatively common, HIV-1 is unusual and so is KS. Col- lectively, these findings lead me to conclude that there is little evidence that HHV-8 is a transforming virus as has been repeatedly suggested and that its role in KS is more likely to be indirect (like HIV-1), perhaps necessary but hardly likely to be sufficient for KS development. [Monogr Natl Cancer Inst 1998;23:55-57] Beginning in 1987, my colleagues and I have studied Kaposi's sarcoma (KS) cells in culture and after xenotransplantation to nude mice in an effort to gain insight into the nature of KS tumor cells and possibly to shed light on its pathogenesis (7,2). Our findings are consistent with the idea that the KS spindle cells (SCs) are chiefly activated vascular endothelial cells (ECs). The vast majority of successfully cultured cells were normal diploid, grew only transiently in nude mice, and induced angiogenesis in these animals by release of cytokines (7,2). We noted that in- flammatory cytokines—interleukin 1 (IL-1), IL-6, oncostatin M, and interferon gamma (IFN vy), known to be elevated in HIV- l-infected persons (3)—promoted growth of these cells (1,4). Interestingly, IFN vy can convert ECs from cuboidal to spindle shape and promotes the migration of these cells into the circu- Journal of the National Cancer Institute Monographs No. 23, 1998 lation, presumably in search of an inflammatory lesion. IFN «y also activates the expression of integrins on EC as well as sev- eral cytokines (5). The “‘culturable’ KS SCs themselves pro- duce high levels of cytokines, such as platelet-derived growth factor (PDGF), IL-6, IL-1, and, notably, basic fibroblast growth factor (bFGF) and vascular endothelial cell growth factor (VEGF), both of which are potent angiogenic molecules (6,7). We have provided data that bFGF is important to sustain growth of the hyperplastic cells (2,4), and Masood and colleagues (8) provided evidence that VEGF is important for neoplastic cells. The inflammatory cytokines also promote increased expression of adhesion molecules that facilitate greater interactions of leu- kocytes and ECs. This includes HIV-1-infected CD4 T cells and macrophages. We showed that the Tat protein of HIV-1 is ac- tively released into the extracellular fluid (9), and others showed that it is taken up by nearby cells (9-11). We showed it is also taken up by KS SCs (/2). Our results indicate that Tat, at levels released by cells, promotes the migration, invasiveness, and growth of KS SC and that these effects are mediated by the RGD motif of Tat and by its basic region (77,12); the RGD region, through its molecular mimicry of fibronectin and vitronectin, interacts with integrins, whereas the basic region, by competing with bFGF for binding to heparin sulfate proteoglycans, allows more free soluble bFGF available for growth promotion (13). ECs require two signals for growth: adhesion and growth pro- motion. Both are augmented by HIV-1 infection. Whether KS is hyperplastic or neoplastic remains debatable. There are results and arguments for both. Reports indicate that many KS patients have tumors which are clonal (/4), and we and others have succeeded in isolating neoplastic clones from a few patients (KS Y-1 and KS SLK, respectively) (15,16). Of possible importance is the result showing similar chromosomal markers in both neoplastic cell lines (/7). At this time, we can conclude that a significant component of cells of a KS lesion (and perhaps all cells in some patients) are hyperplastic, whereas neoplastic cells can be found in some. One interpretation is that all KS lesions have neoplastic cells from the start, but akin to the Reed- Sternberg cells of Hodgkin's disease they recruit normal cells and remain in the minority but in KS are not morphologically distinguishable as are the Reed-Sternberg cells. Alternatively, KS may begin as a hyperplasia, and only in some patients does a neoplastic transformation occur. We may be able to obtain a definitive answer to this problem if we are able to generate a specific probe based on the karyotypic common abnormality *Correspondence to: Robert C. Gallo, M.D., Institute of Human Virology, University of Maryland at Baltimore, 725 W. Lombard St., 3rd Floor, Baltimore, MD 21201-1192. © Oxford University Press 55 found in the short arm of chromosome 3 in the two neoplastic cell lines (17). KS is rare and usually not aggressive except in association with HIV-1 infection. Most investigators suggest that the role of HIV-1 is passive, i.e., in inducing immune suppression. In some unexplained way, this is supposed to increase the incidence of cancer. However, with the exception of lymphomas, this has not been demonstrated. Others, such as R. Weiss (these meetings and report from the October 1995 Pasteur Vaccine meeting), suggested that HIV-1-induced immune suppression may favor HHV-8 replication. However, there is no evidence for increased HHV-8 with increasing immune suppression. Furthermore, it is clear that immune suppression occurs without KS development, e.g., congenital, chemotherapy-induced, radiation-therapy in- duced, or, more to the point, after HIV-2 infection or SIV in monkeys, both of which may produce immune suppression but neither promoting KS. It is of interest, in this regard, that the Tat of SIV and HIV-2 lacks the RGD domain. Finally, KS can occur in the absence of any known immune suppression; e.g., it can be the first sign of HIV-1 infection. Additionally, in non-HIV-1, KS-associated immune suppression is generally not found, e.g., African endemic KS. It is also unlikely that it is a significant feature of classical KS. As noted above, we suggest that the role of HIV-1 in KS is active in its promotion of inflammatory cytokines and by its Tat protein (3), thereby creating micro- vascular inflammatory lesions that I suggest are the pre-KS le- sions. HHV-8 (KSHV) and KS This virus must be involved in the cause of KS if it is not ubiquitous (which seems to be the case), because it is almost always found in KS lesions. Before a final acceptance of this, however, more refined epidemiologic data with HHV-8 are needed. Originally, it was said by Cesarman et al. (/8) to be very rare and virtually limited to KS, but soon after, a higher preva- lence was found (79), which seems to be steadily edging upward. A true understanding of its prevalence is probably not yet known. I believe a coordinated study by polymerase chain re- action analysis of separated blood cell populations in multiple laboratories of multiple populations will be the best answer to this problem. Currently, we can estimate it as at least 5% of most populations and likely substantially higher. Recent serology of Gambians by R. Weiss (reported at the International Meeting on Herpes 6, 7 and 8 in Pisa, Italy, June 1997) showing HHV-8 infection in at least 64% of people is disturbing because KS is unusual in Gambia. Of interest and compatible with our hypoth- esis is that, although HIV-2 is common in Gambia, HIV-1 is unusual. Chang and Moore (these meetings and elsewhere) have ar- gued that HHV-8 is a transforming virus. Mostly, this has been based on indirect suggestions: 1) its presence in KS lesions; 2) its relatedness to Epstein-Barr virus (EBV) and herpes saimiri; and 3) its genome containing homologues of cellular genes, some of which are involved in cell growth. However, EBV very rarely causes cancer and as Fleckenstein noted (these meetings), neither herpes saimiri or any Rhadinovirus (like HHV-8) have ever been shown to cause disease in their own host. Finally, 56 having some genome homology does not make HHV-8 equiva- lent to these viruses. Needless to emphasize, some homology to cellular genes says nothing about the capacity to transform. It only offers leads to a virus known to be transforming. We have been impressed by the fact that, when directly stud- ied, HHV-8 has not only not transformed cells, it has not pro- moted growth of any primary cells and most importantly not of EC. Moore (this meeting) described DNA of HHV-8 transform- ing NIH 3T3 cells. However, the limitations of this assay are well known, and the data are not yet available for scrutiny. Moreover, this artificial assay cannot be used to say a virus is transforming. The NIH 3T3 cells are already immortal, and many kinds of DNA lead to these cells promoting tumors in nude mice. Indeed, DNA from HIV-6 can also do the same as well as DNA of some adenoviruses, neither of which is implicated in human cancer. As already noted, our cultured hyperplastic cells (KS #1-14) and all known cultured neoplastic KS cell lines (KS Y-1 and KS SLK) are HHV-8 negative (20). Furthermore, monkeys infected with SIV (immune deficient as a result) and also infected with HHV-8 do not develop KS (unpublished results of Colombini in our group in collaboration with Biberfeld). These findings and the recent Gambia data (low KS, high prevalence of HHV-8) would lead me to the hypothesis that HHV-8 is not likely to cause KS as a transforming virus. If ultimately found to be ubiquitous, it could still be only a passenger virus. However, this seems unlikely. I favor the idea that it plays an enhancing and likely an essential role in the origin of KS, albeit like HIV-1 likely to be an indirect one, i.e., not simply by infecting a cell and transforming this cell into a neoplastic cell. Instead, I sug- gest that HHV-8 is recruited to microinflammatory lesions and promotes growth of such lesions (by augmentation of cytokine release?). Our preliminary results indicate that the same inflam- matory cytokines which increased after HIV-1 infection and which themselves promote microinflammatory changes also promote HHV-8 replication (unpublished observations of S. Co- lombini). The converse may also be true and needs testing; i.e., does HHV-8 stimulate the production of growth-promoting cy- tokines bFGF, VEGF, IL-6, etc., and/or promote growth by con- tributing, through its own gene products, homologs of cellular inflammatory cytokines? References (1) Salahuddin SZ, Nakamura S, Biberfeld P, Kaplan MH, Markham PD, Larsson L. Gallo RC. Angiogenic properties of Kaposi's sarcoma-derived cells after long-term culture in vitro. Science 1988;242:430-3. Ensoli B, Nakamura S, Salahuddin SZ, Biberfeld P. Larsson L, Beavor B, et al. AIDS-Kaposi's sarcoma-derived cells express cytokines with auto- crine and paracrine growth effects. Science 1989:243:223-6. Samaniego F, Gallo RC. Immunopathogenesis of Kaposi's sarcoma. In: Gupta E, editor. Immunopathogenesis of HIV infection. New York: Plenum Press, 1996:437-50. Samaniego F, Markham PD, Gallo RC, Ensoli B. Inflammatory cytokines induce AIDS-KS-derived spindle cells to produce and release bFGF and to enhance KS-like lesion formation in nude mice. J Immunol 1995:154: 3582-92. Seigneur M, Constans J, Blann A, Renard M, Pellegrin JL, Amiral J, et al. Soluble adhesion molecules in endothelial cell damage in HIV infected patients. Thromb Haemost 1997:77:646-9. Samaniego F, Markham PD, Gendelman R, Gallo RC, Ensoli B. Vascular endothelial growth factor and basic fibroblast growth factor are expressed in both AIDS-associated and classical Kaposi's sarcoma (KS) and syner- (2 (3 4 . ~~ (6 Journal of the National Cancer Institute Monographs No. 23, 1998 (7 (8 = (9) (10) (11) —~ ~ 1°] ~~ (13) gize in inducing angiogenesis, vascular permeability and KS lesion devel- opment. Induction by inflammatory cytokines. AIDS Res Hum Retrovirus- es 1995;11:S95. Samaniego F, Markham PD, Gendelman R, Gallo RC, Ensoli B. Inflam- matory cytokines induce endothelial cells to produce and release basic fibroblast growth factor and to promote Kaposi's sarcoma-like lesions in nude mice. J Immunol 1997;158:1887-94. Masood R, Cai J, Zheng T, Smith DL, Naidu Y, Gill PS. Vascular endo- thelial growth factor/vascular permeability factor is an autocrine growth factor for AIDS-Kaposi sarcoma. Proc Natl Acad Sci US A 1997:94:979-84. Ensoli B, Buonaguro L, Barillari G, Fiorelli V, Gendelman R, Morgan RA, et al. Release, uptake and effects of extracellular human immunodeficiency virus type I Tat protein on cell growth and viral transactivation. J Virol 1993;67:277-87. Frankel AD, Pabo CO. Cellular uptake of the Tat protein from human immunodeficiency virus. Cell 1988;55:1189-93. Westendrop MO, Frank R, Ochsenbauer C, Stricker K, Dhein J, Walezak H. et al. Sensitization of T cells to CD95-mediated apoptosis by HIV-1 Tat and gp120. Nature 1995:375:497-500. Ensoli B, Gendelman R, Markham P, Fiorelli V, Colombini S, Raffeld M, et al. Synergy between basic fibroblast growth factor and HIV-1 Tat protein in induction of Kaposi's sarcoma. Nature 1994:371:674-80. Chang H, Samaniego F, Buonaguro L, Bair BC, Gallo RC, Ensoli B. Extracellular Tat binds heparin-sulfate complexes of the cell surface Journal of the National Cancer Institute Monographs No. 23, 1998 (14) (15) (16) (17) (18) (19) (20) and extracellular matrix (ECM) through the basic region. AIDS. In press. Rabkin CS, Janz S. Lash A, Coleman AE, Musuba E, Liotta L, et al. Monoclonal origin of multicentric Kaposi's sarcoma lesions. N Engl J Med 1997:336:988-93. Lunardi-Iskandar Y, Bryant J, Zeman R, Samaniego F, Larn VH, Thierry A, et al. A pregnancy hormone human chorionic gonadotropin blocks tu- morigenesis and metastasis of Kaposi's sarcoma cells (KS Y-1) in immu- nodeficient mice. Nature 1995;375:64-8. Herndier BG, Werner A, Arnstein P, Abbey NW, Demartis F, Cohen RL, et al. Characterization of a human Kaposi's sarcoma cell line that induces angiogenic tumors in animals. AIDS 1994:8:575-81. Popescu NC, Zimonjie DB, Leventon-Kriss S, Bryant JL, Lunardi-Iskander Y. et al. Deletion and translocation involving chromosome 3 (p14) in two tumorigenic Kaposi's sarcoma cell lines. J Natl Cancer Inst 1996:88: 450-5. Cesarman E, Chang Y, Moore PS, Said JW. Knowles DM. Kaposi's sar- coma-associated herpesvirus-like DNA sequences in AIDS-related body- cavity-based lymphomas. N Engl J Med 1995:332:1186-91. Monini P, deLellis L, Fabris M, Rigolin F, Cassai E. Kaposi's sarcoma- associated herpesvirus DNA sequences in prostate tissue and human semen. N Engl J Med 1996:334:1168-72. Flamand L, Zeman RA, Bryant JL, Lunardi-Iskandar Y, Gallo RC. Ab- sence of human herpesvirus 8 DNA sequences in neoplastic Kaposi's sar- coma lines. J AIDS 1996;13:194-7. 57 Clinical Overview: Issues in Kaposi's Sarcoma Therapeutics Susan E. Krown* Four questions are posed that are critical to the development of improved therapeutic and prophylactic strategies for Ka- posi’s sarcoma (KS). 1) Can we predict who will develop KS? Accurate identification of high-risk factors for KS develop- ment is essential for the development of KS prophylaxis tri- als. 2) Can developing insights into KS pathogenesis be translated into improved therapeutic and/or new prophylac- tic strategies for patients at high risk? Several approaches are being developed that target new blood vessel develop- ment, inflammatory cytokines, and the viruses that are im- plicated in KS pathogenesis. 3) How does the improved prog- nosis for human immunodeficiency virus (HIV)-infected patients affect KS treatment strategy? Improved anti-HIV therapy has implications for the timing of KS therapy, the choice of therapeutic approaches, and the potential for ad- verse drug interactions. 4) How can we best evaluate benefits from KS treatment? More rigorous, standardized criteria are in development and will be essential not only for accu- rate documentation of objective tumor regression, but also for assessment of tumor-associated symptom relief in a quantitative, function-oriented way. [Monogr Natl Cancer Inst 1998;23:59-63] Tremendous progress has been made recently in identifying the factors that contribute to the development of Kaposi's sar- coma (KS), the most common cancer associated with human immunodeficiency virus type 1 (HIV-1) infections, but treatment of KS remains suboptimal. In reviewing the issues in KS man- agement, the following four critical questions must be addressed as attempts are made to improve therapy and develop KS pro- phylaxis strategies over the next several years: 1) Can we predict who will develop KS? 2) Can developing insights into KS pathogenesis be translated into improved therapy and/or new prophylactic strategies for patients at high risk? 3) How do the improved treatments and prognosis for HIV- infected patients affect our overall KS treatment strategy? 4) How can we best evaluate benefits from KS treatment? These issues will be briefly discussed in the succeeding sec- tions of this article. Prediction of KS Development The ability to predict who will develop KS is critical for the development and interpretation of prophylaxis trials. The poten- tial to intervene prior to the development of KS is an exciting new possibility, but some prophylactic strategies may involve the administration of drugs with toxic potential, so accurate Journal of the National Cancer Institute Monographs No. 23, 1998 identification of high-risk individuals is essential. Although we have known for a long time that, in developed countries, HIV- infected gay and bisexual men are most likely to develop KS, only a minority do so. In developed countries, however, it is often not appreciated that HIV-infected individuals from other transmission groups also show a marked increase in KS risk that has recently been estimated in the United States as 10000 times greater than that in the age- and sex-matched general population (1). Therefore, sexual orientation and behavior alone are not sufficient to identify who will develop KS in the United States, and they are not relevant factors in other parts of the world, like Africa, where KS incidence is relatively high in both sexes and is not associated with male homosexuality. Although human herpesvirus 8 (HHV-8; Kaposi's sarcoma- associated herpesvirus) has not been established as the cause of KS (i.e., no published data show that the virus can transform and immortalize cells), the presence of HHV-8 infection has now been unquestionably linked to KS. There is, however, increasing evidence in patients with established KS that no single serologic assay detects evidence for HHV-8 infection in all patients (2,3), whereas certain methods detect antibodies in a substantial mi- nority of unaffected, low-risk individuals (2,3). Clearly, there- fore, the sensitivity and specificity of these tests need to be improved, and the relevant serologic responses must be identi- fied. Studies are also needed to confirm whether antibody titer correlates with subsequent KS risk (4) and to evaluate whether detection or burden of HHV-8 DNA in cells adds predictive value to the serologic assays that are better suited to large-scale screening. If we accept that not all individuals with evidence of HHV-8 infection are at equal risk for KS development, we also need to ask whether other factors interact with HHV-8 either to promote KS development or to protect against it. Those who treat KS have often observed sudden, rapid tumor progression around the time an opportunistic infection develops and improvement when the infection was treated, lending credence to the idea that ex- cess production of inflammatory cytokines as well as increased HIV replication itself may stimulate KS growth. Also observed, on less frequent occasions, has been regression of KS lesions when active antiretroviral therapy was instituted (5). These ob- servations in patients with established KS are the clinical coun- terpart to numerous laboratory observations predating the dis- *Affiliation of author: Clinical Immunology Service, Division of Hematologic Oncology, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY. Correspondence to: Susan E. Krown, M.D., Memorial Sloan-Kettering Can- cer Center, 1275 York Ave., New York, NY 10021. © Oxford University Press 59 covery of HHV-8 that implicated a wide range of cytokines, angiogenic factors, and HIV products in KS pathogenesis (6-9). We now need to look more closely at the role of these factors as codeterminants, with HHV-8, of future KS development. Within the AIDS Malignancy Consortium (AMC), a recently estab- lished, National Institutes of Health-supported multicenter trials group, active discussions are under way to design these predic- tive investigations and to use them to develop prophylaxis and natural history trials. Pathogenesis-Based Therapeutic and Prophylactic Strategies for KS Several cytotoxic agents have been identified that can induce regression, albeit temporary, of established, advanced KS (/0). At the National AIDS Malignancy Conference, the activity of two recent introductions to the therapeutic armamentarium, paclitaxel and vinorelbine, was described (71,12). Paclitaxel’s ability to inhibit angiogenesis (/3) may partially explain its ap- parent high activity and the relatively long duration of response it induces in patients with advanced, refractory disease. Al- though the introduction of these agents increases the options for treatment of aggressive, widespread KS, there is still a need for therapy that induces higher and longer-lasting response rates. There is also a perceived need on the part of many investigators to develop effective therapy that makes better “biologic sense.” This sort of targeted approach, based on the rapidly growing but still incomplete knowledge of KS pathogenesis, has tremendous intellectual appeal and has inspired a variety of clinical trials. With some exceptions, the results thus far have been mostly quite limited. However, for a number of these approaches di- rected against the formation of new blood vessels, inflammatory cytokines, and the viruses implicated either indirectly or directly in KS pathogenesis, there are some glimmers of activity that merit a closer look. Antiviral Strategies Both HIV-1 and HHV-8 are potential targets for anti-KS therapy. Although HIV-1 infection is not a requirement for de- velopment or progression of KS, active infection with HIV-1 is associated with high levels of inflammatory cytokines impli- cated in the development of KS lesions (74,15). In addition, the HIV-1 Tat protein, which is released into the extracellular milieu by productively infected cells, also serves as a mitogen for KS- derived spindle cells in vitro and acts synergistically with growth factors and inflammatory cytokines to increase KS cell proliferation (8,9). It should not be surprising, therefore, that effective control of HIV-1 replication achieved through the use of highly active antiretroviral drug regimens has now been re- ported to induce KS regression in some patients (5) and to ob- viate the need for long-term maintenance chemotherapy in oth- ers (16). It is also possible that the improved KS response rates seen when nucleoside reverse transcriptase inhibitors are added to interferon alfa (/7,18) may be partially attributable to the synergistic antiretroviral activity of these agents. Anecdotal reports that established KS regressed after treatment with fos- carnet (1/9) and several studies that showed a decreased inci- dence of subsequent KS in patients treated with antiherpesvirus agents (20,21) suggest that, under some circumstances, inhibi- 60 tion of HHV-8 may also be of therapeutic or prophylactic value. However, none of the agents tested is specific for HHV-8 (the treatments were directed against cytomegalovirus and the stud- ies were not designed to evaluate effects on HHV-8), so the mechanism of KS inhibition in these cases is entirely specula- tive. It is also possible that part of interferon’s anti-KS activity may involve direct HHV-8 inhibition. Although this possibility has not been specifically tested, interferon has been shown to inhibit a closely related virus, Herpesvirus saimiri (22). Even if KS is the product of viral transformation, a consistent therapeu- tic effect of antiherpesvirus agents against established KS is probably unlikely, since the growth of transformed cells is often independent of the virus. It may be more plausible to envision a prophylactic role for such agents if they are administered prior to a putative transforming event. Cytokine Inhibition Strategies A number of proinflammatory cytokines whose production is reportedly increased in some patients with HIV infection have also been considered targets for KS therapy. Tumor necrosis factor (TNF) and interleukin 1 (IL-1) have each been shown to stimulate KS cell proliferation in vitro, either directly or through induction of interleukin 6 (IL-6) (6,7), and various agents that modify their production or action have either been tested in early clinical trials or proposed as candidate therapeutic agents. Of particular note are the retinoids, which are multifunctional agents among whose many attributes are down-regulation of both IL-6 receptors and IL-6 production in myeloma cells (23), and thalidomide, a specific inhibitor of TNF (24). One retinoid, 9-cis-retinoic acid, which interacts with both the retinoic acid receptor and the retinoid X receptor, has shown anti-KS activity when applied locally to KS lesions in gel form (25). The AMC is now testing this as a systemic, orally administered drug, and there are some early indications of anti-KS activity. In addition, Bernstein et al. (26) have recently presented preliminary evi- dence of anti-KS activity for an intravenously administered li- posomal preparation of all-frans-retinoic acid. Preliminary clini- cal evidence from two studies (27,28) suggests that orally administered thalidomide may also have substantial anti-KS ac- tivity. Interferon alfa, which has demonstrated activity against KS in some patients (29), has been shown in clinical trials in normal volunteers and patients with chronic hepatitis to induce the production of the soluble TNF receptor and the IL-1 receptor antagonist (30,31), which are endogenous down-regulators of these KS stimulatory cytokines. Other potential candidate agents that target cytokines, but which have not yet been tested, include the soluble TNF receptor, the IL-1 receptor antagonist, and the matrix metalloproteinase inhibitor marimistat, which also pos- sesses inhibitory activity against TNF convertase, an enzyme that converts TNF to its active form. Angiogenesis Inhibitory Strategies A large number of strategies aimed at the inhibition of angio- genesis have been proposed. Some of these approaches have not yet been tested clinically, but candidate agents are already in development. These include, for example, inhibitors of matrix metalloproteinases, enzymes that facilitate capillary budding and invasion (32), agents targeted at distinctive endothelial cell sur- face molecules present on proliferating tumor-associated vascu- Journal of the National Cancer Institute Monographs No. 23, 1998 lature (33,34), and agents that interfere with angiogenic poly- peptide signal transduction (35). Agents utilizing other approaches have been tested with mixed results. For example, the fumagillin analogue TNP-470 has been studied in two phase I trials in patients with KS, yield- ing a few partial tumor responses in one study (36) and a sig- nificant reduction in tumor-associated edema in some patients (37). Several heparin-binding compounds that are thought to inhibit angiogenesis by blocking basic fibroblast growth factor (bFGF) receptor binding have also been studied in phase I trials, and while the high hopes and expectations for these agents have not been realized, evidence for biologic activity has been ob- served. For example, in some patients treated with tecogalan, a significant reduction in KS-associated edema was seen (38), perhaps signifying an important biologic effect. Interferon alfa and thalidomide are also capable of acting as angiogenesis in- hibitors: Interferon inhibits bFGF gene expression and produc- tion in certain human tumor cell lines (39) and also inhibits endothelial cell motility, whereas thalidomide (in addition to its inhibitory effects on TNF production) may inhibit angiogenesis by multiple mechanisms, including inhibition of bFGF-induced vascular proliferation, intercellular adhesion, basement mem- brane formation, and overall vascular maturation (40). The rela- tive success of interferon alfa in controlling KS and possibly also thalidomide—although it is probably too early to comment on thalidomide’s usefulness—may be a function of the sheer mul- tiplicity of their potential targets and suggests that some of the other agents with more restricted targets may be successful in controlling KS only when used as part of combination regimens directed at multiple steps in the development of KS lesions. It is disappointing, therefore, to see that the further development of a number of these agents is not being pursued, at least partially, because of their apparent lack of substantial efficacy as single agents. However, as discussed by Judah Folkman at this confer- ence, it may be that newer agents such as angiostatin (4/) or endostatin (42) will be far more active than some of those al- ready shown to have only marginal single-agent activity. Effect of Improved HIV Therapy on KS Treatment Strategy Patients with HIV infection are surviving longer as a result of improved antiretroviral therapy. These advances in HIV man- agement are likely to benefit patients at all stages of HIV infec- tion, including those with KS. The prospect of managing KS over a period of many years, rather than over a much shorter time period, raises several new questions and challenges that will need to be addressed when devising therapeutic plans and designing new therapeutic trials: 1) When in the course of KS should treatment begin and with which approaches or agents? Controversy about this issue preceded the introduction of HIV protease inhibitors, and the apparent benefits of state-of-the-art antiretroviral therapy on KS in some patients have made definition of the optimum KS treatment strategy even more complex. 2) A corollary of the first question is: Has more effective anti- HIV therapy rendered the need to find better KS therapy obsolete? Alternatively, are we merely in a ‘‘honeymoon’’ period that will end—as most honeymoons do—with increas- Journal of the National Cancer Institute Monographs No. 23, 1998 ing numbers of patients becoming affected by KS later in the course of their infection? 3) If we assume that the honeymoon will end, we need to ask: Will better control of HIV and its nonmalignant complica- tions allow pathogenesis-based strategies to work better, and should this prompt a re-examination of some agents that were considered ineffective in the era before protease inhibitors? 4) We also need to ask: Will better control of HIV and its complications also lead to higher response rates, longer re- sponse durations, or a reduced need for maintenance therapy in patients on standard KS therapies? 5) Will the increasingly complex drug regimens of HIV- infected patients lead to untoward interactions with anti-KS therapies? Recently, for example, D. Scadden (personal com- munication, 1997) observed severe toxicity from paclitaxel when a patient who had been tolerating the drug well while on lamivudine, stavudine, and indinavir was switched to a combination of delavirdine, didanosine, and saquinavir. The metabolism of many cytotoxic agents is dependent on hepatic enzymes whose activity may be induced or inhibited by pro- tease inhibitors and non-nucleoside reverse transcriptase in- hibitors. Interferon alfa may also inhibit certain cytochrome P450 enzymes (43), potentially affecting the metabolism of certain protease inhibitors. These potential drug interactions will be studied in several upcoming trials in patients with KS. 6) Do we need to be more concerned about the long-term con- sequences of chemotherapy for KS? There has been a recent trend toward the earlier use of chemotherapeutic agents like Doxil and DaunoXome. These drugs are well tolerated by most patients and rarely induce hair loss, nausea, or vomiting (44,45), making many patients and perhaps even some phy- sicians view them as something other than chemotherapy. Now that the life span of some patients with advanced KS may be measured in years, however, we need to be more concerned about the development of secondary cancers and other late complications of cytotoxic agents. 7) Have we gone as far as we can go with standard chemothera- peutic agents? Currently, few new cytotoxic agents are con- sidered as candidates for testing in KS. The results of studies using combinations of several active drugs, such as the re- cently completed AIDS Clinical Trials Group (ACTG) com- parison trial of Doxil alone or combined with bleomycin and vincristine (46) suggest that we may have reached a point of diminishing returns, where responses are not significantly increased but quality of life is diminished. Therefore, it is more important than ever to develop targeted, pathogenesis- based therapeutic strategies for KS. Methods to Evaluate Benefits From KS Treatment In order to measure the benefits of treatment for KS, it is necessary to document not only whether tumor regression has occurred, but also whether the patient is better and has shown tangible benefits that extend beyond tumor shrinkage. This is true for any tumor but may be especially so for KS, which may not always be directly life-threatening but may adversely affect quality of life by causing pain, disfigurement, or functional dis- ability. The standardized methods of evaluation and criteria for response, developed and later refined by the Oncology Commit- 61 tee of the ACTG (47), were an improvement over the vague and varied criteria used previously. However, these evaluation and response criteria have been criticized because of difficulties in the ability to reproduce and document lesion counts and mea- surements, the lack of standardized assessments of hard-to- measure visceral disease and edema, and the failure of the cri- teria to capture the clinical benefits such as pain relief, improvement in disfiguring lesions, and functional improve- ments induced by successful KS therapy. Some of the problems with the ACTG criteria have stemmed from the failure of some investigators to consistently perform the repeated lesion counts and lesion measurements that were required to document response, but there is also general agree- ment that more rigorous standards of documentation are desir- able. Feigal et al. (48) have described the joint efforts of the National Cancer Institute, the Food and Drug Administration, and the AMC to strengthen the process by which KS therapy is evaluated. This will include documentation not only of lesion size, character, and number in the skin and visceral sites, but also of physician and patient assessment of tumor-associated symp- tom relief in a function-oriented, quantitative way. Rigorous collection of the varied types of documentation required for these assessments will be tedious and time consuming but is an essential step in making effective therapy widely available to patients. Summary and Conclusions Several challenges must be met in order to improve the therapy of AIDS-associated KS over the coming years. The op- portunity now exists to consider strategies to prevent KS devel- opment, but more sensitive and specific methods are needed to accurately identify those patients at highest risk. Increased knowledge of the biology of KS now permits the development of therapeutic strategies that target critical elements in KS patho- genesis. To some extent, these sorts of approaches have already been applied with some moderately successful examples; how- ever, opportunities are expanding as a more sophisticated un- derstanding of the complex processes involved in KS develop- ment unfolds and as an increasing number of agents are developed that influence specific pathogenetic targets. The in- troduction of highly active antiretroviral therapy now presents the opportunity to manage KS over a period of many years. Although more successful anti-HIV therapy may lead to fewer patients affected by KS or to a decrease in KS severity, it is also possible that longer survival with HIV infection may be associ- ated with an increase in KS incidence over time. These possi- bilities raise important yet unanswered questions about optimum treatment strategies, the long-term consequences of cytotoxic chemotherapy, and the potential for adverse drug interactions as treatment regimens become more complex. Rigorous standards are needed to accurately measure the benefits of KS therapy, so that safe and effective new agents can be expeditiously intro- duced into clinical practice. Methods now exist to measure changes in tumor size and number, but development of quanti- tative scales to evaluate functional status, pain, and quality of life presents a more difficult challenge in the drug development process. 62 References (1) Biggar RJ, Rabkin CS. The epidemiology of AIDS-related neoplasms. Hematol Oncol Clin North Am 1996:10:997-1010. Simpson GR, Schulz TF, Whitby D, Cook PM, Boshoff C, Rainbow L, et al. Prevalence of Kaposi's sarcoma associated herpesvirus infection mea- sured by antibodies to recombinant capsid protein and latent immunoflu- orescence antigen. Lancet 1996;348:1133-8. Gao SJ, Kingsley L, Hoover DR, Spira TJ, Rinaldo CR, Saah A, et al. Seroconversion to antibodies against Kaposi's sarcoma-associated herpes- virus-related latent nuclear antigens before the development of Kaposi's sarcoma. N Engl J Med 1996:335:233-41. Ablashi DV, Chatlynne LG, Lapps W, Hoekzema DT, Masood R, Kramar- sky B, et al. Human herpesvirus-8 (HHV-8): epidemiology and character- ization of virus from KS-1 cell line. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A23. Routy JP, Urbanek A, MacLeod J, Thomas R. Significant regression of Kaposi's sarcoma following initiation of an effective antiretroviral combi- nation treatment. J Acquir Immune Defic Syndr Hum Retrovirol 1997:14: A22. Karp JE, Pluda JM, Yarchoan R. AIDS-related Kaposi's sarcoma. A tem- plate for the translation of molecular pathogenesis into targeted therapeutic approaches. Hematol Oncol Clin North Am 1996;10:103 1-49. Miles SA. Pathogenesis of AIDS-related Kaposi's sarcoma. Evidence of a viral etiology. Hematol Oncol Clin North Am 1996;10:1011-21. Ensoli B, Barillari G, Salahuddin SZ, Gallo RC, Wong-Staal F. Tat protein of HIV-1 stimulates growth of cells derived from Kaposi's sarcoma lesions of AIDS patients. Nature 1990:345:84-6. Ensoli B, Gendelman R, Markham P, Fiorelli V, Colombini S, Raffeld M. et al. Synergy between basic fibroblast growth factor and HIV-1 Tat protein in induction of Kaposi's sarcoma. Nature 1994:371:674-80. (10) Krown SE. Acquired immunodeficiency syndrome-associated Kaposi's sarcoma. Biology and management. Med Clin North Am 1997:81:471-94. (11) Gill PS, Tulpule A, Reynolds T, Hadjenberg R, Mocharnuk R, Espina BM, et al. Paclitaxel (Taxol) in the treatment of relapsed or refractory advanced AIDS-related Kaposi's sarcoma [abstract]. Proc ASCO 1996:15:306. (12) Errante D, Spina M, Nasti G, Tavio M, Vacchar E, Landonio G, et al. Evidence of activity of vinorelbine (VNR) in patients with previously treated epidemic Kaposi's sarcoma (KS). J Acquir Immune Defic Syndr Hum Retrovirol 1997:14:A36. (13) Belotti D, Vergani V, Drudis T, Borsotti P, Pitelli MR, Viale G, et al. The microtubule-affecting drug paclitaxel has antiangiogenic activity. Clin Can- cer Res 1996;2:1843-9. (14) Ammann AJ, Palladino MA, Volberding P, Abrams D, Martin NL, Conant M. Tumor necrosis factors alpha and beta in acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. J Clin Immunol 1987.7: 481-5. (15) Breen EC, Rezai AR, Nakajima K, Beall GN, Mitsuyasu RT, Hirano T, et al. Infection with HIV is associated with elevated IL-6 levels and produc- tion. J Immunol 1990;144:480-4. (16) Volm MD, Wernz J. Patients with advanced AIDS-related Kaposi's sar- coma (EKS) no longer require systemic therapy after introduction of ef- fective antiretroviral therapy [abstract]. Proc ASCO 1997;16:46a. (17) Krown SE, Gold JW, Niedzwiecki D, Bundow D, Flomenberg N, Gans- bacher B, et al. Interferon-a with zidovudine: safety, tolerance, and clinical and virologic effects in patients with Kaposi's sarcoma associated with the acquired immunodeficiency syndrome (AIDS). Ann Intern Med 1990: 112: 812-21. (18) Fischl MA, Finkelstein DM, He W, Powderly WG, Triozzi PL, Steigbigel RT, et al. A phase II study of recombinant human interferon-a2a and zidovudine in patients with AIDS-related Kaposi's sarcoma. AIDS Clinical Trials Group. J Acquir Immune Defic Syndr Hum Retrovirol 1996:11: 379-84. (19) Morfeldt L., Torssander J. Long-term remission of Kaposi's sarcoma fol- lowing foscarnet treatment in HIV-infected patients. Scand J Infect Dis 1994:26:749-52. (20) Jones JL, Hanson DL, Chu SY, Ward JW, Jaffe HW. AIDS-associated Kaposi's sarcoma [letter]. Science 1995;267:1078-9. (21) Mocroft A, Youle M, Gazzard B, Morcinek J, Halai R, Phillips AN. Anti- herpesvirus treatment and risk of Kaposi's sarcoma in HIV infection. Royal Free/Chelsea and Westminster Hospitals Collaborative Group. AIDS 1996; 10:1101-5. (22) Barahona HH, Melendez LV. Herpesvirus saimiri: in vitro sensitivity to virus-induced interferon and to polyriboinosinic acid polyribocytidylic acid. Proc Soc Exp Biol Med 1971:136:1163-7. (23) Sidell N, Taga T, Hirano T, Kishimoto T, Saxon A. Retinoic acid-induced growth inhibition of a human myeloma cell line via down-regulation of IL-6 receptors. J Immunol 1991;146:3809-14. (2 ~ (3 ~ (4 = (5 - (6 = (7 ~ (8 = (9 Journal of the National Cancer Institute Monographs No. 23, 1998 (24) (25) (26) (27 ~— (28) (29) (30) {34) (32) (33) (34) (35) (36) (37) Moreira AL, Sampaio EP, Zmuidzinas A, Frindt P, Smith KA, Kaplan G. Thalidomide exerts its inhibitory action on tumor necrosis factor alpha by enhancing mRNA degradation. J Exp Med 1993;177:1675-80. Duvic M, Friedman-Kien AE, Miles SA, Looney DJ. Stevens V, Truglia J, et al. Phase I-1I evaluation of Panretin™ (ALRTI1057; LGDI1057; AGN192013; 9-cis-retinoic acid) topical gel for AIDS-related cutaneous Kaposi's sarcoma [abstract]. Prog/Proc ASCO 1997:16:46a. Bernstein ZP, Cohen P, Rios A, Wagner B, Fischl M, Groopman J, et al. A multicenter, phase II/I1I study of Atragen™ (tretinoin liposomal) in patients with AIDS-associated Kaposi's sarcoma. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A19. Welles L, Little R, Wyvill K, Pluda J, Figg W, Lietzau J, et al. Preliminary results of a phase II study of oral thalidomide in patients with HIV infection and Kaposi's sarcoma (KS). J Acquir Immune Defic Syndr Hum Retrovirol 1997:14:A21. Bower M, Howard M, Gracie F, Phillips R, Fife K. A phase II study of thalidomide for Kaposi's sarcoma: activity and correlation with KSHV DNA load. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A35. Krown SE. Interferon and other biologic agents for the treatment of Ka- posi’s sarcoma. Hematol Oncol Clin North Am 1991:;5:311-22. Tilg H, Mier JW, Vogel W, Aulitzky WE, Wiedermann CJ, Vannier E, et al. Induction of circulating IL-1 receptor antagonist by IFN treatment. J Immunol 1993:150:4687-92. Tilg H, Vogel W, Dinarello CA. Interferon-a induces circulating tumor necrosis factor receptor p55 in humans. Blood 1995;85:433-5. Ennis BW, Matrisian LM. Matrix degrading metalloproteinases. J Neu- rooncol 1993:18:105-9. Brooks PC, Montgomery AM, Rosenfeld M, Reisfeld RA, Hu T, Klier G, et al. Integrin o,f; antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 1994:79:1157. Burrows FJ, Thorpe PE. Vascular targeting—a new approach to the therapy of solid tumors. Pharmacol Ther 1994:64:155-74. Kohn EC, Sandeen MA, Liotta LA. In vivo efficacy of a novel inhibitor of selected transduction pathways including calcium, arachidonate, and ino- sitol phosphates. Cancer Res 1992;52:3208—12. Dezube BJ, vonRoenn JH, Holden-Wiltse J, Remick S, Cooley TP, Cheung TW, et al. Fumagillin analog (TNP-470) in the treatment of Kaposi's sar- coma: a phase 1 AIDS clinical trial group study. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A35. Pluda JM. Wyvill KK. Lietzau J. Figg WD. Whitcup SM, Feigel E, et al. A phase I trial administering the angiogenesis inhibitor TNP-470 (AGM- Journal of the National Cancer Institute Monographs No. 23, 1998 (38) (39) (40) (41) (42) (43) (44) (45) (46) (47) (48) 1470) to patients (pts) with HIV-associated Kaposi's sarcoma (KS). J Ac- quir Immune Defic Syndr Hum Retrovirol 1997;14:A19. Eckhardt SG, Burris HA, Eckardt JR, Weiss G, Rodriguez G, Rothenberg M, et al. A phase I clinical and pharmacokinetic study of the angiogenesis inhibitor, tecogalan sodium. Ann Oncol 1996;7:491-6. Singh RK, Gutman M, Bucana CD, Sanchez R, Llansa N, Fidler 1J. Inter- ferons alpha and beta down-regulate the expression of basic fibroblast growth factor in human carcinomas. Proc Natl Acad Sci U S A 1995.92: 4562-6. D'Amato RJ, Loughnan MS, Flynn E, Folkman J. Thalidomide is an in- hibitor of angiogenesis. Proc Natl Acad Sci U S A 1994:91:4082-5. O'Reilly MS, Holmgren L, Shing Y, Chen C, Rosenthal RA, Moses M, et al. Angiostatin: a novel angiogenesis inhibitor that mediates the suppres- sion of metastases by a Lewis lung carcinoma. Cell 1994;79:315-28. O'Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, et al. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 1997:88:277-85. Mannering GJ, Renton KW, El Azhary R, Deloria LB. Effect of interferon- inducing agents on hepatic cytochrome p-450 drug metabolizing systems. Ann NY Acad Sci 1980:350:314-31. Gill PS, Wernz J, Scadden DT, Cohen P, Mukwaya GM, vonRoenn JH, et al. Randomized phase III trial of liposomal daunorubicin versus doxorubi- cin, bleomycin, and vincristine in AIDS-related Kaposi's sarcoma. J Clin Oncol 1996:14:2353-64. Northfelt DW, Dezube BJ, Thommes JA, Levine R, vonRoenn JH, Dosik GM, et al. Efficacy of pegylated-liposomal doxorubicin in the treatment of AIDS-related Kaposi's sarcoma after failure of standard chemotherapy. J Clin Oncol 1997;15:653-9. Mitsuyasu R, vonRoenn J, Krown S, Kaplan L, Testa M. Comparison study of liposomal doxorubicin (Dox) alone or with bleomycin and vincristine (DBV) for treatment of advanced AIDS-associated Kaposi's sarcoma (AIDS-KS): AIDS Clinical Trial Group (ACTG) protocol 286 [abstract]. Prog/Proc ASCO 1997:16:5a. Krown SE, Metroka C, Wernz JC. Kaposi's sarcoma in the acquired im- mune deficiency syndrome: a proposal for uniform evaluation, response and staging criteria. J Clin Oncol 1989;7:1201-7. Feigal EG, vonRoenn J, Justice R, Yarchoan R, Krown S, Pluda J, et al. Kaposi's sarcoma response criteria: issues identified by the National Can- cer Institute, Food and Drug Administration, and the AIDS Malignancy Consortium. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:A22. 63 Kaposi's Sarcoma-Associated Herpesvirus-Encoded Oncogenes and Oncogenesis Patrick S. Moore, Yuan Chang* Molecular biologic studies of Kaposi’s sarcoma-associated herpesvirus (KSHV) have identified a number of potential viral oncogenes that may contribute to KSHV-related neo- plasia including a D-type cyclin, an IL-6-like cytokine, and a novel member of the interferon regulatory factor family. KSHY is functionally related to other DNA tumor viruses by encoding specific proteins to inhibit pRb, pro-apoptotic, and interferon-signaling tumor suppressor pathways. The virus appears to employ molecular piracy of cellular regulatory genes as a mechanism to avoid cellular antiviral responses. The transparency of the KSHV genome allows ready iden- tification of the cellular regulatory pathways which may be involved in transformation by KSHV. This provides strong support to the notion that some tumor suppressor pathways serve the dual function of being antiviral pathways to induce cell cycle arrest, apoptosis, and enhanced cell-mediated im- munity in response to virus infection. Neoplasia may result from specific viral strategies to overcome these host defense pathways. [Monogr Natl Cancer Inst 1998;23:65-71] The second full year of working with Kaposi’s sarcoma- associated herpesvirus (KSHV) is now finished and scientific studies on this virus are rapidly maturing. With the publication of the full-length genomic sequence (/), new opportunities to investigate its molecular biology and virology are becoming available. This review will try to provide a road map for subse- quent papers in this symposium on the major features of the KSHV genome and its potential oncogenes. Despite the early stage of KSHV molecular biologic studies, the virus is already providing unique insights into tumorigenesis and may serve as an important model for virus-induced oncogenesis. The evidence gathered to date is overwhelming in support of KSHYV as the infectious cause of Kaposi's sarcoma (KS) (2). A review of 21 published studies involving 549 patients, with all forms of Kaposi's sarcoma from various parts of the world, shows that KSHV DNA can be detected on average 95% of the time in KS lesions by polymerase chain reaction. This exceeds the positivity rate for detection of papillomavirus in cervical cancer (3). It is now clearly established using a variety of tech- niques that KSHV are not only localized to KS lesions (4-6), but that viral DNA and RNA are detectable in most KS tumor spindle cells (7-9). Newly developed KSHV serologic assays also support a cau- sal role for the virus in KS. As discussed by Weiss (1/0), pub- lished assays show similar epidemiologic trends, despite impor- tant differences in their sensitivities and specificities. The Journal of the National Cancer Institute Monographs No. 23, 1998 relative KSHV seroprevalence measured by these assays matches the patterns for KS risk groups among patients with acquired immunodeficiency syndrome (AIDS) (//-17) and the patterns for non-HIV-related KS incidence among patients from various countries (/5,16,18). For KSHV to be causally related to KS, infection must precede onset of disease. This is an absolute criteria for causality, and both seroconversion (/3,/8-20) and DNA-based (21,22) detection studies provide evidence for KSHV infection among most AIDS-KS patients prior to disease onset. Taken together, these and other studies (23-26) now indicate that KSHV is not a ubiquitous infection in most human popu- lations, although high rates of infection in some populations [e.g., Italians and Ugandans (/8)] may account for early studies suggesting widespread human KSHV infection (27,28). It is thus likely that the virus is a necessary, but not necessarily sufficient cofactor for KS development. This does not diminish the role of KSHV in the genesis of KS and related neoplasias. With the possible exceptions of rabies and HIV, no infectious syndrome or illness is solely due to infection by a single agent without contributing host or environmental risk factors. Similarly, asymptomatic infection is the norm for most infectious agents and only a minority of persons infected with KSHV should be expected to eventually develop KS. This relatively simple view of KS causation has become complicated by the recent finding of KSHV in cultured multiple myeloma stromal cells (29). Whether or not this association with multiple myeloma, a cancer with a very different epidemiologic presentation from KS, is real re- mains an area for future research. The important questions now become ‘‘How does KSHV cause KS, and what is the relative contribution of the virus to tumorigenesis vis-a-vis immunosuppression and other predis- posing factors?’ As indicated by Blackbourne et al. (30), tech- niques have not yet been developed to allow in vitro high-titered virus transmission to other cell lines (17,31). Therefore, much of the work on KSHV-related tumorigenesis and cell transforma- tion has had to rely on molecular biologic investigation of indi- vidual genes. Like other herpesviruses, KSHV is presumed to replicate as a circular episome in latently infected cells and is *Affiliations of authors: P.S. Moore (School of Public Health), Y. Chang (Department of Pathology, College of Physicians and Surgeons), Columbia Uni- versity, New York, NY. Correspondence to: Patrick S. Moore, M.D., Department of Pathology, Co- lumbia University, 630 W. 168th St., New York, NY 10032. See ‘Notes’ following ‘‘References.”’ © Oxford University Press packaged in a linear genomic form during lytic replication. The importance of this is that standard dogma tells us lytic replica- tion is uniformly fatal to the cell and that lytic phase gene expression cannot contribute to tumorigenesis. However, recent findings on lytic KSHV replication in KS lesions and the effects of viral DNA polymerase inhibitors on clinical disease suggest that we should carefully question this assumption (8,32-34). As indicated by Ganem (35), it is possible that a minority popula- tion of lytically infected cells is important in sustaining the KS lesion by constantly recruiting new, infected tumor cells directly or through paracrine mechanisms. The KSHV genome was first determined by Renne et al. (36) to be approximately 165-kb long using purified virion-banded DNA (36). This has been confirmed through genome mapping and sequencing studies (/). We had previously estimated the genome size to be as large as 270 kb (/17), but this was deter- mined using the BC-1 cell line which has a large genomic du- plication (7). The long unique region (LUR), approximately 140 kb in length, comprises the entire coding region for the virus and encodes at least 81 genes (Fig. 1). Because of the high level of synteny between KSHV and herpesvirus saimiri (HVS, the prototypical member of the genus Rhadinovirus), KSHV genes are named after their corresponding HVS homologs starting from the left hand end of the genome. KSHYV is missing the first three genes found in HVS (37) and the first homolog to a HVS gene begins at ORF 4. Synteny with HVS (except for ORFs 2 and 70) extends throughout the genome up to ORF 75 located on the right end of the LUR. Genes having detectable homology to HVS genes include structural protein and DNA synthetic enzyme genes which are conserved among all herpesviruses. There are a number of genes, however, which are not found in HVS and are given a K designation (ORFs KI1-15). No KSHV genes with sequence homology to the EBV Epstein-Barr nuclear antigens, latent membrane proteins, or the HVS saimiri trans- forming proteins involved in cell transformation are found in KSHV. Over 90% of the LUR (excepting approximately 12 kb on the right-hand portion of the genome) has also been recently sequenced from a KS lesion by Bernard Fleckenstein’s group. While details of this sequencing project have not yet been pub- lished, comparison of the deposited KS lesion Genbank sequence to the BC-1 sequence (/) shows a remarkable degree of sequence conservation between BC-1 and KS isolates (generally <0.1% di- vergence), confirming previous comparisons (//). Sequence diver- gence is most pronounced at the KI locus and in internal repeat regions (e.g., Frnk, Vnct, Zppa, and Moi). The right end sequence has proven particularly difficult to clone into sequencing vectors (/) and it is unknown whether this part of the genome contains a hypervariable region similar to the left hand K1 locus. Fig. 2 shows a simplified schematic diagram of the genome demonstrating the conserved herpesvirus gene blocks as first described by Chee et al. (38) from their sequencing of the cy- tomegalovirus genome. Genes that are conserved among mem- bers of all three herpesvirus subfamilies lie within these blocks. In between conserved gene blocks are regions that are unique to KSHV or are also found only in other rhadinoviruses. The HIP frnk vnct nut-1 KS330 TR CBP ssDBP gB DNA Pol DHFR TS Bcl-2 TK gH yi Ho ligase " I | " n HIP-I ov] 7 Kl 4 6 8 9 101123 70Ka 5 24 aS 262128 20503 a p 10 20 3 © 0 60 Ek] A URS Lotion Ns L dd hay FE ERT Si (N00, WT 0 1 id bot iiitnd | Ea . a 1 b c 3 d a eS es a dissin i isin dos sida dss iid IRR 26 L74 L47 si itn SS ————-——= | L36 L48 LS6 & 2 KSS gX IRFs woke/jwke IRF zppa moi KS63 1 mdsk li = Ra - = R-trons “pene VIRF Se RRG RR _ Yop, protems CycD Adh GCR unseqTR & IC J eal, Nl «a v "an | a> K8 235 wn 57 68 69KI12 K1474 Kil S859 60 61 62 65-67 Ki1372 73 S KIS n 00 9% 100 110 120 1 190 La \ Sibiu 1 1 1 dibs mbeoini ii te biti biiatimiibomibom Yb bisibimiliimilin) f 6 9 7 h | Svem———— | ssn ssisismiincs asim) a is i ita LS6 L8o 2 C J ims ssn simstinsind 28 LS54 Fig. 1. The long unique region (LUR) of Kaposi's sarcoma herpesvirus (KSHV) is bracketted by high G:C content terminal repeat regions and contains the 81 genes thus far identified in the KSHV genome. KSHV genes are named after corresponding genes in herpesvirus saimiri (HVS) starting at the left end of the genome. Those genes without significant sequence homology are given a K prefix. The standard map of KSHV is in the same orientation as HVS (37) and is reversed compared to that of EBV (95). Adapted from (/). 66 Journal of the National Cancer Institute Monographs No. 23, 1998 Long Unique Region A ~ TA 0.7 I » a | LANA 0 Fig. 2. A simplified representa- v-CBP VMIPI vBcl-2 \ VY tion of the Kaposi's sarcoma her- | viL- Nui A VIRF v-Cyc V-GCR pesvirus (KSHV) genome show- ing gene blocks conserved among herpesviruses and intervening re- gions without homology. KSHV Terminal Repeats nl i “ I Terminal Repeats KS330Bam KS631Bam | | | [ homologs to cell regulatory and signaling protein genes are clus- tered within the nonconserved gene blocks. Regions conserved with other herpesviruses [1 Regions unique to KSHV and rhadinoviruses I | [I I I I I I I 50 100 Kilobases unique regions contain a remarkable array of cellular gene ho- mologs, encoding proteins involved in cell cycle regulation or cell signaling (7,39). In addition, some DNA synthetic enzymes, such as dihydrofolate reductase (DHFR) and thymidylate syn- thase (TS), are encoded in these regions. Cellular homologs to the DNA synthetic enzymes encoded by the virus are under the control of the E2F transcriptional factor family (40), suggesting that the virus has pirated those DNA synthesis genes that will allow it to replicate DNA outside of the S phase of the cell cycle. Two additional genes found in unique regions include the T1.1 and T0O.7 (ORF K12) genes encoding polyadenylated transcripts (41-43) used for in situ hybridization studies because of their abundance in infected cells (8). What are the possible reasons for the virus having these cel- lular homologs? Our working hypothesis is that these viral pro- teins provide a defense against stereotypic cellular responses to viral infection. When a cell is infected by a virus it undergoes cell cycle shutdown, induction of apoptosis and in vivo enhance- ment of cell-mediated immunity through increased major histo- compatibility antigen presentation [reviewed in (44)]. These ef- fects may be in part mediated by retinoblastoma (pRb) and p53 tumor suppressor pathways, which are involved in the control of dysregulated cellular division (45) and, possibly, unregulated viral nucleic acid replication. The cell cycle is regulated at the G, checkpoint by the E2F family of transcriptional factors that initiate transcription of a number of enzymes involved in DNA synthesis, allowing the cell to pass into the S phase of the cell cycle. Nonphosphorylated retinoblastoma protein binds to E2F preventing E2F directed transcription and stops the cell from progressing through the S phase of the cell cycle (46,47). This can be abrogated by the activity of D-type cyclins interacting with various cyclin- dependent kinases, such as CDK6, which form a complex that phosphorylates pRb, causing release of active E2F, and allowing transcription of DNA synthesis genes. p21 and related cyclin-dependent kinase inhibitors act as a counterbalance to the cyclin-CDK complex, preventing pRb phosphorylation and inhibition. Under conditions where retino- blastoma protein is mutated or when specific pRb inhibitors are present, such as adenovirus E1A, the cell can progress through the G, checkpoint in an uncontrolled fashion. A feedback mechanism exists (48), however, such that overexpression of active E2F results in p53 activation that induces either transcrip- tion of p21 to reinitiate control of the cell cycle (49) or, failing this, induction of p53-mediated apoptosis (50,51) (Fig. 3). This is a very simplified, and only to some degree correct, explanation of how these two tumor suppressor pathways inter- act with each other, but it is apparent that both tumor suppressor pathways are likely to act as antiviral pathways as well. A num- ber of viruses have evolved specific gene products to inhibit both pRB Pathway p53 Pathway Fig. 3. Schematic diagram of the interaction between pRb and p53 tumor suppressor pathways that may serve the dual function of preventing successful cell colonization by a latently replicating DNA virus. Hypothetical sites of interaction with shaded Kaposi's sarcoma herpesvirus proteins are shown that may lead to dysregulated cell proliferation. Only limited experimental data are currently available to indicate that these viral proteins act as onco- proteins and several of these hypothetical interactions are based entirely on sequence homology to known genes alone. E&> v-IRF (ORF K9) UE G1/S Cell Cycle Arrest V-FLIP v-IL6 oe E (ORF K13)? (DRE (C2)? a —i|— a» v-Bcel-2 Downstream Eo) 16, ye apoptotic pathways @ (ORF72) p53-mediated Apoptosis Journal of the National Cancer Institute Monographs No. 23, 1998 67 the pRb and p53 activation (52-54). Growth suppression im- posed by pRb activity is a probable means of limiting replication of the naked DNA episomes of a latent, infecting virus. Adeno- virus, for instance, encodes EIA which specifically inhibits pRb interactions with E2F to prevent G, arrest. However when E1A is expressed in an unopposed fashion, p53-mediated apoptosis is initiated. Adenovirus also encodes the antiapoptotic E1B,,, and E1Bjs, proteins preventing p53-induced apoptosis. An impor- tant illustration of this counterbalance is seen with engineered adenovirus mutants that lack functional E1B genes and are only capable of replicative growth in pS53-deficient tumor cells (55). A similar, although phylogenetically and mechanistically dis- tinct, interaction occurs for the papillomavirus E7 and E6 pro- teins with pRb and p53 pathways. Latent EBV gene expression may have a similar effect on these tumor suppressor pathways: work from Kieff’s group (56-58) and others (59) demonstrates the potential antiapoptotic role of LMP1 in B cells, and data suggest that EBNA2 and EBNA-LP may interact to induce cy- clin D2 overexpression (60). How does KSHV fit into this pattern? Like adenoviruses and papillomaviruses, KSHV also possesses a specific inhibitor of pRb function. KSHV directly encodes a homolog to D-type cy- clins, v-cyc, on the right end of the genome at ORF72 (61,62). ORF73, which encodes the major latency expressed antigen LANA (9), and ORF74, which encodes an IL-8-like receptor that can induce cellular proliferation (67,63), also lie in this region of the genome. v-cyc has been shown to be functionally active in phosphorylating pRb at authentic sites through inter- actions with cdk6 (62), but it may also have broader specificity than cellular D-type cyclins in that it can mediate phosphoryla- tion of histone HI as well (64,65). The altered substrate speci- ficity of the viral cyclin suggests that it may act at other stages of the cell cycle in addition to the G,/S checkpoint (Fig. 3). With an active viral cyclin inhibiting pRb activity, one could also expect that there should be a counterbalancing anti- apoptotic set of genes expressed by KSHV. Indeed, KSHV v-cyc overexpression in NIH3T3 cells rapidly induces cell death pre- sumably through p53-mediated apoptosis (Boshoff C, Sarid R: unpublished observation). Not surprisingly, there are a number of KSHV genes that can potentially prevent cellular apoptosis. For example, the virus encodes a functional IL-6-like cytokine (ORF K2) which prevents B9 cell apoptosis (39,66,67) and is expressed only in KSHV hematopoietic cells, not KS tumor cells (39). The vIL-6 appears to induce appropriate IL-6 Jak-STAT pathway activation [although its receptor specificity is different from hu-1L-6 (68)] and is thus likely to also induce antiapoptotic bel-x, protein production (69). KSHV also encodes a bcl-2 homologue (ORF16) simulta- neously discovered by our group (70) and Cheng et al. (77) that has functional anti-apoptotic activity in both yeast and mamma- lian cells. There is disagreement as to whether or not v-bcl-2 heterodimerizes with cellular members of the bax-bcl-2 family, but it is clear that the KSHV v-bcl-2 prevents bax-mediated apoptosis. More recently, Thome et al. (72) have identified FLICE inhibiting proteins (v-FLIP) encoded by rhadinoviruses that possess dominant negative DEDD domains to inhibit apop- tosis induced by CD95 pathway activation. No functional studies have been published yet on the corresponding KSHV protein (ORF K13), but it is an intriguing protein since it is expressed 68 during viral latency in PEL and KS tumors, along with ORF72 and ORF73 (Sarid R, Wiezorek J, Moore PS, Chang Y: unpub- lished observation). It is not known whether or not v-FLIP has downstream activity that might cause inhibition of tumor sup- pressor pathway apoptosis. We are at an early stage in under- standing the interplay of these KSHV proteins with known tu- mor suppressor pathways, but it is not unreasonable to suppose that these proteins either singly or in combination are active in preventing apoptosis induced by KSHV-inhibition of pRb. In addition to cell cycle shutdown and apoptosis, enhanced immune recognition is another important cellular defense against virus infection. Many of these antiviral effects are coor- dinated at the cellular level through interferon signal regulation. We have found one KSHV protein involved in immune regula- tion, the v-IRF encoded by ORF K9 (39), which may have a unique mechanism for allowing KSHV to escape immune sur- veillance (73). This gene actually has a relatively low degree of sequence homology with the family of interferon regulatory fac- tors (IRF) involved in positively or negatively regulating inter- feron signal transduction (74). Interferon-f3 binding at its receptor induces activation through the Jak-STAT signaling pathway with the formation of a trimer- ic complex of phosphorylated STATI, STAT2, and p48 (or ISGF3vy) (75,76). This complex, called interferon-stimulated gene factor 3 (ISGF3), translocates to the nucleus and binds to enhancer elements (ISREs) in promoters of interferon-stimu- lated genes. Interferon-induced gene transcription effects the phenotypic changes associated with interferon stimulation of cells, including increased MHC 1 transcription (77,78), shut down of the cell cycle [through transcription of the cyclin- dependent kinase inhibitor p21 (79-81)], and also possibly p53- independent apoptosis (82). Interferon also induces expression of the interferon regulatory factors IRF1 and IRF2 which also bind to ISRE. IRFI probably initiates an amplifying loop that markedly amplifies the interferon signal whereas the more slowly degraded IRF2 shuts off this amplification cascade by competitive inhibition at ISRE sites. Thus, it is apparent that cellular proliferation control is modulated by interferon- mediated immune signaling (74,83). KSHV VIRF inhibits IFN-3 signal transduction in a specific and dose-dependent fashion as measured by an ISRE-containing CAT reporter construct (73). Further, vIRF transfection prevents IFNB-induced transcription of the cyclin-dependent kinase in- hibitor p21 suggesting a possible effect on cell cycle regulation. Expression of VIRF in NIH3T3 cells induces full cellular trans- formation and vIRF-expressing NIH3T3 cells form tumors when injected into nude mice. VIRF appears to have highly reproduc- ible oncogenic activity in the NIH3T3 assay; however, these data should be interpreted to suggest only that this gene is active in cellular proliferation. It is unlikely that vIRF expression alone is responsible for KSHV-related human tumor formation and it is more likely to be only one component contributing to a mul- tigenic process (73). Despite the early stage of molecular studies on KSHV, a functional comparison to other DNA tumor viruses is now rea- sonable (Table 1). Like other DNA tumor viruses, KSHV en- codes specific proteins that are candidates for overcoming pRb- mediated growth arrest and p53-mediated apoptotic activity. Journal of the National Cancer Institute Monographs No. 23, 1998 Table 1. Tumor suppressor pathway inhibition by selected tumor virus proteins pRB pathway Adenovirus EIA Papillomavirus E7 Papovavirus (SV40) LT KSHV v-Cyc (ORF72) p53 pathway IFN signaling pathway EIB, o.E1Bss, EIA (through p300) E6 E7? LT LT (through p300)? v-FLIP (ORF K13) v-IL-6 (ORF K2) v-bcl-2 (ORF16) v-IRF (ORF K9) Further, like the E1A protein of adenoviruses (84,85), vIRF may specifically abrogate interferon-mediated responses, contribut- ing to tumor induction as well. There is a close correlation between the cellular genes induced by EBV and the genes pi- rated by KSHV (/) suggesting that both herpesviruses modify their cellular environments in similar ways: KSHV brings in its own cellular homologs into the cell whereas EBV induces cel- lular genes to modify regulatory pathways essential to its sur- vival. Whether additional common tumor virus pathways con- tributing to cell transformation will be found remains to be seen. By examining the KSHV genome, a number of potential on- cogenes have been identified by sequence homology alone. This does not necessarily mean that they function as transforming oncogenes in vivo and care is needed not to overinterpret the results of isolated gene studies. Primary effusion (or body cav- ity-based) lymphomas are clearly outgrowths of malignant clones (86) and provide a model for cell transformation caused in part, or entirely, by viral oncogene expression. KS, however, is a complex tumor whose origins remain disputed. In one view, KS may result from hyperplastic cellular proliferation driven by exogenous cytokines while an opposing view holds that it may result from KSHV oncogene-driven cellular proliferation. Re- cent HUMARA (87,88) and KSHV terminal repeat analyses (/) provide evidence for a clonal proliferation of tumor cells in KS lesions and support the latter supposition although these findings are not universally accepted (89,90). This is further complicated by the fact that cell clonality does not necessarily result from virus-driven cell proliferation since transformed cells could originate from multiple virus-infected loci. In addition, KSHV encodes secreted cytokines (39) which are likely to play a role in non-neoplastic disorders, such as Castleman’s disease (97,92). This debate need not be polarizing, since it is likely that both paracrine and endogenous viral factors will ultimately be found to contribute to KS pathogenesis. Additionally, in vivo expression of potential KSHV-encoded oncogenes is critical to understanding their contribution to these neoplastic disorders. Some genes, such as vIL-6, are expressed in infected hematopoietic tissues but not KS lesions. Other genes, such as the v-bcl-2 may only be expressed during lytic cycle replication (70) and are less likely to contribute to cellular transformation. Finally, isolated gene studies do not take into consideration immune surveillance, which we already know is likely to be tremendously important for this virus since KS is primarily a disease occurring in severely immunosuppressed pa- tients. Because of the early stage of KSHV, the list of oncogenes and their interactions is almost certainly incomplete. The growth- promoting ability of other viral proteins, such as the GPCR (93), the MIP homologs (94), and K1 (Jung J: personal communica- Journal of the National Cancer Institute Monographs No. 23, 1998 tion), is an active area of investigation and promises a rich source of information on cell-virus interactions. Beyond provid- ing pathogenic insights into KS and primary effusion lympho- mas, KSHV is an important new model virus-induced tumori- genesis. KSHV appears to be another example of a tumor virus which is well-adapted to its host and causes tumors primarily in altered or ‘‘accidental’” environments. Most tumor viruses cause tumors only under conditions where there is immunosuppres- sion, complementing mutations in host cells, or during infection of non-native hosts. It is difficult to imagine the evolutionary benefit for the virus to cause a tumor in which it replicates latently and is therefore nontransmissible, which may kill the host. In the setting of severe AIDS-related immunosuppression, however, KSHV appears to be an extremely potent inducer of tumor formation. Early in the AIDS epidemic, up to 50% of gay male AIDS patients eventually developed KS (93) suggesting an extraordinarily high rate of tumor development among those persons who were both KSHV-infected and immunocompro- mised. The convergent evolution of DNA tumor viruses to inhibit similar tumor suppressor pathways by different mechanisms suggests that these pathways also serve antiviral functions. Tu- mor viruses apparently require a mechanism for inhibiting tumor suppressor pathways if they are to survive as a latent episome in an actively dividing cell. It should not be surprising for cells to use tumor suppressor pathways to control both cancer cell growth and viral replication since both cases involve control of dysregulated DNA replication. Careful examination of KSHV and other tumor viruses will continue to lead to new insights into mechanisms of nonviral carcinogenesis. Ready identification of regulatory gene homologs encoded by KSHYV gives this virus a unique degree of transparency. Despite its complex interaction with cellular regulatory pathways, KSHV’s option to use molecular piracy provides accessible starting points for examining these pathways. The extensive de- gree of symmetry between genes encoded by KSHV and those induced by EBV (1) also indicates that insights gained from one virus may to some degree be transferable to the other. KSHV illustrates in a new fashion that several major cellular pathways preventing tumor outgrowth, such as the p53 and pRb pathways, may also guard against latent virus infection. Virus strategies to overcome these intracellular defenses appear to inadvertently contribute to cell transformation. KSHV may thus provide a new, unique, and tractable model for exploring these and other issues in viral oncogenesis. References (1) Russo JJ, Bohenzky RA, Chien MC, Chen J, Yan M, Maddalena D, et al. Nucleotide sequence of the Kaposi sarcoma-associated herpesvirus (HHVS). Proc Natl Acad Sci U S A 1996:93:14862-7. 69 2 ~ (3 (4) (5 = (0) (7 ~ (8 (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24 hi —~ e) ‘nn ) 70 Olsen SJ, Moore PS. Kaposi's sarcoma-associated herpesvirus (KSHV/ HHVS) and the etiology of KS. In: Freidman H, Medveczky P, Bendinelli M, editors. Molecular immunology of herpesviruses. New York: Plenum Press. In press. IARC: Human Papillomaviruses. In: IARC Working Group on the Evalu- ation of Carcinogenic Risks to Humans (IARC, ed). Lyon, France, 1995: 90-233. Lebbe C, Agbalika F, de CP, Deplanche M, Rybojad M, Masgrau E, et al. Detection of human herpesvirus 8 and human T-cell lymphotropic virus type | sequences in Kaposi sarcoma. Arch Dermatol 1997;133:25-30. Dupin N, Grandadam M, Calvez V, Gorin I, Aubin JT, Harvard S, et al. Herpesvirus-like DNA in patients with Mediterranean Kaposi's sarcoma. Lancet 1995:345:761-2. Moore PS, Chang Y. Detection of herpesvirus-like DNA sequences in Kaposi's sarcoma lesions from persons with and without HIV infection. N Engl J Med 1995:332:1181-5. Boshoff C, Schulz TF, Kennedy MM, Graham AK, Fisher C, Thomas A, et al. Kaposi's sarcoma-associated herpesvirus infects endothelial and spindle cells. Nat Med 1995;1:1274-8. Staskus KA, Zhong W. Gebhard K, Herndier B, Wang H, Renne R, et al. Kaposi's sarcoma-associated herpesvirus gene expression in endothelial (spindle) tumor cells. J Virol 1997;71:715-9. Rainbow L, Platt GM, Simpson GR. Sarid R, Gao SJ, Stoiber H, et al. The 226- to 234-kilodalton latent nuclear protein (LNA) of Kaposi's sarcoma- associated herpesvirus (KSHV/HHVS) is encoded by ORF73 and is a com- ponent of the latency-associated nuclear antigen. J Virol 1997;71:5915-21. Boshoff C, Talbot S, Whitby D, Reeves J, Weiss RA. AIDS-associated tumors and HHVS. In: National AIDS Malignancy Conference. Bethesda (MD), 1997:A13. Neipel F, Albrecht JC, Fleckenstein B. Cell-homologous genes in the Ka- posi’s sarcoma-associated rhadinovirus human herpesvirus 8: determinants of its pathogenicity? J Virol 1977;71:4187-92. Miller G, Rigsby MO, Heston L, Grogan E, Sun R, Metroka C, et al. Antibodies to butyrate-inducible antigens of Kaposi's sarcoma-associated herpesvirus in patients with HIV-1 infection. N Engl J Med 1996;334: 1292-7. Gao SJ. Kingsley L. Hoover DR, Spira TJ, Rinaldo CR, Saah A, et al. Seroconversion to antibodies against Kaposi's sarcoma-associated herpes- virus-related latent nuclear antigens before the development of Kaposi's sarcoma. N Engl J Med 1996:335:233-41. Kedes DH, Operskalski E, Busch M, Kohn R, Flood J, Ganem D. The seroepidemiology of human herpesvirus 8 (Kaposi's sarcoma-associated herpesvirus): distribution of infection in KS risk groups and evidence for sexual transmission. Nat Med 1996;2:918-24. Simpson GR, Schulz TF, Whitby D, Cook PM, Boshoff C, Rainbow L, et al. Prevalence of Kaposi's sarcoma associated herpesvirus infection as measured by antibodies to recombinant capsid protein and latent immuno- fluorescence antigen. Lancet 1996;348:1133-8. Lennette ET, Blackbourne DJ, Levy JA. Antibodies to human herpesvirus type 8 in the general population and in Kaposi's sarcoma patients. Lancet 1996:348:858-01. Kedes DH, Ganem D, Ameli N, Bacchetti P, Greenblatt R. The prevalence of serum antibody to human herpesvirus 8 (Kaposi sarcoma-associated herpesvirus) among HIV-seropositive and high-risk HIV-seronegative women. JAMA 1997:277:478-81. Gao SJ, Kingsley L, Li M, Zheng W, Parravicini C, Ziegler J, et al. KSHV antibodies among Americans, Italians and Ugandans with and without Ka- posi’s sarcoma. Nat Med 1996;2:925-8. Andre S, Schatz O, Bogner JR, Zeichhardt H, Stoffler MM, Jahn HU, et al. Detection of antibodies against viral capsid proteins of human herpesvirus 8 in AIDS-associated Kaposi's sarcoma. J Mol Med 1997;75:145-52. Parravicini C, Olsen SJ, Capra M, Poli F, Sirchia G, Gao SJ, et al. Risk of Kaposi's sarcoma-associated herpesvirus transmission from donor allo- grafts among Italian posttransplant Kaposi's sarcoma patients. Blood 1997; 90:2826-9. Whitby D, Howard MR, Tenant-Flowers M, Brink NS, Copas A, Boshoff C, et al. Detection of Kaposi's sarcoma-associated herpesvirus (KSHV) in peripheral blood of HIV-infected individuals predicts progression to Ka- posi’s sarcoma. Lancet 1997;364:799-802. Moore PS, Kingsley L, Holmberg SD, Spira T, Conley LJ, Hoover D, et al. Kaposi's sarcoma-associated herpesvirus infection prior to onset of Kapo- si’s sarcoma. AIDS 1996:10:175-80. Corbellino M, Bestetti G, Galli M, Parravicini C. Absence of HHV-8 in prostate and semen [letter]. N Engl J Med 1996:335:1238-9. Howard MR, Whitby D, Bahadur G, Suggett F, Boshoff C, Tenant FM, et al. Detection of human herpesvirus 8 DNA in semen from HIV-infected individuals but not healthy semen donors. AIDS 1997;11:FI5-F19. Dupin N, Gorin I, Escande JP, Calvez V, Grandadam M, Huraux JM, et al. (26 (27 (28 (29 (30 (31 (32 (33 (34 (35 (36) (37 (38 (39) (40 (41) (42) (43) (44) (45) (46) (47) (48 (49) ~— Lack of evidence of any association between human herpesvirus 8 and various skin tumors from both immunocompetent and immunosuppressed patients. Arch Dermatol 1997;133:537-7. Boshoff C, Talbot S, Kennedy M, O'Leary J, Schulz T, Chang Y. HHVS and skin cancers in immunosuppressed patients [letter] [published erratum appears in Lancet 1996:348:138]. Lancet 1996;347:338-9. ) Monini P, de Lellis L, Fabris M, Rigolin F, Cassai E. Kaposi's sarcoma- associated herpesvirus DNA sequences in prostate tissue and human semen. N Engl J Med 1996:334:1168-72. ) Monini P, de Lellis L, Cassai E. Absence of HHV-8 in prostate and semen [letter]. N Engl J Med 1996;335:1238-9. Rettig MB, Ma HJ, Vescio RA, Pold M, Schiller G, Belson D, et al. Kaposi's sarcoma-associated herpesvirus infection of bone marrow den- dritic cells from multiple myeloma patients. Science 1997;276:1851-4. ) Blackbourne DJ, Lennette E, Mourich D, Witte M, Way D, Nelson J, et al. Host range studies of human herpesvirus 8 (HHVS8). In: National AIDS Malignancy Conference. Bethesda (MD), 1997:A34. ) Foreman KE, Friborg JJ, Kong WP, Woffendin C, Polverini PJ, Nickoloff BJ, et al. Propagation of a human herpesvirus from AIDS-associated Ka- posi’s sarcoma. N Engl J Med 1997:336:163-71. ) Jones J, Peterman T, Chu S, Jaffe H. AIDS-associated Kaposi's sarcoma. Science 1995:267:1078-9. ) Mocroft A, Youle M, Gazzard B, Morcinek J, Halai R, Phillips AN. Anti- herpesvirus treatment and risk of Kaposi's sarcoma in HIV infection. AIDS 1996:10:1101-5. ) Herndier B, Shiramizu B, Weinstein M, Ganem D, Orenstein JM. KSHV latent and lytic infection in Kaposi's sarcoma lesions and lymphoid tissue. In: National AIDS Malignancy Conference. Bethesda (MD), 1997. Ganem D. Latent and lytic replication of HHV-8/KSHV. In: National AIDS Malignancy Conference. Bethesda (MD), 1997. Renne R, Lagunoff M, Zhong W, Ganem D. The size and conformation of Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) DNA in infected cells and virions. J Virol 1996;70:8151-4. ) Albrecht JC, Nicholas J, Biller D, Cameron KR, Biesinger B, Newman C, et al. Primary structure of the Herpesvirus saimiri genome. J Virol 1992; 66:5047-58. ) Chee MS, Bankier SB, Bohni CM, Brown RC, Horsnell T, Hutchison CA, et al. Analysis of the protein coding content of the sequence of cytomeg- alovirus strain AD169. Curr Top Microbiol Immunol 1990;154:125-69. Moore PS, Boshoff C, Weiss RA, Chang Y. Molecular mimicry of human cytokine and cytokine response pathway genes by KSHV. Science 1996: 274:1739-44. ) Adams PD, Kaelin WI. Transcriptional control by E2F. Semin Cancer Biol 1995;6:99-108. Sun R, Lin SF, Gradoville L, Miller G. Polyadenylated nuclear RNA en- coded by Kaposi sarcoma-associated herpesvirus. Proc Natl Acad Sci US A 1996;93:11883-8. Zhong WD, Ganem D. Characterization of ribonucleoprotein complexes containing an abundant polyadenylated nuclear RNA encoded by Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8). J Virol 1997.71: 1207-12. Zhong W, Wang H, Herndier B, Ganem D. Restricted expression of Kaposi sarcoma-associated herpesvirus (human herpesvirus 8) genes in Kaposi sarcoma. Proc Natl Acad Sci U S A 1996:93:6641-6. Shen Y, Shenk TF. Viruses and apoptosis. Curr Opin Genet Devel 1995: 5:105-11. Weinberg RA. The cat and mouse games that genes, viruses, and cells play. Cell 1997:88:573-5. Sherr CJ. D-type cyclins. Trends Biochemical Sci 1995:20:187-90. Sherr CJ. Cancer cell cycles. Science 1996:274:1672-7. White E. Tumour biology. p53. guardian of Rb. Nature 1994:371:21-2. Michieli P, Li W, Lorenzi MV, Miki T, Zakut R, Givol D, et al. Inhibition of oncogene-mediated transformation by ectopic expression of p21Wafl in NIH3T3 cells. Oncogene 1996;12:775-84. Wu X, Levine Al. p53 and E2F-1 cooperate to mediate apoptosis. Proc Natl Acad Sci U S A 1994:91:3602-6. Hiebert SW, Packham G, Strom DK, Haffner R, Oren M, Zambetti G, et al. E2F-1:DP-1 induces p53 and overrides survival factors to trigger apoptosis. Mol Cell Biol 1995;15:6864-74. White E. Regulation of p53-dependent apoptosis by EIA and EIB. Curr Top Microbiol Immunol 1995;199:34-58. Teodoro JG, Branton PE. Regulation of apoptosis by viral gene products. J Virol 1997:71:1739-46. Farthing AJ. Vousden KH. Functions of human papillomavirus E6 and E7 oncoproteins. Trends Microbiol 1994:2:170-4. Bischoff JR, Kirn DH, Williams A, Heise C, Horn S, Muna M, et al. An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science 1996:274:373-6. = ~~ - Journal of the National Cancer Institute Monographs No. 23, 1998 (56) Kieff E. Viral-mediated mechanisms of cancer pathogenesis. In: National AIDS Malignancy Conference. Bethesda (MD), 1997. (57) Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C, Kieff E. The Epstein-Barr virus transforming protein LMP1 engages signaling proteins for the tumor necrosis factor receptor family. Cell 1995:80: 389-99. (58) Kieff E. Epstein-Barr virus—increasing evidence of a link to carcinoma leditorial]. N Engl J Med 1995:333:724-06. (59) Okan I, Wang Y, Chen F, Hu LF, Imreh S, Klein G, et al. The EBV- encoded LMPI protein inhibits pS53-triggered apoptosis but not growth arrest. Oncogene 1995:11:1027-31. (60) Sinclair AJ, Palmero 1, Peters G, Farrell PJ. EBNA-2 and EBNA-LP co- operate to cause GO to GI transition during immortalization of resting human B lymphocytes by Epstein-Barr virus. EMBO J 1994;13:3321-8. (61) Cesarman E, Nador RG, Bai F, Bohenzky RA, Russo JJ, Moore PS, et al. Kaposi's sarcoma-associated herpesvirus contains G protein-coupled re- ceptor and cyclin D homologs which are expressed in Kaposi's sarcoma and malignant lymphoma. J Virol 1996;70:8218-23. (62) Chang Y, Moore PS, Talbot SJ. Boshoff CH, Zarkowska T, Godden-Kent D, et al. Cyclin encoded by KS herpesvirus. Nature 1996;382:410. (63) Arvanitakis L, Geras RE, Varma A, Gershengorn MC, Cesarman E. Human herpesvirus KSHV encodes a constitutively active G-protein-coupled re- ceptor linked to cell proliferation. Nature 1997;385:347-50. (64) Godden-Kent D, Talbot SJ, Boshoff C, Chang Y, Moore PS, Weiss RA, et al. The cyclin encoded by Kaposi's sarcoma associated herpesvirus (KSHV) stimulates ¢dk6 to phosphorylate the retinoblastoma protein and histone H1. J Virol 1997;71:4193-8. (65) Li M, Lee H, Yoon DW, Albrecht JC, Fleckenstein B, Neipel F, et al. Kaposi's sarcoma-associated herpesvirus encodes a functional cyclin. J Virol 1997:71:1984-91. (66) Nicholas J, Ruvolo VR, Burns WH, Sandford G, Wan X, Ciufo D, et al. Kaposi's sarcoma-associated human herpesvirus-8 encodes homologues of macrophage inflammatory protein-1 and interleukin-6. Nat Med 1997.3: 287-92. (67) Neipel F, Albrecht JC, Ensser A, Huang YQ, Li JJ, Friedman KA, et al. Human herpesvirus 8 encodes a homolog of interleukin-6. J Virol 1997; 71:839-42. (68) Molden J, Chang Y, Yow Y, Moore PS, Goldsmith MA. A Kaposi's sarcoma-associated herpesvirus-encoded cytokine homolog (vIL-6) acti- vates signaling through the shared gp130 receptor subunit. J Biol Chem 1997:272:19625-31. (69) Schwarze MM, Hawley RG. Prevention of myeloma cell apoptosis by ectopic bel-2 expression or interleukin 6-mediated up-regulation of bel-xL. Cancer Res 1995:55:2262-5. (70) Sarid R, Sato T, Bohenzky RA, Russo JJ, Chang Y. Kaposi's sarcoma- associated herpesvirus encodes a functional Bcl-2 homologue. Nat Med 1997:3:293-8. (71) Cheng EH, Nicholas J. Bellows DS, Hayward GS, Guo HG, Reitz MS, et al. A Bcl-2 homolog encoded by Kaposi sarcoma-associated virus, human herpesvirus 8, inhibits apoptosis but does not heterodimerize with Bax or Bak. Proc Natl Acad Sci U S A 1997:94:690-4. (72) Thome M, Schneider P, Hofman K, Fickenscher H, Meinl E, Neipel F, et al. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 1997:386:517-21. (73) Gao SI. Boshotf C, Jayachandra S, Weiss RA, Chang Y, Moore PS. KSHV ORF K9 (vIRF) is an oncogene that inhibits the interferon signaling path- way. Oncogene 1997;15:1979-86. (74) Taniguchi T, Harada H. Lamphier M. Regulation of the interferon system and cell growth by the IRF transcription factors. J Cancer Res Clin Oncol 1995;121:516-20. (75) Schindler C, Fu XY, Improta T, Aebersold R, Darnell JJ. Proteins of transcription factor ISGF-3: one gene encodes the 91-and 84-kDa ISGF-3 proteins that are activated by interferon alpha. Proc Natl Acad Sci U S A 1992:89:7836-9. (76) Schindler C. Cytokine signal transduction. Receptor 1995:5:51-62. (77) Weisz A, Kirchhoff S, Levi BZ. IFN consensus sequence binding protein (ICSBP) is a conditional repressor of IFN inducible promoters. Inter Im- munol 1994:6:1125-31. Journal of the National Cancer Institute Monographs No. 23, 1998 (78) Maudsley DJ, Pound JD. Modulation of MHC antigen expression by vi- ruses and oncogenes. Immunol Today 1991:12:429-31. (79) Hobeika AC, Subramaniam PS, Johnson HM. IFN« induces the expression of the cyclin-dependent kinase inhibitor p21 in human prostate cancer cells. Oncogene 1997:14:1165-70. (80) Chin YE, Kitagawa M, Su WC, You ZH, Iwamoto Y, Fu XY. Cell growth arrest and induction of cyclin-dependent kinase inhibitor p21 WAF1/CIP1 mediated by STATI. Science 1996;272:719-22. (81) Sangfelt O, Erickson S, Einhorn S, Crander D. Induction of Cip/Kip and Ink4 cyclin dependent kinase inhibitors by interferon-alpha in hematopoi- etic cell lines. Oncogene 1997:14:415-23. Tanaka N, Ishihara M, Lamphier MS, Nozawa H, Matsuyama T, Mak TW, et al. Cooperation of the tumour suppressors IRF-1 and p53 in response to DNA damage. Nature 1996;382:816-8. (83) Harada H, Kitagawa M, Tanaka N, Yamamoto H. Harada K, Ishihara M, et al. Anti-oncogenic and oncogenic potentials of interferon regulatory fac- tors-1 and -2. Science 1993;259:971-4. (84) Reich N, Pine R, Levy D, Darnell JJ. Transcription of interferon-stimulated genes is induced by adenovirus particles but is suppressed by EIA gene products. J Virol 1988:62:114-9. (85) Bhattacharya S, Eckner R, Grossman S, Oldread E, Arany Z, D’ Andrea A, et al. Cooperation of Stat2 and p300/CBP in signalling induced by inter- feron-alpha. Nature 1996;383:344-7. (86) Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. Kaposi's sar- coma-associated herpesvirus-like DNA sequences are present in AIDS- related body cavity based lymphomas. N Engl J Med 1995;332:1186-91. (87) Rabkin CS, Bedi G, Musaba E, Sunkutu R, Mwansa N, Sidransky D, et al. AIDS-related Kaposi's sarcoma is a clonal neoplasm. Clin Cancer Res 1995;1:257-60. (88) Rabkin CS, Janz S, Lash A, Coleman AE, Musaba F, Liotta L, et al. Monoclonal origin of multicentric Kaposi's sarcoma lesions. N Engl J Med 1997:336:988-93. (89) Diaz-Cano SJ. Wolfe HJ. Clonality in Kaposi's sarcoma. N Engl J Med 1997:337:571. (90) Gill P, Tsai Y, Rao AP, Jones P. Clonality in Kaposi's sarcoma. N Engl J Med 1997:337:570-1. (91) Soulier J, Grollet L, Oskenhendler E, Cacoub P, Cazals-Hatem D, Babinet P, et al. Kaposi's sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman’s disease. Blood 1995:86:1276-80. (92) Parravicini C. Corbellino M. Paulli M, Magrini U, Lazzarino M, Moore PS, et al. Expression of a virus-derived cytokine, KSHV vIL-6. in HIV sero- negative Castleman’s disease. Am J Pathol. In press. (93) Holmberg S, Buchbinder S, Conley L, Wong L. Katz M, Penley K, et al. The spectrum of medical conditions and symptoms before acquired immu- nodeficiency syndrome in homosexual and bisexual men infected with the human immunodeficiency virus. Am J Epidemiol 1995;141:395-404. (94) Boshoftf C, Endo Y, Collins PD, Takeuchi Y, Reeves JD, Schweickart VL, et al. Angiogenic and HIV-inhibitory functions of KSHV-encoded chemo- kines. Science 1997:;278:290-4. (95) Baer R, Bankier AT, Biggin PL, Deininger PL, Farrell PJ, Gibson TJ, et al. DNA sequence and expression of the B95-8 Epstein-Barr virus genome. Nature 1984:310:207-11. (82 Notes Supported by Public Health Service grant CA67391 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Ser- vices. We thank the members of our laboratory who have contributed to the projects briefly described here, including Roy Bohenzky, Shou-Jiang Gao, Sonja Olsen, Ronit Sarid, and Jeffrey Wiezorek. We would also like to acknowledge the contributions of our collaborators, especially Drs. Robin Weiss and Chris Bo- shoff of the Institute for Cancer Research, James Russo and Izzy Edelman of the Columbia Genome Center, and Thomas Schulz from the University of Liverpool. We also thank Laurie Anderson for her help in preparing this manuscript. 71 Human Herpesvirus 8—the First Human Rhadinovirus Frank Neipel, Jens-Christian Albrecht, Bernhard Fleckenstein Kaposi’s sarcoma (KS)-associated herpesvirus, also known as human herpesvirus 8 (HHV-8), is the first known human member of the genus Rhadinovirus. It is regularly found by polymerase chain reaction in all forms of KS, in certain types of Castleman’s disease, and in body cavity-based B-cell lym- phoma. Other members of this virus group occur in nonhu- man primates, ungulates, rabbits, and mice and cause in part fulminant lymphomas and other neoplastic disorders of the hematopoietic system. Rhadinoviruses share a typical ge- nome structure; most characteristically, they contain numer- ous sequences that appear to be sequestered from cellular DNA. We cloned and sequenced almost the complete genome of HHV-8 from a single KS biopsy specimen. Although this procedure revealed collinear organization and extensive ho- mologies with the open reading frames of herpesvirus saimiri, genes with homology to the known oncoproteins (Stp, Tip) were not identified in the HHV-8 genome. How- ever, HHV-8 reading frame KI, the positional analogue of Stp/Tip, was found to be significantly variable between dif- ferent strains. We found, in addition, the reading frames for homologues of cellular interleukin 6, macrophage inflamma- tory proteins a and 3 (MIP1a and MIP1, respectively), an interferon-responsive factor, and two inhibitors of apoptosis. Several of these cell-homologous genes of HHV-8 have al- ready been shown to code for functional proteins. [Monogr Natl Cancer Inst 1998;23:73-77] Human herpesvirus 8 (HHV-8), also referred to as Kaposi's sarcoma (KS)-associated herpesvirus (KSHV), is the first known human member of the genus Rhadinovirus (1). Members of this group of herpesviruses share a common genome structure: A central segment of low-GC DNA (L-DNA) is flanked by mul- tirepetitive high-GC DNA (H-DNA) (2-6) (Fig. 1). In addition to their common genome structure, which is the cause of physical properties reflected by the term ‘‘rhadinovi- rus” (padwoés [Greek] = fragile), all animal rhadinoviruses known so far share a common epidemiology. They are frequent in their natural host (>50%), where infection is not known to be associated with apparent disease (Table 1). In contrast, infection of closely related species is often the cause of fulminant lym- phoproliferative diseases or overt malignant lymphoma with fa- tal outcome. For example, there is no indication for pathogenic properties of herpesvirus saimiri in its natural host, the squirrel monkey. However, in related species (e.g., common marmosets and several other New World primates), infection by herpesvirus saimiri results in polyclonal T-cell lymphomas (7). Similarly, alcelaphine herpesvirus type | and the related ovine herpesvirus type 2 are nonpathogenic in wildebeest and sheep, respectively, Journal of the National Cancer Institute Monographs No. 23, 1998 whereas a lymphoproliferative syndrome termed ‘malignant ca- tarrhal fever’ is caused upon infection of cattle (4). This finding does not necessarily imply that HHV-8 would be exceptional if it is causing a malignant tumor in its natural host. The incidence of KS, body cavity-based lymphomas, and multicentric forms of Castleman’s disease is so low that such disease associations would have certainly been overlooked in animal models. Like- wise, Epstein-Barr virus (EBV), which is the closest known relative of HHV-8 in humans, is clearly associated with infec- tious mononucleosis and with several cancers. However, pri- mary EBV infection is usually not apparent unless additional factors, such as immunosuppression or delay of primary infec- tion, disturb the delicate balance of virus and host. Infection of species other than the well-adapted natural host is possibly an- other example of an “‘accident’” associated with increased tu- morigenicity. Thus, despite the lack of clear-cut evidence for oncogenesis by nonhuman rhadinoviruses in their natural hosts, the finding that infection of non-natural hosts is frequently associated with lymphoproliferative syndromes hints at the pathogenic potential that these viruses might show even in their natural hosts under certain, albeit unusual, circumstances. It has been shown for herpesvirus saimiri that the gene(s) relevant for malignant trans- formation are located close to the left end of the genome (Fig. 2) (8-11). A single reading frame termed “*STP-A"" is present there in herpesvirus saimiri subgroup A viruses. This reading frame is required for the transforming phenotype of herpesvirus saimiri subgroup A, as shown by deletion mutants (/2). Two reading frames, termed **STP-C’* and **TIP,”” are present at the equiva- lent position of herpesvirus saimiri subgroup C viruses, and both are obviously related to the oncogenic phenotype (10,13). Ex- pression of STP-A in transgenic mice results in lymphoid tumors (14), whereas animals transgenic for STP-C develop epithelial tumors (/5). In addition, subgroup C strains of herpesvirus saimiri can transform human T lymphocytes (16). These cells still depend on interleukin 2, but the respective T-cell antigen is no longer required for continuous growth in cell culture [re- viewed in (7)]. Sequencing the complete genome of HHV-8 from both a body cavity-based lymphoma cell line (5) and a single KS biopsy specimen has revealed that no obvious homo- logue of Stp/Tip or other viral oncogenes is encoded by HHV-8 *Affiliation of authors: Institut fiir Klinische und Molekulare Virologie, Uni- versitit Erlangen-Niirnberg, Schlofigarten 4, Erlangen, Germany. Correspondence to: Bernhard Fleckenstein, M.D., Institut fiir Klinische und Molekulare Virologie. Universitit Erlangen-Niirnberg, Schlofigarten 4, D-91054 Erlangen, Germany. E-mail: fleckenstein@viro.med.uni-erlangen.de See “*Note’” following ‘‘References.” © Oxford University Press 73 K2 K4 19 26 K1 02 K4.1 K5 K7 20 27 04 06 07 08 09 10 11 K3 70 K4.2 K6 16 17 18 21 22 23 24 25 28 IX 2 AAD AA> X > AEb< Fig. 1. Human herpesvirus 8 (HHV-8) -) KIKI JU & J DD [> i : . . UT KKK are termed “KI to K14."" Regions of d & CRRA Wd RIT 4 = repetitive sequence are denoted by 5 & oo po 70 - 5 os 0 os 100 shaded boxes. TR = terminal repeat: LIR and LIR" = long inverted repeat; DRI-DRS = internal direct repeat: 66 T1.1 = 1.1-kb transcript; T0.7 = 0.7- 67 68 7 K14 kb transcript. 62 63 64 65 67.5 69 72 73 74 75 KI 100 105 110 (5,17). Instead, a reading frame, termed “*K1,”” is present at the same position in HHV-8 (Figs. 1 and 2). This reading frame has no detectable sequence homology to STP or other transforma- tion relevant genes. However, KI has both a transmembrane region and a short intracytoplasmic tail like the transforming genes of other rhadinoviruses that are encoded at equivalent genomic positions. K1 is thus certainly a candidate for a trans- forming gene. Comparison of the two available complete HHV-8 genomic sequences reveals a striking degree of conser- vation. With the possible exception of the very right end of the unique region, reading frame Kl is by far the most divergent reading frame. The high degree of interstrain variability is an- other feature that K1 shares with the STP genes of herpesvirus saimiri. Although there is less than 0.1% divergence between the two complete sequences, 6% of the amino and nucleic acids are different between the K1 sequences derived from a KS and a B-cell line. This makes K1 a prime candidate for studies of HHV-8 molecular epidemiology. Acquisition of genes from the host cell genome is a common feature of most herpesviruses and of rhadinoviruses in particular. So far, there are least 14 reading frames of HHV-8 that are clearly homologous to known cellular genes (Table 2). In con- trast to their cellular counterparts, these viral genes are usually transcribed into an unspliced messenger RNA. This implies that not genomic DNA fragments but complementary DNA (cDNA) molecules were integrated into the viral genome. Reverse tran- scriptase activity must therefore have been present. One may speculate that not only was reverse transcription enabled by co-infection of the same cell by a retrovirus, but also cellular genes were transferred to the herpesvirus via retroviral genomes with their capability of integration. Although different members of the genus Rhadinovirus acquired different genes from the host cell, these genes are usually found at strikingly similar genomic positions (Fig. 2). These sequestered genes are frequently clus- tered around areas of repetitive sequence: close to the genomic termini and—most notably—in two genomic areas about 20-30 kilobase pairs apart from the ends of the unique region (Fig. 2). The pattern of repetitive elements in these areas includes stretches rich in AT (adenosine—thymidine) flanked by elements of dyad symmetry. These are the typical hallmarks of origins of DNA replication, although experimental data to confirm this are not available for any of the rhadinoviruses. Therefore, one sce- nario for the acquisition of foreign genes by a herpesvirus is linked to DNA replication. It could well be that, during replica- tion by the rolling circle mechanism, fragments arise and, upon re-circularization, cDNA molecules are inserted. Table 1. Pathogenic properties of rhadinoviruses in natural and foreign hosts Natural host Virus type Human herpesvirus type 8 Human Putative: Kaposi's Herpesvirus saimiri Squirrel monkey Herpesvirus ateles Spider monkey Pathogenic properties in natural host Pathogenic properties in foreign hosts Foreign hosts sarcoma, B-cell lymphomas Marmosets, etc. Polyclonal malignant T-cell lymphomas Polyclonal malignant T-cell lymphomas Marmosets, etc. Alcelaphine herpesvirus type | Wildebeest Cattle, buffalo Malignant catarrhal fever Ovine herpesvirus type 2 Sheep Cattle Malignant catarrhal fever Bovine herpesvirus type 4 Cattle Herpesvirus sylvilagus Cottontail rabbit Benign lymphoproliferation Equine herpesvirus type 2 Horse Foals: transient immunosuppression Murine herpesvirus-68 Bank vole Mouse Benign splenomegaly 74 Journal of the National Cancer Institute Monographs No. 23, 1998 IL-0 FGAM bck-2 AHV-A P bel-2 coc 4 FGAM FGAM ea ge [lop gl [llneamm oo [ ™» [[w] Go) am) [TTT] wom OOK, Whvoy | oramaman [[[] #98 00 C0 pull HVA od Cre i | —- BCT or Ts S.C 1 hem [Jef aE oi; vs ive (I 00 real io CCPH FGAM pry oo hwnd] FAM MBP BB pl gr | 1G loft terminal L-DNA Saas eras Maas es eo borer bea a baa Fig. 2. Positions of cell-homologous genes in circular Rhadinovirus genomes. The termini of the L-DNA segments of genomes of HHV-8 (human herpesvirus- 8), HVS-C (herpesvirus saimiri subgroup C), HVA (herpesvirus ateles), AHV-1 (alcelaphine herpesvirus type 1) (4), EHV-2 (equine herpesvirus type 2), and MHV-68 (murine herpesvirus type 68) (3) are shown as oriented in circularized latent genomes. Cell-homologous genes are symbolized by dark gray arrows. Open arrows indicate reading conserved herpesvirus reading frames (mDBP = major DNA binding protein; tp = transport protein; gB = glycoprotein B; pol = DNA polymerase). Most reading frames with homology to known cellular genes are located close to the terminal repeats and in a nonconserved region flanked by ORF 11 on one side and ORF17 on the other side. Repetitive elements (indicated by hatched boxes) are also present in this area. In addition to the genes shown here, an interferon response factor homologue K9 is encoded by HHV-8 Although for most of these captured genes it is not very likely that they are essential for virus replication in cell culture, they certainly have important functions in the viruses’ natural habitat. Although different rhadinoviruses acquired different host-cell genes, their putative functions apparently converge to achieve three common goals: 1) to enhance DNA replication indepen- dently from the cell cycle, 2) to expand the pool of infectable cells, and 3) to counteract the host’s responses to infection (/8) (Table 3). The first function, enhancement of DNA replication indepen- dent from the status of the infected cell, is achieved by enzymes of the nucleotide metabolism, i.e., thymidylate synthase, dihy- drofolate reductase, and formylglycinoamidine synthase. Virus- encoded cyclins, interleukin 6, and the interleukin 8 receptor may enhance cell proliferation and may thus expand the pool of infectable cells. The three macrophage inflammatory proteins encoded by HHV-8 might work in a similar way by attracting susceptible cells. Apoptosis is a typical response of the host to infection by a virus. HHV-8 carries two genes, ORF16 (vbcl-2) and ORF71 (vFLIP), both of which could extend the life span of infected cells through the inhibition of apoptosis by two differ- ent mechanisms (79-21). Similarly, the complement control pro- tein homologues present in most rhadinoviruses counteract the Journal of the National Cancer Institute Monographs No. 23, 1998 in a nonconserved area between open reading frames 57 and 58 (Fig. 1). CCPH = complement control protein homologue; sag = open reading frame 14 of herpesvirus saimiri and herpesvirus ateles with homology to mouse mammary tumor virus superantigen; CD59 = viral CD59 homologue; DHFR = dihydro- folate reductase; TS = thymidylate synthase; cyc = viral cyclin D homologue; bel-2 = viral bel-2 homologue; Flip = Flice inhibitory protein; vIL-6 = viral interleukin 6; MIPla = macrophage inflammatory protein a; MIPI = open reading frame with homology to macrophage inflammatory protein If3 and mac- rophage chemoattractant protein; IL-8R = viral interleukin 8 receptor; IL-17 = viral interleukin 17; ger = G-protein coupled receptor o/f3; FGAM = phos- phoribosylformylglycinamide synthase; HSUR = herpesvirus saimiri U-like RNA; HAUR = herpesvirus ateles U-like RNA. host’sresponse (2,22). The virus-encoded interferon-response fac- tor homologue (VIRF) might fit into this scenario at two different places: 1) It could counteract interferon-mediated suppression, and 2) it could mimic the proliferative effect of human interferon response factor 2. One can easily imagine how these genes can contribute to malignant growth transformation. Increasing the pool of available nucleotides not only enhances viral DNA rep- lication but also facilitates the proliferation of transformed cells (Table 3). Genes that extend the pool of infectable cells or that prolong their life span, i.e., the virus-encoded cyclins, apoptosis inhibitors, cytokines, cytokine receptors, and interferon response factors, could also contribute to dysregulated growth and favor the development of cancers. Although under normal circum- stances the functions of these viral genes are well balanced with the host, malignant growth does not occur. Like several other rhadinoviruses, HHV-8 has a host of genes that could acciden- tally contribute to tumor development. Examples of such acci- dents that favor malignant growth are infection of a non-natural host (Table 1), infection that occurs usually late in the host’s life, and certainly infection of an immunosuppressed host. The ini- tially mentioned finding that rhadinoviruses are not usually pathogenic in their natural host is therefore still compatible with the likelihood of HHV-8 being associated with at least two can- 75 Table 2. Cellular homologues of rhadinoviruses and the vy, herpesvirus Epstein-Barr virus (EBV)* HHV-8 HVS HVA AHV-1 BHV-4+ EHV-2 MHV-68 EBV CCPH + + + er - oe + — vIL-6 + - - - — — = = DHFR + + - - - ~ - - TS + + + - - + - - CC-chemokines 3 - - — — = = _ bel-2 family protein + + + + + - + + Interferon + - — - - oe — _ Responsive factor vFLIP + + + om + + — = D-type cyclin + + + - - 4 - N-CAM family protein + — - - — - _ _ IL-8R/ger + + + + — 3 + _ FGAM + 2 2 2 + 2 3 + U-RNAs - <7 2 - - - — - Tyrosin kinase interacting protein - + + - —- - — — Protein with collagen motifs - + + - - - = = IL-17 - + - - - - _ _ CD59 - + - - — = — _ IL-10 - - = - n.k. + — + Semaphorin - - - + — - — _ Serpin - - - - = - 3 = *CCPH = complement control protein homologue; vIL-6 = viral interleukin 6; DHFR = dihydrofolate reductase; TS = thymidylate synthase; VFLIP = viral Flice inhibitory protein; N-CAM family = neuronal cell-adhesive membrane protein family; IL-8R/gcr = interleukin 8 receptor or other G-protein-coupled receptor homologue; FGAM = phosphoribosylformylglycinamidine synthase; IL-17 = interleukin 17 homologue; CD59 = CD59 homologue: IL-10 = interleukin 10 homologue; HHV-8 = human herpesvirus-8: HVS = herpesvirus saimiri; HVA = herpesvirus ateles: AHV-1 = alcelaphine herpesvirus type 1: BHV-4 = bovine herpesvirus type 4; EHV-2 = equine herpesvirus type 2; MHV-68 = murine herpesvirus 68. +/= = homologous reading frame present/absent or number of reading frames if more than one is present. n.k. = not known, not present at equivalent position. Numbers in columns represent multiple copies within the genomes. fPositional conserved regions sequenced. cers in humans. Diseases as infrequent as KS, body cavity-based tion, and infection of a non-natural host by animal rhadinovi- lymphomas, and Castleman’s disease would likely have been ruses can be seen as one example of such an accident, revealing overlooked in animal models. All rhadinoviruses are highly the facultative pathogenic potential inherent to the rhadinovi- pathogenic only if certain rare ‘‘accidents’” coincide with infec- ruses. Table 3. Putative functions of cell-homologous human herpesvirus 8 (HHV-8) genes* ORF Gene product In natural lytic or persistent infection In Kaposi's sarcoma, B-cell lymphoma and CD A) Cytokines and cytokine receptors K2 VIL-6 Growth stimulation of infected cells Paracrine/autocrine growth stimulation of spindle/B cells K4 VMIPl« Chemotaxis of persistently infected Chemoattraction of persistently infected cells hematopoietic cells K4.1 vMIP1B ! " K5 vMIPl« ! " 74 vGCR Amplification of natural habitat Dysregulated growth stimulation by constitutive action B) Genes involved in regulation of cell growth or survival 9 Kl KI Positional homologue of rhadinoviral transforming oncoproteins 04 CCPH Protection against elimination by Protection against elimination by complement complement 16 vbel-2 Extension of life span of infected cells Extension of life span of growth- transformed cells K9 VvIRF Counteracting IFN-mediated virus Mimicking transforming effect of IRF-2 or suppression interfering with antiproliferative action of IFN 71 vFLIP Extension of life span of infected cells Extension of life span of transformed cells 72 k-cyclin Amplification of natural habitat Dysregulated growth stimulation C) Genes involved in nucleotide metabolism 02 vDHFR Increasing nucleotide pool for virus Increasing nucleotide pool for cell proliferation replication 70 vTS ? ’ 75 VFGAM-synthase " " *ORF = open reading frame; vIL-6 = viral interleukin 6; vMIPla/B = viral macrophage inflammatory protein o/f3; vVGCR = viral G-protein-coupled receptor: CCPH = complement control protein homologue; VIRF = viral interferon response factor homologue; vFLIP = viral Flice inhibitory protein; k-cyclin = cyclin D homologue of HHV-8/Kaposi's sarcoma-associated herpesvirus; vDHFR = viral dihydrofolate reductase; vTIS = viral thymidylate synthase; VFGAM-synthase = viral phosphoribosylformylglycinamidine synthase; IRF-2 = interferon response factor-2; IFN = interferon. 76 Journal of the National Cancer Institute Monographs No. 23, 1998 References (1) ro ~ (3 — (4 (5 — (6 (7) (8 — (9 — (10) (11) (12) Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Ka- posi’s sarcoma. Science 1994:266:1865-9. Albrecht JC, Nicholas J, Cameron KR, Newman C, Fleckenstein B, Honess RW. Herpesvirus saimiri has a gene specifying a homologue of the cellular membrane glycoprotein CD59. Virology 1992:190:527-30. Telford EA, Watson MS. Aird HC, Perry J, Davison AJ. The DNA se- quence of equine herpesvirus 2. J Mol Biol 1995:249:520-8. Ensser A, Pflanz R, Fleckenstein B. Primary structure of the alcelaphine herpesvirus 1 genome. J Virol 1997:71:6517-25. Russo JJ, Bohenzky RA, Chen MC, Chen J, Yan M, Maddalena D, et al. Nucleotide sequence of the Kaposi's sarcoma-associated herpesvirus (HHVS). Proc Natl Acad Sci US A 1996:93:14862-7. Neipel F, Albrecht JC, Fleckenstein B. Cell-homologous genes in the Ka- posi’s sarcoma-associated rhadinovirus human herpesvirus 8: determinants of its pathogenicity? J Virol 1997;71:4187-92. Meinl E. Fickenscher H, Fleckenstein B. Chemokine receptors and che- mokine-inducing molecules of lymphotropic herpesviruses. Immunol To- day 1996:17:199. Biesinger B, Trimble JJ, Desrosiers RC, Fleckenstein B. The divergence between two oncogenic herpesvirus saimiri strains in a genomic region related to the transforming phenotype. Virology 1990;176:505-14. Albrecht JC, Nicholas J, Biller D, Cameron KR, Biesinger B, Newman C, et al. Primary structure of the herpesvirus saimiri genome. J Virol 1992; 66:5047-58. Jung JU, Trimble JJ, King NW, Biesinger B, Fleckenstein BW, Desrosiers RC. Identification of transforming genes of subgroup A and C strains of herpesvirus saimiri. Proc Natl Acad Sci US A 1991:88:7051-5. Biesinger B, Tsygankov AY. Fickenscher H, Emmrich F, Fleckenstein B, Bolen JB, et al. The product of the herpesvirus saimiri open reading frame I (tip) interacts with T cell-specific kinase pS6lck in transformed cells. J Biol Chem 1995;270:4729-34. Desrosiers RC, Silva DP, Waldron LLM, Letvin NL. Nononcogenic deletion mutants of herpesvirus saimiri are defective for in vitro immortalization. J Virol 1986;57:701-5. Journal of the National Cancer Institute Monographs No. 23, 1998 (13) Duboise SM, Guo J, Czajak S, Desrosiers RC, Jung JU. STP and Tip are essential for herpesvirus saimiri oncogenicity. J Virol 1998;72:1308-13. (14) Kretschmer C, Murphy C, Biesinger B, Beckers J, Fickenscher H, Kirchner T. et al. A herpes saimiri oncogene causing peripheral T-cell lymphoma in transgenic mice. Oncogene 1996:12:1609-16. (15) Murphy C, Kretschmer C, Biesinger B, Beckers J, Jung J, Desrosiers RC, et al. Epithelial tumours induced by a herpesvirus oncogene in transgenic mice. Oncogene 1994:9:221-6. (16) Biesinger B, Miiller-Fleckenstein I, Simmer B, Lang G, Wittmann S, Platzer E, et al. Stable growth transformation of human T lymphocytes by herpesvirus saimiri. Proc Natl Acad Sci US A 1992;89:3116-9. (17) Neipel F, Albrecht JC, Ensser A, Huang YQ, Li JJ, Friedman Kien AE, et al. Primary structure of the Kaposi's sarcoma associated human herpesvirus 8. Genbank accession No. U93872. Moore PS. Boshoff C, Weiss RA, Chang Y. Molecular mimicry of human cytokine and cytokine response pathway genes by KSHV. Science 1996; 274:1739-44. (19) Cheng EH, Nicholas J. Bellows DS, Hayward GS, Guo HG, Reitz MS, et al. A Bcel-2 homolog encoded by Kaposi sarcoma-associated virus, human herpesvirus 8, inhibits apoptosis but does not heterodimerize with Bax or Bak. Proc Natl Acad Sci US A 1997:94:690-4. (20) Thome M, Schneider P, Hofmann K. Fickenscher H, Meinl E, Neipel F, et al. Viral FLICE-inhibitory proteins (FLIPs) prevents apoptosis induced by death receptors. Nature 1997;386:517-21. (21) Sarid R, Sato T, Bohenzky RA, Russo JJ, Chang Y. Kaposi's sarcoma- associated herpesvirus encodes a functional bcl-2 homologue. Nat Med 1997:3:293-8. Fodor WL, Rollins SA, Bianco-Caron S, Rother RP, Guilmette ER, Burton WV, et al. The complement control protein homolog of herpesvirus saimiri regulates serum complement by inhibiting C3 convertase activity. J Virol 1995:69:3889-92. (18 22 Note Supported by the Ria Freifrau von Firtsch Stiftung, the ‘‘Deutsche Kreb- shilfe—Dr. Mildred Scheel Stiftung’ grant No. W134/94/FL2, and European Union grant BMH4-CT95-1016. 71 Novel Organizational Features, Captured Cellular Genes, and Strain Variability Within the Genome of KSHV/HHVS John Nicholas, Jian-Chao Zong, Donald J. Alcendor, Dolores M. Ciufo, Lynn J. Poole, Robert T. Sarisky, Chuang-Jiun Chiou, Xiaoqun Zhang, Xiaoyu Wan, Hong-Guang Guo, Marvin S. Reitz, Gary S. Hayward* Strong serologic and molecular probe correlations indicate that the newly discovered gamma herpesvirus KSHV or HHVS is the likely etiologic agent of all forms of Kaposi's sarcoma as well as BCBL/PEL and MCD in patients with acquired immunodeficiency syndrome (AIDS). Two large segments of HHV8 DNA from an AIDS-associated BCBL tumor covering genomic positions 0-52 kilobase [kb] and 108-140 kb have been cloned, mapped, and partially se- quenced. Our studies have focused on novel viral proteins encoded within a 13-kb divergent locus (DL-B) by nine cap- tured homologues of cellular genes, including vIL-6, vDHFR, vTS, vBcl-2, three C-C beta chemokines (vMIP-1A, vMIP- 1B, and vBCK), and two LAP/PHD subclass zinc finger pro- teins (IE1A and IE1B). The HHV-8 vIL-6, vDHFR, vTS, and vBcl-2 proteins have all been shown to be active in a variety of appropriate functional assays, and transcripts from vIL-6, vMIP-1B, vIE1-A, vIE1-B, and vDHFR genes are all ex- pressed as abundant single messenger RNA species after bu- tyrate or phorbol ester (TPA) induction of the lytic cycle in HHV8-positive BCBL cell lines. All of these genes lie within a divergent transcriptional domain that contains a single central enhancer and associated untranslated leader region plus seven distinct proximal promoters, some of which are negatively regulated through AP-1 and ZRE motifs by the EBV ZTA transactivator. This region also encompasses a predicted complex oriLyt domain of 1050 bp that is dupli- cated in inverted orientation adjacent to the T(.7 latency RNA in another large divergent locus (DL-E). We have pre- viously described three distinct subtypes of the HHVS ge- nome that differ by 1.0%-1.5% at the nucleotide level within the ORF26 and ORF75 genes. Certain strains or clades ap- pear to have preferential geographic distributions, but it is not known as yet whether there are any specific disease as- sociations. Interestingly, the A, B, and C subtypes of HHV-8 also proved to differ dramatically in coding content at both the extreme left and right ends of the unique segment of the genome as well as in the positions of the junctions with the terminal repeats. On the left-hand side, the receptor-like ORF-K1 protein is highly variable with A-strain subtypes displaying 15% amino acid differences from C strains and up to 30% differences from B strains. On the right-hand side, two unrelated alternative types of the putative multiple Journal of the National Cancer Institute Monographs No. 23, 1998 membrane spanning ORF-K15 protein are found. [Monogr Natl Cancer Inst 1998:23:79-88] Introduction Genome Characteristics of Gamma Herpesviruses The complete genomic DNA sequences of five distinct gamma herpesviruses are now available. A comparison of the organization and gene content of Epstein-Barr virus (EBV) (7), herpesvirus saimiri (HVS) (2), EHV-2 (3), HHV-8 (4), and MHV68 (5) is shown in Fig. 1. Major blocks (I, II, III, and IV) of conserved genes present in all herpesviruses are shown and individual conserved genes that have residual homology among all of the five gamma herpesvirus genomes are given as solid bars. Hatched or open bars show those genes that are present in only a subset of these viruses or are unique to individual viruses. These genes tend to group within six regions referred to as divergent loci (DL) A, B, C, D, E, and F. Prior to the discovery of HHV-8 (6), the human B-cell trophic virus EBV and the squirrel monkey T-cell lymphoma-associated virus HVS were designated as the prototypes of two subgroups of gamma her- pesviruses, namely, yl and y2. A number of old-world primate viruses similar to EBV are also clearly yl viruses and several other new-world primate y2 viruses (also known as Rhadinovi- ruses) are closely related to HVS. More recently, several equine (EHV-2), bovine (BHV-4, AHV-1), and murine (MHV68) vi- ruses have also been tentatively assigned to the y2 subclass based on data about their gene content and genome organization, although a case could be made for assigning EHV-2 to a separate subclass because of its higher GC content and the presence of *Affiliations of authors: J. Nicholas, J.-C. Zong, D. J. Alcendor, D. M. Ciufo, L. J. Poole, R. T. Sarisky, C.-J.- Chiou, X. Zhang, X. Wan, G. S. Hayward, Molecular Virology Laboratories, Oncology Center, The Johns Hopkins School of Medicine, Baltimore, MD; H.-G. Guo, M. S. Reitz, Institute of Human Vi- rology, University of Maryland, Baltimore. Correspondence to: Gary S. Hayward. Ph.D., Department of Pharmacology and Molecular Sciences, The Johns Hopkins University, 725 N. Wolfe St., WBSB 317, Baltimore, MD 21205. E-mail: Gary.Hayward @qmail.bs.jhu.edu See “Notes following ‘ ‘References.’ © Oxford University Press 79 ABC Dt D2 D3 EF 3 i tae Lady EBY HVS EHV-2 HHV-8 IT 73 io i eam tema — 100 120 130 160 180 Fig. 1. Comparative organization of the genomes of five gamma herpesviruses. The scale diagram shows the relative size, location, and leftward or rightward orientation of all of the known or predicted genes of each virus based on com- plete DNA sequence information for human y1 Epstein-Barr virus [EBV: (/)]: the prototype simian y2 virus of squirrel monkeys, herpesvirus saimiri [HVS: (2)]; equine herpesvirus two [EHV-2: (3)]; Kaposi's sarcoma-associated herpes- virus or human herpesvirus eight [HHV-8; (4)]; and murine herpesvirus one IMHV68: (5)]. Genes within conserved blocks labeled 1, II, 111, and IV are common to all mammalian and avian herpesviruses of the alpha. beta, and gamma classes. Each individual DNA genome is drawn as a solid horizontal line with open boxes representing large terminal and internal direct or tandemly repeated domains. Individual genes with evolutionary homologues in all se- large direct terminal repeats rather than the otherwise typical multicopy short terminal repeats (3). The genome organization of EBV differs from that of HVS primarily by the presence of the unique B-cell latency genes and other features, including ori-P, EBNA-1, EBNA-2, EBNA-3abc, LPNA, EBERs, and LMP-1 and 2 as well as the B-cell receptor glycoprotein, the lytic cycle triggering DNA-binding bZip pro- tein ZTA, the large (10 x 3.1 kilobase [kb]) internal IR repeats, and the duplicated ori-Lyt domains (DS-L and DS-R). In their place, HVS has genes encoding the TIP, STP, and vDHFR pro- teins together with the HSUR small RNAs at the left-hand end (DL-A region) and the vTS(ORF70), vFLIP(ORF71), vCycD(ORF72), ORF73, and an IL8R-like vVGCR (ORF74) gene in DL-E. HHV-8 is clearly not a y1 virus and it has some features that closely resemble HVS, such as the VFLIP (K13) (7), vCycD, LANA(ORF73), and vGCR gene block in DL-E (8), and the presence of short multicopy terminal tandem repeats, but it also differs significantly from HVS by its higher overall GC content (55%), lack of CpG-suppression, insertion of the IRF gene block (DL-D3), and major additions or alterations in sev- eral of the divergent loci, especially in DL-A, DL-B, and DL-F. At the present time, equally compelling arguments can be made 80 quenced gamma herpesviruses are shown as solid bars and those that are present in only one or several of the five gamma herpesviruses are given as open or hatched arrows. The standard orientation of EBV used by Baer et al. (7) has been inverted to match that of the other genomes. Size scales in kilobase pairs are given for EBV at the top and for EHV-2 at the bottom. The major divergent loci (DL-A, B, C, DI, D2, D3. E. and F) are indicated by gray shading and vertical dashed lines denote positions of homologous genes within these loci. Small solid or hatched boxes and arrows depict short GC-rich repeats and inverted palin- dromic areas, whereas solid and open circles indicate structurally complex do- mains with known or presumed origin-like features, e.g., ori-Lyt(L) or ori-Lyt(R) (@). and ori-P (O) in EBV and the duplicated proposed ori-(L) and ori-(R) domains in HHV-8 (@). either for classifying HHV-8 along with all of the other non- EBV-like gamma herpesviruses together in a super y2 subgroup or for giving both HHV-8 and EHV-2 separate status as y3 and v4 viruses. Alternatively, it might be wise to abandon any at- tempt at rational subdivisions within the gamma herpesviruses. Novel Features of KSHV/HHV-8 Major features of the HHV-8 genome were described previ- ously by Moore et al. (9), Cesarman et al. (8), Russo et al. (4), and Nicholas et al. (7/0). These include the lack of all EBV- specific features and the absence of most HVS specific features, except for the presence of vTS and vVDHFR genes (both at new locations) and retention of the vFLIP, vCyc-D, LANA, and vGCR block. Several vIRF-like genes (e.g., ORF-K9) (/7) and other potential novel genes (ORF-K8, K8.1, and K9-K11) lie in the DL-D2 and D3 regions, and a vOX-2 homologue (ORF-K 14) has been inserted adjacent to the vIL8R-like vGCR gene in DL-E. A series of novel viral-encoded cytokine genes map within DL-B e.g., vIL-6 (K2), vMIP-1A (MIP-1, K6), vMIP-1B (MIP-1I, K4), and vBCK (K4.1), together with vTS, vDHFR, and two LAP/PHD class zinc finger protein genes referred to as IE1-A (KS) and IE1-B (K3) (10-13). Two novel abundant short Journal of the National Cancer Institute Monographs No. 23, 1998 RNAs (T1.1/Nut/K7 and TO0.7/K12), which are unlike the HSURs of HVS or the EBERs of EBV, have been described (14,15) that also map within DL-B and DL-E, respectively. Fi- nally, two small novel open reading frames (ORFs), KI and K15, occur at the left and right ends of the HHV-8 genome (DL-A and DL-F) and are positional equivalents of the latency and transformation-related LMP-1 and LMP-2 membrane gly- coprotein genes of EBV and the strain variable TIP and STP genes of HVS (16,17). Many of these genes appear to be likely to contribute to the unique biology of HHV-8, but evidence for any direct involve- ment in disease processes must await extensive analysis of the patterns of gene expression in different cell types in Kaposi's sarcoma (KS) and other tumors, etc., and detailed examination of the functions of their protein products. Materials and Methods The sources of the body cavity-based lymphoma (BCBL) and KS DNA samples and cell lines used have all been described previously, as well as the procedures used for library prepara- tion, subcloning, and genomic or PCR sequencing of HHV-8 DNA (10,13,18-20). Procedures used for transcriptional and functional analysis of vIL-6 were also described previously (13). Details of the functional analysis of vTS and vDHFR, the tran- scriptional evaluation of the DL-B region, and the sequence data for the LHS and RHS ends of HHV-8(BCBL-R) will be pre- sented elsewhere. New HHV-8 phage lambda clones not previously described include N\D-S1 (—1.8/14.5-kb), NE-A2 (108.3/120.5-kb), and \E- C2 (115.7/129.0-kb), all in the ADASHII background, and AB3-2 (120.6/137.4) in the AEMBL3 background (Fig. 2, A). These were all isolated from either of two phage libraries gen- erated from size selected Sau3A partial digests of a BCBL tumor DNA sample described previously that we refer to as BCBL-R (10,19). AD-S1 was identified by hybridization to a probe rep- resenting the right-hand terminus of AD3-80(ORF6), AE-A2 by hybridization to a T0.7 gene probe generated by polymerase chain reaction (PCR) based on sequence data from Zhong et al. A HHV-8(BCBL-R) PHAGE LAMBDA CLONES 0 20 40 60 80 100 120 140 TTR ORI-L ORI-R TTR a . o—— TH K2 -K7 K8 K9-Ki1 K12-K14 K15 Fig. 2. (A) Diagram illustrating the map locations of two sets of 0 [] [ [] [] 0 overlapping phage lambda clones of HHV-8(BCBL-R) relative to DL-A DL-B DL-D2 DL-D3 DL-E DL-F the positions of the divergent loci (DL) and other key features of the genome. The six divergent loci containing gamma-2 specific or D-51 4 > E-A2 — HHV-8 specific genes (K numbered ORFs) are shown by open D3-80 ¢— EC2 —b boxes. The relative positions (in kbp) of 11 selected phage lambda D-VR3a ¢—» 0 clones characterized in this study are shown as horizontal arrows. D-VR4a ¢— TTR indicates the 801-bp tandem repeats at the termini (4). (B) G7 B22 « ’ Structural organization of the genes and repeated regions in the HHV-8 divergent loci DL-B and DL-E relative to predicted pro- C21 ’ moter control elements and origin-like features. The diagram shows MZ =—4 preliminary transcript maps of the 13-kb divergent transcription B domain in DL-B (LHS). and the area around the inverted duplica- | PREDICTED CONTROL ELEMENTS AND ORIGINS tion of the 1050-bp ori-like domain in DL-E (RHS). Map coordi- nates are given in parentheses. Open arrows show the size and ORF11 orientation of ORFs and solid arrows show the two abundant RNA > <0 species referred to as T1.1 (lytic) and T0.7 (latent). Multiple adja- A S—\ Se 1, =r cent vertical lines indicate four sets of short GC-rich tandem repeats viL-6 \DHFR 1E1-B(Zn) VMIP-1B (BCK) (ORF-X) ORI-(L) DL-B and solid circles represent recognized promoter domains containing K2 ~~ ORF2 K3 a Ka Kadai K42 US closely packed TATAAA-like, ATE, SP-1, AP-1, CTF, or ZRE (K7) ORF16 ifs sed di nt enhancer (c z ifs i T1.1 RNA vBCL-2 motifs. The proposed divergent enhancer (core SRE motifs in DL . . (17,050- B) and the origin-like domains are shown as open boxes. Open 1 > i > 30,750) circles indicate the core origin (TA), motifs and solid arrowheads 23x20 bp IE1-AZn) wMIP-1A ENH Hx1abp ORF17 indicate three copies of an approx 90-bp region, one representing 5x30 bp Ks Ke6 (SRF) the leftwards upstream promoter and leader associated with the DL-B ENH domain, and the others are partially conserved versions of the same sequence included in the ori domain duplications. ORF69 R J ZR Ee ORI-(R . TO7RANA 274 231p (R) K13 ORF72 DL-E (K12) VFLIP vCYC-D RHS K14 ORF74 vOX-2 vIL-8R (117,430- (Complex Repeats) 00 —— > 130,500) LT ——————————————————————— < ORF73 (ENH?) ORF75 LANA Journal of the National Cancer Institute Monographs No. 23, 1998 81 (14), N\E-C2 by hybridization to a probe generated from the right-hand terminus of AB6-1(ORF71), and AB3-2 by hybrid- ization to a probe representing the left-hand terminus of AB6-1 (ORF75). The map coordinates of those and our previously de- scribed BCBL-R phage lambda clones (70,19) were confirmed by restriction enzyme cleavage mapping and terminal sequenc- ing to correspond to genome positions in the HHVS(BC-1) ge- nomic DNA sequence of Russo et al. (4) as defined in paren- thesis as follows: AD-VR3A (14.7/26.7 kb) and AD-VR4A (18.5/31.2 kb) in the ADASHII background and AD3-80 (5.4/ 22.9 kb), AC7-1 (19.3/37.9 kb), NC2-1 (27.8/47.2 kb), NA12-1 (35/51 kb), and AB6-1 (121.7/134.4-kb), all in the AEMBL3 background. New plasmid subclones used for additional se- quencing included pDJA6O (BamHI/EcoR]1 2.9 kb encompassing ori-L), pDJA61 (BamHI/BamHI, 3.5 kb encompassing the LHS TTR, ORF-KIA, and part of ORF4), pDJA62 (HindIll/Sall, 3.3 kb encompassing ORF-K15A and the RHS TTR), plus pDJA63 (EcoRI/EcoRl, 6.0 kb encompassing T0.7 and ori-R). Results Extension of the BCBL-R Phage Library We have described previously the identification and sequenc- ing of nine novel captured cellular genes mapping between ORFI1 and ORFI17 within DL-B on the LHS of the HHV- 8(BCBL-R) genome, as well as vCycD (ORF72) and an IL8R- like VGCR(ORF74) within DL-E on the RHS (10,13,19-21). In the course of that work, two sets of contiguous phage lambda clones encompassing ORF6 to ORF31 (now defined as map coordinates 5.4-51 kb) and ORF72 to ORF75 (map coordinates 120.6-134.4 kb) were assembled. These two blocks have sub- sequently been extended both internally between ORF64 and ORF72 to completely cover the remainder of the DL-E region as well as toward the ends of the genome to encompass DL-A at the extreme LHS boundary and DL-F at the extreme RHS boundary, including proximal copies of the terminal tandem repeats (TTR). New phage clones AD-S1, AE-A2, NE-C2, and AB3-2 were iden- tified by plaque hybridization with probes representing the left- hand side (LHS) end of AD3-80 (70), the TO0.7 transcript (14), and the right-hand side (RHS) end of AB6-1 (79), and short regions at the termini of each insert were sequenced. A summary of the map locations of a total of seven characterized phage lambda clones covering 53 kb at the left end (positions —1.8 to 51 kb), and four other selected lambda clones covering 30 kb at the right end (positions 108 to 138 kb) of the HHV-8(BCBL-R) genome are shown in Fig. 2, A. An Inverted Duplication Contains Potential Origin Features Our analysis has revealed the presence of a 1050-bp domain in DL-B between ORF-K4.2 and IE1-A (KS) that has typical features of a herpesvirus lytic cycle DNA replication origin, including clustered consensus binding motifs for AP-1 (JUN/FOS) and CTF, multiple short repetitive motifs contain- ing Xcal and Pyull sites that resemble features of EBV ori-Lyt, two Fspl/Sphl motifs matching sites found within CMV ori-Lyt, and two long AT-rich palindromes, including ATATATAT- ATATATATAT, which resemble the core loop structures of HSV ori-L and ori-S and are also found in the HHV6 and HHV7 82 lytic origins. Most remarkably, the entire 1050 bp is duplicated as an almost identical inverted copy between T0.7 (K12) and vELIP (ORF71) toward the other end of the genome within DL-E. Notably, the positions of the two putative ori-Lyt do- mains in HHV-8 closely match those of the two directly repeated 800-bp ori-Lyt domains in EBV and, as in EBV, both also lie directly adjacent to high GC-content tandem repeats. In HHV-8, these repeats are represented as both 20-bp and 30-bp tandem arrays in the LHS region (DL-B) and as two types of 23-bp tandem repeats in the RHS region (DL-E), and they are posi- tionally analogous to the 102-bp PstI and 125-bp Norl tandem repeats in EBV, respectively. Interestingly, MHV68 also has equivalent sets of 40-bp and 100-bp GC-rich tandem repeats at these locations, and EHV-2 has a total of four sets of both direct and inverted duplicated palindromic regions with (AT), core loops and multiple AP-1 binding sites on the stems within or adjacent to the DL-B and DL-E regions. Only HVS appears to lack these features. The locations of the predicted LHS ori-Lyt domains within DL-B(ori-L) and DL-E (ori-R) are shown in Fig. 2, B. Evaluation of Transcriptional Control Elements We recognize eight distinct promoter domains within the 13- kb DL-B region, including a 500-bp divergent promoter/ enhancer (ENH) between vMIP-1A and T1.1 that functions as a relatively weak constitutive plus butyrate and TPA-inducible promoter in both orientations in transient reporter gene assays in Vero, HeLa, and U937 cells. This domain contains a predicted rightwards noncoding leader exon of 45 bp that together with its upstream TATA-box region is duplicated in a partially con- served manner within the two proposed ori-Lyt domains. In fact, within the DL-B region, the arrangement of the leftwards- oriented leader (ENH version) together with the vMIP-1A and IE1-A genes, followed by the leader (ori-L version) plus vMIP- IB and IE1-B genes in the same orientation, suggests another ancient duplication that may perhaps have occurred prior to the acquisition of the other inserted cellular complementary DNAs (cDNAs) in DL-B (Fig. 3). Note that the 13-kb DL-E domain also appears to represent a divergent transcriptional unit with a single potential ENH-like domain associated with several rec- ognizable transcription motifs between ORF73 (LANA) and ORF-K14 (vOX-2), but no functional data are available here as yet. Curiously, both orientations of the divergent ENH-like control domains in DL-B and DL-E contain consensus 7-bp ZRE bind- ing motifs (22,23) either very close to or slightly downstream from the TATA motifs, and both orientations of the DL.-B ENH proved to be down-regulated by cotransfection with the EBV ZTA transactivator in Vero cells (Chiou C-J, Hayward GS: data not shown). Many of the proximal promoters also contain adja- cent ZRE motifs either upstream or downstream of their TATA motifs, and several of the potential ZREs in DL-B are also classical JUN/FOS binding AP-1 sites. These findings raise the possibility that some key HHV-8 lytic cycle promoters may be regulated negatively by EBV in dually infected BCBL cell lines and tumors, and they also reinitiate old unsolved questions about whether there may be a cellular homologue of the EBV lytic cycle triggering protein ZTA, which HHV-8 may use. Expression of messenger RNAs (mRNAs) from most of the Journal of the National Cancer Institute Monographs No. 23, 1998 POTENTIAL DUPLICATED ORI-LYT DOMAINS Fig. 3. Detailed comparison and structural features of the duplicated HHV-8 ori-like domains in DL-B (inverted ori- [GC-Rich Fun Li) entation) and DL-E (forward orientation). Arrows, boxes, Repeats] 3x AP-1 ¢ . RR 6x CTF Dra and symbols in the upper segment denote various subre- 0 ° ° op O ros] ; hy 4: JH 0 [wowroy O=@p 0 Oce——> gions and characteristic motifs, and the solid lines in the I - a 0 SP 0 2xSP-1 Apt | oxate SPT| 0 ORFKa2 lower segment indicate the positions of significant nonho- TATA y TATA x TATA mologous blocks within the left-hand copy (above the line) 18 bp 16 bp compared with the right-hand copy (below the line). The o 30 AT PAL AT PAL: he 36-bp insertion in ori-(R) compared with ori-(L) represents S 2 part of a 7 x 14-bp set of diverged repeats of the Xca/Pvull a 1,053 bp LHS (INV) & motif. To the right of the insertion, the two copies in both Bi ; s irtuz antics oN 1,076 bp RHS (F) © HHV-8(BC-1) and HHV-8(BCBL-R) are virtually identical S 5 over 500 bp, whereas to the left of the insertion the left and © ins 36 2 I 2 25 w= right copies are approximately 95% identical outside the - 7x14 bp diverged blocks near the left end. ps = od Gl fr fo fot (sm Cm) tm PW em em fm m= Ee Ee = > 90% homology 100% homology genes in DL-B (e.g., vDHFR, vTS, vMIP-1B, IE1-B, IE1-A, and T1.1) has been found to occur only after butyrate induction of the HHV-8 lytic cycle in the HBL-6/BC-1 (EBV-positive) cell line and after either butyrate or TPA induction in the BCBL-1 (EBV-negative) cell line. In contrast to TPA induction, butyrate induction appears to be abortive (to the extent that the viral mRNAs are induced only transiently and disappear after be- tween 24 and 48 hours), and curiously TPA does not induce the lytic cycle genes efficiently in the dual EBV- and HHV-8- infected HBL-6 cell line (10,24,25). vIL-6 gives low-level con- stitutive expression in the HBL6/BC1 cell line but is also greatly increased by induction (//,13). Both the predominant vIL-6 (1.0-kb) and vMIP-1B (0.8-kb) mRNAs produced after induc- tion have been found by primer extension analysis to represent short monocistronic unspliced mRNA species (Poole L, Ciufo D, Hayward GS: data not shown). However, both the IEI-A and IEI-B genes, which lack identifiable proximal promoters, con- tain excellent splice-acceptor motifs, and the vIL-6 and vMIP- 1A genes also appear to have the potential to occur as either spliced or unspliced versions (Fig. 3). Functional Activity of the Viral IL-6, Dihydrofolate Reductase (DHFR), and Thymidylate Synthetase (TS) Genes Considering that the multistep captures of those cellular cDNAs are likely to have been relatively ancient evolutionary events, most such genes that are retained are expected to be functionally active, although they may have altered their func- tions in subtle ways either by broadening the substrate or target specificity or by acquiring constitutive rather than conditional activity that may lead to interference with normal host mecha- nisms or apoptotic or immune responses, etc. We have previ- ously shown that the HHV-8-encoded vIL-6 produced from an SV2-vIL-6 plasmid in transfected rat embryo fibroblasts substi- tutes for human IL-6 in promoting growth of B9 mouse my- eloma cells in a manner that uses the interleukin 6 (IL-6) recep- tor (13). It also mimics human IL-6 by inducing specific mRNA and DNA-binding transcription factors associated with the acute-phase response in HEP-3B cells [(/3); Wan X.-Y., Nicho- las J: manuscript in preparation)]. Some properties and functions of the HHV-8-encoded Cyc-D (ORF72), the IL-8R-like vGCR Journal of the National Cancer Institute Monographs No. 23, 1998 (ORF74), vMIP-1A (K6), and vIL-6 (K2) from HHV-8(BCI) have also been described by others (8,11,26,27). We have also more recently cloned the HHV-8-encoded vDHFR and vTS genes into bacterial glutathione S-transferase (GST) and mammalian Flag-tagged vector systems (Sarisky R, Ciufo D, Zhang X, Chiou C.-J, Hayward GS: manuscript in preparation). Stable DHFR-negative CHO cell lines constitu- tively expressing the tagged VDHFR have been generated by Neo co-selection and these proved to be susceptible to cell kill- ing with either 10 or 100 mM methotrexate. Furthermore, a methotrexate-resistant high copy number subclone efficiently bound to a fluorescent tagged dihydrofolate substrate as assayed by fluorescence-activated cell sorter analysis. Similarly, a trans- fected GST/VTS fusion protein plasmid proved both to rescue a thymidylate synthetase-negative Escherichia coli strain in Thy™ medium and to make the host cells susceptible to killing by fluorouracil. All of these properties indicate that the vIL-6, vDHFR, and vTS genes retain the normal functional activities of their cellular homologues, although we have yet to discover whether or not their substrate or target specificities have been altered. Evidence that the HHV-8 encoded vBcl-2 (ORF16) pro- tein functions similarly to human Bcl-2 to block apoptosis has also been presented (20,28). Terminal Heterogeneity on the Right-Hand Side Selective sequencing of relevant terminal segments of HHV- 8(BCBL-R) DNA subcloned into plasmids from the phage lambda clones AD-S1, AE-A2, and AB3-2 has also been carried out (Alcendor D, Zhong J.-C, Wan X, Nicholas J, Guo H, Reitz M, et al: manuscript in preparation). At the extreme RHS end of the unique segment of the HHV-8(BCBL-R) genome a 3.3-kb HindIII to Sall insert from AB3-2 subcloned in plasmid pDJA62 was completely sequenced by standard M13 shotgun procedures to extend our previous UPS75 DNA sequence (Genbank acces- sion no U85269). The results revealed a match to the published HHV-8(BC-1) sequence across the N-terminus of ORF75, but with a pattern of increasing nucleotide differences for 400 bp between positions +120 to —280 relative to the initiator codon, followed by 2300 bp of completely different sequence toward the end of the unique region, and then returning to a perfect match (for at least 700 bp) with the 801-bp HHV-8(BC-1) TTR 83 unit sequence (Fig. 4). The published HHV-8(BC-1) sequence contains 3067 bp of non-TTR sequence upstream of the initiator codon for ORF75 but is incomplete at the junction between the unique and TTR domains because of an unstable (apparently unclonable) G plus A-rich variable repeat region estimated to be 2-3 kb in size (4). In contrast, our HHV-8(BCBL-R) sequence contains 2579 bp between the ORF75 initiator codon and the junction with the first copy of the TTR, which begins at position 652 within the complete 801-bp TTR sequence given by Russo et al. (4). There are no G plus A-repeats at the TTR junction in BCBL-R, but we believe, nevertheless, that this represents the complete undeleted form because PCR amplification with prim- ers that span the junction from the unique region into the RHS TTR gave identical-sized products from the original BCBL-R tumor DNA sample, the AB3-2 phage DNA, and the pDJA62- subcloned plasmid DNA. On the basis of our previous identification of three distinct subtypes of HHV-8 in the ORF26 and ORF75 gene regions (1/8), the published sequence for HHV-8(BC-1) (4) indicates that it represents a C subtype at the RHS end but an A subtype else- where, whereas the HHV-8(BCBL-R) genome is an A subtype throughout. DNA from the RHS of our prototype C strain proved to be identical to that in both the BC-1 and HBL6 cell lines in two distinct KS lesions from this patient (ASM72/76) for as far as at least 1200-bp upstream from the ORF75 initiator codon. In contrast, all three other A strains tested (C282, AKSI, and BCBL-1) proved to be nearly identical to BCBL-R between ORF75 and the TTR, although our prototype B strain (431 KAP) differed by the presence of both a 10-bp insertion and a 55-bp deletion across the unique sequence/TTR boundary. Therefore, whereas the HHV-8 C strains are dramatically different from the A and B strains over at least 2500 bp (and perhaps up to 5000 bp) at the extreme RHS end of the unique region, B strains proved to be only slightly different from the A strains in this region (Fig. 4). Interestingly, both the A/B and C strain unique regions up- stream and to the right of ORF75 have the potential to encode complex multiply spliced leftwards-oriented mRNAs for hydro- phobic membrane proteins that have an equivalent position and orientation to that of EBV LMP-2. Attempts to characterize the very different ORF-KI5A/B and ORF-KI15C transcripts and gene products are in progress. Terminal Heterogeneity on the Left-Hand Side We also carried out plasmid and PCR primer-based sequenc- ing over a 1500-bp block from genomic nucleotide positions 40-1540, encompassing the entire ORF-K1 and the C-terminal segment of ORF4 for BCBL-R (pDJA61) and eight other HHV-8 samples from various sources. Although this segment of the conserved ORF4 coding region showed only three nucleotide variations over 400 bp among A, B, and C strains, which is equivalent to the level found for ORF26 and ORF75 (i.e., ap- proximately 1% differences: 18), the 840-bp (289 amino acid [aa]) ORF-K1 coding region proved to display much greater variability (Figs. 4, 5, and 6). Most dramatically, our prototype B strain HHV-8(431 KAP) differed by 14.6% (127 positions) at the DNA level and 29% (84 amino acids) at the protein level from the published data for HHV-8(BC-1). Similarly, our pro- totype C strain HHV-8(ASM72/76) differed by 5.8% (49 posi- tions) at the DNA level and 14% (39 amino acids) at the protein level from HHV-8(BC-1). The B and C strains also differed from each other at 125 nucleotide positions (and 83 amino ac- ids), although again the subtype C DNA genomes present in two distinct lesions from the same patient (ASM72 and ASM76) were identical throughout. The HBL6 and BC-1 cell lines were derived independently from the same patient and both are A subtype sequences throughout (except at the RHS or ORF-K15 region) and as expected they proved to have identical ORF-K1 DNA sequences. Among the other A strains tested, BCBL-R exhibited 19-bp (2.0%) and 13 amino acid (4.5%) changes, BCBL-1 gave 27-bp (2.8%) and 22 aa (8%) changes, and C282 gave 16-bp (1.7%) and 11 aa (4%) changes relative to HHV- 8(BC-1). However, the ORF-KI regions in BCBL-R, C282, AKSI1, AKS2, and AKS4 differed by only 2-5 bp relative to one another and all are considered to be members of the Al substrain or clade. Even the junctions between the LHS unique sequences and the TTR differed significantly between BCBL-R(AI), BCBL-1(A3), and BC1(A2), but no data about the LHS junc- Fig. 4. Locations of major heterogeneous domains in the DL-A and DL-F regions at the extreme left and right geno- HHV-8 TERMINAL HETEROGENEITY mic termini of the A, B, and C Substrains of HHV-8. ORFs are depicted as open arrows and the most proximal complete LHS RHS and partial (where known) terminal repeat units are indi- a Soahy large ope ; boxes. Multiple potential S¥e88 in ORE: BoBL1 801 KIA ORF4 ORFs K14.1 KisA 148 801 5 are shown as small open boxes. Differences in orga- A goELR [| [==> / _ Da Ia BCBL1 nization and coding content of strains A, B, and C at the BC-1 Cd 0000 BIBLA 289aa 2579 termini include the following: 1) the LHS junctions of the 20) KiB 7oaa ASE 114 (801) TR and unique regions differ even between the Al, A2, and B 431KAP i > / Po jm an [7] 431KAP B A3 substrains; 2) the protein coding content of ORF-K1A 289aa - 75aa 2534 and ORF-KIC differ by 15% at the amino acid identity K1C KIEC 801 level, whereas ORF-KIB differs from both of the other C ASM72 [7] ——o / {ome alo (R) Aur Cc subtypes by nearly 30%. ORF-KIC genes also have a con- 282aa - 110 aa? sor GIA sistent five amino acid deletion; 3) a potential short right- (15-30%aa) HT (+2000?) 801 wards ORF-K 14.1 coding region differs in size and coding wr content in the C strains compared with A and B strains; 4) Boundaries I3 W890 #145 +980 +360 -50 -280 the entire unique 2.5- to 5-kb region between ORF75 and ?) | | [| | the TTRs, which encompasses the spliced ORF-K15 genes, Nucleotide 114 8 6 14 42 2300 2 0 is different in some C strains compared with A and B Variations 14% 1.5% 0.6% 4% 15% (100%) 0% strains; and 5) the predicted splicing patterns of ORF-K15 ¢ > < > and the positions of the unique/TTR junctions on the RHS DL-A DL-F differ between A and B strains. 84 Journal of the National Cancer Institute Monographs No. 23, 1998 HBL6 / BC1 BCBL1 A BCBLR A BCBLR C282 AKS4 AKS4 EKS1 AKS2 EKS1 AKS1 AKS5 AKS1 1 AKS5 ST1 AKS2 BCBL1 C282 Fig. 5. Summary of the levels of nucleotide differences ob- 4 4 6 16 served between sequenced HHV-8 A, B, and C strains within B C B four distinct coding regions of the genome. The first three lo- C cations, namely, ORF26 and two adjacent blocks within ORF75 oa 2 STH oa 14 referred to here as ORF75(N) and ORF75(C) (18), display typi- ASM72/76 cal 0.8%—1.5% nucleotide variations between the A, B, and C 1 431KAP ASM72/76 1 a3tkap HBL6 / BC1 strains with 10-fold less among isolates from within the same strain designation. However, the ORF-K1 region displays 10- to 20-fold greater variation, with C strains differing by 15% at the ORF26 ORF75(N) DNA tevel an op to oe the protein level trom A and (500 bp) (750 bp) strains, and with C strains differing from A strains by up to 8% at the DNA level and 15% at the protein level. Even within the (116 aa) (238 aa) A strains, distinct Al, A2, and A3 ORF-KI substrain patterns can be discerned, with differences of 2%-3% at the DNA level and 4%-8% at the protein level between them, but falling to A BCBL1 Al between 0.2% and 0.5% nucleotide differences at the DNA BCBLR level within the Al subgroup. Individual patient samples tested ARSE AKS4 o 7 ings ward . C282 AKS4 3 Cog represent the following: U.S. AIDS-associated BCBL tumors EKG! AKS2 SOE ANSE (BCI/HBL6, BCBL-1, and BCBL-R); African classical KS STI AKSt A3 8% AKST 40, AD (431 KAP) and African AIDS-associated KS (STI, ST2, and BCBL1 16 ° ST3), U.S. classical KS (EKSI1); East Coast U.S. AIDS- 27 HBL6 / BC1 associated KS (C282, AKS1, AKS2, AKS4, and ASM72/76). Note that three of the 13 genomes shown (namely, BC-1/HBL6, STI, and EKS1) represent chimeras with the structures A/A/C, 28% 15% B/C/A, and C/A/A, respectively, within their LHS (ORF-K1), 7 13 ASM72/ 76 central (ORF26), and RHS (ORF75N and ORF75C) blocks. B ST2 10 C B 1 573 oe EKS1 #BIRAP ASM72/76 £31KAP HBL6 / BC1 8 ST2 ORF75(C) ORF K1 (760 bp) (960 bp) (2565 aa) (289 aa) tions are available as yet for the prototype B and C strains (Fig. 4). Summary diagrams comparing these ORF-K1 variation lev- els with those in the three previously analyzed more conserved regions in ORF26, ORF75(N), and ORF75(C) (18) are presented in Fig. 5. The potential 289 aa membrane protein encoded by the pre- dicted unspliced ORF-K1 gene is a positional and orientation counterpart to LMP-1 of EBV (Fig. 1), but it lies in an inverted orientation relative to the highly variable STP (and TIP) trans- forming proteins of HVS, and it does not closely resemble any of these known gamma herpesvirus transformation related pro- teins in structure or amino acid content. The fact that all versions tested remain in frame, despite the presence of both a 15- and a 6-bp deletion in the prototype C strain, attests to the presumed functional importance of the ORF-K1 gene product. Interest- ingly, the major features of 12 Cys residues and 8 N- glycosylation sites within the N-terminal 175 amino acids of ORF-K1 are nearly fully conserved and hydrophobic blocks be- tween positions 1-22 and positions 227 and 251, which are not altered significantly by the strain variations, probably serve as a signal peptide and a membrane anchoring transmembrane do- main, respectively (Fig. 6, A). Therefore, the bulk of the ORF- Journal of the National Cancer Institute Monographs No. 23, 1998 K1 protein has the appearance of an extracellular receptor that displays a similar level of variation to that encountered in the HIV ENV protein. In contrast, the smaller C-terminal cytoplas- mic domain is identical between most A and C strains, but differs by 30% at the amino acid level in the B strains. An illustration of the extent of the amino acid changes over the two most variable extracellular domains of the ORF-K1 protein be- tween amino acids 52-92 and between 191 and 228 is shown in Fig. 6, B. Discussion and Conclusions Although the recently characterized herpesvirus KSHV or HHV-8 (6) has not yet been demonstrated directly to be trans- missible from tumor cells to other target cells or from patient to patient, most of the epidemiologic, serologic, and molecular bi- ology evidence currently available suggests that it is likely to be an essential infectious causative agent of both classical and AIDS-associated KS (29-34). Herpesvirus particles can be in- duced from cultured HHV-8-positive EBV-negative BCBL cell lines (25) and both the lytic and latent HHV-8-encoded abundant RNAs (T1.1 and T0.7) can be detected in spindle cells and mononuclear cells in KS lesions (8,14,35,36). The genome of 85 , Extracellular Cytopl Signal Domain ™ “Fail 1 21 227 251 289 “V Ne N-gly AA asm a \ A Deleti cvs © eee @ eee ee CON 52 V1 gp 191 V2 208 y: | 162771 1 [ rm | | | I I I A/B 27% | 60% | 15% | 60% | 30% A/C 10% I 30% | 5% I 33% | 0% i I | A ORF-K1 Protein B | STRAIN VARIABILITY IN KSHV ORF K1 52 92 CNNTRLFRPJTEfTLFPVTIACNFTCVEQSGHRQSIWITWHA BCBL-R Al {cunTrirRLTERTRFBN TI ACKF TCU EGS GHROS TW TWA C282 CNNTRLFRLTERTIFPVTIACNFTCVEQSGHRQSIWITWHA AKS 1 A2 CNNTRLUUORLTERRVILD] IACNFTCVEQSGHRQSIWITWRA BCI/HBLS6 A3 CDONTRLER LTP LIIVIJICNF TCVEQSGHRQS IWI TWHA BCBL- 1 { CONTR LFRLTHBHFTY vNFIEN FiSlc vias GH RES LWHIT WEE] AsM 72176 cBINT R LR LTKDITEIV vVNFIICN FSlc vices 6 HRHS [WMT WET] ES 1 {ROBIN Ta NF TCHS GPITS 1 w 1 BWI] 43 1kAP CENT R LER(TAENLTVEEILITIcN F 7 CHIT GPITS 1w 1 Twi sT1 191 231 TGFRTFSTNSLVNITHATTHDVVVVKEAKS THF HI ELHF LV Al § TGFRTFSTNSLVNITHATTHDVVYVKEAKSTHFH I EHF LV TGFRTFSTNRILVNI THATTHDVVV[OKEAKS TSF H 1 EHF LV A2 TGFRTUSTNS LVKII IHATTRIDVVVVKEAKS THF HI EVHF LV A3 TGFRTFSTNSLVNITHATTHKVVVVKEAKSTNEHIEV[AF Lv BCBL-R C282 AKS 1 BC1/HBL6 BCBL-1 ASMT2176 EKS 1 [1OFRTFSTNSE== RATT ADVE KEAGETNEN 1 EVPF LV TGFRTFSTNSE-———HATTDVVVVKEAKFITNPHIEVPF LY oi NGLLK]T 1PJAT THARVAVEEMKS TNFHIEVPFLY 43 1KAP MAVIGVILRIT NGIL[LK]T 1[PIAT T HARNVAVEEMKS TNFHT EVP FLY ST Fig. 6. Organization of the HHV-8 ORF-K1 protein and illustration of the level of amino acid divergence observed within and between the A, B, and C strains. A) Domain structure diagram of the predicted 289 amino acid ORF-K1 protein, illustrating the size and locations of the N-terminal signal peptide and C-terminal transmembrane domain and cytoplasmic tail. The positions of N glycosylation sites (NXS/T) and fully conserved Cys residues in the extracellular domain are denoted by solid triangles and circles, respectively. Overall amino acid identity values (%) between the prototype A(C282), B(431KAP), and C(ASM72) ge- the virus is essentially intact and probably circular in the BCBL cell lines, but these points are less clear for the KS tumor ver- sions (9,36,37). Whether virus-infected cells only provide cytokine support for the other cells in KS lesions or are themselves capable of proliferating is also the subject of considerable debate at present. However, the presence, expression, and functional integrity of the many captured cellular cDNA homologues identified en- hance the plausibility that this virus, like EBV and HVS, prob- ably has the potential to directly influence the proliferation state of its host cells whilst in a latent or nonpermissive mode. Ex- pression of the viral-encoded MIP-1-like inflammatory chemo- kines (even if it occurs only during lytic cycle infection) implies that infected cells may be able to profoundly alter neighboring cells, even to the extent of influencing patterns of HIV infectiv- ity through blocking of the CCRS5 co-receptor (11,13,39,40). Furthermore, it has long been known that IL-6 and related cy- tokines play a major role in the biology of KS spindle cells (41-43), as well as in multiple myeloma (44,45) and MCD (46), and recent evidence by in situ hybridization (Cannon J, Ambinder R: personal communication) and immunohistochem- istry (11) that vIL-6 is expressed in vivo in most BCBL tumor cells greatly increases the likelihood that such scenarios may be valid. Currently, first generation serologic assays for the level of seropositivity to HHV-8 antigens have indicated that the virus 86 nomes are given for the two defined major variable domains (V1 and V2). V* denotes the hypervariable region between amino acids 62-71. B) Amino acid sequences of two 41-residue blocks representing the major variable regions (V1 and V2) are shown for nine distinct patient samples, together with isolate names and substrain designations. Boxed amino acid symbols represent deviations from the commonest residues found at each position and dashes represent deleted residues. may not be widespread outside the AIDS epidemic, except in isolated pockets in areas associated with classical and endemic KS, including some Mediterranean and middle east countries, such as Italy and Greece, and in equatorial Africa, including Uganda and Zaire (24,26,30,31,35,47). Such a pattern would be very atypical for a well-established herpesvirus, whereas one usually expects a ubiquitous distribution of a highly infectious agent with nearly universal asymptomatic primary infection at a very young age. Therefore, many intriguing questions about the origins and mode of transmission of the virus arise. If infection with the virus in humans was actually an ancient and widespread event, then numerous distinct isolates or strains would be ex- pected compared with just a single narrowly diverged genome pattern if it had been recently acquired from an exogenous pri- mate source. Similarly, the patterns of genome heterogeneity observed might be expected to reflect whether the virus in pa- tients with AIDS represents a new horizontally transmitted agent, rather than multiple reactivated endogenous latent infec- tions. Our evidence clearly shows that there are at least three major subdivisions of HHV-8 genomes present in human populations (all three of which are represented in patients with AIDS) and that (with few exceptions) each patient isolate studied to date can be distinguished from that of other patients. Importantly, no differences have yet been found between the genomes present in multiple lesions from the same patient (/0) or in subcloned Journal of the National Cancer Institute Monographs No. 23, 1998 BCBL cell lines with different passaging history from the same tumor (i.e., HBL6 and BC). Our collection of genomes that were tested came from a wide variety of sources, but interpre- tation and extrapolation of the current strain variability data are greatly complicated by large differences in the rates of evolu- tionary divergence in some genes (e.g., ORF-K1) compared with the others and the existence of two distinct alternative RHS ends. The possible presence of chimeric or recombinant genomes con- taining gene blocks derived from more than one strain lineage is also implied by mismatches between the ORF-K1 subtype and the constant region (ORF26/ORF75) subtypes in some genomes. However, it is not clear as yet whether the low level of variabil- ity in the constant region genes (especially ORF26) is sufficient to provide a valid measure of subtype category (Fig. 5). Re- markably, the pattern and degree of amino acid variability in the HHV-8 ORF-KI1 protein quite closely resemble those seen in both the HIV ENV gene and in immunoglobulin variable re- gions, although obviously in this case the source of the variabil- ity and the nature of the selective pressures involved are at present a complete mystery. Recent direct PCR screening of HHV-8-positive samples from 22 different U.S. patients has revealed only five with the C-type of RHS compared with 17 of the A type. In contrast, eight were of the C type in the ORF-K1 region on the LHS and 14 were A type, but only three genomes were found to be C type at both the LHS and RHS ends. Sur- prisingly, no B-strain types have been found as yet amongst either KS or BCBL samples from U.S. patients. Overall, the results strongly imply that HHV-8 is an ancient human virus with variability and subtype patterns that closely resemble those seen in both EBV and HVS. Whether the A, B, and C strain groupings represent ancient human population bottlenecks, current biologic niche or disease associated selec- tion, or current (pre-AIDS) geographic isolation remains to be determined. However, the fact that five of six KS samples tested from Africa have B strain ORF-K1 regions and that four of our five New York/Baltimore/Washington area AIDS KS samples are closely related Al strains, whereas the fifth, which was derived from a classic non-AlIDS-associated case (EKS1), had a C strain ORF-K1 gene, suggests that we might be able to define clades of the virus similar to the situation in HIV. Obviously, the surprisingly high genetic divergence identified in ORF-KI among HHV-8 genomes will now provide opportunities to ad- dress these questions of origins, geographic preferences, trans- missibility, and disease association in considerable detail. References (1) Baer R, Bankier AT, Biggin MD, Deininger PJ, Farrell PJ, Gibson TI, et al. DNA sequence and expression of the B95-8 Epstein-Barr virus genome. Nature 1984:310:207-11. Albrecht JC, Nicholas J, Biller D, Cameron KR, Biesinger B, Newman C, et al. Primary structure of the herpesvirus saimiri genome. J Virol 1992; 606:5047-58. Telford EA, Watson MS, Aird HC, Perry J, Davison Al. The DNA se- quence of equine herpesvirus 2. J Mol Biol 1005:249:520-8. Russo JJ, Bohenzky RA, Chien MC. Chen J, Yan M, Maddalena D, et al. Nucleotide sequence of the Kaposi sarcoma-associated herpesvirus (HHVS). Proc Natl Acad Sci US A 1996:93:14862-7. Virgin HW, Latreille P, Wamsley K. Hallsworth K, Week KE, Canto AJ, et al. Complete sequence and genomic analysis of murine gammaherpes- virus 68. J Virol 1997,71:5894-904. (6) Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Ka- posi’s sarcoma. Science 1994:266:1865-9. (2 — (3 (4 XZ (3 Journal of the National Cancer Institute Monographs No. 23, 1998 (7) Thome M, Schneider P, Hofman K, Fickenscher H, Melnl E, Neipel F, et al. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 1997:386:517-21. Cesarman E, Nador RG, Bai F, Bohenzky A, Russo JJ, Moore PS, et al. Kaposi's sarcoma-associated herpesvirus contains G protein-coupled re- ceptor and cyclin D homologs which are expressed in Kaposi's sarcoma and malignant lymphoma. J Virol 1996:70:8218-23. Moore PS, Gao SJ, Dominguez G, Cesarman E, Lungu O, Knowles D, et al. Primary characterization of a herpesvirus agent associated with Kaposi's sarcoma. J Virol 1996:70:549-58. (10) Nicholas J, Ruvolo V. Zong JC, Ciufo D, Guo HG. Reitz MS, et al. A single 13 kilobase divergent locus in Kaposi sarcoma associated herpesvi- rus (HHVS) contains at least nine genes encoding homologues of cellular proteins. J Virol 1997:71:1963-74. (11) Moore PS, Boshoff C, Weiss RA, Chang Y. Molecular mimicry of human cytokine and cytokine response pathway genes by KSHV. Science 1996; 274:1739-44. (12) Neipel F, Albrecht JC, Ensser A, Huang YQ. Li JJ, Friedman-Kien AE, et al. Human herpesvirus 8 encodes a homolog of interleukin-6. J Virol 1997; 71:839-42. (13) Nicholas J, Ruvolo VR, Burns WH, Sandford G, Wan X, Ciufo D, et al. Kaposi's sarcoma-associated human herpesvirus-8 encodes homologues of macrophage inflammatory protein-1 and interleukin-6. Nature Med 1997; 3:287-92. (14) Zhong W, Wang H, Herndier B, Ganem D. Restricted expression of Kaposi sarcoma-associated herpesvirus (human herpesvirus 8) genes in Kaposi sarcoma. Proc Natl Acad Sci US A 1996:93:6641-6. (15) Sun R, Lin SF, Gradoville L, Miller G. Polyadenylated nuclear RNA en- coded by Kaposi sarcoma-associated herpesvirus. Proc Natl Acad Sci US A 1996;93:11883-8. (16) Jung JU, Trimble JJ, King NW, Beisinger B, Fleckenstein BW, Desrosiers RC. Identification of transforming genes of subgroup A and C strains of herpesvirus saimiri. Proc Natl Acad Sci US A 1991:88:7051-5. (17) Biesinger B, Tsygankov AY, Fickenscher H, Emmrich F, Fleckenstein B, Bolen JB, et al. The product of the Herpesvirus saimiri open reading frame I (tip) interacts with T cell-specific kinase pS6™* in transformed cells. J Biol Chem 1995:270:4729-34. (18) Zong JC, Metroka C, Reitz MS, Nicholas J. Hayward GS. Strain variability amongst Kaposi sarcoma associated herpesvirus (HHV-8) genomes: evi- dence that a large cohort of U.S.A. AIDS patients may have been infected by a single common isolate. J Virol 1997:71:2505-11. (19) Guo HG, Browning P. Nicholas J, Hayward GS. Jiang YW, Sadowska M. (8 = (9 et al. Characterization of a chemokine receptor-related gene in human herpesvirus 8 and its expression in Kaposi's sarcoma. Virology 1997:228: 371-8. (20) Cheng EH, Nicholas J. Bellows DS, Hayward GS. Guo HG, Reitz MS, et al. A Bcl-2 homolog encoded by Kaposi sarcoma-associated virus, human herpesvirus 8, inhibits apoptosis but does not heterodimerize with Bax or Bak. Proc Natl Acad Sci US A 1997:94:690-4. (21) Davis MA, Sturzl M, Blasig C, Schreier A, Guo HG, Reitz M, et al. Expression of human herpesvirus 8-encoded cyclin D in Kaposi's sarcoma spindle cells. J Natl Cancer Inst 1997:89:1868-74. (22) Chang YN, Dong D, Hayward GS, Hayward SD. The EBV Zta transacti- vator: A member of the B/Zip family with unique DNA-binding specificity and a dimerization domain that lacks the characteristic heptad leucine zip- per motif. J Virol 1990:64:3358-69. (23) Lieberman PM, Hardwick JM, Sample J, Hayward GS, Hayward SD. The Zta transactivator involved in induction of lytic cycle gene expression in Epstein-Barr virus-infected lymphocytes binds to both AP-1 and ZRE sites in target promoter and enhancer regions. J Virol 1990:64:1143-55. (24) Miller G, Heston L, Grogan E, Gradoville L, Rigsby M, Sun R, et al. Selective switch between latency and lytic replication of Kaposi's sarcoma herpesvirus and Epstein-Barr virus in dually infected body cavity lym- phoma cells. J Virol 1997:71:314-24. (25) Renne R, Zhong W, Herndier B, McGrath M, Abbey N, Kedes D, et al. Lytic growth of Kaposi's sarcoma-associated herpesvirus (human herpes- virus 8) in culture. Nature Med 1996:2:342-6. Chang Y, Ziegler J, Wabinga H, Katagole-Mbidde E, Boshoff C, Schulz T, et al. Kaposi's sarcoma-associated herpesvirus and Kaposi's sarcoma in Africa. Arch Intern Med 1996:156:202-4. (27) Arvanitakis L, Geras-Raaka E, Varma A, Gershengorn MC, Cesarman E. Human herpesvirus KSHV encodes a constitutively active G-protein- coupled receptor linked to cell proliferation. Nature 1997:385:347-50. (28) Sarid R. Sato T, Bohenzky A, Russo JJ, Chang Y. Kaposi's sarcoma- associated herpesvirus encodes a functional Bel-2 homologue. Nature Med 1997;3:293-8. (29) Ambroziak JA, Blackbourn DJ, Herndier BG. Glogan RC, Gullett JH. (26 - 87 (30) (31) (33) (34) (33) (36) (37) (38) (39) (40) 88 McDonald AR, et al. Herpes-like sequences in HIV-infected and uninfected Kaposi's sarcoma patients. Science 1995;268:582-3. Gao SJ, Kingsley L, Li M, Zheng W, Parravicini C, Ziegler J, et al. KSHV antibodies among Americans, Italians and Ugandans with and without Ka- posi’s sarcoma. Nature Med 1996;2:925-8. Gao SJ, Kingsley L, Hoover DR, Spira TJ, Rinaldo CR, Saah A, et al. Seroconversion to antibodies against Kaposi's sarcoma-associated herpes- virus-related latent nuclear antigens before the development of Kaposi's sarcoma. N Engl J Med 1996:;335:233-41. Dupin N, Grandadam M, Calvez V, Gorin I, Aubin JT, Havard S. et al. Herpesvirus-like DNA sequences in patients with Mediterranean Kaposi's sarcoma. Lancet 1995;345:761-2. Beral V, Peterman TA, Berkelman RC, Jaffe HW. Kaposi's sarcoma among persons with AIDS: a sexually transmitted infection? Lancet 1990; 335:123-8. Boshoff C, Whitby D, Hatziioannou T, Fisher C, Walt J, Hatzakis A, et al. Kaposi's sarcoma-associated herpesvirus in HIV-negative Kaposi's sar- coma. Lancet 1995:345:1043-4. Boshoft C, Schulz TF, Kennedy MM, Graham AK, Fisher C, Thomas A, et al. Kaposi's sarcoma-associated herpesvirus infects endothelial and spindle cells. Nat Med 1995;1:1274-8. Staskus KA, Zhong W, Gebhard K, Herndier B, Wang H, Renne R, et al. Kaposi's sarcoma-associated herpesvirus gene expression in endothelial (spindle) tumor cells. J Virol 1997:71:715-9. Decker LL, Shankar P, Khan G, Freeman RB, Dezube BJ, Lieberman J, et al. The Kaposi sarcoma-associated herpesvirus (KSHV) is present as an intact latent genome in KS tissue but replicates in the peripheral blood mononuclear cells of KS patients. J Exp Med 1996;184:283-8. Renne R, Lagunoff M, Zhong W, Ganem D. The size and conformation of Kaposi's sarcoma-associated herpesvirus (humanherpesvirus 8) DNA in infected cells and virions. J Virol 1996;70:8151-4. Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, Lusso P. Identification of RANTES, MIP-la, and MIP-18 as the major HIV- suppressive factors produced by CD8+ T cells. Science 1995:270:1811-5. Dragic T, Litwin V, Allaway GP, Martin SR, Huang Y, Nagashima A , et al. HIV-1 entry into CD4" cells is mediated by the chemokine receptor CC-CKR-5. Nature 1996;381:667-73. (41) Bailer RT, Lazo A, Ng-Bautista CL, Hout BL, Ness GM, Hegtvedt AK, et al. Comparison of constitutive cytokine release in high and low histologic grade AIDS-related Kaposi's sarcoma cell strains and in sera from HIV+/ KS+ and HIV+/KS— patients. J Inter Cytok Res 1995:5:473-83. Cai J, Gill PS, Masood R, Chandrasoma P, Jung B, Law RE, et al. On- costatin-M is an autocrine growth factor in Kaposi's sarcoma. Am J Pathol 1994;145:74-9. Miles SA. Pathogenesis of HIV-related Kaposi's sarcoma. Curr Opin Oncol 1994:6:497-502. Kawano M, Hirano T, Matsuda T, Taga T, Hori Y, Iwato K, et al. Auto- crine generation and requirement of BSF-2/IL-6 for human myelomas. Nature 1988:;332:83-4. Zhang XG, Klein B, Bataille RR. Interleukin-6 is a potent myeloma cell growth factor in patients with aggressive multiple myeloma. Blood 1989; 74:11-3. Soulier J, Grollet L, Oksenhendler E, Cacoub P, Cazals-Hatem D, Babinet P, et al. Kaposi's sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman’s disease. Blood 1995:86:1276-80. Whitby D, Howard MR, Tenant-Flowers M, Brink NS, Copas A. Boshoff C, etal. Detection of Kaposi sarcoma associated herpesvirus in peripheral blood of HIV-infected individuals and progression to Kaposi's sarcoma. Lancet 1995:346:799-802. (42) (43) (44) (45 (46) (47) Notes Supported by Public Health Service grants CA73585 (G. S. Hayward) and CA06973 (G. S. Hayward and J. Nicholas) from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services. L. J. Poole is a member of the Biochemistry, Cellular and Molecular Biology training program at The Johns Hopkins University School of Medicine (T32 GM07445). R. T. Sarisky and I. Nicholas were supported in part by a Postdoctoral Fellow- ship and a Junior Faculty Award, respectively, from the American Cancer So- ciety. We thank Sarah Heaggans for her assistance in the preparation of the manu- script and diagrams and Skip Virgin and Sam Speck for providing MHV68 sequence data prior to publication. Journal of the National Cancer Institute Monographs No. 23, 1998 Immunotherapy for Epstein-Barr Virus-Associated Cancers Cliona M. Rooney, Marie A. Roskrow, Colton A. Smith, Malcolm K. Brenner, Helen E. Heslop Epstein-Barr virus (EBV)-associated lymphoproliferative disease (EBV-LPD) is a frequently fatal complication of or- gan transplantation and human immunodeficiency virus (HIV) infection. We have studied the safety and efficacy of adoptively transferred, gene-marked virus-specific cytotoxic T lymphocytes (CTLs) as prophylaxis and treatment of EBV-LPD in recipients of T-cell-depleted allogeneic bone marrow. In 42 patients treated prophylactically, no toxicity was experienced. None of these patients developed EBV- LPD, in contrast with eight of 53 (15%) patients who did not receive prophylactic CTL. Three patients who had not re- ceived CTL developed aggressive disease and received CTL as treatment. Gene-marked CTL homed to tumor sites and selective accumulation of marker gene was detected in tumor tissues. Tumors regressed completely in two patients, but the third died of respiratory failure. Infused CTLs persisted for up to 3 years in vivo, they rapidly reconstituted EBV-specific immune responses to levels seen in normal individuals, and they reduced high viral titers by two to three logs. We are now using autologous EBV-specific CTL to treat patients with relapsed EBV-positive Hodgkin’s disease and we are developing methods for the generation of antigen-specific lines. This approach could be applied to patients with HIV who develop EBV-LPD, using CTL derived early in the course of HIV infection. [Monogr Natl Cancer Inst 1998;23: 89-93] More than 90% of individuals are infected with Epstein-Barr virus (EBV). Primary infection occurs through the oropharynx, where the virus replicates in epithelial cells and infects circulat- ing B lymphocytes (/,2). The virus then persists for the life of the infected individual, in epithelial cells and B lymphocytes. EBV can efficiently transform B lymphocytes in vitro into per- manently growing lymphoblastoid cell lines (LCLs) if immune T cells are first removed or inactivated (/). Similarly, in the absence of T-cell function in vivo, EBV may transform B cells causing lymphoproliferative disease (EBV-LPD) or lymphoma (3). Recipients of stem cells from human leukocyte antigen (HLA)-mismatched family members or unrelated donors are particularly at risk of developing this complication if T cells are removed from the allograft to prevent graft-versus-host disease (GVHD) (4,5). Unlike disease caused by herpesviruses, such as CMV and HSV, EBV-LPD is associated with the latent phase of the virus. Therefore, drugs such as acyclovir, which target the viral DNA polymerase, are ineffective in prevention or treatment. The nine latency-associated EBV proteins expressed in the tumor cells Journal of the National Cancer Institute Monographs No. 23, 1998 provide potent antigenic targets for immunotherapeutic ap- proaches (/). Donor T cells have been used effectively to treat EBV-LPD (6,7), but are associated with GvHD and when used to treat active disease, they can cause significant pathology due to inflammation. We show here that donor-derived, EBV-specific cytotoxic T lymphocytes (CTLs) can safely be given prophylacti- cally and will prevent EBV-LPD without causing GVHD. EBV is also found in the malignant cells of 50% of patients with Hodgkin's disease (HD) (8-10), offering another potential target for CTL therapy. Although the long-term survival for the majority of patients with HD is good, the outlook for the 10% who fail initial therapy is poor. Furthermore, primary therapy for HD is aggressive and the incidence of second cancers and other long-term treatment-related toxic effects is high. Targeted im- munotherapy may provide a less toxic alternative primary treat- ment, which should improve the quality of life of survivors and could offer salvage therapy to those who fail conventional therapy. Although EBV gene expression in the malignant cells of HD is more limited than in immunoblastic lymphoma cells, the three latency-associated proteins that are expressed (EBNAI, LMPI, and LMP2a) should provide antigenic targets for immu- notherapeutic approaches (2,71). We are currently analyzing the potential of EBV-specific CTL as therapy for relapsed HD. Methods Generation of EBV-Transformed B-Cell Lines LCLs were prepared from bone marrow donors or patients with HD by infection of peripheral blood mononuclear cells (PBMCs) with concentrated viral supernatant from the B95-8 cell line (12). Cells were cultured in flat-bottomed wells in the presence of 1 pg/mL of cyclosporin A until clumps of virus- transformed cells began to expand. After establishment, LCLs were maintained in acyclovir to prevent the production of infec- tious virus in the cultures (/2). CTL Generation EBV-specific CTL lines were generated by co-culture of 2 x 10° PBMCs with 5 x 10% irradiated autologous *Affiliations of authors: C. M. Rooney, M. A. Roskrow, C. A. Smith (Depart- ment of Virology and Molecular Biology), M. K. Brenner, H. E. Heslop (De- partment of Hematology and Oncology), St. Jude Children’s Research Hospital, Memphis, TN. Correspondence to: Cliona M. Rooney, Ph.D., Department of Pediatrics- Hematology/Oncology, Texas Children’s Hospital, Baylor College of Medicine, 6621 Fannin St., Houston, TX 77030-2399. E-mail: cmrooney @msmail.his. tch.tme.edu See ‘Note’ following **References.” © Oxford University Press 89 EBV-transformed B cells in 2-mL wells. CTLs were expanded by weekly restimulation with the irradiated autologous LCL and twice weekly stimulation with 20 U/mL of recombinant inter- leukin 2 after day 14. CTLs were gene marked with a retrovirus vector containing the bacterial neomycin resistance (neo) gene as described previously (72). Safety Testing of CTL Lines When sufficient CTL numbers were achieved, the line was safety tested and frozen. Release criteria mandated that the line should possess the donor HLA type; should contain less than 1% B cells; should be free of contaminating bacteria, fungi, and mycoplasma; should have low endotoxin; and should have mini- mal cytotoxic activity against recipient-derived phytohemagglu- tinin-induced lymphoblasts (PHA blasts), a measure of potential GVHD activity. Supernatants from gene-marked lines were saved to test for replication-competent retrovirus. CTL Assay Target cells, which included autologous HLA class I- mismatched LCLs, a lymphokine-activated killer-sensitive T- cell lymphoma, HSB-2, and PHA blasts derived from the recipi- ents prior to bone marrow transplant (BMT), were labeled with 100 mCi "Cr in a small volume of medium for | hour at 37 °C. They were then washed extensively and incubated with CTLs at the required effector : target ratio for 5 hours. Half the superna- tant was then removed from each well and counted. The percent specific lysis was calculated using the standard formula, experimental release — medium release percent specific lysis = z 7 maximum release — medium release PCR for neo and EBV DNA was prepared from PBMC using QIAmp Blood PCR kits (Qiagen, Chatsworth, CA). For quantitation of EBV, dilu- tions of DNA from 1 to 0.1 pg were subjected to a 25-cycle polymerase chain reaction (PCR) using 25 mM MgCl,, 1 pM of each primer, and 200 uM deoxynucleotides. The primers, first described by Boyle et al. (/3), amplified a unique region of the EBNAZ2 gene that distinguishes between the two strains of EBV. The amplified products were electrophoresed on a 2.5% agarose gel, transferred to nylon MSI membranes, and detected by DNA hybridization using nonradiolabeled EBV probes according to the GENIUS (Boehringer Mannheim Corp, Indianapolis, IN) system (/4). Quantitation of neo sequences was carried out as described previously (7/5). PCR products were electrophoresed, Southern blotted and probed with a radiolabeled neospecific probe. PCR products were compared with known standards and the sensitivity of detection was 0.01% neomarked cells. Precursor CTL Assay Patient PBMCs were plated in flat-bottomed 96-well plates at six to ten doubling dilutions from 5 x 10” cells per well with 10* cells from the irradiated autologous LCL and 10° irradiated au- tologous PBMC as feeders. Replicates ranged from 6 to 12 de- pending on the PBMC numbers available. Cells were restimu- lated weekly with the autologous irradiated LCL and fed twice weekly with 20 U per mL of interleukin 2 from day 14. Cyto- toxicity against the autologous and an HLA-mismatched LCL was tested after 6 weeks of culture. 90 Dendritic Cell Preparation Dendritic cells (DC) were enriched from peripheral blood by adhering 10° PBMC to 2 mL culture wells in 1 mL of medium for 2 hours. Nonadherent cells were removed gently and | mL of medium containing 800 U per mL granulocyte—macrophage colony-stimulating factor (GM-CSF) and 500 U per mL inter- leukin 4 was added to the adherent cells (16). These cells were cultured for 3-5 days with the addition of extra GM-CSF on day 3. Between 20% and 50% of these cells had DC morphology and phenotype (CD3-, CD14-, CD19-, DR+, and CDla+). The crude population was used for transduction. Dendritic Cell Transduction Cells (10°) that were enriched for DC were transduced with a retrovirus vector containing the LMP2a gene, using a flow- through transduction method. Retrovirus supernatant (20-40 mL) was concentrated on an Anocell 25 membrane by applying a pressure of 40-60 mm Hg. DC in 1 mL were then gently adhered to the virus on the membrane by applying a pressure of 20 mm Hg for 5-10 minutes. The membrane was then placed in a 6-mL well and cultured in 5 mL of complete medium contain- ing GM-CSF for 16 hours before analysis for retrovirus inte- gration or for irradiation and use as stimulator cells. Results and Discussion Safety Forty-five recipients of T-cell-depleted marrow have received donor-derived, EBV-specific CTL, generated by stimulation of donor PBMC with the autologous irradiated, EBV-transformed B cell line. Twenty-six EBV-specific CTL lines were gene marked prior to infusion to determine the in vivo fate of the infused cell line. All patients were treated on protocols approved locally by our Institutional Review Board and federally by the Recombinant DNA Advisory Committee of the National Insti- tutes of Health and the Food and Drug Administration (17). CTL as prophylaxis. Forty-two patients received EBV- specific CTL as prophylaxis and in these patients there was no short- or long-term toxicity. Eleven patients died, eight from relapse of the original leukemia, two from bacterial sepsis un- related to CTL infusion, and one from pneumonitis. One patient developed a recurrence of GvHD, with no accumulation of marked T cells in biopsy tissue, and three developed skin rashes attributed to drug sensitivity. CTL as treatment. Three patients who did not receive pro- phylactic CTL (two were ineligible and one refused prophylaxis) received CTL as treatment for fulminant disease (5,18). Disease regressed completely in two patients, one without incident, but one with severe morbidity as a result of the inflammatory re- sponse to his bulky and extensive disease. The third patient died of overwhelming disease, 25 days after CTL infusion. Persistence Twenty-six patients received CTL that had been gene marked with efficiencies ranging from 0.1% to 10%. Marked cells could be tracked by PCR analysis of peripheral blood DNA prepared from patients for up to 18 weeks after CTL infusion (Fig. 1) (19). Semiquantitative analysis of marker DNA indicated an expan- sion of CTL in vivo of up to three logs in some patients. If Journal of the National Cancer Institute Monographs No. 23, 1998 Fig. 1. Marked cells de- tected for up to 17 months in patient-derived EBV- ~ Control ‘7 UPN 214 . specific CTL lines. South- wo» ern blot of neospecific PCR Ss 2 of DNA prepared from S 5 BE EBV-specific CTL lines ° X s E E 2 initiated from patient UPN o X = — OM I~ — -— 0 © i y- - 214 at the indicated times after CTL infusion. Signals are compared with control cells in which from 0.01% to 10% of the cells contain one copy of the marker gene. $ } ¢ ¢ ' EBV-specific CTLs were selectively expanded from patients, marked T cells could be detected for up to 3 years postinfusion (19). The generation of an immune response to the marker gene and resultant destruction of marked cells did not appear to be a problem in our patient group, since marked cells persisted long term in all but two assessable patients. Immune Reconstitution Evidence for immune reconstitution came from comparisons of the precursor frequency of EBV-specific T cells before CTL infusion and 1 and 4 weeks after. In 15 patients, there was a 35-fold median increase in precursor frequency (range, 1.5 to >500-fold) (19). An example of the increase in precursor fre- quency is demonstrated in Fig. 2. Antiviral Effects We had previously shown that in stem cell transplant recipi- ents, more than 2000 EBV genomes per 10° peripheral blood mononuclear cells were highly predictive of EBV-LPD (7/4). Seven of 63 individuals who had high EBV DNA levels and did not receive CTL prophylaxis developed EBV-LPD. By contrast, while six patients in the prophylaxis group also developed high EBV DNA, in all of these individuals, EBV DNA levels dropped by two to four logs within 3 weeks of infusion, and none sub- sequently developed lymphoma (79). Antitumor Effects Three lines of evidence indicated that infused EBV-specific CTLs had antitumor effects. First, none of 42 patients who re- ceived prophylactic CTL developed EBV-LPD compared with seven of 63 who did not receive CTL (either because they were transplanted before the study opened or because they were in- eligible or refused). Second, none of the patients who developed high EBV DNA and received CTL developed EBV-LPD. The third line of evidence came from patients who responded to infusion of EBV-specific CTL as treatment for bulky disease. In one of these patients, we obtained biopsy specimens and found evidence for accumulation or expansion of infused CTL in tu- mor biopsy tissue. In this case, in situ PCR analysis demon- strated marked cells in tumor tissue biopsied 10 days after CTL infusion. Comparative analysis of the marker gene in peripheral blood and tumor tissue demonstrated selective accumulation or amplification of marked cells in the tumor tissue. About 1% of cells in the tumor tissue were marked by comparison with less than 0.01% in peripheral blood. Although this patient recovered completely and has been disease free for more than 1 year, he had severe morbidity as a result of the CTL-mediated inflam- matory response to his diffusely invasive disease. Prophylaxis is Better Than Cure Although we successfully treated two of three patients with extensive lymphoma, a third patient who received CTL as treat- ment of advanced pulmonary disease died of respiratory failure with overwhelming tumor burden. Together with the morbidity experienced by the second patient, our results show that, while prophylactic use of 35 2 --=--Auto LCL Fea ——MM LCL 2 25 ’ Ss c S = 20 E @ Qa 9 15 o @ 3 oO So 5 0 T T EBV-specific CTL is safe and effective, treatment of this rapidly progressive and invasive disease by infusion of EBV-specific CTL can cause damage and may be ineffective in advanced cases. Treatment of EBV-Positive HD To expand on this successful prevention of EBV-lymphoma in allogeneic stem cell transplant recipients, we decided to apply CTL to a second EBV-associated tumor, HD. Two protocols were designed and approved, one for the treatment of patients who have relapsed after therapy and have active disease and the second as adjuvant therapy for patients who receive autologous BMT as treat- ment for relapsed disease. EBV-specific CTL lines were activated and expanded in vitro, using the Pre-CTL 1week post CTL 4 weeks post CTL patient-derived EBV-transformed B cell line as an- tigen-presenting cells. CTLs were derived success- fully from 9 of 13 patients, although lines grew Fig. 2. The frequency of EBV-specific CTL precursors increases after CTL infusion. Here shown is an example of one patient in whom CTL precursors that kill the autologous LCL increase over the 4 weeks following infusion of one dose of 2 x 10” CTL/m?. Dotted line shows the number of precursors that kill the autologous LCL and solid line shows killing of an HLA class I- mismatched LCL. Journal of the National Cancer Institute Monographs No. 23, 1998 more slowly from patients with HD than from nor- mal donors and usually required additional mito- genic stimulation (20). A drawback to the approach of using LCL as the 91 Antigen-specificity of TU CTL line 70 Fig. 3. EBV-specific CTLs gener- ated by co-culture with the autolo- 60 | gous EBV-transformed B cell line have limited antigen specificity. The specificity of the TU CTL line 50 | /] was tested against the autologous TU LCL and an HLA-mismatched LCL (MM LCL). The antigen wo specificity was tested using autolo- gous TU fibroblasts that had been infected with vaccinia recombi- nants expressing the eight known 301 EBV latency-associated proteins that are expressed in LCL. Only fi- ” Percent specific lysis broblasts expressing EBNAs 3b and 3c are recognized. There is no significant killing of fibroblasts expressing EBNAI, LMPI. or 10 |- LMP2a. Auto-LCL MM-LCL EBNA-1 fmnansiain -—Vaccinia-infected fibroblasts-----------eemevr| ww WW mx EBNA-2 EBNA-3a EBNA-3b EBNA3c EBNA-LP LMP1 LMP2a beta-gal antigen-presenting cell for the generation of CTL for HD is that these CTL may not contain clones with specificity for the three EBV proteins expressed in Hodgkin's tumor cells. These three, EBNA-I, LMPI, and LMP2, are poorly immunogenic by com- parison with EBNAS 3a, 3b, and 3c, which are immunodominant on the majority of HLA backgrounds (2,77). Fig. 3 shows that CTLs generated by co-culture with LCL contain clones with specificity for a limited range of EBV proteins. If the CTL lines generated from patients with HD contained no clones with speci- ficity for EBNA1, LMPI, or LMP2, then they would have no effect on disease. To generate CTL with specificity for the proteins expressed in HD tumor cells from patients in whom these proteins may be poorly immunogenic, we are using DCs to present proteins in- dividually. DCs are the most potent antigen-presenting cells known and have been shown both to induce primary immune responses in vitro and to overcome nonresponsiveness to tumor antigens (2/7). Since EBNAL is not processed appropriately for class I recognition (22) and LMP1 is heterogeneous between EBV isolates (23), we have chosen to express LMP2a in DC for stimulation of autologous CTLs, using retrovirus vector contain- ing the LMP2a complementary DNA. As shown in Fig. 4, LMP2a-transduced DCs are able to induce autologous CTLs that kill not only autologous fibroblasts expressing LMP2a from a vaccinia virus vector but also the autologous LCL that is a biologic target for EBV. Future studies will assess the efficacy of LMP2a-specific CTL in the treatment of EBV-positive HD. In summary, we provide clinical evidence that infusion of EBV-specific CTL provides safe and effective prophylaxis for EBV-associated lymphoproliferative disease in stem cell trans- plant recipients. CTLs persist for up to 3 years in vivo, they rapidly restore cellular immune responses to EBV, and they reduce high EBV load to nonthreatening levels. As treatment for advanced and rapidly progressive disease, CTL may be less 92 ~ o N WA OO O oO O oO oOo oOo Percent specific lysis ah o o Auto-LCL MM-LCL Fib-LMP1 Fib- Fib-beta- LMP2a gal Fig. 4. CTLs induced by stimulation with autologous dendritic cells transduced with a retrovirus-LMP2a construct kill fibroblasts infected with a vaccinia LMP2a recombinant. Autologous fibroblasts are also killed strongly. HLA- mismatched LCL and fibroblasts infected vaccinia recombinants expressing LMPI or 3-galactosidase are not killed. effective and may cause pathology. However, our studies pro- vide good rationale for applying cellular immunotherapy to other cancers with identifiable target antigens. References (1) Rickinson AB, Kieff E. Epstein-Barr virus. In: Fields BN, Knipe DM, Howley PM, editors. Fields virology. Philadelphia: Lippincott-Raven, 1996, pp 2397-446. (2) Rickinson AB, Lee SP, Steven NM. Cytotoxic T lymphocyte responses to Epstein-Barr virus. Curr Opin Immunol 1996:8:492-7. (3) Straus SE. Cohen JI, Tosato G, Meier J. Epstein-Barr virus infections: biology, pathogenesis and management. Ann Intern Med 1992;118:45-58. (4) O'Reilly RJ, Small TN, Papadopoulos E, Lucas K. Lacerda J, Koulova L. Biology and adoptive cell therapy of Epstein-Barr virus-associated lym- phoproliferative disorders in recipients of marrow allografts. Immunolog Rev 1997:157:195-216. Journal of the National Cancer Institute Monographs No. 23, 1998 (5) (6) (7 = (8) (9) (10) (11) (13) (14) (15) (16) Heslop HE, Rooney CM. Adoptive immunotherapy of EBV lymphoprolif- erative diseases. Immunolog Rev 1997;157:217-22. Papadopoulos EB, Ladanyi M, Emanuel D, MacKinnon S, Boulad F, Car- abasi MH, et al. Infusions of donor leukocytes to treat Epstein-Barr virus- associated lymphoproliferative disorders after allogeneic bone marrow transplantation. N Engl J Med 1994;330:1185-91. Heslop HE, Brenner MK, Rooney CM. Donor T cells to treat EBV- associated lymphoma. N Engl J Med 1994:331:679-80. Herbst H, Steinbrecher E. Niedobitek G, Young L, Brooks L, Muller- Lantzsch N, et al. Distribution and phenotype of Epstein-Barr virus- harboring cells in Hodgkin's disease. Blood 1992:80:484-91. Pallesen G, Hamilton-Dutoit SJ, Rowe M, Young LS. Expression of Ep- stein-Barr virus latent gene products in tumour cells of Hodgkins disease. Lancet 1991:337:320-2. Gulley ML, Eagan PA, Quintanilla-Martinez L, Picardo AL, Smir BN, Childs C, et al. Epstein-Barr virus DNA is abundant and monoclonal in the Reed-Sternberg cells of Hodgkins disease: association with mixed cellu- larity subtype and hispanic American ethnicity. Blood 1994;83:1595-602. Khanna R, Burrows SR, Moss DJ. Immune regulation in Epstein-Barr virus-associated diseases. Microbiolog Rev 1995;59:387-405. Smith CA, Ng CY, Heslop HE, Holladay MS, Richardson S, Turner EV, et al. Production of genetically modified EBV-specific cytotoxic T cells for adoptive transfer to patients at high risk of EBV-associated lymphoprolif- erative disease. J] Hematother 1995:4:73-9. Boyle MJ. Sewell WA, Sculley TB, Apolloni A, Turner JJ, Swanson CE, et al. Subtypes of Epstein-Barr virus in human immunodeficiency virus- associated non-Hodgkin lymphoma. Blood 1991:;78:3004—11. Rooney CM, Loftin SK, Holladay MS, Brenner MK, Krance RA, Heslop HE. Early identification of Epstein-Barr virus-associated post-transplant lymphoproliferative disease. Br J Haematol 1995;89:98-103. Brenner MK, Rill DR, Holladay MS, Heslop HE, Moen RC, Buschle M, et al. Gene marking to determine whether autologous marrow infusion re- stores long-term haemopoiesis in cancer patients. Lancet 1993;342: 1134-7. Romani N, Gruner S, Brang D, Kampgen E, Lenz A, Trockenbacher B, et Journal of the National Cancer Institute Monographs No. 23, 1998 al. Proliferating dendritic cell progenitors in human blood. J Exp Med 1994:180:83-93. (17) Heslop HE, Brenner MK, Rooney CM, Krance RA, Roberts WM, Roch- ester R, et al. Administration of neomycin-resistance-gene-marked EBV- specific cytotoxic T lymphocytes to recipients of mismatched-related or phenotypically similar unrelated donor marrow grafts. Hum Gene Ther 1994;5:381-97. (18) Rooney CM. Smith CA, Ng C, Loftin SK, Li C, Krance RA, et al. Use of gene-modified virus-specific T lymphocytes to control Epstein-Barr virus- related lymphoproliferation. Lancet 1995;345:9-13. (19) Heslop HE, Ng CY, Li C, Smith CA, Loftin SK, Krance RA, et al. Long- term restoration of immunity against Epstein-Barr virus infection by adop- tive transfer of gene-modified virus-specific T lymphocytes. Nat Med 1996:2:551-5. (20) Roskrow MA, Suzuki N, Heslop HE, Hudson M, Brenner MK, Rooney CM. Genetically modified EBV-specific cytotoxic T cells for adoptive transfer to patients with EBV-positive Hodgkin disease. Blood 1996:88: 673a. (21) Steinman RM, Witmer-Pack M, Inaba K. Dendritic cells: antigen presen- tation, accessory function and clinical relevance. Adv Exp Med Biol 1993; 329:1-9. (22) Levitskaya J, Coram M, Levitsky V, Imreh S, Steigerwald-Mullen PM, Klein G, et al. Inhibition of antigen processing by the internal repeat region of the Epstein-Barr virus nuclear antigen-1. Nature 1995;375:685-8. (23) Khecht H, Bachmann E, Brousset P, Sandvej K, Nadal D, Bachmann F, et al. Deletions within the LMP1 oncogene of Epstein-Barr virus are clustered in Hodgkin's disease and identical to those observed in nasopharyngeal carcinoma. Blood 1993;82:2937-42. Note Supported in part by Public Health Service grants CA21765, CA81364, and CA71426 from the National Cancer Institute, National Institutes of Health, De- partment of Health and Human Services; by the Assisi Foundation of Memphis: and by the American Lebanese Syrian Associated Charities. Genetic Basis of Acquired Immunodeficiency Syndrome-Related Lymphomagenesis Gianluca Gaidano, Antonino Carbone, Riccardo Dalla-Favera® The molecular pathogenesis of systemic acquired immuno- deficiency syndrome (AIDS)-related non-Hodgkin’s lympho- mas (AIDS-NHL) is a complex process involving both host factors and the accumulation of genetic lesions within the tumor clone. On the basis of the pattern of molecular lesions involved in these tumors, several distinct pathogenetic path- ways can be presently identified in AIDS-related lymphoma- genesis. These pathways selectively associate with the differ- ent clinicopathologic variants of AIDS-NHL. AIDS-related Burkitt’s lymphoma is characterized by activation of c-MYC in all cases, disruption of p53 in 60% of the cases, and in- fection by Epstein-Barr virus (EBV) in 30% of the cases. AIDS-related diffuse large-cell lymphoma harbor frequent EBYV infection (80%) and, in 20% of the cases, BCL-6 rear- rangements. Finally, the pathogenesis of AIDS-related body cavity-based lymphoma involves infection by human herpes- virus-8 in all cases and frequently also the co-infection by EBV. [Monogr Natl Cancer Inst 1998;23:95-100] Non-Hodgkin's lymphoma (NHL) is the second most frequent cancer associated with acquired immunodeficiency syndrome (AIDS) after Kaposi's sarcoma, and in some risk groups, namely, the hemophiliacs, NHL represent the most common AIDS-related neoplasm (/,2). The incidence of AIDS-related NHLs (AIDS-NHL) has continued to rise since the outburst of the AIDS epidemic and, in 1985, the Centers for Disease Control (CDC) has recognized NHL as an AIDS-defining condition (1-3). Because AIDS—-NHL are frequently a late complication of AIDS, it is thought that their incidence will continue to increase steadily with the prolongation of life expectancy of human im- munodeficiency virus (HIV)-infected individuals (4). All AIDS-NHL share a number of similarities (2,5). They all derive from B cells, are characterized by extreme clinical ag- gressiveness, and display a predilection for extranodal sites, in- cluding unusual locations otherwise rarely implicated by B-cell NHL of the immunocompetent host. Despite these apparent similarities, however, it is now well established that AIDS-NHL display a marked clinicomorphologic heterogeneity. The de- tailed pathologic classification of AIDS—NHL has been a matter of controversy and has been recently updated by the World Health Organization (6). First of all, AIDS—-NHL may present as a systemic disease (systemic AIDS—-NHL) or may originate and locate selectively in the central nervous system (primary central nervous system lymphomas, PCNSL). Depending on histology, AIDS-NHL are generally distinguished into two major catego- ries, which include Burkitt’s lymphoma (BL) and diffuse large cell lymphoma (DLCL). AIDS-related BL (AIDS-BL) and AIDS-related DLCL (AIDS-DLCL) account for approximately Journal of the National Cancer Institute Monographs No. 23, 1998 one third and two thirds of systemic AIDS-NHL, respectively (1). Conversely, AIDS-PCNSL are consistently represented by the DLCL histology. An additional, although rare, pathologic category of AIDS-NHL is constituted by body cavity-based lymphoma (AIDS-BCBL), a peculiar type of lymphoma grow- ing in liquid phase in the serous cavities of the body (7-9). The pathogenesis of AIDS-NHL is a complex process impli- cating different phases (2). Initially, factors contributed by the immunocompromised host, and mainly represented by disrupted immunosurveillance and chronic antigen stimulation, lead to polyoligo-clonal B-cell hyperstimulation and hyperplasia, thus configurating the clinicopathologic picture known as persistent generalized lymphadenopathy (PGL) (2,10). The role of dis- rupted immunosurveillance is exemplified by the close associa- tion between low levels of CD4-positive T cells in the host’s peripheral blood and increased risk of AIDS-NHL development (11). This association is particularly marked in the case of AIDS-DLCL and AIDS-PCNSL, whereas AIDS-BL may also develop in the context of a relatively preserved immune func- tion. The contribution of chronic antigen stimulation and the selection to AIDS-related lymphomagenesis is documented by the high rate of somatic mutations accumulating in the hyper- variable regions of immunoglobulin (Ig) genes expressed by AIDS-NHL cells as well as by the preferential usage of Ig- variable genes belonging to Ig gene families, such as V4, which are frequently implicated in the generation of B-cell autoreactive clones (12). Within this context of B-cell hyperstimulation and hyperpla- sia, the emergence and progressive expansion of the AIDS-NHL clone are driven by the clonal accumulation of genetic lesions within the neoplastic population. The search for these genetic lesions has been the focus of intense efforts during the last 5 years of research in this field. Here we report on the identifica- tion of the molecular alterations most frequently associated with the major categories of AIDS-NHL. *Affiliations of authors: G. Gaidano, Division of Internal Medicine, Depart- ment of Medical Sciences, University of Torino at Novara, Italy; A. Carbone, Division of Pathology, .N.R.C.C.S., Centro di Riferimento Oncologico, Aviano, Italy; R. Dalla-Favera, Division of Oncology, Department of Pathology, College of Physicians & Surgeons, Columbia University, New York, NY Correspondence to: Riccardo Dalla-Favera, M.D., Division of Oncology, De- partment of Pathology, College of Physicians & Surgeons, Columbia University, 630 W. 168th St., New York, NY 10032. E-mail: rd10@columbia.edu. See “Note” following ‘‘References.”’ © Oxford University Press 95 Identification of Genetic Alterations in AIDS-NHL The very first molecular studies performed on AIDS-NHL suggested that the genetic lesions associated with this group of neoplasms are markedly heterogeneous. These same studies demonstrated that, as in other human cancers, genetic lesions of AIDS-NHL may cause different consequences, including the activation of cellular proto-oncogenes, the disruption of tumor suppressor genes, or, alternatively, the insertion of foreign genes in the neoplastic cells, as exemplified by the case of tumor infection by oncogenic viruses. To define with precision the patterns of genetic lesions asso- ciated with the various categories of AIDS-NHL, we have in- vestigated extensively the molecular features of large panels of systemic AIDS-NHL derived from different geographic back- grounds, including the United States and Northern Italy. We have chosen to focus our attention on the genetic lesions most frequently involved in aggressive B-cell NHL of the immuno- competent host, including the alterations of ¢c-MYC and BLC-6 among proto-oncogenes, the inactivation of p53 and 6q among tumor suppressor loci, and the infection by oncogenic herpesvi- ruses. Viral Infection Several studies (2,5) have focused on the involvement of EBV in AIDS-NHL. These studies have demonstrated that the virus infects the tumor clone of a large proportion of systemic AIDS-NHL, including 30% AIDS-BL and 70%-80% AIDS-DLCL (/3-15). EBV infection is also detectable in 100% AIDS-PCNSL and approximately 90% AIDS-BCBL (7-9,16). In most AIDS-NHL tested, infection by EBV has been demon- strated to be monoclonal, consistent with a model of infection preceding, and putatively contributing to, clonal expansion of the neoplastic population (73,14). Among EBV-positive AIDS-NHL, the expression pattern of the EBV-encoded trans- forming antigens EBNA-2 and LMP-1 varies substantially (/7). Cases of AIDS-BL are consistently negative for expression of both antigens. whereas systemic AIDS-DLCL express LMP-1 in a substantial fraction of cases, approximately 50% (17,18). Some rare cases of systemic AIDS-DLCL also express EBNA-2. In addition to EBV, another herpesvirus, namely, human her- pesvirus type-8 (HHV-8), is involved in AIDS-NHL pathogen- esis, although at a substantially lower frequency (7-9). Infection of the tumor clone by HHV-8 is restricted to cases of AIDS- BCBL, whereas it is consistently absent in all other AIDS-NHL types (7-9,19) (Fig. 1, A-C). Other viruses, including HIV and HTLV-I, do not appear to be directly implicated in AIDS-NHL pathogenesis, since they do not infect the tumor clone (2). c-MYC Since the initial phases of the AIDS-NHL epidemic, cytoge- netic studies had revealed substantial similarities between AIDS-BL and BL of the immunocompetent host, based on the presence of chromosomal translocations affecting band 8q24, the site of the ¢-MYC proto-oncogene, and an Ig locus (20). These initial suggestions were later confirmed by molecular analysis of ¢-MYC in AIDS-NHL samples (/4). These genetic 96 studies aimed at defining the frequency and distribution of c- MYC alterations throughout the spectrum of AIDS-NHL and at characterizing the precise mechanism of activation of the proto- oncogene in these tumors. Analysis of large panels of cases have revealed that activation of c-MYC associated selectively with all cases of AIDS-BL, whereas it is absent among AIDS-DLCL and AIDS-BCBL (7-9,14). Although rare cases of AIDS- DLCL have been reported to harbor an activated c-MYC, these samples presumably represent cases of AIDS-BL that have been misdiagnosed because of the well-known histologic pleomor- phism characteristic of AIDS-NHL (2/7). Regarding the precise mechanism of ¢-MYC activation in AIDS-BL, it generally reflects that typical of sporadic BL, as opposed to endemic BL (7/4). Thus, in the case of t(8;14)(q24:q32), the most frequent translocation type affecting ¢-MYC, the translocation breakpoints on chromosome 8 fall within sequences internal or immediately 5’ to the c-MYC gene, whereas the chromosome 14 breakpoints map to the Ig switch region (22). As a consequence of the translocation, the expres- sion of the c-MYC gene undergoes transcriptional deregulation (22). In addition to being truncated, the ¢-MYC gene in AIDS-BL is also frequently mutated within its exon 2 (23). These mutations lead to amino acid substitutions of the c-MYC protein and alter the physiologic interactions between ¢-MYC and its partner proteins, namely p107 (24). Whereas in normal conditions p107 is able to suppress the transactivation properties of c-MYC, mutated c-MYC alleles are no longer responsive to pl107 and thus escape one of the main physiologic mechanisms regulating ¢-MYC activity (24). BCL-6 The BCL-6 gene has been originally cloned by virtue of its involvement in chromosomal translocations affecting chromo- somal band 3q27 in the DLCL of the immunocompetent host (25). Among NHL of the immunocompetent host, BCL-6 is rearranged in approximately 40% of B-cell DLCL (26). Among AIDS-NHL, rearrangements of BCL-6 cluster selectively with a fraction of AIDS-DLCL, whereas they are absent among AIDS-BL and AIDS-BCBL (27,28). The frequency of BCL-6 rearrangements in AIDS-DLCL is significantly lower than that detected among DLCL of the immunocompetent host, confirm- ing the notion that the pathogenesis of these two groups of neoplasms is different (see also the p53 section below). Rearrangements involving BCL-6 are promiscuous, in that several distinct chromosomal sites may serve as chromosomal partners juxtaposing to 327 (25). In contrast to c-MYC trans- locations, the chromosomal partners of 3q37 include, but are not restricted to, the Ig loci. The rearrangement breakpoints cluster within a 4-kb region spanning the BCL-6 promoter sequences and the first noncoding exons and result in the fusion of BCL-6 coding sequences (exons 2-10) to heterologous promoters de- rived from the partner chromosome. This mechanism of BCL-6 alteration is termed promoter substitution and is thought to cause deregulated expression of the BCL-6 protein (29). In addition to chromosomal rearrangements, the BCL-6 gene in AIDS-NHL may be altered by an alternative type of genetic alteration. Recent studies of NHL of the immunocompetent host have shown that in approximately 70% DLCL and 50% follic- ular lymphomas the BCL-6 gene is altered by multiple mutations Journal of the National Cancer Institute Monographs No. 23, 1998 Fig. 1. A) AIDS-related HHV-8" body-cavity-based lymphoma. In a cell block from pleural effusion, the tumor cells are characterized by nuclei that are irregu- larly shaped and variably chromatic with prominent, often multiple nucleoli. Most tumor cells have moderately abundant cytoplasm with plasmacytoid ap- pearance. Hematoxylin—eosin stain, original magnification x630. B) AIDS- related HHV-8* body-cavity-based lymphoma-derived cell line (HBL-6). HHV-8 in situ hybridization signal is present as dense grains over the nuclei of most tumor cells. Cytospin preparation; digoxigenin-labeled KS330 probe; in situ hybridization, nuclear fast red counterstain, original magnification x630. C) AIDS-related body-cavity-based lymphoma. In a cell block from pleural effu- sion, EBER in situ hybridization signal is present as dense grains over the nuclei of most tumor cells. In situ hybridization, nuclear fast red counterstain, original clustering in its 5’ noncoding region (30). These mutations fre- quently occur in the absence of any recognizable chromosomal abnormality affecting band 3¢27 or molecular rearrangement of the BCL-6 locus. The genomic sequences most frequently in- volved by these mutations are adjacent to the BCL-6 promoter region and overlap with the major cluster of chromosomal break- Journal of the National Cancer Institute Monographs No. 23, 1998 magnification x250. D) HBL-2/EBV obtained by EBV in vitro infection of HBL-2, an EBV-negative AIDS-BL cell line. Nuclear positivity for BCL-6 is restricted to a small number of tumor cells. Cytospin preparation, APAAP method, original magnification x400. E) HBL-2/EBV obtained by EBV in vitro infection of HBL-2, an EBV-negative AIDS-BL cell line. LMP-1 cytoplasmic positivity is detected on two tumor cells of large size. Cytospin preparation, APAAP method, original magnification x400. F) HBL-2/EBV obtained by EBV in vitro infection of HBL-2, an EBV-negative AIDS-BL cell line. Double la- beling for nuclear BCL-6 (blue) and cytoplasmic LMP-1 (brown). Against a background of unstained cells, individual tumor cells express either BCL-6 (arrows) or LMP-1 (asterisks). No coexpression of both proteins by the same tumor cell is detectable. Cytospin preparation, original magnification x400. points, suggesting that mutations and rearrangements may be selected for their ability to alter the same region. The combined frequency of mutations and rearrangements approaches 100% of DLCL cases arising in the immunocompetent host, suggesting that structural alterations of the 5 noncoding region of the BCL-6 gene are necessary for the development of these tumors 97 (30). To verify the involvement of BCL-6 mutations among AIDS-NHL, we have performed an extensive analysis of this genetic lesion by a combination of PCR-SSCP and DNA direct sequencing in a panel representative of the major pathologic categories of these lymphomas. These data show that mutations of the 5" noncoding regions of BCL-6 are detectable in approxi- mately 60% of systemic AIDS-NHL (28). According to histol- ogy. mutations were found to cluster with approximately 70% AIDS-BL and AIDS-DLCL, whereas they were restricted to 20% AIDS-BCBL (28). Within the AIDS-DLCL group, BCL-6 5" mutations occurred both in the presence and in the absence of BCL-6 rearrangements, thus mimicking the pattern reported for DLCL of the immunocompetent host. With respect to other ge- netic lesions frequently encountered in AIDS-NHL, the pres- ence of BCL-6 5" mutations in a given AIDS-NHL sample was independent of the concomitant presence of ¢-MYC rearrange- ments, pS3 mutations, and EBV infection (28). The extreme frequency of BCL-6 mutations renders these alterations the most frequent genetic lesion of systemic AIDS-NHL. Since BCL-6 mutations are not only frequent, but are also specific for a given tumor sample, they may be exploited as a molecular marker for monitoring the course of the disease. In addition to exploring the genetic features of BCL-6 in AIDS-NHL, we have recently investigated the expression status of the BCL-6 protein in these tumors. The BCL-6 protein be- longs to the family of transcription factors containing zinc-finger motifs (25). Functional studies have indicated that BCL-6 can function as a transcriptional repressor that can bind a specific DNA sequence and repress transcription from linked promoters (31). Thus, the physiologic function of BCL-6 may be to repress the expression of genes carrying its specific DNA binding motif. Several observations indicate that BCL-6 is involved in the de- velopment and function of the germinal center (GC). First, the BCL-6"" phenotype associates with lack of GC formation, as demonstrated by animal models in which the expression of the physiologic BCL-6 protein has been constitutively abrogated in the germline (32). In addition, since the BCL-6 gene is expressed in GC B cells, but not in their differentiated cell progenies (plasma cells and memory B cells), it is conceivable that BCL-6 may be involved in the induction and sustainment of GC- associated functions and that BCL-6 down-regulation may be necessary for B cells to progress toward further differentiation into memory B cells or plasma cells (33). In NHL carrying a rearranged BCL-6, the down-regulation of the BCL-6 protein may be prevented by the juxtaposition of the gene to heterolo- gous promoters. Among AIDS-NHL, expression of the BCL-6 protein was found in 100% AIDS-BL and in approximately 50% AIDS— DLCL (18). Intriguingly, among AIDS-DLCL, expression of the BCL-6 protein was mutually exclusive with expression of the EBV-encoded antigen LMP-1 (78). The precise molecular mechanism for this phenomenon is presently unclear, although it is interesting to note that BCL-6"/LMP-1- AIDS-DLCL gener- ally display a large noncleaved cell morphology, whereas cases of BCL-6 /LMP-1" AIDS-DLCL are classifiable as immuno- blastic plasmacytoid lymphomas (7/8). We are currently inves- tigating whether this phenotypic heterogeneity of AIDS-DLCL reflects a different clinical behavior of these two AIDS-DLCL variants. 98 The concept that LMP-1 and BCL-6 are mutually exclusive antigens was also based on experimental data derived from the comparison of the AIDS-BL cell line HBL-2 and its EBV- infected counterpart HBL-2/EBV. The EBV-negative HBL-2 cell line expresses high levels of the BCL-6 protein, whereas the cell line HBL-2/EBV, which expresses LMP-1 in a fraction of cells, exhibits a marked reduction in the number of BCL-6- positive tumor cells. Notably, no co-expression of BCL-6 and LMP-1 can be detected in a single tumor cell of HBL-2/EBV (Fig. 1, D-F). The results of genetic and protein studies of BCL-6 may also help us to understand the histogenesis of AIDS-NHL. On one hand, in fact, mutations of BCL-6 5’ regulatory regions are generally regarded as a marker of transition of a given B cell through the GC (30). Since a large fraction of these tumors does carry BCL-6 mutations (28), it is conceivable that they derive from B-cell subsets that have already transited through (and potentially reside in) the GC. Furthermore, expression of the BCL-6 protein is also considered as a marker of a B-cell differentiation state corresponding to B cells residing in the GC (33). In this respect, all AIDS-BL and a fraction of AIDS- DLCL may thus be derived from GC-proliferating B cells, i.e., the centroblasts (/8). According to this same model, the fraction of AIDS-DLCL that do not express BCL-6, but conversely ex- press LMP-1 when EBV infected, may derive from more dif- ferentiated B cells that have already exited from the GC (18). p53 Inactivation of the p53 tumor suppressor gene represents one of the most frequent genetic alterations in human cancers (22). Among B-cell neoplasms of the immunocompetent host, muta- tions of p53 are virtually restricted to the case of BL and DLCL transformed from a previous follicular phase (22). Inactivation of p53 in human neoplasia most commonly occurs through mu- tation of one allele and deletion of the other allele, although biallelic deletions, or, alternatively, bi-allelic mutations, may also occur in some cases. In the case of AIDS-NHL, our analysis of more than 50 cases by PCR-SSCP and DNA direct sequenc- ing has demonstrated that p53 inactivation clusters selectively with AIDS-BL, whereas it is consistently absent among cases of AIDS-DLCL and AIDS-BCBL (8,14,28). The pattern of p53 inactivation in AIDS—-NHL reflects that of human neoplasia in general. The frequency of pS3 mutations among AIDS-BL (60%) far exceeds that of non-AIDS-related BL, including both sporadic and endemic BL (30%) (14). The reason for this excess of p53 mutations in AIDS-BL is presently unknown, although one hypothesis is that it may be related to the host's immuno- deficit. Finally, the absence of p53 mutations among AIDS— DLCL, combined to the negativity for BCL-2 rearrangements of these tumors, is consistent with the de novo origin of AIDS— DLCL. In fact, among DLCL of the immunocompetent host, cases arising de novo are devoid of BCL-2 and p53 lesions, which, conversely, denote cases arising from a preceding follic- ular lymphoma (22). 6q Deletions Deletions of the long arm of chromosome 6 (6q) represent a frequent genetic alteration in B-cell NHL of the immunocom- Journal of the National Cancer Institute Monographs No. 23, 1998 petent host (22). Although the relevant tumor suppressor gene is presently unknown, it is thought to map to a small region of chromosomal band 6q27. Among AIDS-NHL, deletions of 6q27 cluster with a fraction of AIDS-DLCL (20%), whereas deletions are consistently absent in other AIDS-NHL types (34). Dele- tions of 6q27 in AIDS-DLCL may occur in combination with other genetic alterations typical of this lymphoma category, in- cluding BCL-6 rearrangements and EBV infection. Conclusions The data summarized in this report demonstrate that AIDS— NHL is a strikingly heterogeneous disease. At present, four ma- jor molecular pathways can be identified. Each of these molecu- lar pathways associates with peculiar clinical features and is restricted to a given AIDS—-NHL histologic type. The first path- way associates with AIDS-BL and is characterized by relatively mild immunodeficiency of the host and multiple genetic lesions of the tumor, including activation of c-MYC, disruption of p53, and, although less frequently, infection by EBV. Typically, EBV-infected AIDS-BL fail to express the viral transforming antigens LMP-1 and EBNA-2. Two distinct pathways associate with AIDS-DLCL, a type of AIDS-NHL generally characterized by a marked disruption of immune function. Whereas the majority of AIDS-DLCL carry EBV infection, only a fraction of cases express the viral antigen LMP-1. Expression of LMP-1 and BCL-6 segregate the two pathways associated with AIDS-DLCL. On the one hand, in fact, LMP-1-positive AIDS-DLCL fail to express the BCL-6 protein and display features consistent with immunoblastic- plasmacytoid differentiation, suggesting a derivation from post- GC cells. On the other hand, LMP-1-negative AIDS-DLCL ex- press BCL-6 and display a large noncleaved cell morphology, suggesting an origin from the GC. Finally, the fourth pathway associates with AIDS-BCBL. This rare lymphoma type consistently harbors infection by HHV-8 and, frequently, also by EBV. All other genetic lesions commonly detected among AIDS-NHL are consistently nega- tive in AIDS-BCBL. References (1) Beral V, Petterman T, Berkelman R, Jaffe H. AIDS-associated non- Hodgkin's lymphoma. Lancet 1991;337:805-9. Gaidano G, Dalla-Favera R. Molecular pathogeneis of AIDS-related lym- phomas. Adv Cancer Res 1997:67:113-53. Center for Disease Control. Revision of the case definition of acquired immunodeficiency syndrome for national reporting—United States. Ann Intern Med 1985:103:402-8. Gail MH, Pluda JM, Rabkin CS, Biggar RJ, Goedert JJ, Horm JW, et al. Projection of the incidence of non-Hodgkin's lymphoma related to acquired immunodeficiency syndrome. J Natl Cancer Inst 1991:83:695-700. Gaidano G, Carbone A. AIDS-related lymphomas: from pathogenesis to pathology. Br J Haematol 1995;90:235-43. (6) IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Human Immunodeficiency Viruses and Human T-cell Lymphotropic Vi- ruses. France, 1-18 June 1996. IARC Monogr Eval Carcinog Risks Hum 1996:67:1-424. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. Kaposi's sar- coma-associated herpesvirus-like DNA sequences in AIDS-related body- cavity-based lymphomas. N Engl J Med 1995:332:1186-91. Carbone A, Globhini A, Vaccher E, Zagonel V, Pastore C, Dalla Palma P, et al. Kaposi's sarcoma-associated herpesvirus DNA sequences in AIDS- related and AIDS-unrelated lymphomatous effusions. Br J Haematol 1996:94:533-43. Carbone A, Gaidano G. HHV-8-positive body-cavity-based lymphoma: a novel lymphoma entity. Br J Haematol 1997;97:515-22. (2 Ps (3 (4 ~ (3 Rr (7 (8 = (9 -— Journal of the National Cancer Institute Monographs No. 23, 1998 (10) Pelicci PG, Knowles DM, Arlin ZA, Wieczorek R, Luciw P, Dina D, et al. Multiple monoclonal B cell expansions and c-myc rearrangements in acquired immune deficiency syndrome-related lymphoproliferative disorders: implications for lymphomagenesis. J Exp Med 1986;164: 2049-76. (11) Pluda JM, Venzon D, Tosato G, Lietzau J, Wyvill K, Nelson DL, et al. Parameters affecting the development of non-Hodgkin's lymphoma in pa- tients with severe human immunodeficiency virus infection receiving an- tiretroviral therapy. J Clin Oncol 1993;11:1099-107. (12) Riboldi P, Gaidano G, Schettino EW, Steger TG, Knowles DM, Dalla- Favera R, et al. Two AIDS-associated Burkitt's lymphomas produce spe- cific anti-i IgM cold agglutinins utilizing somatically mutated V,;4-21 seg- ments. Blood 1994:83:952-61. (13) Neri A, Barriga F, Inghirami G, Knowles DM, Neequaye J, Magrath IT, Dalla-Favera R. Epstein-Barr virus infection precedes clonal expansion in Burkitt's and acquired immunodeficiency associated-lymphoma. Blood 1991;77:1092-5. (14) Ballerini P, Gaidano G, Gong JZ, Tassi V, Saglio G, Knowles DM, Dalla- Favera R. Multiple genetic lesions in acquired immunodeficiency syn- drome-related non-Hodgkin’s lymphoma. Blood 1993;81:166-76. (15) Carbone A, Gloghini A, Zanette I, Canal B, Volpe R. Demonstration of Epstein-Barr virus genomes by in situ hybridization in acquired immune deficiency syndrome-related high grade and anaplastic large cell CD30" lymphomas. Am J Clin Pathol 1993:99:289-97. (16) MacMahon EME, Glass JD, Hayward SD, Mann RB, Becker PS, Charache P, McArthur JC, Ambinder RF. Epstein-Barr virus in AIDS-related primary central nervous system lymphoma. Lancet 1991:338:969-73. (17) Carbone A, Tirelli U, Gloghini A, Volpe R, Boiocchi M. Human immu- nodeficiency-virus-associated systemic lymphomas may be subdivided into two main groups according to Epstein-Barr viral latent gene expression. J Clin Oncol 1993;11:1674-81. (18) Carbone A, Gaidano G, Gloghini A, Pastore C, Saglio G, Tirelli U, et al. BCL-6 protein expression in AIDS-related non-Hodgkin's lymphomas. Am J. Pathol 1997;150:155-65. (19) Gaidano G, Pastore C, Gloghini A, Cusini M, Nomdedeu J, Volpe G, et al. Distribution of human herpesvirus-8 sequences throughout the spectrum of AIDS-related neoplasia. AIDS 1996:10:941-9. (20) Chaganti RS, Jhanwar SC, Koziner B, Arlin Z, Mertelsmann R, Clarkson B. Specific translocations characterize Burkitt's like lymphoma of homo- sexual men with the acquired immunodeficiency syndrome. Blood 1983:61:1265-8. (21) Delecluse HJ, Raphael M, Magaud JP, Felman P, the French Study Group of Pathology for Human Immunodeficiency Virus-associated tumors, Alsamad IA, et al. Variable morphology of human immunodeficiency virus-associated lymphomas with ¢-MYC rearrangements. Blood 1993:82:552-63. Gaidano G, Dalla-Favera R. Molecular biology of lymphomas. In: DeVita VT, Hellman S, Rosenberg SA, editors. Cancer: principles and practice of oncology. Philadelphia: Lippincott, 1997:2131-45. (23) Bhatia K, Spangler G, Gaidano G, Hamdy N. Dalla-Favera R, Magrath I. Mutations in the coding region of c-myc occur frequently in acquired immunodeficiency syndrome-associated lymphomas. Blood 1994:84: 883-8. (24) Gu W, Bhatia K, Magrath IT, Dang CV, Dalla-Favera R. Binding and suppression of the Myc transcriptional activation domain by p107. Science 1994:264:251-4. (25) Ye BH, Lista F, LoCoco F, Knowles DM, Chaganti RS, Dalla-Favera R. Alterations of blc-6, a novel zinc-finger gene, in diffuse large cell lym- phoma. Science 1993:262:747-50. (26) LoCoco F, Ye BH. Lista F, Corradini P, Offit K, Knowles DM, et al. Rearrangements of the bcl-6 gene in diffuse large cell non-Hodgkin's lym- phoma. Blood 1994:83:1757-9. (27) Gaidano G, LoCoco F, Ye BH, Shibata D, Levine A, Knowles DM, et al. Rearrangements of the bcl-6 gene in acquired immunodeficiency syn- drome-associated non-Hodgkin's lymphoma: association with diffuse large-cell subtype. Blood 1994:84:397-402. (28) Gaidano G, Carbone A, Pastore C, Capello D, Migliazza A, Gloghini A, et al. Frequent mutation of the 5’ noncoding region of the BCL-6 gene in acquired immunodeficiency syndrome-related non-Hodgkin's lymphomas. Blood 1997:89:3755-62. (29) Ye BH, Chaganti S, Chang CC, Niu H, Corradini P, Chaganti RS, et al. Chromosomal translocations cause deregulated BCL6 expression by promoter substitution in B cell lymphoma. EMBO J 1995:14: 6209-17. (30) Migliazza A, Martinotti S, Chen W, Fusco C, Ye BY, Knowles DM, et al. Frequent somatic hypermutation of the 5’ noncoding region of the bel-6 gene in B-cell lymphoma. Proc Natl Acad Sci USA 199592: 125204. (22 hs 99 (31) Chang CC, Ye BH, Chaganti RS, Dalla-Favera R. BLC-6, a POZ/zinc- finger protein, is a sequence-specific transcriptional repressor. Proc Natl Acad Sci USA 1996:93:6947-52. (32) Ye BH, Cattoretti G, Shen Q, Zhang J, Hawe N, de Waard R, et al. The BCL-6 proto-oncogene controls germinal-centre formation and Th2-type inflammation. Nat Genet 1997;16:161-70. (33) Cattoretti G, Chang CC, Cechova K, Zhang J, Ye BH, Falini B, et al. BCL-6 protein is expressed in germinal-center B cells. Blood 1995:86:45-53. (34) Pastore C, Carbone A, Gloghini A, Volpe G, Saglio G, Gaidano G. Asso- 100 ciation of 6q deletions with AIDS-related diffuse large cell lymphoma. Leukemia 1996:10:1051-3. Note Supported by Public Health Service grant CA37295 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Ser- vices; by Ministero della Sanita, Istituto Superiore di Sanita, AIDS project 1997, Rome, Italy; and by Fondazione *‘Piera, Pietro e Giovanni Ferrero,”” Alba, Italy. Journal of the National Cancer Institute Monographs No. 23, 1998 Clinical Management of Human Immunodeficiency Virus-Associated Non-Hodgkin’s Lymphoma Lawrence D. Kaplan The human immunodeficiency virus (HIV)-associated non- Hodgkin's lymphomas are best divided into three categories: 1) primary central nervous system (CNS) lymphomas, 2) systemic lymphomas, and 3) primary effusion lymphomas. As described elsewhere in this monograph, these lymphomas differ in many of their molecular characteristics and presumed mechanisms of pathogenesis. These factors may be relevant in terms of devel- opment of future therapeutic approaches that take advantage of some of these unique molecular characteristics. However, of more immediate relevance is the fact that these lymphomas oc- cur in different patient populations and are associated with dif- ferent clinical outcomes. Systemic lymphomas are observed across a broad range of levels of immune function, with a me- dian CD4" cell count of approximately 100/mm”’ (7,2). The me- dian survival among patients with systemic lymphoma is ap- proximately 5-8 months, and 10%-20% survive disease free for longer than 2 years (2-4). In contrast, primary CNS lymphoma occurs in the most severely immunocompromised group, 75% of whom have CD4 counts less than 50/mm” (7). CNS lymphoma in the setting of this profound immunosuppression is associated with a median survival of only 3 months. The diagnosis and management of primary CNS lymphoma therefore present unique clinical management problems, and it is here that I will begin the discussion of clinical management. Primary CNS Lymphomas Diagnosis The difficulty in diagnosing primary CNS lymphoma is re- lated to the fact that lymphoma in the brain is difficult to dis- tinguish from intercerebral toxoplasmosis, a common opportu- nistic infection in patients with HIV disease. Although the radiographic appearance on computed tomography or magnetic resonance imaging scan may provide clues as to the diagnosis, these studies are rarely diagnostic in acquired immunodeficiency syndrome (AIDS) patients with CNS lesions. Since magnetic resonance scanning in individuals with toxoplasmosis reveals a solitary lesion in only 21% of cases (5), the finding of a solitary lesion is more suggestive of a diagnosis of lymphoma. However, magnetic resonance scanning in individuals with CNS lym- phoma demonstrates multiple lesions in 50% of cases (5), and this finding is less likely to be helpful. The diagnosis of lym- phoma may also be suggested by a periventricular location or a lesion that crosses the midline. However, these ‘classic’ radio- graphic findings occur infrequently in HIV-associated CNS lym- phoma. Journal of the National Cancer Institute Monographs No. 23, 1998 Recently, thallium-201 single-photon emission computed to- mography (SPECT) scanning has been used to distinguish be- tween these two entities. In one study (6), all 24 individuals with toxoplasmosis had negative scans, whereas all 12 with CNS lymphoma had positive scans. In another study (7), nine patients with CNS lymphoma were all positive on thallium-201 SPECT scanning. Of concern, however, was the fact that three of 10 patients with toxoplasmosis also had positive scans. Although these series of patients are small and data from larger numbers of patients are needed, these observations do suggest a potential utility for this nuclear medicine study in the diagnosis of CNS lymphoma. Of even greater interest are observations made with the use of polymerase chain reaction (PCR) for detecting Epstein-Barr vi- rus (EBV) DNA sequences in cerebrospinal fluid. De Luca et al. (8) showed that seven of eight individuals with documented CNS lymphoma had positive PCR for EBV in cerebrospinal fluid. All 11 patients with brain lesions and no lymphoma were negative, and 21 individuals with AIDS but no CNS lesions were also negative. In a second series (9), all 17 individuals with CNS lymphoma were positive for EBV by PCR and 67 of 68 indi- viduals with HIV and no lymphoma were negative. Data from a much larger series of patients have been reported elsewhere in this monograph and confirm these results. These data strongly suggest that this technique is a useful means of noninvasive diagnosis of primary CNS lymphoma in patients with HIV dis- ease. This may be particularly powerful when used in conjunc- tion with thallium-201 SPECT scanning and the results of Toxo- plasma serologic studies. At the present time, the standard of care for diagnosis of CNS lymphoma is brain biopsy. Since treatment outcome strongly depends on early diagnosis and institution of therapy, it is strongly recommended that individuals who are Toxoplasma se- ronegative, and therefore unlikely to have a diagnosis of toxo- plasmosis, be referred for early brain biopsy if cerebrospinal fluid cytology is negative. Once the diagnosis of CNS lymphoma has been established, all patients should undergo a slit lamp examination before ini- tiation of therapy to rule out the presence of ocular lymphoma. *Affiliations of author: Department of Medicine, San Francisco General Hos- pital, and Department of Medicine, University of California. Correspondence to: Lawrence D. Kaplan, M.D., University of California, San Francisco, San Francisco General Hospital/AIDS Program, Oncology Division, Ward 84, 995 Potrero Ave., San Francisco, CA 941110. © Oxford University Press 101 Treatment In most series reported in the literature, patients have been treated with a course of whole-brain radiotherapy. In three series reporting the occurrence of clinical improvement with therapy (10-12), 62%-79% had improvement in their neurologic symp- toms. Overall response rates of 52%-70% have been reported, with complete response rates of 40%-50% (1,10-14). Median survival has been 2-4.8 months, with most patients dying as a result of nonlymphoma complications from their HIV disease (1,10-14). However, the proportion of treated patients dying as a result of lymphoma progression ranged from 13% to 55%, suggesting that improvement in treatment of the lymphoma itself is clearly needed (/7,10-14). Two small pilot studies (15,16) have investigated the use of combined modality therapy with chemotherapy and radiotherapy being administered. In one series (7/5), 10 individuals treated in this fashion had a complete response rate of 88%, and none of these patients died as a result of their CNS lymphoma. However, the reported median survival of 3.5 months was no better than that reported in studies using whole-brain radiotherapy alone. It is clear then that improvement in the management of CNS lym- phoma will require not only improvement in management of the lymphoma, but also better treatment for the underlying HIV disease. The current use of multiagent antiretroviral therapies, which have been shown to be highly effective, may prolong survival from the underlying immunodeficiency disease long enough so that better management of the lymphoma may soon become a more critical factor. Systemic Lymphomas Although primary CNS involvement is associated with the worst prognosis of all presentations of AIDS-associated lym- phoma, a number of factors have been associated with clinical outcome in systemic disease. Early retrospective studies (7,3) demonstrated factors associated with the underlying immunode- ficiency disorder such as CD4 count, presence or absence of a prior AIDS-defining diagnosis, and performance status to be the most important predictors of outcome. Presence or absence of extranodal disease, particularly bone marrow involvement, was the only tumor-related factor associated with prognosis. Data from more recent studies, such as the prospective data on 192 patients enrolled in the ACTG 142 study (1/7), a study of low-dose versus standard-dose chemotherapy, more closely re- semble those factors identified in the International Prognostic Index (18) for non-immunodeficiency-associated aggressive non-Hodgkin's lymphomas. In this study (/7), factors associated with poor prognosis included age older than 35 years, CD4 count of less than 100/mm®, history of intravenous drug use, and, for the first time, tumor bulk as measured by stage of disease. Ad- vanced stage Ill or IV disease was associated with a poor out- come. Similarly, a recent retrospective review of 96 patients treated for aggressive HIV-associated lymphomas (7/9) indicated that bulky tumor, as measured by elevation of serum lactate dehydrogenase levels, was again associated with a poor clinical outcome. Treatment A number of therapeutic approaches have been utilized in the management of AIDS-associated lymphoma. Initially, dose- 102 intensive regimens were used in an effort to improve outcome. These regimens included high-dose methotrexate and high-dose cytarabine (20) or regimens containing high-dose cyclophospha- mide (3). In general, these treatments were associated with a high risk of death due to opportunistic infection and survival times of 5-6 months, which were not better than those seen with more standard-dose therapy. Patients treated in these studies had median CD4 counts in the mid-100 range. A large clinical trial of the aggressive LNH-84 regimen (217) targeted individuals with higher CD4 counts (median count, 227/ mm?) and was associated with a high complete remission rate, a median survival time of 9 months, and a 2-year disease-free survival of 42%. However, it cannot be demonstrated in this trial whether the improved clinical outcome is due to the aggressive chemotherapy regimen or to the fact that patients with higher CD4 counts are more likely to survive longer. Because of the disappointing results initially observed with higher dose regimens, studies of dose-reduced chemotherapy regimens were subsequently conducted. The AIDS Clinical Tri- als Group (22) utilized an mBACOD regimen (methotrexate, bleomycin, doxorubicin, cyclophosphamide, vincristine, and dexamethasone) in which the dosages of cyclophosphamide and doxorubicin were reduced by approximately 50% in 35 patients who had aggressive AIDS-associated lymphomas. The complete response rate of 46% and the median survival of 5.6 months were not significantly different from those observed in previous trials of standard-dose chemotherapy, but they were achieved at the expense of significantly less hematologic toxicity. In this study (22), the median CD4 count was 150/mm>. Subsequently, an Italian study of a similar low-dose regimen in 37 patients with lymphoma and a median CD4 count of 25/mm?® (23) achieved only a 14% complete remission rate and a median survival of only 3.5 months. Taken together, these data imply a more im- portant effect of immune function on survival than choice of therapeutic regimen. In view of these observations, the AIDS Clinical Trials Group designed a clinical trial to compare directly the importance of chemotherapy dose intensity for clinical outcome in AIDS- associated lymphoma. In this trial, patients were randomly as- signed to receive either standard-dose mBACOD chemotherapy with granulocyte-macrophage colony-stimulating factor (GM- CSF) support or reduced-dose mBACOD with GM-CSF admin- istered only as required for neutropenia (2). The results (Table 1) demonstrated no significant difference in complete response rate, response duration, time to progression, and overall or dis- ease-free survival. And the difference for response duration very nearly reached statistical significance in favor of the low-dose Table 1. Standard- versus low-dose chemotherapy for acquired immunodeficiency syndrome-associated non-Hodgkin’s lymphoma Dose Parameter Standard Low P* Complete response 42/81 (52%) 39/94 (41%) NS Time to progression 30 wk 39 wk NS Time to recurrence 106 wk 190 wk .06 Median survival 31 wk 35 wk NS Toxicity = grade 3 70% patients 51% patients .008 Grade 4 neutropenia 39% cycles 24% cycles .001 Journal of the National Cancer Institute Monographs No. 23, 1998 regimen. When subjects were divided into those with greater than or equal to or less than 100 CD4 cells/mm?, those with better immune function did survive longer than those with poor immune function. However, within each of these two CD4 co- horts, no significant differences were observed with respect to treatment assignment, indicating that these results may apply to the majority of individuals with HIV-associated lymphoma and that the most appropriate choice of chemotherapy regimens for most individuals with AIDS-associated lymphoma should be a dose-reduced treatment regimen. However, an insufficient num- ber of individuals with CD4 counts greater than or equal to 200/mm?* were enrolled in this clinical trial to determine whether some of those patients may have benefited from the use of a stan- dard-dose regimen. For this reason, it is recommended that some patients with relatively intact immune function (CD4=200/ mm?) be considered for standard-dose therapy. Recently, the use of continuous infusion of cyclophospha- mide, doxorubicin, and etoposide has been studied for the man- agement of AIDS-associated lymphoma (24). This treatment ap- proach, piloted by Sparano et al., used a 96-hour continuous infusion of these three agents with granulocyte colony- stimulating factor. Patients assigned to group A received adjunc- tive therapy with didanosine in cycles 1, 2, 5, and 6; those assigned to group B received this antiretroviral agent in cycles 3, 4,5, and 6. The complete response rate observed in this study of 56% was not significantly different from that observed in pre- vious trials; however, the median survival time of 18.4 months reported in this study appeared to be significantly longer than that previously reported. It remains to be seen whether this longer survival time is due to improvement in antiviral therapy, more recent improvement in the overall management of HIV disease, or a regimen that truly produces more durable re- sponses. A large phase II study of this chemotherapy regimen is currently under way by the Eastern Cooperative Oncology Group. For those individuals who have refractory lymphoma, treat- ment outcome is particularly poor. Little data have been reported in the literature. Tirelli et al. (25) recently reported on the use of a combination of etoposide, mitoxantrone, and prednimustine in 21 patients with either primary resistance or relapse after having complete responses. In this study group, the complete response rate was 26%, and the median survival time was only 2 months; the majority of patients ultimately died of refractory lymphoma. Data from San Francisco General Hospital (Kaplan LD: un- published data), in which 14 patients were treated with escalat- ing doses of infusional ifosfamide and etoposide, demonstrated a relatively similar outcome. The overall complete response rate in this series of patients was 43% with a relatively short median response duration of only 79 days. Levine et al. (26) recently reported on the treatment of 35 patients with refractory AIDS-associated lymphoma with the single agent mitoguazone; although this agent appeared to be extremely well tolerated and non-myelosuppressive, the overall response rate of 23% suggests that mitoguazone may be better suited for use in combination regimens than as a single agent. Meningeal Lymphomas The reported incidence of lymphomatous meningitis at the time of diagnosis of systemic lymphoma ranges from 3% to Journal of the National Cancer Institute Monographs No. 23, 1998 25%. Most of these cases are reported from relatively small retrospective series of patients. In the largest prospective series reported, the ACTG 142 study of 192 patients, the incidence of meningeal lymphoma at diagnosis was only 3%, suggesting that the frequency of this complication may be lower than initially suspected (2). The frequency of meningeal relapse is a much more difficult number to determine because most reported series in the literature routinely treat their patients prophylactically with the use of intrathecal chemotherapy. However, in one small, early study, Gill et al. (20) treated 22 patients with a high-dose methotrexate, high-dose cytarabine treatment regi- men. Despite the fact that these are agents that should have adequate CNS penetration, the frequency of meningeal relapse was very high (35%). However, it is noteworthy that seven of the eight relapses occurred in individuals who had bone marrow involvement at the time of their initial diagnosis of lymphoma. Similarly, in the series by Lowenthal et al. (27), two of three individuals developing meningeal relapse had bone marrow in- volvement at diagnosis. At the San Francisco General Hospital, we no longer routinely treat prophylactically all patients with systemic AIDS-associated lymphoma for meningeal recurrence. Rather, we target those individuals with risk factors for meningeal disease identified in the non-immunodeficiency-associated lymphoma population. These risk factors include bone marrow involvement, epidural disease, paranasal sinus involvement, and small, non-cleaved histologic pattern. Antiviral Therapy At this point in time, there is extensive experience in com- bining chemotherapy with nucleoside analogues. With the ex- ception of concerns regarding overlapping myelosuppression when zidovudine (AZT) is used (and its use is therefore not recommended), there is little concern regarding the safety of these antiviral agents while chemotherapy is being administered. There are, however, little data concerning potential drug inter- actions between the protease inhibitors and agents such as cy- clophosphamide and the anthracyclines. Clinical trials are cur- rently investigating these potential drug interactions. Despite this lack of data, we believe that it is important to continue patients on their multiagent antiretroviral therapies while they are being administered chemotherapy. Physicians are urged to follow their patients closely for evidence of excessive toxicity when such combinations are used. Future Therapeutic Approaches There are undoubtedly a variety of mechanisms by which lymphoma arises in individuals with HIV disease. Lymphoma appears to arise out of a background of polyclonal B-cell hy- perproliferation, which may arise through clonal integration of HIV in macrophages, resulting in overexpression of interleukin 6 or as a result of chronic antigenic stimulation from other sources (including viruses or other mitogens) (28). Overexpres- sion of interleukin 10 also appears to be involved in an autocrine fashion (28,29). Other events that may involve viruses, such as EBV or human herpesvirus type 8 (HHVS), or genetic events, such as P53 mutations or c-myc gene rearrangements, may ul- timately give rise to frank lymphoma. At the same time, the cellular immune response to lymphoma 103 may be blocked by a variety of mechanisms, including high concentrations of interleukin 6 and interleukin 10, which may prevent CDS8 cytotoxic lymphocytes from becoming activated (28). The responsiveness of EBV-specific cytotoxic lympho- cytes and natural killer (NK) cells appears to be blunted as well. In vitro data as well as some preliminary in vivo data suggest that cytokines such as interleukin 2 (30) and interleukin 12 may be capable of improving CDS and NK cell responses (37), and clinical trials investigating the use of each of these agents as potential therapeutic modalities or as an adjunct to standard chemotherapy are currently under way. In addition, expansion of EBV-specific cytotoxic lymphocytes used in adoptive immuno- therapy is another therapeutic modality that is currently being explored. Unique molecular characteristics of AIDS-associated lympho- mas can also be taken advantage of with the use of monoclonal antibodies and immunotoxins. Substantial antitumor responses to an anti-CD22-ricin-A-chain immunotoxin have been reported (32), suggesting that this may be a potentially viable approach to therapy in some patients. Anti-B4-blocked ricin has also been studied in individuals with refractory HIV-associated lymphoma with some observed responses, and clinical trials are currently in development for the use of anti-CD20 monoclonal antibodies, including the chimeric anti-CD20 antibody C2B8 which has been used successfully in the management of refractory low-grade lymphoma. Primary Effusion Lymphomas The primary effusion lymphomas constitute only approxi- mately 5% of all HIV-associated lymphomas. These lymphomas present as body cavity effusions (pleural, pericardial, or ascites) that are generally not associated with a contiguous mass lesion and tend to remain localized in the body cavity of origin (33). Morphologically, these lymphomas have a unique histologic pat- tern—one that is midway between that of large-cell immuno- blastic and large-cell anaplastic lymphomas (33). These lympho- mas are unique in their characteristic association with HHVS (33,34). In addition, EBV genome was reported to be present in 100% of 15 tumors analyzed in one series (33) and in 50% of those evaluated in a second series of eight patients (35). Both B- and T-cell-associated antigens are usually absent (33). Clinically, the majority of reported cases have occurred in homosexual or bisexual men (33). The median CD4 lymphocyte count in one study (n = 19) was 84/mm”; in a second study, it was 78/mm® (35). Median survivals have been short (2-5 months), and no long-term disease-free survivors have been re- ported (33,35). However, the number of such cases reported in the medical literature is small, and there may be individuals whose clinical outcome is significantly better than those out- comes reported. We are beginning to learn a great deal about the biology and pathogenesis of the HIV-associated non-Hodgkin's lymphomas. It is hoped that this new knowledge will lead to the development of more pathogenesis-based approaches to therapy and to a greater ability to take advantage of some of the unique molecular features of these lymphomas so that therapies can be identified but will allow us to reduce the amount of cytotoxic therapy required for the management of this disease. 104 References (1) Levine AM, Sullivan-Halley J, Pike MC, Rarick MU, Loureiro C, Bern- stein-Singer M, et al. Human immunodeficiency virus-related lymphoma: prognostic factors predictive of survival. Cancer 1991:68:2466-72. (2) Kaplan LD, Straus DJ, Testa MA, Von Roenn J, Dezube BJ, Cooley TP, et al. Randomized trial of standard-dose vs. low-dose mBACOD chemo- therapy for HIV-associated non-Hodgkin's lymphoma. N Engl J Med 1997; 336:1641-8. Kaplan L, Abrams DI, Feigal E, McGrath M, Kahn J, Neville P, et al. AIDS-associated non-Hodgkin's lymphoma in San Francisco. JAMA 1989: 261:719-24. Kaplan LD, Northfelt D, Kahn J, Volberding P. Degree of immunodefi- ciency at diagnosis and treatment outcome in HIV-associated non- Hodgkin's lymphoma (HIV-NHL) [abstract]. Presented at 7th International Conference on AIDS, Amsterdam, July 19-24, 1992. Vol 2, p B105:# PoB3113. Ciricillo S, Rosenblum M. Use of CT and MR imaging to distinguish intracranial lesions and to define the need for biopsy in AIDS patients. J Neurosurg 1990;73:720-4. Ruiz A, Ganz WI, Post MJ, Camp A, Landy H, Mallin W, et al. Use of thallium-201 brain SPECT to differentiate cerebral lymphoma from toxo- plasma encephalitis in AIDS patients. Am J Neuroradiol 1994;15:1885-94. Barker DE, Trepashko D, DeMarais P, Egofske P, Petry M, Langer B, et al. Utility of thallium brain SPECT in the exclusion of CNS lymphoma in AIDS [abstract]. Presented at 4th Conference on Retroviruses and Oppor- tunistic Infections, Washington, DC, January 22-26. 1997, p 195:# 708. (8) De Luca A, Antinori A, Cingolani A, Larocca LM, Linzalone A, Ammas- sari A, et al. Evaluation of cerebrospinal fluid EBV-DNA and IL-10 as markers for in vivo diagnosis of AIDS-related primary central nervous system lymphoma. Br J Haematol 1995:90:844-9. Cinque P, Brytting M, Vago L. Castagna A, Parravicini C, Zanchetta N, et al. Epstein-Barr virus DNA in cerebrospinal fluid from patients with AIDS- related primary lymphoma of the central nervous system. Lancet 1993:342: 398-401. (10) Baumgartner J, Rachlin J, Beckstead J, Meeker TC, Levy RM, Wara WM, et al. Primary central nervous system lymphomas: natural history and re- sponse to radiation therapy in 55 patients with acquired immunodeficiency syndrome. J Neurosurg 1990:73:206-11. (11) Nisce L, Kaufmann T, Metroka C. Radiation therapy in patients with AIDS-related central nervous system lymphomas. JAMA 1992:267: 1921-2. (12) Lee J, Kaplan LD, Conant M, Northfelt DW. Natural history, treatment outcome and causes of death in AIDS-related primary central nervous system lymphoma (PCNSL). Proc ASCO 1994,18:54. (13) Rubio R, Rubio M, Grauss F, Moreno S, Ribera JM, Pintado V, et al. Primary central nervous system lymphoma (PCNSL) in AIDS. A multi- centric clinical study [abstract]. Presented at 11th International Conference on AIDS, Vancouver, July 7-12, 1996. Vol 2, p 291:# Th.B.4232. (14) Ling SM, Roach M 3rd, Larson DA, Wara WM. Radiotherapy of primary central nervous system lymphoma in patients with and without human immunodeficiency virus. Ten years of treatment experience at the Univer- sity of California, San Francisco. Cancer 1994:73:2570-82. (15) Forsyth PA, Yahalom J, DeAngelis LM. Combined-modality therapy in the treatment of primary central nervous system lymphoma in AIDS. Neurol- ogy 1994:44:1473-9. (16) Chamberlain MC. Long survival in patients with acquired immune defi- ciency syndrome-related primary central nervous system lymphoma. Can- cer 1994:73:1728-30. (17) Straus D, Huang J, Testa M, Levine A, Kaplan L. Prognostic factors in the treatment of HIV-associated non-Hodgkin's lymphoma: analysis of ACTG 142 (low-dose vs. standard-dose mBACOD + GM-CSF). Seattle (WA): American Society of Hematology, December 1-5, 1995, p 604a:# 2404. A predictive model for aggressive non-Hodgkin's lymphoma. The interna- tional non-Hodgkin's lymphoma prognostic factors project. N Engl J Med 1993:329:987-94. (19) Vaccher E, Tirelli U, Spina M, Talamini R, Errante D, Simonelli C, et al. Age and serum lactate dehydrogenase level are independent prognostic factors in human immunodeficiency virus-related non-Hodgkin's lympho- mas: a single-institute study of 96 patients. J Clin Oncol 1996;14:2217-23. (20) Gill P, Levine AM, Krailo M, Rarick MU, Loureiro C, Deyton L, et al. AIDS-related malignant lymphoma: results of prospective treatment trials. J Clin Oncol 1987:5:1322-8. Gisselbrecht C, Oksenhendler E, Tirelli U, Lepage E, Gabarre J, Farcet J, et al. Human immunodeficiency virus-related lymphoma treatment with intensive combination chemotherapy. Am J Med 1993:95:188-96. (3 (4 (5 (6 (7 (9 (18 ~~ (21 Journal of the National Cancer Institute Monographs No. 23, 1998 (22) (23) (24) (25 ic (26) (27) (28) Levine AM, Wernz JC, Kaplan L, Rodman N, Cohen P, Metroka C, et al. Low-dose chemotherapy with central nervous system prophylaxis and zi- dovudine maintenance in AIDS-related lymphoma. JAMA 1991:266: 84-8. Errante D, Tirelli U, Oksenhendler E, Vaccher E, Gastaldi R, Rizzardini G, et al. Prospective study with combined low-dose chemotherapy and zi- dovudine for 37 patients with poor prognosis HIV-related non-Hodgkin's lymphoma. Proc ASCO 1992:11:44. Sparano JA, Wiernik PH, Hu X, Sarta C, Schwartz EL, Soeiro R, et al. Pilot trial of infusional cyclophosphamide, doxorubicin, and etoposide plus didanosine and filgrastim in patients with human immunodeficiency virus-associated non-Hodgkin's lymphoma. J Clin Oncol 1996:14: 3026-35. Tirelli U, Errante D, Spina M, Gastaldi R, Nigra E, Nosari AM, et al. Second-line chemotherapy in human immunodeficiency virus-related non- Hodgkin's lymphoma: evidence of activity of a combination of etoposide. mitoxantrone, and prednimustine in relapsed patients. Cancer 1996:77: 2127-31. Levine AM, Tulpule A, Tessman D, Kaplan L, Giles F, Luskey BD, et al. Mitoguazone therapy in patients with refractory or relapsed AIDS-related lymphoma: results from a multicenter phase II trial. J Clin Oncol 1997:15: 1094-103. Lowenthal DA, Straus DJ, Campbell SW, Gold JW, Clarkson BD, Koziner B. AIDS-related lymphoid neoplasia: the Memorial Hospital experience. Cancer 1988;61:2325-37. Herndier BG, Kaplan LD, McGrath MS. Pathogenesis of AIDS lymphoma. AIDS 1994:8:1025-49. Journal of the National Cancer Institute Monographs No. 23, 1998 (29) (30) (31) (33) (34) (35) Emilie D, Touitou R, Raphael M, Peuchmaur M, Devergnee O, Rea D, et al. In vivo production of interleukin-10 by malignant cells in AIDS lym- phomas. Eur J Immunol 1992;22:2937-42. Bernstein ZP, Porter MM, Gould M, Lipman B, Bluman EM, Stewart CC, et al. Prolonged administration of low-dose interleukin-2 in human immu- nodeficiency virus-associated malignancy results in selective expansion of innate immune effectors without significant clinical toxicity. Blood 1995; 86:3287-94. McFarland EJ, Harding PA, Schooley RT, Kuritzkes DR. Interleukin-12 (IL-12) enhances HIV-specific cytotoxic T-lymphocyte (CTL) activity [ab- stract]. Presented at 3rd Conference on Retroviruses and Opportunistic Infections, Washington, D.C., January 28-February 1, 1996. Vol I, p 95:# 228. Kaplan LD, Robles R, Kahn JO, Schindler J, Ghetie V, Vitetta ES. Phase I trial of anti-CD22 ricin-A-chain immunotoxin (IgG-RFB4-dgA) for treat- ment of AIDS- lymphoma [abstract]. The National AIDS Malignancies Conference, April 28-30, 1997, Bethesda, MD. Nador RG, Cesarman E, Chadburn A, Dawson DB, Ansari MQ, Sald J, et al. Primary effusion lymphoma: a distinct clinicopathologic entity associ- ated with the Kaposi's sarcoma-associated herpes virus. Blood 1996:88: 645-56. Nador RG, Cesarman E, Knowles DM, Said JW. Herpes-like DNA se- quences in a body-cavity-based lymphoma in an HIV-negative patient. N Engl J Med 1995:33:943. Komanduri KV, Luce JA, McGrath MS, Herndier BG, Ng VL. The natural history and molecular heterogeneity of HIV-associated primary malignant lymphomatous effusions. JAIDS 1996:13:215-26. 105 U. C. BERKELEY LIBRARIES C0L093007e >