key: cord-312702-fruzsn26 authors: Finch, Courtney L.; Crozier, Ian; Lee, Ji Hyun; Byrum, Russ; Cooper, Timothy K.; Liang, Janie; Sharer, Kaleb; Solomon, Jeffrey; Sayre, Philip J.; Kocher, Gregory; Bartos, Christopher; Aiosa, Nina M.; Castro, Marcelo; Larson, Peter A.; Adams, Ricky; Beitzel, Brett; Di Paola, Nicholas; Kugelman, Jeffrey R.; Kurtz, Jonathan R.; Burdette, Tracey; Nason, Martha C.; Feuerstein, Irwin M.; Palacios, Gustavo; Claire, Marisa C. St.; Lackemeyer, Matthew G.; Johnson, Reed F.; Braun, Katarina M.; Ramuta, Mitchell D.; Wada, Jiro; Schmaljohn, Connie S.; Friedrich, Thomas C.; O’Connor, David H.; Kuhn, Jens H. title: Characteristic and quantifiable COVID-19-like abnormalities in CT- and PET/CT-imaged lungs of SARS-CoV-2-infected crab-eating macaques (Macaca fascicularis) date: 2020-05-14 journal: bioRxiv DOI: 10.1101/2020.05.14.096727 sha: doc_id: 312702 cord_uid: fruzsn26 Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is causing an exponentially increasing number of coronavirus disease 19 (COVID-19) cases globally. Prioritization of medical countermeasures for evaluation in randomized clinical trials is critically hindered by the lack of COVID-19 animal models that enable accurate, quantifiable, and reproducible measurement of COVID-19 pulmonary disease free from observer bias. We first used serial computed tomography (CT) to demonstrate that bilateral intrabronchial instillation of SARS-CoV-2 into crab-eating macaques (Macaca fascicularis) results in mild-to-moderate lung abnormalities qualitatively characteristic of subclinical or mild-to-moderate COVID-19 (e.g., ground-glass opacities with or without reticulation, paving, or alveolar consolidation, peri-bronchial thickening, linear opacities) at typical locations (peripheral>central, posterior and dependent, bilateral, multi-lobar). We then used positron emission tomography (PET) analysis to demonstrate increased FDG uptake in the CT-defined lung abnormalities and regional lymph nodes. PET/CT imaging findings appeared in all macaques as early as 2 days post-exposure, variably progressed, and subsequently resolved by 6–12 days post-exposure. Finally, we applied operator-independent, semi-automatic quantification of the volume and radiodensity of CT abnormalities as a possible primary endpoint for immediate and objective efficacy testing of candidate medical countermeasures. Finch et al. 5 29). However, in both human disease and animal models, the temporal and mechanistic 90 relationship between viral replication, subsequent immunopathology (30, 31) , and clinical 91 disease remains uncertain. Furthermore, in the available NHP models, all of which are 92 sublethal, markers of clinical disease (cage-side scoring, chest X-ray) have been of 93 limited sensitivity. More concerningly, both metrics are subject to observer bias (32-35). 94 Reliable animal models needed for rapid development and evaluation of candidate 95 medical countermeasures (MCMs) require an unbiased reproducible and quantifiable 96 metric of disease that mirrors key aspects of COVID-19. Based on the rather limited X-97 ray findings in the lungs of reported NHP models of SARS-CoV-2 infection with either 98 mild or no clinical signs (11, 25, 27-29), we turned to high-resolution chest CT and 99 PET/CT to characterize lung abnormalities in infected NHPs toward longitudinal 100 quantitative comparison. 101 We used direct bilateral primary intrabronchial instillation in a 1-day-apart 102 staggered design to expose two groups of three crab-eating macaques (Macaca 103 fascicularis) to medium (mock group macaques M1-3) or medium including 3.65x10 6 104 pfu/macaque of SARS-CoV-2 (virus group macaques V1-3) (Supplementary Table 1) . 105 All macaques were observed daily for 11 days prior to exposure (day [D] 0) and for 30 106 days post-exposure. Physical examination scores and blood, conjunctival, 107 nasopharyngeal, oropharyngeal, rectal, fecal, and urine specimens were collected at 108 identical timepoints. Virus-exposed macaques were indistinguishable from mock group 109 macaques during the pre-exposure time period. Two pre-exposure chest CT and whole-110 previously published results (11, (25) (26) (27) (28) (29) , none of the macaques developed any major 113 clinical abnormalities (including by cage-side assessment and clinical scoring or physical 114 examination) throughout the study and clinical laboratory results were not significantly 115 different between the mock-exposed and virus-exposed groups (Supplementary Tables 116 2-3) . SARS-CoV-2 RNA could not be detected by RT-qPCR in any sample from mock-117 exposed macaques but was variably present during the early days post-exposure in 118 conjunctival, fecal, nasopharyngeal, oral, and rectal swabs, but never in plasma or urine 119 (Figure 2a) . Anti-SARS-CoV-2 IgG antibodies were not detectable by ELISA in mock-120 exposed macaques but were detectable at D10 post-exposure and continued to rise in all 121 virus-exposed macaques to at least D19 (Figure 2b) . Consistent with ELISA results, 122 fluorescent neutralization titers generated from sera were undetected until D10 and were 123 detected only in virus-exposed macaques (Figure 2c) . Longitudinal measurement of 124 selected peripheral cytokines revealed between-and within-group differences with 125 marked abnormalities noted in macaque V3, which also had the highest IgG antibody 126 titers (Supplementary Figure 1) . 127 With the exception of minor and transient abnormalities on baseline imaging, CT 128 scans of all mock-exposed macaques appeared generally normal over the entire study 129 period (Supplementary Figure 2) . However, all virus group macaques developed lung 130 abnormalities clearly visible by chest CT as early as D2. Qualitatively, the distribution 131 morphology, and duration of abnormalities described a spectrum similar to mild-132 moderately ill humans with COVID-19. Characteristic CT findings in all virus group 133 macaques included bilateral peripheral GGOs variably in association with intra-or 134 interlobular septal prominence (so-called "crazy paving"), reticular or reticulonodular 135 SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 7 opacities, peri-bronchial thickening, subpleural nodules, and, in one macaque, dense 136 alveolar consolidation with air bronchograms (Figures 3-4a, Videos 1-3 Increased FDG uptake detected by PET (Figure 6, Supplementary Figures 4-5 ) 146 corresponded well to the structural changes in the lungs observed by CT, and regional 147 lymph node uptake was seen in all virus group macaques at D2. In macaque V1, FDG 148 uptake decreased in the lungs at D6 but increased in mediastinal lymph nodes, and new 149 FDG uptake was identified in the spleen. The two macaques (V2, V3) with persistent or 150 progressive structural abnormalities on chest CT had variable changes in FDG uptake 151 associated with the structural abnormalities in the lungs (some markedly increased, some 152 improved) with an accompanying marked increase in FDG uptake in regional lymph 153 nodes and spleens on D6. PET scan on D12 revealed normalization of previous areas of 154 increased FDG uptake in the lung parenchyma in all three virus-exposed macaques, and 155 persistent increased FDG update in regional lymph nodes and spleen. Mock-exposed 156 macaques did not have similar increased FDG uptake with the exception of transient 157 increased uptake in regional lymph nodes after mock-exposure in a single macaque (M1). 158 SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 8 Quantification of the SUVmax in selected regions of interest (ROI) in the lung, specific 159 regional lymph nodes, and spleen mapped well to the qualitative findings in both mock-160 exposed and virus-exposed macaques (Figure 7) . 161 CT images can be used for quantification of lung abnormalities using measures of 162 volume or radiodensity, i.e., total lung volume (LV); average radiodensity in the total 163 lung volume (LD); hyperdense volume (HV), a volume of lung in which radiodensity 164 (Hounsfield units, HU) is above a pre-defined threshold; and average radiodensity in the 165 hyperdense volume (hyperdensity, HD). Normalized changes from a pre-exposure 166 baseline can be longitudinally measured as the percent change in the volume of lung 167 hyperdensity (PCLH). Toward standardization across lung volumes, PCLH can be also be 168 expressed as a percent of total lung volume (PCLH/LV). Increases in PCLH or PCLH/LV 169 were not seen in the mock-exposed macaques over the entire study (Figure 8a A key advantage of quantifiable CT chest imaging readout over serial euthanasia 212 studies, in addition to potentially reduced experimental animal numbers, is the ability not 213 only to evaluate between-group differences, but also to compare severity and duration of 214 disease at higher resolution in single animals and even in isolated parenchymal areas 215 sequentially. This approach can reduce the error inherent in cross-sectional sampling of 216 individual animals at single timepoints. Imaging does, however, introduce its own 217 experimental complexities and limitations. As we aimed to evaluate whether PCLH (or 218 other CT imaging readouts presented in Figure 8 ) could be a useful quantitative readout 219 for radiographic progression in the SARS-CoV-2 infected lung, we chose not to include 220 irradiated inactivated SARS-CoV-2 in the mock inoculum to avoid antigen-induced 221 inflammation and related radiographic changes. For similar reasons, and to avoid 222 artificial dissemination of SARS-CoV-2, we specifically did not perform bronchoalveolar 223 lavage (BAL) to obtain lung samples for downstream cellular, molecular, and virologic 224 analysis (45, 46) and did not perform lung biopsies. The frequency of anesthesia and 225 instrument availability pragmatically limit imaging to carefully chosen timepoints during 226 SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 11 a study. In particular, the extended time required to perform PET imaging resulted in 227 logistical limitations of the number of macaques that could be included in the study. 228 Finally, with complete resolution of radiographic abnormalities by the end of the study 229 period, we opted not to euthanize these macaques to be able to perform a re-exposure 230 study in the future. Thus, we cannot correlate radiographic with histopathologic findings. 231 Future studies should extend our initial findings in several directions. First, 232 follow-up confirmation of these pilot results in this model of mild-moderate COVID-19 233 is needed to further establish quantifiable lung CT as a reliable disease readout and to 234 forge imaging-pathologic correlates in macaques euthanized at peak radiographic 235 abnormality. Confirmation should enable proof-of-concept evaluation of whether a 236 candidate MCM will indeed significantly decrease peak or AUC of PCLH or PCLH/LV 237 compared to untreated infected control macaques. Data from additional macaques will be 238 used to confirm the sensitivity and relevance of the AUC0-8 and AUC0-30 for PCLH or 239 PCLH/LV as robust measures of lung changes from CT evaluation. 240 In parallel, disease severity could possibly be increased in the crab-eating 241 macaque model by optimizing delivery of SARS-CoV-2 to the most vulnerable lung (via 242 aerosol or more distal bronchoscopic delivery), with the ultimate goal of using the CT-243 quantifiable volume or radiodensity readouts to model the sick hospitalized human. 244 Other groups are already evaluating NHPs of diverse species as possible COVID-245 19 models. In these models, serial chest CT imaging after intrabronchial instillation of 246 SARS-CoV-2 could be used to establish a meaningful and quantifiable COVID-19-like 247 disease readout that will enable objective evaluation of medical countermeasures and also 248 a comparison of SARS-CoV-2-induced lung abnormalities in different NHP models. The macaques were split into 2 groups of 3 animals each (Supplementary Table 1) . 315 Mock group (M) macaques received 2 ml of DMEM + 2% heat-inactivated FBS into each 316 bronchus by direct bilateral primary post-carinal intrabronchial instillation, followed by a 317 1-ml normal saline flush and then 5 ml air. Virus group (V) macaques were exposed the SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 15 same way with each 2-ml instillate containing 9.13x10 5 pfu/ml (i.e., a total exposure dose 319 of 3.65x10 6 pfu) of SARS-CoV-2 followed by 1-ml saline flush and then 5 ml air. All 320 macaques were sedated prior to instillation. Prior to administering anesthesia, 321 glycopyrrolate (0.06 mg/kg) was delivered intramuscularly to reduce saliva secretions. 322 Next, each macaque received 10 mg/kg ketamine and then 35 μg/kg dexmedetomidine 323 intramuscularly. To reverse anesthesia, 0.15 mg/kg atipamezole was administered 324 intravenously. All macaques were evaluated daily for health and were periodically 325 Table 4 ). Cage-side and physical exam scoring 336 criteria were developed in collaboration with National Primate Research Centers 337 (NPRCs) to standardize disease assessment and compare disease outcomes between NHP 338 models. Heart rate was not incorporated into the physical exam scores until D2 because 339 heart rate score was determined as beats per minute over baseline. Baseline heart rate was 340 determined as an average over three timepoints, D-11, D-5/D-6, and D0 (except for 341 Normalized with baseline Density/radiodensity difference with baseline A new coronavirus associated with human respiratory disease in A pneumonia outbreak associated with a new coronavirus of 769 probable bat origin A novel coronavirus from patients with pneumonia in China World Health Organization, Coronavirus (COVID-19 Clinical course and outcomes of critically ill patients with SARS-775 CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, 776 observational study Clinical characteristics of coronavirus disease 2019 in China Clinical findings in a group of patients infected with the 2019 780 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case 781 series Covid-19 in critically ill patients in the Seattle region -case 783 series Clinical features of patients infected with 2019 novel coronavirus 785 in Wuhan SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 36 The incubation period of coronavirus disease 2019 (COVID-787 19) from publicly reported confirmed cases: estimation and application Age-related rhesus macaque models of COVID-19 COVID-19): A Critical 792 Review of the Literature to Date Epidemiological and clinical characteristics of 26 asymptomatic 794 SARS-CoV-2 carriers Incidental findings suggestive of Covid-19 in asymptomatic 796 patients undergoing nuclear medicine procedures in a high prevalence region Histopathologic changes and SARS-CoV-2 immunostaining in 799 the lung of a patient with COVID-19 Pathological findings of COVID-19 associated with acute respiratory 801 distress syndrome Multiscale 3-dimensional pathology findings of COVID-19 diseased 803 lung using high-resolution cleared tissue microscopy. bioRxiv Pulmonary and cardiac pathology in Covid-19: the first autopsy 806 series from New Orleans. medRxiv Autopsy findings and venous thromboembolism in patients 808 with COVID-19: a prospective cohort study SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 37 The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice Susceptibility of ferrets, cats, dogs, and other domesticated animals to Infection and rapid transmission of SARS-CoV-2 in ferrets Simulation of the clinical and pathological manifestations of 816 coronavirus disease 2019 (COVID-19) in golden Syrian hamster model: 817 implications for disease pathogenesis and transmissibility Susceptibility of tree shrew to SARS-CoV-2 infection. bioRxiv Ocular conjunctival inoculation of SARS-CoV-2 can cause mild COVID-19 in Rhesus macaques. bioRxiv Comparative pathogenesis of COVID-19, MERS, and SARS in a 823 nonhuman primate model Infection with novel coronavirus (SARS-CoV-2) causes 825 pneumonia in the rhesus macaques Comparison of SARS-CoV-2 infections among 3 species of non-828 human primates. bioRxiv Respiratory disease in rhesus macaques inoculated with 830 SARS-CoV-2 SARS-CoV-2 pulmonary abnormalities in macaques Finch et Dysregulation of immune response in patients with COVID-19 in COVID-19: immunopathology and its implications for therapy Frequency and distribution of chest radiographic findings in COVID-19 positive patients Interobserver reliability of the chest radiograph in 838 community-acquired pneumonia Portable chest X-ray in coronavirus 841 disease-19 (COVID-19): a pictorial review Chest x-ray findings in 636 ambulatory patients with COVID-19 presenting to an urgent care center: a normal chest x-ray is no 844 guarantee Bias in radiology: the how and 846 why of misses and misinterpretations Error and discrepancy in radiology: inevitable or avoidable? Insights 848 Imaging Observer bias 850 in lung nodule detection with spiral CT Interpretive error in radiology SARS-CoV-2 pulmonary abnormalities in macaques Finch et al. 39 Cytokine release syndrome in severe COVID-19 On the alert for cytokine storm: immunopathology in 856 COVID-19 Clinical and immunological features of severe and moderate 858 coronavirus disease 2019 Clinical predictors of mortality 860 due to COVID-19 based on an analysis of data of 150 patients from Wuhan Risk factors for severity and mortality in adult COVID-19 inpatients 863 in Wuhan Bronchoalveolar lavage affects computed tomographic and 868 radiographic characteristics of the lungs in healthy dogs Inactivation and safety testing of Middle East Respiratory 871 Syndrome Coronavirus Panmicrobial oligonucleotide array for diagnosis of infectious 873 diseases SARS-CoV-2 pulmonary abnormalities in macaques Finch et Influenza A and methicillin-resistant Staphylococcus aureus 877 co-infection in rhesus macaques -A model of severe pneumonia The nonhuman primate in nonclinical drug development and 880 safety assessment Time course of lung changes on chest CT during recovery from 2019 883 novel coronavirus (COVID-19) pneumonia Mass preserving image registration for lung CT 3D Slicer as an image computing platform for the Quantitative 890 Imaging Network A rhesus macaque model of Asian-lineage Zika virus 892 infection Anti-SARS-CoV-2 S1 Subunit IgG ELISA