key: cord-290041-zxlq63n5
authors: Srivastava, Arnav; Patel, Hiren V.; Kim, Sinae; Shinder, Brian; Sterling, Joshua; Tabakin, Alexandra L.; Polotti, Charles F.; Saraiya, Biren; Mayer, Tina; Kim, Isaac Y.; Ghodoussipour, Saum; Patel, Hiten D.; Jang, Thomas L.; Singer, Eric A.
title: Delaying Surgery for Clinical T1b-T2bN0M0 Renal Cell Carcinoma: Oncologic Implications in the COVID-19 Era and Beyond
date: 2020-10-20
journal: Urol Oncol
DOI: 10.1016/j.urolonc.2020.10.012
sha: 
doc_id: 290041
cord_uid: zxlq63n5

PURPOSE: During COVID-19, many operating rooms were reserved exclusively for emergent cases. As a result, many elective surgeries for renal cell carcinoma (RCC) were deferred, with an unknown impact on outcomes. Since surveillance is commonplace for small renal masses, we focused on larger, organ-confined RCCs. Our primary endpoint was pT3a up-staging and our secondary endpoint was overall survival (OS). MATERIALS AND METHODS: We retrospectively abstracted cT1b-T2bN0M0 RCC patients from the National Cancer Database (NCDB), stratifying them by clinical stage and time from diagnosis to surgery. We selected only those patients who underwent surgery. Patients were grouped by having surgery within 1 month, 1-3 months, or >3 months after diagnosis. Logistic regression models measured pT3a up-staging risk. Kaplan Meier curves and Cox proportional hazards models assessed OS. RESULTS: 29,746 patients underwent partial or radical nephrectomy. Delaying surgery >3 months after diagnosis did not confer pT3a up-staging risk among cT1b (OR=0.90; 95%CI: 0.77–1.05, p=0.170), cT2a (OR=0.90; 95%CI: 0.69–1.19, p=0.454), or cT2b (OR=0.96; 95%CI:0.62–1.51, p=0.873). In all clinical stage strata, non-clear cell RCCs were significantly less likely to be up-staged (p<0.001). A sensitivity analysis, performed for delays of <1, 1-3, 3-6, and >6 months, also showed no increase in up-staging risk. CONCLUSIONS: Delaying surgery up to, and even beyond, 3 months does not significantly increase risk of tumor progression in clinically localized RCC. However, if deciding to delay surgery due to COVID-19, tumor histology, growth kinetics, patient comorbidities, and hospital capacity/resources, should be considered.

The emergence and rapid spread of the 2019 Novel Coronavirus (COVID-19) has disrupted healthcare systems and highlighted new perspectives on healthcare delivery. As a result of the pandemic's exponential growth, hospitals, in conjunction with professional body recommendations, have deferred non-emergent surgeries. 1 These delayed surgeries include many potentially curative urologic oncologic surgeries 2, 3 , such as partial and radical nephrectomies for renal cell carcinoma (RCC), which remain the preferred curative treatment for localized kidney cancer. 4 As a result of the COVID-19 pandemic, many patients with known renal masses will experience surgical treatment delays. For oncology patients, delay of definitive treatment may increase the risk of progression to metastatic disease and consequently allow some neoplasms to transform from curable to uncurable. 5 Thus, urologists and kidney cancer patients must understand the effects of treatment delay, both during the COVID-19 pandemic and in the coming months. 6 Active surveillance (AS) literature has demonstrated that urologists can safely monitor most cT1a renal masses. 4, 7, 8 However, for cT1b-cT2b masses this is less clear. The effect of surgical delay on kidney cancer outcomes has not been robustly evaluated. 9, 10, 11 In this current study, we examine oncologic outcomes as a function of time to surgery using the National Cancer Database (NCDB). Specifically, we selected only those patients who underwent surgery to best model the effects of surgical delay on patients scheduled to receive surgery.

The NCDB, a national oncology registry, records approximately 70% of the incident cancer cases in the United States. It contains information regarding patient demographics, neoplasm staging (both clinical and pathologic staging), surgical treatment, and survival outcomes. 12, 13, 14 Study Population Exploratory analysis identified 422,258 patients with kidney cancer between 2004 and 2014. We excluded patients with missing staging data or histology. NCDB captures the first course of treatment for patients. Patients receiving percutaneous ablation, active surveillance (AS), or who never underwent surgery (partial or radical nephrectomy) were excluded. The study population only contains cT1b-T2bN0M0 patients. Lastly, we identified and excluded 23 ,065 patients whose time from diagnosis of kidney cancer to surgery was coded as "0 days."

Patients were stratified into 3 groups by clinical stage (cT1b, cT2a, and cT2b). Within each clinical stage strata, patients were grouped by their time from diagnosis to surgery: <1 month, 1-3 months, and >3 months. Means and medians assessed continuous variables.

We examined how surgical delay may increase risk of pathologic T3a (pT3a) up-staging using logistic regression. We defined up-staging in patients with clinically localized disease (cT1-cT2) who had pT3a disease after receiving partial or radical nephrectomy. For example, we did not classify a patient with a cT1b lesion who is found to have pT2a disease as upstaged. We excluded those up-staged to pT3b tumors as these constituted only 1.7% of the entire cohort.

Survival analyses conducted within each clinical stage strata grouped patients by time from diagnosis to surgery. Only overall survival (OS) was assessed, as the NCDB does not contain information regarding cancer-specific survival (CSS) nor recurrence-free survival.

Kaplan-Meier curves illustrated OS among these patient groups and produced survival percentages at one, three, five and ten years. Log rank tests quantified statistical differences in survival. Cox proportional hazards modeled how delays in surgery correlated to OS in both univariate and multivariable models. Multivariable models were constructed to adjust for various clinical factors, such as histology, tumor size, insurance status, and ethnicity.

Lastly, we conducted a sensitivity analysis for pT3a up-staging and OS, by substratifying surgical delay into <1 month, 1-3 months, 3-6 months, and >6 months. All analyses were conducted using SAS statistical software package (v. 9.4; SAS Institute Inc., Cary, North Carolina).

The final analytical cohort encompassed 29,746 patients with cT1b (N=22,719), cT2a (N=4,670) and cT2b (N=2,357) tumors. While median time from diagnosis to surgery was similar between clinical stage groups, we found cT1b (13.5%) had a higher proportion of patients who had surgery >3 months after diagnosis compared to cT2a (8.0%) and cT2b (5.3%) (Table 1) .

We observed up-staging rates of 9.1%, 24.4%, and 32.3% for cT1b, cT2a, and cT2b tumors, respectively (Supplemental Figure 1 ). As clinical stage increases, overall rates of upstaging increase regardless of surgical waiting time (Figure 1) Figure 2ABC ).

Supplemental Tables 4-6 re-examine pT3a up-staging risk by further dividing the surgical delay >3 months category in to 3-6 months and >6 months. Multivariable logistic regression still did not show correlation between surgical delay and up-staging for cT1b-T2b RCC. In the sensitivity analysis for OS, we find again that surgical delay of 1-3, 3-6, and >6 months all portend poorer survival in cT1b tumors (Supplemental Table 7 ). In cT2a masses, the new stratification finds that patients with a delay of surgery >6 months have worse survival (Multivariable HR=2.21 (95% CI: 1.33-3.66), p<0.001). However, delays of 1-3 and 3-6 months had similar survival to those without surgical delay (Supplemental Table 8 ). Lastly, the expanded stratification still showed no relationship between surgical delay and survival of patients with cT2b tumors (Supplemental Table 9 ).

As COVID-19 continues to stress healthcare systems in many countries, hospitals have required the delay of time-sensitive, non-emergent procedures. This has included partial and radial nephrectomy for RCC. Prospective AS data shows that urologists can observe most cT1a tumors for extended periods of time, without detriment to patient survival. 7 However, for larger localized tumors, cT1b, the effects of surgical delay are unclear. Our retrospective analysis using the NCDB is the largest study to date to examine the risks of surgical delay on up-staging and OS in cT1b-T2b RCC. Within each clinical stage strata, surgical delay did not increase the risk of pT3a up-staging for cT1b-T2b RCC (Tables 2-4 ). The up-staging rates by clinical stage (cT1b=9.10%, cT2a=24.39%, cT2b=32.26%) seen here are corroborated by prior multiinstitutional and population-based studies (Figure 1, Supplemental Figure 1 ). 15 , 16 We also find that patients with non-clear cell histology (excluding sarcomatoid) have reduced risk of upstagingsuggesting that histologic information may help surgeons triage patients, as those with non-clear cell histology appear to be less impacted by surgical delay. Regarding OS, we note that surgical delay predicts worse survival for cT1b tumors, but not cT2 tumors (Table 5, Figure   2 ).

To date, one systematic review and a few narrative reviews have described how delayed surgery affects oncologic outcomes in localized RCC; there is, however, a relative paucity of data for T1b masses. 9 Modelling surgical waiting time as a continuous variable, they found no association with surgery delay and pT3a up-staging, 5-year recurrence, or CSS. However, they did note poorer OS (HR=1.15, p=0.015). Importantly, older age, high comorbidity index, non-White ethnicity, nonclear cell histology, and partial nephrectomy correlated with being placed in the delayed surgery group. 19 Mehrazin et al. examined a series of 72 cT1b masses initially selected for AS, 23 of which underwent surgery. In this cohort no patient progressed to metastatic disease and linear growth rates were slow (median=0.34 cm/year), suggesting a delay of several months unlikely precludes these masses from complete resection. 20 Our analysis closely matches these findings, particularly among cT1b RCC. We find no association between surgical delay and pT3a up-staging at 1-3 months of delay or >3 months (Table 2 ). pT3a up-staging often results in worse oncologic outcomes, compared to those patients with concordant clinical and pathologic staging. 13, 15, 21 Also as these tumors may remain confined to the renal capsule and exhibit favorable growth rate, complete resection is likely even after a surgical delay. 20 Additionally, we find that surgical delay of 1 month or more (1-3 months or >3 months) confers worse survival among cT1b masses. However, these patients still experience favorable OS (5-year OS: surgical delay 1-3 months=80.1%, >3 months=70.9%) ( Figure 2A ). As surgical delay does not seem to predict up-staging in our study or CSS in prior studies-factors such as comorbidity index (Charlson Index ≥2 Multivariable HR=2.24, 95% CI: 2.02-2.47; p<0.001) and unmeasured confounding, rather than tumor progression, likely drive differential OS. For example, older patients or those with more comorbidities, may require more deliberate coordination and optimization prior to their surgery. Consequently, patients at higher risk for surgery may have delayed operations, reflected in a reduced OS.

In our analysis of cT2 lesions surgical delay does not portend poor OS nor risk of pT3a up-staging. However, substantially fewer patients underwent surgery >3 months after diagnosis for cT2a (N=370) and cT2b (N=123) tumors, thus limiting our statistical power. Kim et al.

(N=309) and Shiff et al. (N=391) examined surgical delay in cT2 masses. Neither retrospective analysis found worse up-staging rates, recurrence-free survival. CSS, or OS. 10, 18, 22 We must also consider more deleterious in comorbidities for patients with cT2 renal masses, which likely require careful medical optimization and coordination prior to surgery. Though our data does not show obvious harm from surgical delay, confounding patient variables and tumor growth rates make it difficult to definitively state whether surgical delay in cT2 renal masses is prudent. 23 Our study also has some key differences from prior data, specifically the recent publication by Shiff et al. In addition to a shorter follow-up (median=22.8 months), the Canada-based population also had longer median wait times compared to our cohort for cT1b (2.6 vs. Our study has several limitations to consider including those inherent to the NCDB such as its retrospective nature. 24 Our data has relatively short follow-up, especially considering OS was our survival endpoint. Additionally, we could not assess CSS or progression/recurrencefree survival as oncologic endpoints, which is why OS was a secondary endpoint. However, our primary end point was the rate of= pT3a up-staging to gauge the risk of neoplasm progression associated with surgical delay, for which the NCDB is well suited as it captures both clinical and pathologic staging. We did not include those patients up-staged to pT3b disease after surgery.

While this represents an important subset of our patients, we focused on pT3a up-staging to be in concordance with prior literature. Furthermore, as only 1.7% were up-staged to pT3b, it is unlikely that including this subset of patients would change our clinical interpretation. We also note that pT3a up-staging may also represent limitations in clinical staging, rather than true oncologic progressionin particular cT2 tumors, of which 20% may be up-staged. 16 Additionally, our grouping points of <1 month, 1-3 months, >3 months were selected based on prior literature and our clinical practice during the COVID-19 pandemic where local institutions delayed most kidney cancer cases by 1-3 months. Furthermore, this selection of surgical delay strata allows us to preserve sample size, particularly for cases delayed >3 months in T2 masses. However, the authors acknowledge that different time points may alter regression results. We have mitigated this limitation by including a sensitivity analysis, dividing the >3 month time to surgery group into 3-6 months, and >6 monthswe note similar results.

Linear growth rates of renal mass, which are not provided in the NCDB, are often variable and may provide rationale for either prompt surgery or AS. Similarly, we do not have granular tumor pathology data. Thus tumor nephrometry scores are not available to quantify mass complexity, we cannot differentiate hilar, endophytic tumors from polar, exophytic ones. We also are not able to sub-stratify pT3a tumors (i.e. renal vein, sinus fat, perinephric fat invasion) to best understand which patients require more urgent treatment. Additionally, while the NCDB lists Charlson-Deyo comorbidity index, which we controlled for in multivariable analysis, it does not provide detailed information regarding comorbidities which may complicate or delay surgery.

Lastly, the analyzed data did not contain granular data regarding why a patient had delayed surgery. Regardless, by focusing on cT1b masses, not typically selected for AS, we reduce this limitation.

Despite these limitations, our study builds significantly on prior work. We recommend that for most patients with cT1b tumors, surgery can be safely delayed up to 3-6 months without significantly reducing survival. Among cT2 tumors, our analysis did not find a significant effect of surgical delay on survival or up-staging. However, we caution providers and patients, that this may represent data limitations. For patients with these larger clinically localized masses, a thorough understanding of a tumor's biology, histology, and patient comorbidities are required when discussing delayed intervention. Although the authors do not recommend delaying resection of tumors, our manuscript aims to guide and reassure providers as they treat kidney cancer during the COVID-19 crisis. Even beyond the current pandemic, urologists, patients, and hospitals must make challenging decisions on how to allocate care and resources while maximizing health outcomes as the healthcare system slowly recovers.

During the current COVID-19 pandemic and subsequent recovery, urologists and their patients can expect delays in radical and partial nephrectomy for clinically localized RCC. In most patients with clinically localized cT1b tumors, surgery may be safely delayed for up to 6 months without significant sacrifices in overall survival. In patients with cT2 disease, we must carefully weigh tumor characteristics and patient comorbidities when discussing surgical delay.

However, our data suggests that most patients experiencing a delay of 3 months due to the COVID-19 pandemic will not experience worse oncological outcomes.

This work is supported by a grant from the National Cancer Institute (P30CA072720).

Services, results and/or products in support of the research were generated by Rutgers Cancer Institute of New Jersey Biometrics Shared Resource P30CA072720-5918. 

Considerations in the Triage of Urologic Surgeries During the COVID-19

Delayed Radical Prostatectomy Is Not Associated with Adverse Oncological Outcomes: Implications for Men Experiencing Surgical Delay Due to the COVID-19 Pandemic

Renal Mass and Localized Renal Cancer: AUA Guideline

Collateral damage: the impact on outcomes from cancer surgery of the COVID-19 pandemic

Urologic oncology surgery during COVID-19: a rapid review of current triage guidance documents

Comparative effectiveness of management options for patients with small renal masses: a prospective cohort study

Small Renal Mass Surveillance: Histology-specific Growth Rates in a Biopsy-characterized Cohort

Risks from Deferring Treatment for Genitourinary Cancers: A Collaborative Review to Aid Triage and Management During the COVID-19 Pandemic

Urologic Oncology Practice During COVID-19 Pandemic: A Systematic Review on What can be Deferrable vs

Waiting time from initial urological consultation to nephrectomy for renal cell carcinoma--does it affect survival?

The National Cancer Data Base: A Powerful Initiative to Improve Cancer Care in the United States

Impact of Pathologic Lymph Node-Positive Renal Cell Carcinoma on Survival in Patients without Metastasis: Evidence in Support of Expanding the Definition of Stage IV Kidney Cancer

Factors Associated With Receipt of Partial Nephrectomy or Minimally Invasive Surgery for Patients With Clinical T1a and T1b Renal Masses: Implications for Regionalization of Care

Outcomes and predictors of clinical T1 to pathological T3a tumor up-staging after robotic partial nephrectomy: a multi-institutional analysis

Incidence of T3a up-staging and survival after partial nephrectomy: Size-stratified rates and implications for prognosis

Impact of the length of time between diagnosis and surgical removal of urologic neoplasms on survival

Impact of Time-To-Surgery and Surgical Delay on Oncologic Outcomes for Renal Cell Carcinoma

The effect of delaying nephrectomy on oncologic outcomes in patients with renal tumors greater than 4cm

Growth kinetics and short-term outcomes of cT1b and cT2 renal masses under active surveillance

Multi-institutional Survival Analysis of Incidental Pathologic T3a Upstaging in Clinical T1 Renal Cell Carcinoma Following Partial Nephrectomy

The impact of delaying radical nephrectomy for stage II or higher renal cell carcinoma

Outcomes of Robot-assisted Partial Nephrectomy for Clinical T2 Renal Tumors: A Multicenter Analysis (ROSULA Collaborative Group)

Using the National Cancer Database for Outcomes Research: A Review