Thompson, R. Houston

Abstract INTRODUCTION AND OBJECTIVES: Patient responses to cancer immunotherapy such as checkpoint inhibitor drugs vary greatly between patients and there remains no biomarker to predict sensitivity to these drugs. We describe a patient derived xenograft (PDX) avatar system that predicts cancer immunotherapy responses in the context of tumor heterogeneity within 8-10 days. This avatar model preserves the original tumor microenvironment and tumor infiltrating lymphocytes (TILs) at a high tumor take rate for evaluation of various immunotherapy drugs. This prediction model also involves tumor infiltrating lymphocyte isolation and primary cell generation for confirming the immunotherapy response in vitro. This PDX avatar platform generates rapid actionable information and readily identifies patients that would maximally benefit from checkpoint inhibitor therapies in a phenotype-based manner. METHODS: Tissue from patients that underwent a nephrectomy for RCC or metastasectomy were engrafted into the chorioallantoic membrane of avian embryos with a portion used for TIL isolation and primary cell generation. PD-1 and PD-L1 checkpoint inhibitors were intravenously injected and high frequency ultrasound imaging and immunophenotyping was used to quantitate changes in tumor volume, vascularity, and TIL expansion in all treatment groups. Isolated TILs were expanded and co-cultured with respective primary cells to confirm and examine immunotherapy response at a molecular level. RESULTS: Patient PDXs demonstrated heterogeneous responses to immunotherapy that relates to presence and expansion of TILs after checkpoint inhibitor injection. Patient #1 demonstrated the greatest response with a >50% reduction in tumor volume and 4X expansion of T-cells over 8 days of treatment. Patient #2 demonstrated intratumoral heterogeneities in terms of tumor volume change after checkpoint inhibitors treatment despite TIL expansion in all treatment groups. Patient #3 showed no response that correlated to lack of TILs in the PDX. In vitro expansion of TILs and co-culture with primary cell line demonstrated a specific immune response via INF-gamma release and changes in immunomodulatory molecules. CONCLUSIONS: This PDX system preserves TILs and permits rapid large scale prediction of tumor specimens to various systemic therapy agents such as checkpoint inhibitors without the need for humanized models. This avatar system is a useful tool to rapidly predict and examine response to checkpoint inhibitors on a per-patient basis to improve the assignment of cancer immunotherapies to specific RCC patients. Citation Format: Mario Cepeda, Matthew Lowerison, Cindy Liu, Yarolslav Fedyshyn, Vidhu Joshi, Paras Shah, John Cheville, R. Houston Thompson, Boorjian Stephen, Bradley Leibovich, Haidong Dong, Hon S. Leong. Rapid patient derived xenograft avatars that consider tumor heterogeneity for prediction of cancer immunotherapy responses in metastatic renal cell carcinoma patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-155.

Objective To investigate the prognostic significance of various patterns of extrarenal extension that comprise pathological stage T3a clear‐cell renal cell carcinoma (ccRCC) amongst patients undergoing nephrectomy for non‐metastatic disease. Patients and Methods A retrospective review of 563 patients who underwent radical nephrectomy for pathologically confirmed T3aN0/NxM0 ccRCC between 1970 and 2011 was performed. All pathological slides were re‐reviewed by one urological pathologist. Associations of patterns of extrarenal extension (perinephric fat [PF], renal sinus fat [SF], and renal vein [RV], in isolation or in any combination) with disease progression, cancer‐specific mortality (CSM), and all‐cause mortality were evaluated on multivariable analyses. Results Overall, PF invasion, renal SF invasion, and RV tumour thrombus were present in 144 (26%), 51 (9%), and 163 (29%) patients, respectively, with multiple patterns of extrarenal extension identified in 205 (36%) patients. There were no significant differences in survival outcomes for isolated involvement of PF, renal SF, or RV. However, patients with multiple patterns of extrarenal extension were at significantly increased risk of disease progression (hazard ratio [HR] 1.31, 95% confidence interval [CI] 1.04–1.65; P = 0.020), CSM (HR 1.64, 95% CI 1.27–2.12; P < 0.001), and all‐cause mortality (HR 1.32, 95% CI 1.08–1.61; P = 0.008). Conclusions The presence of multiple patterns of extrarenal extension is associated with a higher risk of disease progression and cancer‐related death after radical nephrectomy compared to isolated involvement of the PF, renal SF, or RV, which carry similar prognostic weight. If validated, these findings may help refine risk stratification of non‐metastatic T3a RCC by distinguishing patients with multiple vs one pattern of extrarenal extension.

Background Glomerular volume increases when demand exceeds nephron supply, which may lead to glomerulosclerosis. It is unclear if determinants of glomerular volume are consistent between populations that differ by severity of comorbidities. Methods We studied kidney biopsy specimens from living kidney donors ( n =2453) and patients who underwent radical nephrectomy for a renal tumor ( n =780). We scanned specimen sections into high-resolution digital images, manually traced glomerular profiles, and calculated mean glomerular volumes using the Weibel–Gomez stereologic formula (separately for nonsclerosed glomeruli and globally sclerosed glomeruli). We then assessed the relationship of glomerular volume with age, clinical characteristics, and nephrosclerosis on biopsy specimen. Results Compared with kidney donors, patients with tumors were older and more frequently men, obese, diabetic, or hypertensive, had more glomerulosclerosis and interstitial fibrosis on biopsy specimen, and had 12% larger nonsclerosed glomeruli ( P <0.001). In both populations, male sex, taller height, obesity, hypertension, and proteinuria associated with larger nonsclerosed glomeruli to a similar extent. In patients with tumors, diabetes, glomerulosclerosis >25%, and interstitial fibrosis >25% also associated with larger nonsclerosed glomeruli. Independent clinical predictors of larger nonsclerotic glomeruli were family history of ESRD, male sex, taller height, obesity, diabetes, and proteinuria. After adjustment for these characteristics, nonsclerotic glomerular volume did not differ between populations and was stable up to age 75 years, after which it decreased with age. Many of these findings were also evident with globally sclerotic glomerular volume. Conclusions Characteristics associated with glomerular volume are consistent between patient populations with low and high levels of comorbidity.

Objectives To evaluate the peri‐operative and renal functional outcomes of patients undergoing synchronous bilateral partial nephrectomy (PN) or percutaneous cryoablation (PCA). Patients and Methods We retrospectively reviewed our institutional nephrectomy and renal mass ablation registries to identify all patients with synchronous bilateral renal masses who underwent simultaneous bilateral PN ( n = 76) or PCA ( n = 13) between 1974 and 2013. Changes in estimated glomerular filtration rate (eGFR) as well as peri‐operative complications are descriptively reported for each procedure. Results The number of treated renal masses in the 76 patients in the PN group and the 13 patients in the PCA group was 249 and 28, respectively. The median (interquartile range [IQR]) age at treatment was 62 (50, 71) years for the PN group and 67 (56, 72) for the PCA group. The median (IQR) maximum tumour sizes were 4.6 (3.4, 6.5) cm and 2.6 (2.4, 3.2) cm for the PN and PCA groups, respectively. The median (IQR) length of hospital stay was 7 (5, 8) days for the PN group and 1 (1, 10) days for the PCA group. The median (IQR) change in eGFR from baseline to discharge was −32 (−46, −15)% for the PN group and −17% (−33, −3) for the PCA group. By 3 months, median (IQR) renal function improved, with changes of −9 (−19, 0)% and −8 (−11, 15)%, respectively, compared with baseline. No patient in either group required renal replacement therapy in the peri‐operative period. Early postoperative complications (within 30 days) occurred in 16 patients (21.6%) in the PN and four patients in the PCA group. In particular, angioembolization for bleeding was required in the postoperative period in two patients (2.7%) in the PN and one patient in the PCA group. Conclusions Our experience suggests that synchronous bilateral PN or PCA are feasible treatment options for select patients presenting with bilateral renal masses. In select cases, both approaches appear to have reasonable rates of peri‐operative complications and effects on renal function.

704 Background: While the probability of benign versus malignant histology based on renal tumor size has been described, this alone does not sufficiently inform decision-making in the modern era since indolent malignant tumors can be surveilled. Thus, we sought to characterize the probability of indolent versus aggressive histology based on radiographic tumor size. Methods: We evaluated patients who underwent radical or partial nephrectomy at Mayo Clinic for a pT1-2, pNx/0, M0 solid renal tumor between 1990-2010. Pathology was reviewed by one genitourinary pathologist. Benign tumors, low grade (1-2) clear cell and papillary renal cell carcinoma (RCC), and any chromophobe, clear cell papillary, mucinous tubular and spindle cell, SDH-B deficient, and tubulocystic RCC were considered indolent. All other histologies were considered aggressive, as were any malignancies with necrosis or sarcomatoid differentiation. Cancer-specific survival (CSS) was estimated using the Kaplan Meier method. Logistic regression models were used to estimate the probability of malignant and aggressive histology based on tumor size. Sex-stratified analyses were also performed. Results: Of the 2650 patients included, there were 1773 patients with indolent tumors (303 benign; 1470 malignant) and 877 with aggressive tumors. Ten-year CSS was 96% for indolent malignant tumors and 82% for aggressive tumors. The predicted probabilities of any malignant histology and aggressive malignant histology increased with tumor size (Table; 1-7cm point estimates shown). For example, a 3 cm tumor had an 87% probability of malignancy and a 27% probability of being aggressive. For any given tumor size, men had a greater probability of aggressive histology than women. Conclusions: We present tumor size-based estimates of the probability of aggressive histology for renal masses. This information should be useful for patient counseling and treatment decision-making. [Table: see text]

703 Background: Clinical and pathologic factors alone have limited prognostic ability in patients with metastatic clear cell renal cell carcinoma (ccRCC). We sought to determine if tissue biomarkers, along with our previously reported clinical metastases score, can be used to predict cancer specific survival (CSS) in patients with metastatic ccRCC. Methods: Patients with metastatic ccRCC who underwent nephrectomy between 1990-2004 were identified using the Mayo Clinic Nephrectomy Registry. Sections from paraffin-embedded primary tumor tissue blocks were used for immunohistochemistry staining for PD-1, B7-H1 (PD-L1), B7-H3, Bim (downstream pro-apoptotic signaling molecule in PD-1 pathway), CA-IX, IMP3, Ki67, and survivin. CSS was the primary outcome. Biomarkers that were significantly associated with CSS after adjusting for the metastases score were used to develop a biomarker-specific multivariable model using a bootstrap resampling approach and forward selection. Predictive ability was summarized using a bootstrap-corrected c-index. Results: The cohort included 602 patients, 192 (32%) with metastases at diagnosis and 410 (68%) who developed metastases after nephrectomy. Median follow-up among survivors was 9.6 years (IQR 4.2,12.8) and 504 patients died of RCC. Bim, IMP3, Ki67, and survivin expression were significantly associated with CSS after adjusting for the metastases score and were used to develop a biomarker-specific model. High Bim (HR 1.44; 95%CI 1.16-1.78; p < 0.001), high survivin (HR 1.35; 95%CI 1.08-1.68; p = 0.008), and the metastases score (HR 1.13 per one point; 95%CI 1.10-1.16; p < 0.001) were retained as independent predictors in the final multivariable model (c-index 0.69). Conclusions: We created a prognostic model combining the clinical metastases score and two primary tissue expression biomarkers, Bim and survivin, for patients with metastatic RCC who underwent nephrectomy. External validation will be required prior to clinical use.