Search

722 Background: Previous large scale genomic studies have described somatic alterations that affect tumor progression. However, metabolic and immune pathway changes that occur with advancing disease stage are not well characterized. Herein, we performed transcriptomic profiling of clear-cell renal carcinoma (ccRCC) tumors across clinical stages to define these changes. Methods: TCGA Firehose Legacy cohort was used to determine differential gene expression between stage 3 ccRCCs from patients who remained disease free (S3DF) vs. those who relapsed (S3RL) within two years of nephrectomy. Single cell RNA-seq (scRNAseq) analysis was performed on primary ccRCCs from nine patients (stage 1 n=4, stage 3 n=4, metastatic n=1). Seurat V4 package and module score was used to analyze patterns of gene expression associated with tumor progression at the single cell level. Results: Transcriptomic comparison of ccRCCs from TCGA Firehose Legacy cohort revealed induction of genes encoding mitochondrial respiratory complexes and mitochondrial ribosomal proteins (MRPs) and decrease in expression of immune genes in S3RL as compared to S3DF tumors. Using a set of 23 differentially regulated mitochondrial and immune genes we established a transcriptomic signature that stratified patients into those at risk of relapse vs. not. ScRNA-seq data analysis determined increased expression of mitochondrial respiratory complexes’ gene-sets, and decreased levels of glycolytic genes and HIF targets during tumor progression from stage-I to metastatic. This was supported by spatial transcriptomics of cancer cells, which revealed clusters of cells with high mitochondrial activity and decreased MHC class I and II gene expression. Conclusions: The data support a model of progressive metabolic reprogramming from glycolysis and HIF-regulated pathways to mitochondrial respiration in parallel with diminished immune expression of antigens on cancer cells with advancing disease stage in ccRCC. Identification of targetable mitochondrial vulnerability could be a therapeutic option in patients who progress on immune checkpoint inhibitors.

562 Background: Recent data suggest that patients with liver metastases (mets) are resistant to immune checkpoint inhibition (CPI) independent of historical biomarkers of CPI efficacy, raising the question of whether liver mets may be associated with an immunosuppressive tumor microenvironment (TME). We investigated the immune TME of hepatic and non-hepatic mets compared to primary tumors in advanced urothelial carcinoma (UC) tissue samples. Methods: NextGen sequencing (NGS) of DNA (592-gene/whole exome) and RNA (whole transcriptome) from UC tissue samples (N=4746) was performed at Caris Life Sciences (Phoenix, AZ). Immune cell infiltration was estimated by RNA expression deconvolution (MCP-counter). PD-L1 expression (SP142: immune cell stain ≥ 5%; 22c3: CPS ≥ 10) was assessed by immunohistochemistry (IHC). Deficient mismatch repair/high microsatellite instability (dMMR/MSI-H) was tested by IHC/NGS. Real-world overall survival (OS) information was obtained from insurance claims data and Kaplan-Meier estimates were calculated. Mann–Whitney U and X2/Fischer-Exact tests were applied where appropriate, with p-values adjusted for multiple comparisons (Benjamini-Hochberg). *P+ T and B cells (0.55* and 0.29-fold*) but increased monocytic lineage cells (1.23-fold*), while non-hepatic mets had increased CD8+ T, NK, and monocytic lineage cells (1.28*, 1.27*, 1.31-fold*) with no difference in B cells (1.05-fold). Hepatic mets had decreased expression of integrin LFA-1/ ITGAL (0.77-fold*), as well as hyaluronic acid (HA) receptor CD44 and synthase HAS2 (0.61 and 0.61-fold*), compared to primary tumors, whereas expression of these genes and LFA-1 ligand ICAM1 was increased in non-hepatic mets (1.08 to 1.30-fold*). Hepatic mets had increased expression of immunosuppressive cytokines CCL2 and CXCL2 (1.72* and 2.32-fold*) and decreased expression of pro-inflammatory cytokines CCL5 and CXCL10 (0.63* and 0.79-fold*) compared to primary tumors. PD-L1+ IHC was less frequent in hepatic mets compared to primary tumors and non-hepatic mets. TMB-High (≥10 mut/MB) and dMMR/MSI-H rates were similar across tumor sites. Hepatic mets (N=40) were associated with worse OS from the start of pembrolizumab compared to non-hepatic mets (N=177) (19.6 vs 4.4 months, HR 3.01, 95% CI 1.91-4.75, p

378 Background: Patients with localized or locally advanced ccRCC undergo surgical resection by partial or radical nephrectomy. Approximately 50% of tumors recur within five years after surgery. Currently the only FDA approved drug in the adjuvant setting is sunitinib. However, the absence of an overall survival benefit and the prevalence of side effects limit widespread use of sunitinib. Accurate identification of patients at high risk of recurrence would enhance targeted adjuvant therapy, while sparing low-risk patients from side effects. A single scoring system, transcriptomic signature or genomic classifier, is unlikely to accurately define prognosis. Here, we describe an integrated transcriptomic-metallomic classifier to assess the risk of relapse in localized ccRCC. Methods: We investigated the transcriptomic landscape of Caucasian male patients with stage III ccRCCs who remained disease free (S3DF) or relapsed (S3RL) within 24 months after surgery using the TCGA Firehose Legacy cohort. In a separate cohort of S3RL and S3DF ccRCCs, copper (Cu) content and molecular distribution were analyzed by size exclusion chromatography coupled to inductively coupled plasma mass spectrometry (SEC-ICP-MS). Results: Transcriptomic analysis identified that S3RL tumors were enriched for genes encoding subunits of mitochondrial electron transport chain (ETC) and mitochondrial ribosomal proteins (MRPs), an indication that mitochondrial activity may contribute to the relapse. In contrast, S3DF tumors were enrichedfor genes related to immune responses, including genes encoding members ofMHC-II, suggesting tumor control by the immune system. We established a setof 23 genes (23G) as a signature that stratified ccRCCsinto distinct groups characterized by low, intermediateand high risk of relapse after surgery. Importantly, SEC-ICP-MS analysis identified augmented distribution of Cu to the Cu cytochrome c oxidase complex (Cu-COX) in tumor mitochondria from S3RL, consistent with the prediction of increased ETC activity in the transcriptomic analysis. Conclusions: Integration of the 23G transcriptomic signature and Cu-COX complex measurements in locoregional ccRCC can serve as a prognostic biomarker for recurrence, and as a predictive classifier for novel treatments, including inhibitors of ETC for the S3RL tumors. In particular, a combination classifier using both measurements may have a stronger prognostic/predictive power as compared to either biomarker alone.

287 Background: RCC has a distinct pattern of metastatic spread with common sites of metastasis including the lung, bone, and liver. Less common sites include the brain, adrenal gland, and pancreas. While the pattern of metastatic spread has prognostic significance, the biology driving tropism to specific organ sites has not been fully characterized. We utilized a multi-institutional real-world dataset to examine genomic alterations and transcriptional signatures across the spectrum of metastatic sites in patients with RCC. Methods: RCC tissue specimens derived from the kidney and distant metastatic sites were sequenced utilizing a commercially available Clinical Laboratory Improvement Amendments (CLIA)-certified assay by Caris Life Sciences. Whole exome and transcriptome sequencing was performed. Molecular subgroups were defined according to the IMmotion151 metastatic RCC subtypes, with subgroups determined by a weighted average of gene expression levels. Molecular analysis and PD-L1 expression (SP142) were described by metastasis site. Results: 657 RCC samples from 653 patients underwent molecular profiling. The median age was 62 years (range 14-90) and the majority were male (70.6%). The most common histology was clear cell RCC (n = 509, 77.5%), followed by papillary (n = 63, 9.6%), chromophobe (n = 30, 4.6%), medullary (n = 8, 1.2%), collecting duct (n = 6, 0.9%), and mixed (n = 41, 6.2%). Specimen source included the kidney (n = 340, 51.8%), lung (n = 75, 11.4%), bone (n = 45, 6.8%), lymph nodes (n = 34, 5.2%), liver (n = 28, 4.3%), endocrine glands (adrenal, pancreas, and thyroid; n = 23, 3.5%), brain/CNS (n = 16, 2.4%), and other metastatic sites (n = 96, 14.6%). Compared to kidney, several genes were mutated at higher rates for select metastatic sites, including PBRM1 (59.5% bone, 59.1% endocrine, and 45.9% lung vs 33.8% kidney, p< 0.05) and KDM5C (27.8% endocrine, 29.2% lymph nodes, and 35.3% soft tissue vs 9.3% kidney, p< 0.05). When evaluating metastatic specimens versus kidney specimens, bone metastases had a significantly higher proportion of tumors classified as ‘Angio/stromal’ (n = 19, 42.2%; vs n = 52, 15.4%; p< 0.0001), while liver metastases had a higher proportion of the ‘complement/Ω-oxidation’ subgroup (n = 17, 60.7%; vs n = 48, 14.1%; p< 0.0001). PD-L1 expression in metastatic sites (range 6.8%-21.7%, with exception of 0% in GI; p= 0.09 to 0.99) was not significantly different from the kidney (16.6%). Conclusions: In our contemporary real-world analysis, we demonstrate differential patterns of molecular alterations among sites of metastasis in RCC. Our observations elucidate the biology underlying heterogeneous disease outcomes associated with site of metastasis. Application of predictive signatures by site of metastasis may help inform personalized therapy strategies in advanced RCC. Further studies are warranted to validate our findings.

4529 Background: Oncology patients experience more severe disease outcomes from COVID-19 infection than the general population. BCG is a live bovine tuberculosis bacillus with immunotherapeutic effects in urothelial cancers; it is also used as vaccination against Mycobacterium tuberculosis in parts of the world. As BCG vaccination has been associated with broad protection against viral pathogens, BCG exposure through vaccination or intravesical therapy may modulate host immunity and reduce the severity of COVID-19 infection. We report the effect of BCG exposure on COVID-19 severity in oncology patients from the CCC19 registry. Methods: The CCC19 registry (NCT04354701) was used to identify patients with prior BCG exposure. Cohort A received intravesical treatment for bladder carcinoma, and cohort B received prior BCG vaccination. Each cohort was matched 3:1 to non-BCG-exposed controls by age, sex, race, primary cancer type, cancer status, ECOG performance status (PS) and calendar time of COVID-19 infection. The primary endpoint was COVID-19 severity reported on an ordinal scale (uncomplicated, hospitalized, admitted to ICU +/- ventilated, died within 30 days) of patients exposed to prior BCG compared to matched non-exposed controls. 2-sided Wilcoxon rank-sum tests were used. Results: As of 6-Feb-2021 we included 124 patients with BCG exposure, 68 patients with bladder carcinoma who had received intravesical BCG (Cohort A), and 64 cancer patients with prior BCG vaccination (Cohort B). Median age was 76 years, IQR 69-83 (Cohort A) and 67 years, IQR 62-74 (Cohort B). Bladder cancer pts were predominately male (78%) vs 55% for Cohort B. Patients with PS 2+ were uncommon, 18% in Cohort A and 16% in Cohort B. COVID-19 illness severity was no different in patients exposed to prior intravesicular BCG (p=0.87). COVID-19 illness severity was no different in patients exposed to prior intradermal BCG vaccination (p=0.60). Conclusions: Despite this being the largest such cohort reported to date, we failed to demonstrate an association of prior BCG exposure with modulation of severity of COVID-19 illness. Prospective trials evaluating the protective effect of BCG vaccination are ongoing and will add further insight into the effect of BCG on COVID-19 illness.[Table: see text]

4578 Background: Nivolumab in combination with cabozantinib (N+C) has demonstrated significantly improved progression-free survival (PFS), objective response rate (ORR), and overall survival (OS), compared with sunitinib as a first-line (1L) treatment for aRCC in the phase 3 CheckMate (CM) 9ER trial. As there are no head-to-head trials comparing N+C with pembrolizumab in combination with axitinib (P+A), this study compared the efficacy of N+C with P+A as 1L treatment in aRCC. Methods: An MAIC was conducted using individual patient data on N+C (N = 323) from the CM 9ER trial (median follow-up: 23.5 months) and published data on P+A (N = 432) from the KEYNOTE (KN)-426 trialof P+A (median follow-up: 30.6 months). Individual patients within the CM 9ER trial population were reweighted to match the key patient characteristics published in KN-426 trial, including age, gender, previous nephrectomy, International Metastatic RCC Database Consortium risk score, and sites of metastasis. After weighting, hazards ratios (HR) of PFS, duration of response (DoR), and OS comparing N+C vs. P+A were estimated using weighted Cox proportional hazards models, and ORR was compared using a weighted Wald test. All comparisons were conducted using the corresponding sunitinib arms as an anchor. Results: After weighting, patient characteristics in the CM 9ER trial were comparable to those in the KN-426 trial. In the weighted population, N+C had a median PFS of 19.3 months (95% CI: 15.2, 22.4) compared to a median PFS of 15.7 months (95% CI: 13.7, 20.6) for P+A. Using sunitinib as an anchor arm, N+C was associated with a 30% reduction in risk of progression or death compared to P+A, (HR: 0.70, 95% CI: 0.53, 0.93; P = 0.015; table). In addition, N+C was associated with numerically, although not statistically, higher improvement in ORR vs sunitinib (difference: 8.4%, 95% CI: -1.7%, 18.4%; P = 0.105) and improved DoR (HR: 0.79; 95% CI: 0.47, 1.31; P = 0.359). Similar OS outcomes were observed for N+C and P+A (HR: 0.99; 95% CI: 0.67, 1.44; P = 0.940). Conclusions: After adjusting for cross-trial differences, N+C had a more favorable efficacy profile compared to P+A, including statistically significant PFS benefits, numerically improved ORR and DoR, and similar OS.[Table: see text]

5063 Background: ATM mutations, one of a family of DNA repair defects prevalent in prostate cancer, have been included in a list of actionable mutations for PARP inhibitor (PARPi) therapeutic trials. Despite preclinical evidence, PARPi have shown minimal clinical activity in ATM mutant prostate cancer (ATMmPCa). The present analysis explores co-occurring genomic alterations that may drive outcomes of metastatic PCa (mPCa) patients with tumors harboring ATM mutations and provide clues for understanding therapy resistance and potential targets. Methods: This study included molecular profiling analysis of 1375 cases of mPCa. Tumors were analyzed using next-generation sequencing (NGS), whole transcriptome sequencing (WTS), and immunohistochemistry (IHC) (Caris Life Sciences, Phoenix, AZ). dMMR/MSI-H status was determined by IHC, NGS, and fragment analysis and tumor mutational burden (TMB) was calculated based on somatic nonsynonymous missense mutations. We performed differential gene expression analysis of HR-associated transcripts such as ATR, PARP1-3, RAD50, RAD51A/B/C/D and RAD54. Significance was determined using the ꭓ2 test and Benjamini-Hochberg method. Results: Fifty-nine (4.2%) cases harbored pathogenic ATM mutations, 84 (6.2%) harbored BRCA2 mutations. 1018 tumors (74%) were deemed homologous recombination proficient (HRP) and 155 tumors (11.3%) were HR Deficient (HRD); harboring one or more mutation in HR-related genes excluding ATM and BRCA2. The mutation rate of TP53 was significantly lower in ATMmPCa (12.0%) compared to BRCA2mPCa (35%), HRD (35%) and HRP (46.6%) tumors. ATMmPCa showed higher rates of SMAD2 (3.7%/1%) and FLCN (5.2%/0.3%) alterations compared to HRP cases. PARP1 and RAD51D gene expression was reduced in ATMmPCa compared to HRP (p < 0.05) and BRCA2mPCa ( p < 0.05) tumors, respectively. No differences in gene expression levels were detected for ATR, PARP2, PARP3, RAD50, and RAD54. Chromosomal segments demonstrating differential CNA in ATMmPCa vs HRP, HRD, or BRCA2mPCa included FGF19, FGF4, PTPN11, ALDH2, DAXX, BCL7A, CCND1, BMPR1A and MEF2B (Q-value < 0.05 determined by ꭓ2). The most common CNA in ATMmPCa was CCND1, present in approximately 13% (7/55) of cases. Compared to BRCA2mPCa and HRD cases, ATMmPCa cases are less likely to display markers of immunotherapy response such as dMMR/MSI-H or TMB ≥10 mutations/MB. Conclusions: ATMmPCa demonstrated several differences in co-occurring alterations compared to BRCA2mPCa, HRD and HRP mPCa. ATMmPCa tumors were less likely to harbor alterations in TP53 compared to BRCA2, HRD or HRP tumors. CNA in ATMmPCa occurred in 9 genes across distinct mPCa molecular subtypes and were enriched for those associated with the 11q13 amplicon harboring Cyclin D1. The FGF and PTPN11 related pathways are potentially targetable pathways in ATMmPC and may merit further study.

6031 Background: Salvage surgery for locally recurrent head and neck squamous cell carcinoma (rHNSCC) results in local control rates of 33-50% but only 20-40% of patients achieve long-term survival necessitating additional therapy (Haque et al., Oral Oncol. 2019). Many patients are ineligible for re-irradiation and chemotherapy alone after salvage surgeryhas shown no survival benefit. The clinical activity and tolerability of immune checkpoint inhibitors has been demonstrated in metastatic HNSCC, but the benefit after salvage surgery (SS) has not been studied. Here we report the results of a multi-center phase II investigation of nivolumab, a PD-1 inhibitor, after SS in recurrent HNSCC (NCT03355560). Methods: HNSCC patients undergoing curative-intent SS were enrolled to receive 6 months of nivolumab beginning 4-11 weeks after surgery. All received radiation with or without chemotherapy as prior definitive therapy and had no other curative treatment options at the time of surgery. Key exclusion criteria included: distant metastatic disease, gross residual disease, or a history of immunodeficiency, autoimmunity, or pneumonitis. The primary endpoint was 2-year disease-free survival (DFS) measured by Kaplan Meier curves. Safety was evaluated by CTCAE v5.0. Results: 39 patients were enrolled. Median age was 68 years (range, 49-85). 12/39 (31%) were female. 34/39 (87%) were white. Disease sites included oropharynx 9/39 (23%), oral cavity 14/39 (36%), and larynx 16/39 (41%). P16 status was 26% (+), 48% (-), and 26% (unknown). 17/39 (44%) had high risk pathologic features (positive margins or extranodal spread) at time of SS. 28/39 (72%) patients experienced treatment-related adverse events (TRAE), the most common of which were fatigue (26%), hypothyroidism (10%) and acneiform rash (13%). Grade 3-4 TRAEs were rare, occurring in 3/39 (8%) patients and included diarrhea, oral pain, neck pain, productive cough, stridor, and COPD exacerbation. 3/39 (8%) required treatment discontinuation and there were no grade 5 events. The 2-year DFS was 60% (95%CI 0.39-0.91). 2-year overall survival was 74% (95% CI 0.54-1). In single-cell multiplex cytokine analysis, patients who relapsed following adjuvant nivolumab had a significantly higher proportion of peripheral blood CD8 T cells which displayed a polyfunctional cytokine profile. IFN-γ and Granzyme were the dominant CD8 cytokines in both responders and non-responders, however CD8 expression of MIP1a and TNF-α were significantly higher in patients who ultimately relapsed. Conclusions: Nivolumab after salvage surgery in rHNSCC is well tolerated and shows promising antitumor activity in this high-risk patient population with unmet need. Immunotherapy after salvage surgery should be studied in randomized clinical trials. Clinical trial information: NCT03355560.

e18788 Background: Patients (pts) with cancer have a high risk of venous thromboembolic (VTE) complications, further enhanced by anti-cancer treatments, specifically hormonal therapies, targeted therapies (VEGF inhibitors, other TKIs) and immune checkpoint inhibitors (ICIs). We hypothesized that high-risk therapies would predispose pts with cancer and COVID-19 to higher risk of VTE complications. Methods: CCC19 is the largest international registry (NCT04354701) recording outcomes of pts with cancer and COVID-19. The registry was queried for hospitalized pts who developed VTE and received systemic cancer treatment in the year prior to COVID-19. Incidence of VTE was analyzed as the primary endpoint; 30-day any cause mortality & need for ICU admission at baseline were secondary endpoints in pts with and without VTE respectively. Pts were stratified by treatment type and time from last treatment dose

465 Background: UTUC is a rare genitourinary malignancy and a number of studies, limited by small sample sizes, have attempted to characterize its mutational landscape. Because immunotherapy is commonly used for this disease type, we evaluated the prevalence of microsatellite instability and characterized the mutational landscapes of UTUC in a large contemporary patient cohort. Methods: UTUC tumor samples were analyzed using next generation sequencing (NGS) (NextSeq, 592 gene panel) or whole exome sequencing (WES) (NovaSeq) (Caris Life Sciences, Phoenix, AZ). Mismatch repair status (deficient [dMMR] or proficient [pMMR]) and microsatellite instability status (MSI-high or stable [MSS]) were detected by immunohistochemistry (IHC), fragment analysis, and NGS. Tumor mutational burden (TMB) was measured by counting all somatic mutations found per tumor (high cutoff ≥ 10 mutations per MB). PD-L1 expression was tested by IHC using PD-L1 antibody clones 22c3 (Agilent; positive cutoff CPS ≥ 10) and SP142 (Ventana; positive cutoff ≥ 5% IC). Pathogenic fusion events were detected using whole transcriptome sequencing (NovaSeq). Statistical significance was determined using the Chi-square test and adjusted for multiple comparison. Results: 538 patients with included – median (range) age 71.5 (30-89) years and 37.5% female/62.5% male. Prevalence of dMMR/MSI-H was 3.9% (21/538) and TMB-high was 22.7% (96/423). Significant molecular differences were not detected in primary vs metastatic disease or in male vs female cases. dMMR/MSI-H tumors had higher frequency of TMB-high compared to MSS tumors (100% vs. 19%, p = 0.00003). dMMR/MSI-H tumors also had a higher frequency than MSS tumors for mutations in genes involved in chromatin remodeling (ASXL 82.4%, CREBBP 60%, SMARCA4 40%, KMT2D 95%, ARIDIA 100%, KMT2A 20%, KMT2C 35.3%, NSD1 20%), DNA-damage repair (FANCG 10%, ATM 45%, ATRX 40%) and other biological pathways (RNF43 10%, PTCH1 21.4%, ERBB3 30%, CDKN2A 25%, TSC2 15%, FLNC 15%, HNF1A 20%, CIC 15%, DNMT3A 17.6%); all adjusted p < 0.05. Pathogenic fusions were detected in 3.8% (17/443) cases, with FGFR3 fusion being the most common, occurring in 2.7% (12/443) cases. PD-L1 positivity was identified in 33.2% (133/400) cases tested by 22c3 antibody and 28.4% (89/313) cases tested by SP142 antibody. No difference was seen in PD-L1 positivity between MSI-H/dMMR vs. MSS tumors. Conclusions: In the largest analysis to date, we found a 3.9% prevalence of dMMR/MSI-high rate in UTUC. All dMMR/MSI-H tumors displayed TMB-high. PD-L1 positivity was comparable between dMMR/MSI-H and MSS tumors. dMMR/MSI-H tumors had a significantly higher rate of mutations in genes involved in chromatin remodeling and DDR biological pathways. These results could inform design of targeted therapy trials in UTUC.

343 Background: With the emergence of multiple active treatment options in RCC, predictive biomarkers for optimal treatment selection are lacking. Gene expression data from IMmotion151 and Javelin Renal 101 clinical trials generated anti-angiogenic and immune signatures that warrant further validation. We aimed to describe the genomic and gene expression profiles in a multi-institutional database of patients with ccRCC, and its association with other biomarkers of interest. Methods: Whole transcriptome sequencing was performed for ccRCC patient samples submitted to a commercial CLIA-certified laboratory (Caris Life Sciences, Phoenix, AZ) from February 2019 to September 2020. Tumor GEP and hierarchical clustering based on the validated 66-gene signature (D’Costa et al, 2020) were used to identify patient subgroups. Samples from both primary tumors and metastatic sites were included. Results: A total of 316 patients with ccRCC, median age 62 (range 32-90), 71.8% men, were included. Tissue samples were obtained from primary tumor (46.5%), lung (12.3%), bone (9.5%), liver (4.7%) and other metastatic sites (27%). Gene expression analysis identified angiogenic, mixed and T-effector subgroups in 24.1%, 51.3% and 24.7%, respectively. Patients with angiogenic subgroup tumors compared to those with T-effector subgroup tumors were more likely to be older (63 versus 60 years, p=0.035), female (40.8% versus 16.7%, p=0.0009) and more frequently found in pancreatic/small bowel metastases (75% versus 12.5%, p=0.0103). Biomarkers of potential response to immunotherapy such as PD-L1 (p=0.0021), TMB (not significant), and dMMR/MSI-H status (not significant) were more frequent in the T-effector subgroup. PBRM1 mutations were more common in the angiogenic subgroup (62.0% vs 37.5%, p=0.0034) while BAP1 mutations were more common in the T-effector subgroup (18.6% versus 3.0%, p= 0.0035). Immune cell population abundance (e.g. NK cells, monocytes) and immune checkpoint gene expression (TIM-3, PD-L1, PD-L2, CTLA4) were also increased in the T-effector subgroup. Conclusions: Our hierarchical clustering results based on the 66-gene expression signature were concordant with results from prior studies. Patient subgroups identified by evaluation of angiogenic and T-effector signature scores exhibit significantly different mutations and immune profiles. These findings require prospective validation in future biomarker-selected clinical trials.

355 Background: Management of patients with localized or locally advanced renal cell cancer (RCC) involves surgical resection. However, 20-40% of all localized kidney cancer patients experience a recurrence. Adjuvant treatment in high-risk kidney cancer has not proven to successfully improve overall survival (OS) and the only drug approved so far in this setting is vascular endothelial growth factor (VEGF) inhibitor, sunitinib. As multiple trials are evaluating drugs including immune checkpoint inhibitors in patients with high risk localized clear cell RCC, there is an urgent need to identify biomarkers to help with therapeutic decisions as well as for risk stratification. Here we present analysis pertaining to genes involved in metabolic reprogramming in kidney cancer. Methods: A deep transcriptomic analysis of patients with stage III ccRCC in the TCGA KIRC Firehose Legacy cohort was undertaken. Caucasian males with stage III ccRCC in the TCGA cohort for whom recurrence data was available with a minimum follow-up of 2 years were identified for the analysis. Expression profiles of differentially expressed genes were clustered using the Bayesian infinite mixture model and samples clustered using average linkage hierarchical clustering based on pairwise Pearson’s correlations as the measure of similarity. Enrichment analysis of up- and down-regulated genes was performed using logistic regression. R package was used to generate Kaplan-Meier curves and assess statistical significance of differences in overall and disease-free survival using log-rank test. Results: The cohort evaluated for this study included Caucasian male patients that remained disease free for at least 24 months after surgery (n = 22) and patients whose cancer recurred within 24 months (n = 20). We identified metabolic genes, encoding subunits of mitochondrial electron transport chain as well as malate aspartate shuttle (MAS), where loss of coordinated co-expression between these genes identified patients at risk of recurrence after surgery. The gene signature stratified the 42 patients into three subtypes: significantly enriched for a) recurrence (subtype 1), for b) disease free status (subtype 2) and b) intermediate (subtype 3). Work is underway to combine the use of individual gene expression and ratios of gene expression within the signature to optimize prognostic biomarkers for localized ccRCC. Conclusions: In this analysis, we have identified a novel metabolic gene signature which can help identify patients with localized ccRCC at high risk of recurrence after surgery to guide aggressive therapy. These findings require further validation in tumors collected from patients with stage III ccRCC.

6576 Background: Immune checkpoint inhibitors (ICIs) are FDA approved for use in head and neck squamous cell cancer (HNSCC), however, only ~20% patients achieve a response. Identification of biomarkers of response or toxicity remains a challenge. Polyfunctional T-cells, or T-cells producing multiple cytokines, have been recognized as contributors to durable immunity against various cancers. However, their role has not been studied prospectively in HNSCC patients receiving ICIs. To look for an early predictor of response, we used single-cell functional proteomic profiling (IsoPlexis) on blood samples pre- and post- first dose of nivolumab (nivo) in patients on our phase-II study of locally recurrent HNSCC (NCT03355560). Methods: HNSCC patients who failed definitive radiation +/-chemotherapy and were subsequently treated with curative intent salvage resection were enrolled to receive 6 months of nivo beginning 4 to 11 weeks after surgery. Blood samples were collected before and after the first dose of nivo. Peripheral blood mononuclear cells were isolated, enriched for CD8+ T cells and using the 32-plex IsoCode technology, single-cell cytokine signals were captured and polyfunctional strength of CD8+ T cells was evaluated across four groups (effector, stimulatory, regulatory, inflammatory). A comparison analysis was performed between pre- and post- nivo treatment and between patients who relapsed (non-responders) vs those who did not (responders). Results: Thirty-three of 39 planned patients have been enrolled, of which 28 are evaluable and 5/28 (18%) developed recurrence. Median age is 68 years (range 51-85), 9/28 (32%) patients are female, 26/28 (93%) are white, disease sites include oropharyngeal 6/28 (21%), oral cavity 11/28 (39%) and larynx 11/28 (39%). Samples were evaluated at a median follow up of 5.9 months from enrollment. Single-cell analysis demonstrated a strong upregulation of polyfunctional human CD8+ T cell subsets in responders. Polyfunctional Strength Index (PSI) was enhanced in CD8+ T cells across the responders’ samples, composed largely of effector cytokines (granzyme-β, IFN-γ, MIP-1α, perforin, TNF-α). Conclusions: Single-cell functional proteomic analysis revealed significantly upregulated polyfunctional profiles and an increase in effector cytokines in patients who responded to nivo. This data provides important insights into PD-1 inhibitor triggered T-cell activity and may be used to predict response to ICIs in HNSCC patients using a blood test. Clinical trial information: NCT03355560 .

TPS587 Background: Patients (pts) with cisplatin ineligible metastatic urothelial carcinoma (mUC) who progress after first-line PD1/PDL1 inhibition have limited treatment options. The concept of maintenance therapy with targeted agents and adding onto it at time of progression is a proven effective strategy. Preclinical data indicate that carboplatin+gemcitabine have immunomodulatory effect to potentially augment immune response. We hypothesize that in pts with cisplatin ineligible mUC, the use of atezolizumab+chemotherapy after progression on single-agent PD1/PDL1 inhibitor will result in clinical benefit. Methods: Multi-center, single arm, open label phase 2 trial of atezolizumab+carboplatin+gemcitabine in pts with cisplatin ineligible mUC. Eligible pts are adults with mUC (mixed histology allowed) who progressed after first line PD1/PDL1 inhibitor. Pts should be cisplatin ineligible based on consensus criteria. Neoadjuvant/adjuvant chemotherapy completed ≥12 months prior to enrollment is allowed. Treatment with atezolizumab will continue until disease progression or unacceptable toxicity while carboplatin+gemcitabine can be stopped after 4-6 cycles. Primary objective is progression-free survival per RECIST and secondary objectives are overall response rate, clinical benefit rate, and overall survival (OS). Exploratory endpoints include to compare OS of atezolizumab+carboplatin+gemcitabine in this trial compared to a virtual control arm of carboplatin+gemcitabine in mUC after progression on first-line PD1/PDL1 inhibitors. Other exploratory endpoints include to compare PD-L1 status at time of diagnosis and at time of enrollment (after progression on PD1/PDL1 inhibitor). Using an alternate hypothesis that atezolizumab+carboplatin+gemcitabine will have a median PFS of 9 months compared to historical control of 5 months with a platinum regimen in 2nd line setting, we plan to enroll 33 patients. This study is currently enrolling pts. A protocol amendment is under way that will allow pts with prior platinum-based chemotherapy to enroll on this trial. Clinical trial information: NCT03737123.

e14273 Background: Cetuximab is a recombinant chimeric monoclonal IgG1 antibody which binds specifically to the epidermal growth factor receptor (EGFR) and stimulates an innate immune response by promoting natural killer (NK) cell mediated antibody-dependent cell-mediated cytotoxicity (ADCC). Cetuximab is approved as a single agent in relapsed/metastatic head and neck squamous cell carcinoma (R/M HNSCC). PD-1 check-point inhibitors which release the inhibition of the adaptive immune response, are also approved as single agents in this setting. However the response rates with these drugs, when used individually range from 10-20%. We hypothesized that adding a PD-L1 inhibitor, durvalumab to cetuximab would cause anti-tumor synergy by activating the innate as well as adaptive immune systems without compromising safety in this phase-2 trial in R/M HNSCC patients who have progressed on platinum based therapy. Methods: Blood samples were collected from the first six enrolled patients prior to starting treatment and 4 weeks after the first combined dose of cetuximab and durvalumab. PBMCs were isolated, stained with a live/dead stain as well as CD3, CD4, CD8, CD56, CD16 and NK2GD (natural killer group 2 member D activation receptor) antibodies and analyzed by flow cytometry. Cytokine levels in plasma were measured using standardized ELISA assay kits. Results: Compared to pre-treatment levels, post-treatment samples showed an increase in activated cytokine producing NK cells (CD56bright/CD16-) in all but one patient. Activated cytotoxic NK cell subpopulations (CD56dim/CD16+) showed variable results post-treatment. CD8+ T cells were similar pre and post-treatment in 5 patients. TGF-b levels increased in 5 patients and decreased in 1 patient post-treatment. Interestingly, the patient with decreased TGF-b levels post-treatment had an almost doubling of CD8+ T-cells and an increase in activated cytokine producing NK cells (CD56bright/CD16-). Conclusions: The clinical trial is ongoing and therefore, comparison to clinical response has not yet been analyzed. However, these findings support the combination of cetuximab and durvalumab in R/M HNSCC given the activation of an NK-cell mediated innate immune response in these patient samples. Clinical trial information: NCT03691714.

6037 Background: Metformin is a biguanide, widely used oral hypoglycemic agent. Metformin has also shown to inhibit tumor growth and progression in a wide variety of cancers including Head and Neck Squamous Cell Carcinoma (HNSCC). Metformin activates AMP protein kinase (AMPK) related pathways leading to inactivation of mammalian target of rapamycin (mTOR) and suppression of its downstream effectors. In addition, metformin is postulated to alter immune regulation in the tumor microenvironment leading to increased tumor cell killing. Here, we report our findings on the impact of metformin on T cells, NK cells and cytokines from patient peripheral blood mononuclear cells (PBMCs) from a phase I open-label single site dose escalation study combining metformin and chemoradiation (CRT) in HNSCC (NCT02325401). Methods: In this study, we evaluated the immune cell phenotypes and cytokine profiles of peripheral blood in patients before and after metformin treatment on trial by using flow cytometry and cytokine magnetic bead assays (Luminex). Cytokine profiles were further studied in co-culture experiments combining PBMCs, HNSCC cell lines, and metformin. Results: Patients who received metformin developed expanded NK cell populations, increased NKG2D expression, and a shift in their CD8+ T-cell memory phenotypes. Patient serum ELISA examination revealed increased anti-tumorigenic cytokine profiles. Metformin treatment of HNSCC cell lines in vitro as well as HNSCC PBMCs ex vivo resulted in downregulation of STAT3 compared to healthy controls. Downregulation of STAT3 may be a potential mechanism in which metformin stimulates NK cells. Conclusions: Here we show evidence that metformin treatment has a direct effect on the innate immune system in patients with HNSCC, inducing an anti-tumorigenic immune response suggesting that metformin continues to be a good candidate to yield improved clinical outcomes in patients with advanced stage HNSCC. Clinical trial information: NCT02325401.

6074Background: Up to 60% of HNSCC present as locally advanced disease (LAHNSCC). Although prognosis has improved significantly, 3 year PFS and OS remain at 62%, and 73% respectively (RTOG 0522) de...

e17500 Background: Metformin is hypothesized to suppress tumor cell growth by mTOR pathway inhibition, which mediates the phosphoinositide 3-kinase/Akt signaling pathway (frequently deregulated in human cancers). This effect is mediated by activation of the AMP protein kinase (AMPK). Despite knowledge of the mechanism, various retrospective studies show conflicting results about effect of metformin on survival in head and neck squamous cell carcinoma (HNSCC) patients (pts). Methods: We analyzed a retrospective cohort of 1500 consecutive pts (between 2009-2015) with HNSCC. We added about 1000 pts to previously presented data (e17514-ASCO 2016). These pts included those with diabetes (DM) on metformin (MET) or other anti-diabetic agents. Descriptive statistics were used for demographic and pt data. Kaplan-Meier method was used to calculate median overall survival (OS) and progression free survival (PFS), 95% CI. Cox model was used to test association between MET (or DM) and OS (or PFS) (using SPSS-24 (IBM)). Results: Of the 1500 pts with HNSCC, 329 had advanced cancer (stage III/ IV), and received definitive chemo-radiation (CRT). Of these, majority were Caucasian (90%), males (90%) and smokers (70%). Mean age at diagnosis was 57 years (SD 10). 53 pts had DM, of which 37 were on MET. Statistical results are shown in Table 1. Conclusions: This retrospective study showed no statistically significant difference between various subgroups described in the table. Given the low number of diabetics, and pts on MET, data is limiting. However, further subset analyses are ongoing to determine whether a difference in treatment (RT vs. cetuximab vs. cisplatin) had an impact on outcomes. Future experiments will utilize tissue microarrays to evaluate for association with mTOR pathway aberrations. [Table: see text][Table: see text]

e15624Background: The ABC-02 trial defined gemcitabine plus cisplatin (GC) as the standard regimen for advanced biliary tract cancers (BTC). This single institution retrospective review evaluates t...

e17514Background: Head and neck squamous cell carcinomas (HNSCCs) have been shown to exhibit activation of the PI3K–mTOR pathway, which is important for cellular proliferation and survival. The hyp...

e22237 Background: Recent preclinical data through a systematic drug-repurposing computational approach querying a large number of gene expression datasets identified TCAs as having a survival bene...

e19150 Background: Low Molecular Heparin (LMWH) is approved for treatment of and secondary prophylaxis against thromboembolic events in patients with cancer. However, its role in improving Overall ...