Utilizing the autoantibody immune response to tumor antigens for kidney cancer early detection.

369 Background: Kidney cancer (renal cell carcinoma, RCC), the 8th most common U.S. cancer, is in need for better cure rates through early detection (5-year relative survival for stage I RCC: ̃95%; for stage IV RCC ̃19%). Autoantibodies are common in cancer and result from the altered expression, localization, or post-translational modification of endogenous proteins in tumor cells (autoantigens) and from the expression of mutated genes that give rise to new proteins (neoantigens). In contrast to cellular immune responses in cancer, autoantibodies are less well characterized, yet hold promise to enable cancer early detection by immune amplification of the ‘cancer signal’ while retaining specificity to cancer types including RCC. Autoantibodies may therefore be useful for kidney cancer early detection and diagnosis. Our goal was to profile the autoantibody repertoire in blood from patients with clear cell RCC (ccRCC), the most common form of RCC, in order to: 1) determine if autoantibodies can be detected in patients with early-stage and late-stage ccRCC; 2) identify common epitopes amongst ccRCC patients that could suggest common RCC antigens; and 3) determine specificity and sensitivity of potential autoantibody biomarkers for ccRCC vs. other non-cancer conditions. Methods: We use the SERA platform (https://serimmune.com/publications/) to compare putative autoantibody signal in blood from 177 patients with ccRCC, 23 with benign kidney lesions, and ̃800 healthy controls. SERA utilizes a random bacterial display 12mer peptide library of 1010 diversity in conjunction with next-generation sequencing to ascertain epitope enrichment across the entire human proteome. Results: We find significant differences in epitope repertoires in ccRCC compared to the healthy human cohort. Patients with ccRCC exhibit a rich repertoire of rare, enriched epitopes which may comprise putative autoantibody signal. This epitope signal is present with high abundance in all ccRCC stages, including stage I ccRCC. In contrast, healthy controls and patients with benign kidney lesions demonstrate more restricted repertoires. However, we do not find evidence of common ccRCC antigens: epitopes are not conserved across large subsets of ccRCC patients. Conclusions: Our initial results suggest that each patient may develop an individualized tumor-associated antibody response. Whether assessing a select epitope panel in a patient’s blood could be useful for ccRCC early detection, or even epitope diversity without needing to identify specific epitopes, warrants further study.