Abstract 2184: Dynamics between germline immune checkpoint variants and susceptibility for cancer and autoimmunity

Abstract Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of cancer and are approved for a wide range of malignancies. ICIs block the interactions of checkpoint proteins that limit T cell-mediated immune responses, thereby boosting cytotoxic activity against cancers. The most widely used ICIs act through CTLA-4, PD-1, or PD-L1 pathway blockade, and numerous agents are at various stages of development. Due to increased T cell activity, ICIs are associated with immune-related adverse events, and little is known about their long-term impacts. Individual-level genetic variation mimics the effects that are seen with larger magnitude when the same pathways are perturbed through checkpoint modulation. Leveraging large-scale biobanks with genomics data, proteomics data, and rich health registries can enable us to understand the broad impacts of checkpoint modulation on adverse events and its potential for treatment of common diseases. We utilize FinnGen, a biobank study of 520, 210 Finnish individuals, to investigate germline variants affecting immune checkpoint proteins and their associations with cancer and autoimmunity. The Finnish population, being a genetic isolate, has an unusual enrichment of disease-associated variants compared to many other ancestry groups, enabling variant-disease associations not detectable in many other populations. Using phenome-wide association analysis, we identify several coding variants in checkpoint protein-encoding genes with opposite associations to autoimmune diseases and cancer or other tumors. Such associations are supported by Olink proteomic quantitative trait loci for respective proteins (cis-pQTLs). For example, rs149722682 (LAG3 missense variant; 440x enriched in Finns) is protective of colon adenomas (OR 0.26, p=8.4x10-5) but increases the risk of asthma (OR 1.5, p=1.1x10-4) and hypothyroidism (OR 1.9, p=1.7x10-9). We also identify multiple variants with associations to a range of conditions beyond classical autoimmune diseases, pointing to potentially new or rare adverse events of ICIs. For instance, rs781221749 (LAG3 missense; 2x enrichment) increases the risk of manic episode (OR 23.3, p=1.6x10-4) and COVID-19 infection (OR 4.9, p=1.2x10-4). Finally, we develop checkpoint protein-specific genome-wide polygenic risk scores to quantify the cumulative impacts of cis and trans checkpoint protein-associated variants on different cancer- and autoimmune-related endpoints. Our findings offer insight into the short- and long-term toxicities associated with checkpoint inhibition and provide a framework for detecting common and rare adverse events in current and emerging therapies. Citation Format: Zade Akras, Benjamin M. Neale, Nina Mars. Dynamics between germline immune checkpoint variants and susceptibility for cancer and autoimmunity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2184.