Autoantibody (AAb) profiling in metastatic cisplatin-ineligible metastatic urothelial carcinoma (mUC) treated with ipilimumab (IPI), nivolumab (NIVO), and sacituzumab govitecan (SG).

863 Background: We previously reported that the combination of IPI-NIVO plus SG shows high efficacy with 83.3% objective response rate (ORR) as first-line treatment for cisplatin-ineligible mUC patients in a Phase I/II trial (NCT04863885). However, two grade 5 immune-mediated myocarditis events attributed to IPI-NIVO occurred, leading to early trial termination after accruing 25 patients. We evaluated serum autoantibodies (AAbs) to predict response and immune-related adverse events (irAEs) in patients receiving IPI-NIVO + SG on this trial. Methods: Using a customized 96-antigen NavigAID panel potentially linked to heart, muscle, autoimmune myositis, myasthenia gravis (MG) and checkpoint inhibitors (CPI)-induced irAEs in mUC, IgG AAbs were measured via SeroTag in pre- and post-treatment serum from mUC patients enrolled in NCT04863885. Comparative analyses included baseline vs. healthy controls, association with irAEs, best overall response (BOR), and time to progression (TTP). Top associated AAbs were selected by |log2 fold change| > 1 and |ΔMFI| > 500 between groups. P-values were calculated post-hoc using logistic regression risk modeling, ANOVA or SAM analysis. Results: Risk model analysis showed moderate evidence linking cardiac and muscle-related antigens L1CAM (OR 4.43, p = 0.08) and KCNJ8 (OR 4.31, p = 0.086) to myocarditis. KCNJ8 (KIR6.1) mutation has been previously associated with cardiac arrest. While no clear association was observed between total AAb burden and specific irAEs or response types, individual AAbs such as HMGCR (p = 0.66/0.22), TRIM28 (p = 0.43/0.058), and MYL4 (p = 0.44/0.61) showed potential predictive value for BOR/TTP. TRIM28 (TIF1β) and HMGCR AAbs are myositis-specific AAbs (MSA). HMGCR AAbs are a marker for necrotizing autoimmune myopathy (NAM), a rare but severe muscle disease. MYL4 is expressed in adult cardiac atrial tissue. Treatment-induced AAb variation was higher in responders (CR/PR). The most prominent post-treatment changes in AAbs were observed for anti-TONSL (p = 0.07). AAbs to several tumor associated antigens were elevated in mUC vs. controls including CTAG1B (NY-ESO-1) (p = 0.03) and BCL2L1 (p = 0.02). Conclusions: In patients with mUC receiving first-line IPI-NIVO+ SG, pre-treatment AAbs to cardiac and muscle-related antigens L1CAM and KCNJ8 showed a trend for association with immune myocarditis, while AAbs to HMGCR, TRIM28 and MYL4 were associated with better outcomes with the treatment. Together, these observations support further exploration of AAb profiling as a biomarker to predict efficacy and toxicities in those receiving CPI based therapies.