Abstract 7126: Divergent molecular features associated with immune checkpoint blockade response

Abstract The advent of immune checkpoint blockade (ICB) has revolutionized cancer management. However, not all patients will benefit from this powerful therapy, necessitating clinical biomarkers and combination treatment strategies to enhance therapeutic efficacy. Recently, high tumor mutation burden (TMB-H) was FDA-approved as a pan-cancer biomarker for treatment with ICB based on a limited sample size and diversity of tumors. In systematic analysis across cancer types, we observed that immune cell infiltration was only associated with TMB in a subset of cancers. Based on this, we divided cancers into “Category I” cancers where TMB is associated with increased immune infiltration, and “Category II” cancers where TMB is not associated with immune infiltration. Using these classifiers, we analyzed the ability of TMB to predict ICB outcomes across over 2, 000 patients. We found that TMB-H predicted ICB outcomes in Category I, but not Category II, cancers. As Category II cancers represent many of the most common cancers in the United States, including those of the breast, prostate, kidney, and brain, we next sought to identify biomarkers relevant to this subset of cancers. We identified that defects in DNA replication stress response predicts ICB response in 12 cohorts of Category II cancers, but not Category I cancers. Furthermore, induction of replication stress response defects sensitized otherwise resistant tumors to ICB. Based on these divergent molecular correlates of ICB response, we hypothesized that different mechanisms may underlie ICB activity in these tumors. Using single-cell RNA sequencing, we identified a non-canonical PD1+ dendritic cell population uniquely in mouse and human Category II cancers, and that this population undergoes extensive transcriptional rewiring following treatment with ICB. Together, this work illustrates the need for context-specific biomarkers for treatment with immune checkpoint blockade. Citation Format: Travis D. Kerr, Patrick G. Pilie, Hui Dai, Amy B. Heimberger, Renata Ferrarotto, Jeffrey M. Rosen, Shiaw-Yih Lin, Daniel J. McGrail. Divergent molecular features associated with immune checkpoint blockade response [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 7126.