Abstract 578: Prediction of PD-1 treatment outcome in mismatch repair deficient metastatic colorectal cancer using a multi-omic approach

Abstract Background: Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of patients with deficient DNA mismatch repair (d-MMR) cancers, yet less than half of patients with metastatic colorectal cancer (mCRC) receive clinical benefit. We analyzed heterogeneity within d-MMR tumors using a multi-omics approach inclusive of percent unstable microsatellite loci (MS). Methods: We studied consecutive patients with d-MMR mCRC with available primary tumors (N=32) that were treated with an anti-PD-1 antibody at Mayo Clinic (2015-2022). Tumors were profiled using a whole exome and transcriptome platform (ImmunoID NeXT®). MSIsensor was used to calculate percent unstable microsatellite loci from whole exome sequencing data. CIBERSORT and ssGSEA were used to analyze immune gene expression in relationship to percent unstable microsatellite loci. An immune checkpoint gene expression signature was calculated as the mean expression of CD274, CTLA-4, HAVCR2, LAG-3, PDCD1, PDCD1LG2, TIGIT. Aforementioned variables were analyzed in relationship to objective tumor response (RECIST version 1. 1). A Cox proportional hazards model was fit to predict progression-free survival (PFS). Results: Among 32 patients with d-MMR mCRC treated with PD-1 blockade, 16 (50%) received first-line anti-PD-1 therapy; others had 1 prior chemotherapy regimen. Eighteen (56. 3%) tumors harbored BRAFV600E, 4 (12. 5%) had mutant KRAS, and median PFS was 24. 6 months (95% CI: 8. 4, NR) with 18 events. Neither tumor mutation burden (TMB) nor PD-L1 mRNA expression predicted PFS. Median percent unstable MS loci was 14. 38 (IQR: 2. 6-23. 2). Patients whose tumors had higher percent unstable MS loci had significantly better PFS (Q2-4 vs Q1; HR: 0. 22, 95%CI: 0. 06, 0. 62; P=0. 003), and were associated with increased objective response (P=0. 049). Tumors with higher unstable loci also had higher neoantigen clonality (R= 0. 64; P= 0. 0003) and increased M0 (resting) macrophages (P=0. 02). Nonresponders (vs responders) to PD-1 blockade had an increased immune checkpoint expression signature (P= 0. 005). Conclusion: A higher percentage of unstable microsatellite loci, associated with higher neoantigen clonality and resting macrophages, was predictive of anti-PD-1 efficacy. Furthermore, an immune checkpoint gene expression signature was associated with objective response. Citation Format: Bahar Saberzadeh-Ardestani, Oluwadunni E. Emiloju, Rondell P. Graham, Jason T. Lewis, Charles W. Abbott, Sean M. Boyle, Frank A. Sinicrope. Prediction of PD-1 treatment outcome in mismatch repair deficient metastatic colorectal cancer using a multi-omic approach abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 578.