Time to target TIM-3 in colorectal cancer: TIM-3 expression to predict survival outcomes in patients in CALGB (Alliance)/SWOG 80405.

3583 Background: HAVCR2 encodes TIM3, an inhibitory checkpoint glycoprotein expressed on both innate and adaptive immune cells, with particularly heightened expression (exp) on tumor-infiltrating lymphocytes. TIM3 plays a pivotal role in immune regulation and is implicated in the pathogenesis and progression of colorectal cancer (CRC). This study aims to examine the prognostic and predictive significance of TIM3 in CRC. As TIM3 inhibitors undergo evaluation for various solid tumors, they present a promising strategy to overcome resistance to other immune checkpoint inhibitors in microsatellite-stable CRC. Methods: 433 pts with mCRC treated with either bevacizumab (Bev, n = 226) or cetuximab (Cet, n = 207) in combination with first-line chemotherapy within the CALGB/SWOG 80405 trial were included in the analysis. HAVCR2RNA was isolated from FFPE tumor samples and sequenced on the HiSeq 2500 (Illumina). Overall survival (OS) and progression-free survival (PFS) in months (mo) were compared between groups of pts categorized by tertiles of HAVCR2 exp into low (L), medium (M) and high (H). Sensitivity analyses were conducted after stratifying by side. Logrank P-values describe differences without adjustment for patient (pt) characteristics. Transcriptome-wide gene association analysis was performed using linear regression, adjusting for age, sex, ethnicity, ECOG, location, number of metastases, KRAS, MSI status, treatment, with FOLFOX or FOLFIRI, and the first three principal components from the RNA-seq data. Gene set enrichment analysis was performed with FDR adjusted p-value < 1e-5 (Benjamini-Hochberg method). Results: TIM3 exp is associated with genes involved in antigen presentation and processing, T cell and myeloid cell differentiation, activation and immunoregulation. The most highly associated genes were LAPTM5, FCGR3A, MSR1, IFI30, CD84 and LILRB4. TIM3-H showed worse PFS (9.2 vs 11.4 vs 12.9 mo, p=0.028) and mOS (23.6 vs. 30.9 vs 35.5 mo, p=0.0049) in the entire cohort. In Cet treated pts, TIM3-H had significantly shorter mOS (19.5 vs 35.8 vs 36.3 mo, p=0.00052) and in those treated with FOLFOX (23.6 vs 31.1 vs 34.1 mo, p=0.031). Pts with liver metastases and TIM3-H had worse PFS (8.9 vs 11.1 vs 12.7 mo, p=0.00094) and OS (20.9 vs 31.3 vs 35.9 mo, p=5.1e-05). Conclusions: Our results suggest that TIM3 plays an important role in the immune microenvironment, immunoregulation and may be involved in EGFR resistance pathways in CRC. CRC KRAS WT tumors with liver metastases have worse prognosis with high TIM3 exp. TIM3 exp is also predictive of survival in pts treated with FOLFOX and Cet-based treatment. These findings provide rationale for the continued investigation of TIM3 in CRC. Trial Identifier NCT00265850. Support: P30CA014089, U10CA180821, U10CA180882, U10CA180820 and UG1CA233277; U10CA180888; Pfizer, Genentech. https://acknowledgments.alliancefound.org