Single nucleotide polymorphisms (SNPs) in MHC class I and II genes to predict outcome in patients (pts) with metastatic colorectal cancer (mCRC): Data from FIRE-3, MAVERICC, and TRIBE trials

3578 Background: The immune system is alerted for virally infected cells in the body by the antigen presentation pathway, which is in turn mediated by the major histocompatibility complex (MHC) class I and II molecules. Cancer cells overcome immune evasion as a major hallmark by downregulation of antigen presentation pathway molecules. Therefore, the present study aimed to explore the effect of genetic variants in MHC class I and II pathways on first-line treatment outcome in mCRC pts. Methods: Genomic DNA from blood samples of 775 pts enrolled in three independent, randomized, first-line trials: TRIBE (FOLFIRI-bevacizumab [bev], N = 215), FIRE-3 (FOLFIRI-bev, N = 107; FOLFIRI-cetuximab [cet], N = 129) and MAVERICC (FOLFIRI-bev, N = 163; FOLFOX-bev, N = 161) was genotyped through OncoArray, a custom array manufactured by Illumina including approximately 530K SNP markers. The impact on outcome of 40 selected SNPs in 22 genes of MHC class I and II pathways ( ERAP1, ERAP2, TAP1, TAP2, TAPBP, B2M, HLA-A, HLA-B, HLA-C, HLA-E, HLA-F, HLA-G, CIITA, HLA-DMA, HLA-DMB, HLA-DOA, HLA-DOB, HLA-DPA1, HLA-DPB1, HLA-DQA1, HLA-DQB1 and HLA-DRA) was analyzed. Results: We identified several SNPs in multiple genes associated with targeted treatment benefit across different treatment arms in our study population ( P < .05). Germline variants in ERAP1 rs2287987 were associated with worse PFS in pts receiving FOLFIRI-bev in TRIBE (10.4 vs 8.8 months, HR 1.48, 95%CI 1.04-2.12, P = .035) and FIRE-3 trials (11.1 vs 9.9 months, HR 3.46, 95%CI 1.18-10.12, P = .049), while rs26653 SNP in the same gene was associated with better PFS in MAVERICC (10.1 vs 14.5 months, HR 0.54, 95%CI 0.35-0.84, P = .0062). TAP1 rs1135216, TAP2 rs1800454 and rs1044043, HLA-B rs2770, HLA-G rs1610696, CIITA rs4774, HLA-DRA rs7192 were associated with longer OS and/or PFS in cet-treated pts in FIRE-3; whereas TAP2 rs241447, TAPBP rs3106191, HLA-DMB rs10751, HLA-DOB rs11244, HLA-DPB1 rs3097671 showed worse PFS and/or OS. SNPs in TAP2 (rs1800454), HLA-C (rs1049281), HLA-G (rs1063320), CIITA (rs1139564), HLA-DMB (rs1042337), HLA-DOA (rs375256, rs3129303), HLA-DPA1 (rs1042190) and HLA-DRA (rs7192) were specifically associated with clinical outcomes in the FOLFOX-bev arm of MAVERICC but not in control cohorts of pts treated with FOLFIRI-bev. Treatment-SNP interaction analyses with targeted agents (bev vs cet) and chemotherapy backbone (FOLFIRI vs FOLFOX) confirmed a significant treatment interaction for HLA-G, TAP2, CIITA, and HLA-DMB SNPs ( P < .05). Conclusions: Our results highlight an important role for MHC SNPs as prognostic and predictive biomarkers for first-line treatment in mCRC, with differential effects based on biologic agent and chemotherapy backbone. These biomarkers, when further validated, may contribute to personalized treatment strategies for mCRC patients.