Genome-wide association study (GWAS) and novel prognostic markers in metastatic colorectal cancer (mCRC).

Yan Yang,Francesca Battaglin,Shivani Soni,Sandra Algaze,Jae Ho Lo,Pooja Mittal,Lesly Torres-Gonzalez,Federico Innocenti,Alan P. Venook,Sebastian Stintzing,Aparna Raj Parikh,Christoph Mancao,Volker Heinemann,Chiara Cremolini,Wu Zhang,Joshua Millstein,Heinz-Josef Lenz

DOI: https://doi.org/10.1200/jco.2024.42.3_suppl.174

IF: 45.3

2024-01-20

Journal of Clinical Oncology

Abstract:174 Background: Few genome-wide investigations have examined germline variants associated with mCRC outcomes following primary treatment. Here, we performed a GWAS meta-analysis to explore the impact of germline genetic variation on progression-free survival (PFS) and overall survival (OS) in four clinical trials of mCRC. Methods: The GWAS comprised 1,324 mCRC patients (pts) enrolled in four randomized Phase II/III trials for first line treatment: FIRE-3 (FOLFIRI-bevacizumab [bev]; FOLFIRI-cetuximab[cet]), MAVERICC (FOLFIRI-bev; FOLFOX6-bev), TRIBE (FOLFIRI-bev; FOLFOXIRI-bev), and TRIBE2 (mFOLFOX6-bev; FOLFOXIRI-bev). DNA isolated from blood was genotyped using the OncoArray and imputed against the Haplotype Reference Consortium panel. In each trial arm, associations between SNPs and PFS and OS were assessed using an additive coding in a Cox proportional hazard framework, adjusting for trial specific patient characteristics and principal components. Meta-analysis was carried out using an inverse-variance weighted fixed effect model. All reported GWAS p-values were adjusted for genomic inflation. To further investigate significant hits, we evaluated associations between gene expression measured from FFPE tumor samples and PFS and OS in 433 pts enrolled in an independent randomized first-line Phase III trial, CALGB/SWOG 80405. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated in multivariable Cox models for expression levels in the top (Q4, high) vs the bottom quartile (Q1, low). Results: Two SNPs in the MLPH intronic region and one SNP in the TYR intronic region surpassed the GWAS significance threshold for PFS and OS, respectively. The top ranked SNPs associated with PFS and OS, rs13409241 (A>G, PFS HR: 1.35, 95% CI [1.21, 1.51], P=4.34E-8) and rs10765199 (T>A, OS HR: 1.45, 95%CI [1.27, 1.65], P=4.69E-8), showed similar effects across all eight treatment arms ( I 2 = 0% and 38.5%, respectively). No significant heterogeneity variant was found. In the CALGB/SWOG 80405 cohort, pts with MLPH-low tumors (N=109) showed significantly longer PFS (median 14.8 vs 7.6 months, low vs. high (as reference) adjusted HR 0.53, 95% CI [0.38, 0.74], P=2.1E-4) and OS (median 35.8 vs. 18.2 months, adjusted HR 0.65, 95% CI [0.46, 0.92], P=0.015) compared to MLPH-high (N=107). No significant associations for TYR expression and outcomes were found. MLPH and TYR are involved in melanosome transport and melanogenesis. Melanin has been shown to reduce vitamin D levels, which can affect CRC prognosis. MLPH has also been reported to accelerate epithelial-mesenchymal transition and metastasis in lung and prostate cancers. Conclusions: These results suggest that germline genetic variation in MLPH and TYR may affect mCRC prognosis via a mechanism that involves melanin and possibly the epithelial-mesenchymal transition. Note: Authors Millstein and Lenz contributed equally to this work.

oncology