Variations in Y chromosome-related genes and clinical outcome in metastatic colorectal cancer.

Abstract

634 Background: Males have a higher cancer risk and worse clinical outcome in metastatic colorectal cancer compared to females, which has been related to loss of Y chromosome. In this study, we investigated single nucleotide polymorphisms (SNPs) in Y chromosome-related genes and their association with clinical outcome in patients from the FIRE-3 study (NCT00433927). Methods: Three SNPs (FOXL2 rs11924939 C>T, DMRT1 rs755383 T>C and DMRT1 rs3739583 A>T) were investigated in genomic DNA isolated from tissue samples of 295 patients (KRAS exon 2 wild-type) from the bevacizumab-arm of the FIRE-3 study. The median age of 195 (66.6%) male and 98 (33.4%) female patients was 65 months (range 27-76). Seventy-four (25.6%) patients had right-sided tumors and 215 (74.4%) had left-sided tumors. Results: The C/C genotype of DMRT1 rs755383 T>C was associated with significantly longer progression-free survival (PFS) in univariable (C/C 14.3 months, T/T or T/C 10.0 months, hazard ratio (HR) 0.58; P=0.033) and in multivariable Cox regression analysis adjusted for baseline characteristics (HR 0.63; P=0.026). This difference was confirmed in male patients (C/C 15.9 months, T/T or T/C 10.1 months, HR 0.58; P=0.038 and HR 0.53; P=0.023) and patients with left-sided tumors (C/C 14.6 months, T/T or T/C 10.4 months, univariable HR 0.61; P=0.044 and multivariable HR 0.60; P=0.046). The T/T genotype of FOXL2rs11924939 C>T was associated with longer PFS in patients with right-sided tumors (T/T 13.4 months, C/C or C/T 7.8 months, univariable HR 0.47; P=0.043 and multivariable HR 0.30; P=0.031) and shorter PFS in left-sided tumors (T/T 6.9 months, C/C or C/T 11.3 months, univariable HR 1.83; P=0.017 and multivariable HR 1.99; P=0.009). Conclusions: In this study, we demonstrated for the first time that SNPs in Y chromosome-related genes are associated with outcome in metastatic colorectal cancer in a gender- and location-specific way. These results highlight the value of the SNPs as potential biomarkers.