Abstract

394 Background: Beyond clinicopathologic stage, there are few established markers of prognosis in colorectal cancer (CRC). Recent genome-wide association studies have identified 17 germline single nucleotide polymorphisms (SNPs) significantly associated with incident CRC. However, it is unclear if these CRC susceptibility SNPs influence survival after CRC diagnosis. Although the functionality of many of these SNPs remains unknown, a few, including rs4939827 in SMAD7, map to genes with plausible biological mechanisms associated with both cancer risk and prognosis. We examined 17 CRC susceptibility SNPs in relation to survival after CRC diagnosis. Methods: We genotyped 2,611 men and women enrolled in five prospective cohort studies who were diagnosed with invasive CRC during study follow-up: the Physicians’ Health Study (N=281), Health Professionals Follow-up Study (N=268), Nurses’ Health Study (N=367), Vitamins and Lifestyle Study (N=281), and the Women’s Health Initiative (N=1414). Analyses were limited to Caucasians with known vital status, cause of death, and survival time. We used Cox proportional hazards regression to assess associations between each SNP and CRC-specific and overall survival in study-specific models adjusted for age, sex, and stage; SNPs were modeled additively to reflect associations per copy of the minor allele. Study-specific results were combined via fixed-effects meta-analysis. Results: The G allele in rs4939827 was associated with poorer CRC-specific survival [hazard ratio (HR)=1.16, p=0.02] and overall survival (HR=1.13, p=0.03) in CRC patients. The A alleles in rs10795668 and in rs4925386 were associated with a 1.14-fold increased risk of overall mortality (both p-values=0.03) but not CRC-specific mortality. Other evaluated SNPs were not associated with survival. Conclusions: Genetic variation in rs4939827 (SMAD7) is associated with CRC-specific and overall survival. These results suggest that SMAD7 may have a role in CRC progression, and provide proof-of-principle that common germline variation may provide prognostic information beyond traditional considerations such as stage.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call