Recent genome-wide association studies (GWAS) have identified several genetic loci associated with the risk of developing colorectal cancer (CRC), including SMAD7 (18q21). SMAD7 regulates the Wnt signaling pathway, interacts with transforming growth factor-β (TGF-β), and is involved in other functions of the inflammatory process. Pre-diagnostic use of non-steroidal anti-inflammatory drugs (NSAID) has been shown to reduce the risk of developing colorectal cancer and may affect survival following diagnosis. It is unclear what role, if any, SMAD7 may have on survival. We considered two candidate single nucleotide polymorphisms (SNP) of SMAD7 identified from GWAS: one where the minor allele is associated with a decreased risk of developing colorectal cancer (rs4939827) and one where the minor allele is associated with an increased risk of developing colorectal cancer (rs4464148). A population-based registry within the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program was used to identify incident cases of invasive colorectal cancer occurring from 1997 to 2002 in women between the ages of 50 and 74 that were residents of 13 counties in the Puget Sound area of Washington State. A total of 731 cases of colorectal cancer were genotyped, including 352 with tumors in the proximal colon and 379 with tumors in the distal colon or rectum. Through December 31, 2009, a total of 161 deaths due to colorectal cancer occurred (77 proximal and 84 distal/rectal). The median duration of follow-up was 6.4 years. Proportional hazards regression was used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for CRC-specific survival using a co-dominant model, and P-values for a linear trend in effect were calculated assuming a log-additive model. Each SNP was considered individually in models that adjusted for race, age at diagnosis, and stage at diagnosis. Additionally, models were stratified by tumor subsite and pre-diagnostic NSAID use. Among all cases, there was no significant association with CRC-specific survival for either polymorphism (for rs4939827 TT is the reference genotype; adjusted HR = 1.03 for TC, 95% CI: 0.71 — 1.49; adjusted HR for CC = 1.21, 95% CI: 0.78 — 1.88, with P for trend = 0.41; for rs4464148 TT is the reference genotype; adjusted HR for TC = 0.90, 95% CI: 0.64 — 1.26; adjusted HR for CC = 1.17, 95% CI: 0.69 — 1.98, with P for trend = 0.92). For those with distal/rectal tumors, we observed evidence of effect modification for rs4939827 by pre-diagnostic NSAID use (P for interaction = 0.04), with the strongest association occurring among those who ever used NSAIDs regularly (adjusted HR = 3.12, 95% CI: 1.20 — 8.11 for the CC genotype, with P for trend = 0.02, based on 43 deaths among 188 at risk). In contrast, no modification of the rs4464148 effect on CRC-specific survival by pre-diagnostic NSAID use was observed. Previous studies indicate that the risk of developing colorectal cancer may be reduced among individuals with the rs4939827 variant. Our results do not show evidence of an overall association between this variant and CRC-specific survival. In stratified analysis, however, we noted a potential association between the C allele of rs4939827 and poorer prognosis, specifically in patients that have tumors in the distal colon or rectum and have ever used NSAIDs regularly prior to diagnosis.

Citation Information: Cancer Prev Res 2010;3(12 Suppl):B71.

Copyright © 2010, American Association for Cancer Research
2010