Abstract PR-05: Biomarkers of inflammation predict colorectal cancer risk among women: Results from the Women's Health Initiative Observational Study (WHI-OS) cohort

Introduction : There is strong evidence that chronic inflammation plays a role in colorectal carcinogenesis. Despite this, epidemiological studies that have investigated this association using C-reactive protein (CRP) as an inflammatory biomarker have reported conflicting results, especially among women. Limitations of previous studies include the small number of cases and the fact that CRP measurements were performed at only one point in time. Likewise, no previous study has investigated the association of serum amyloid A (SAA), a related marker of inflammation, and colorectal cancer (CRC) risk. We investigated the associations of CRP and SAA with CRC risk, using repeat assessments, in a case-control study nested within the Women9s Health Initiative Observational Study cohort (WHI-OS). We also explored the impact of changes in the two biomarkers (from baseline to 3rd year of follow-up) on CRC risk. Methods : The WHI-OS is a prospective cohort study that enrolled 93,676 post-menopausal women between 1993 and 1998 at 40 U.S. clinical institutions. We identified 988 women with CRC and matched 988 cancer-free control women based on age (±3 years), clinical center, race/ethnicity, and time of blood draw (±6 months). CRP and SAA measurements were performed at baseline and during the 3rd year of follow-up. Conditional logistic regression models were used to estimate the multivariate-adjusted odds ratios and 95% confidence intervals (OR, 95% CI) of CRC (adjusted for age, body mass index, post-menopausal hormone use (HRT) and previous history of colonoscopy). Results : Elevated CRP, but not SAA concentrations were positively associated with CRC risk. Women in the highest quintiles of CRP and SAA concentrations had an odds ratios of 1.28 (95% CI 0.94–1.76, p-trend=0.03) and 1.19 (95% CI 0.88–1.62, p-trend=0.17) respectively, compared to women in the lowest quintiles. Women who had high levels of both biomarkers had increased risk of colorectal cancer (OR=1.35, 95% CI 1.06–1.73, p-value=0.02), particularly colon cancer (OR=1.49, 95% CI 1.13–1.97, p-value=0.005) compared to those with low concentrations. Neither CRP nor SAA was positively associated with rectal cancer risk in any of the analyses. There was a non-significant increased risk of CRC (OR 1.31, 95% CI 0.91–1.89, p-trend=0.14) among women who had the largest increase in CRP concentration (from baseline to 3rd year of follow-up), while the risk associated with changes in SAA concentration was close to unity (OR 1.05 95% CI 0.73–1.49, p-trend 0.94). The associations between CRP and CRC risk were not modified by HRT use. Conclusions : Our study supports the hypothesis that inflammation is associated with colorectal carcinogenesis and that CRP and SAA used together may better predict CRC risk than either used alone. Citation Information: Cancer Prev Res 2011;4(10 Suppl):PR-05.