Abstract A107: Predicted vitamin D status and pancreatic cancer risk in two prospective cohort studies

Abstract

Abstract Background: Experimental data indicate that vitamin D has anticancer properties. However, epidemiologic studies on vitamin D and pancreatic cancer risk have yielded inconsistent results. We previously developed a model to predict long-term plasma 25-hydroxyvitamin D3 [25(OH)D] levels from multiple determinants of vitamin D status (dietary and supplementary vitamin D, skin pigmentation, adiposity, geographic residence, and leisure-time physical activity as a surrogate of sunlight exposure). Here, we examine the association between predicted vitamin D status and pancreatic cancer risk in the Nurses' Health Study and Health Professionals Follow-up Study. Methods: We prospectively followed 118,597 participants in these two cohorts from 1986–2006. We calculated a 25(OH)D score from six predictors of vitamin D status for each individual using a multiple linear regression model and then examined the predicted 25(OH)D levels in relation to pancreatic cancer risk with Cox proportional hazards models. The multivariate models were adjusted for age, sex, race, height, smoking, and diabetes. Results: During 20 years of follow-up, we identified 575 incident pancreatic cancer cases. Higher 25(OH)D score was associated with a significant reduction in pancreatic cancer risk; compared with the lowest quintile, participants in the highest quintile of 25(OH)D score had an adjusted RR of 0.65 (95% CI, 0.50–0.86; P for trend = 0.001). The results remained unchanged after further adjusting for BMI and physical activity or after excluding the first 2 years of follow-up. This inverse association was also consistent across gender and strata of other covariates, including age, region, smoking, BMI, and physical activity. Conclusion: Higher 25(OH)D score was associated with a lower risk of pancreatic cancer in these two prospective cohort studies. Citation Information: Cancer Prev Res 2010;3(1 Suppl):A107.