Abstract:<h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Objective</h3><p>The relationship between vitamin intake and occurrence of cancer is controversial. To address this issue, this study examined the genetic associations between vitamins D, E, B12 and five cancers, i.e., colorectal cancer, breast cancer, prostate cancer, malignant melanoma and squamous cell carcinoma.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Methods</h3><p>This study started from genome-wide association data for three vitamins (<em>n</em> = 11,238) and five cancers (<em>n</em> = 373,316), and analyzed their associations using Mendelian randomization (MR) methods. In addition, survival analysis was performed using the Cancer Genome Atlas (TCGA) data to further evaluate some MR results.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Results</h3><p>MR analysis indicated that intake of vitamins D, E, B12 is irrelevant with the risk of five cancers (P<sub>MR</sub> > Bonferroni-corrected P = 0.02). Some of the results were supported by epidemiological observations; some were further supported by survival analysis using TCGA data.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Conclusion</h3><p>There is no genetic evidence to support the associations between intake of vitamins D, E, B12 and risk of five cancers.</p>

Are vitamins relevant to cancer risks? A Mendelian randomization investigation