Aim: To estimate the association between serum 25-hydroxyvitamin D (25(OH)D) levels and survival among colorectal and breast cancer patients.
Methods: We performed a comprehensive literature search of prospective cohort studies assessing the association of serum 25(OH)D levels with survival in colorectal and breast cancer patients. Study characteristics and results were extracted and dose-response relationships were graphically displayed in a standardised manner. Meta-analyses using random effects models were performed to estimate pooled hazard ratios.
Results: The systematic search yielded five studies including 2330 colorectal cancer patients and five studies including 4413 breast cancer patients all of which compared mortality across two to five categories of 25(OH)D levels. Among colorectal cancer patients, pooled hazard ratios (95% confidence intervals) comparing highest with lowest categories were 0.71 (0.55-0.91) and 0.65 (0.49-0.86) for overall and disease-specific mortality, respectively. For breast cancer patients, the corresponding pooled estimates were 0.62 (0.49-0.78) and 0.58 (0.38-0.84), respectively. No significant evidence of heterogeneity between studies was observed.
Conclusion: Higher 25(OH)D levels (>75nmol/L) were associated with significantly reduced mortality in patients with colorectal and breast cancer. Randomised controlled trials are needed to evaluate whether vitamin D supplementation can improve survival in colorectal and breast cancer patients with low vitamin D status (25(OH)D<50nmol/L) at diagnosis and before treatment.
Keywords: Breast cancer; Colorectal cancer; Meta-analysis; Patients; Vitamin D.
Copyright © 2014 Elsevier Ltd. All rights reserved.