Serum 25‐Hydroxyvitamin‐D and Prostate Cancer U‐Shaped Risk Curves‐Does Latitude Play A Role? A Meta‐Analysis of Case‐Control Studies

Abstract

Data from some epidemiological studies indicate that both low and high levels of serum 25-hydroxyvitamin-D (25(OH)D) can increase the risk for developing prostate cancer (PC). This has caused concern that higher vitamin D nutrition may pose a cancer risk. The purpose of this meta-analysis is to test the premise that a J or U-shaped risk exists, and whether the phenomenon is specific to populations living at high latitudes. We searched the databases SCOPUS and PUBMED for relevant articles. Case-control studies fulfilling pre-determined criteria were included. A meta-regression of nested case-control studies which examined the odds ratios (OR) for PC versus the average latitude for the study region were determined for the highest two quantiles of serum 25(OH)D. We identified 18 case-control studies which collectively included 12,471 cases and 14,472 controls. For all studies combined, there was no difference in OR of PC incidence between the highest two quantiles of serum 25(OH)D (OR 1.058; 95% CI: 0.982, 1.140; p=0.140). To examine whether latitude has an effect on modifying the OR for PC we conducted a fixed-effect meta-regression and found there was no significant effect of latitude on OR for PC (p=0.07), with a slight trend towards a higher OR for PC for those living at higher latitudes. Overall, the hypothesis of J or U-shaped risk curves for PC is not supported by our analysis. There is a suggestion that J or U-shaped relationships do tend to occur at higher latitudes, but there is a lack of epidemiological data for regions south of 21 degrees– latitudes that are natural for primate habitation. Therefore, the latitudinal risk hypothesis cannot be fully tested.