Vitamin D and Cancer Mortality: Systematic Review of Prospective Epidemiological Studies
Anti-Cancer Agents in Medicinal Chemistry
Accumulating evidence from experimental and epidemiological studies suggests that vitamin D deficiency might be a causal risk factor for cancer and therewith associated mortality. We performed a systematic review in Medline up to February 2012 to identify prospective studies on 25-hydroxyvitamin D (25[OH]D) and cancer mortality as well as on 25(OH)D and survival in cancer patients. Our search retrieved 13 studies on cancer-specific mortality and 20 studies on overall mortality in cancer
... y in cancer patients. Data on 25(OH)D and cancer mortality were mainly derived from general populations. The results were inconsistent and yielded either no, inverse or positive associations. By contrast, the majority of studies in cancer patients showed that patients with higher 25(OH)D levels had a decreased risk of mortality. This relationship was particularly evident in cohorts of colorectal cancer patients. In contrast, there was no indication for increased mortality risk with higher vitamin D levels in any cancer cohort. In conclusion, the relationship of vitamin D status and cancerspecific mortality is still unclear and warrants further studies. Our results provide a strong rationale to perform prospective randomized controlled studies to document a potential effect of vitamin D supplementation on survival in cancer patients. Pilz et al. the association of vitamin D status and total mortality in cancer patients. Finally, we critically discuss our findings. Vitamin D Metabolism Vitamin D exists in two main isoforms, i.e. vitamin D 3 and vitamin D 2 , but unless stated otherwise we do not differentiate between these two isoforms and refer to vitamin D (meaning both isoforms) in general [1, 34] . Vitamin D can be endogenously obtained by synthesis in the skin, where sunlight (i.e. UV-B) induces the conversion of the liver derived precursor 7dehydrocholesterol to vitamin D 3. This is the main vitamin D source accounting for ~80% of vitamin D [1, 14] . Because vitamin D production in the epidermis is mainly the result of UV-B radiation, its production is modulated by various factors such as latitude, time of day, season, weather or air pollution as well as by skin parameters such as melanin content  . Natural foods contain relatively minor amounts of vitamin D 3 (e.g. in fish) or vitamin D 2 (plant sources such as mushrooms). For this reason we did not include nutritional vitamin D intake in this review. Further vitamin D sources are vitamin D fortified food (e.g. milk in the US) and vitamin D supplements. Vitamin D itself does not exert significant biological actions. Two hydroxylation steps are required to form the most active vitamin D metabolite 1,25-dihydroxyvitamin D (1,25[OH]2D). In a first step vitamin D is hydroxylated in the liver to 25(OH)D which is the main circulating vitamin D metabolite that is usually measured to assess and categorize vitamin D status. Subsequently, 25(OH)D is further hydroxylated by the 1-alpha-hydroxylase to 1,25(OH)2D in the kidney. The 1-alpha-hydroxylase activity is tightly regulated by parathyroid hormone (PTH) and by the Fibroblast growth factor-23 (FGF-23)/klotho axis. Apart from such main biosynthesis in the kidney, almost all organs express 1-alphahydroxylase as well. [32, 34, 36] . This extra-renal synthesis of 1,25(OH)2D seems to be significantly dependent on circulating 25(OH)D levels which underlines that 25(OH)D serum levels are a good parameter of whole-body vitamin D status, whereas circulating 1,25(OH)2D levels mainly reflect synthesis of 1,25(OH)2D in the kidney [32, 34, 36] . It should also be acknowledged that serum levels of 25(OH)D are up to 1000 fold higher concentrated compared to 1,25(OH)2D levels. The affinity of 1,25(OH)2D to the almost ubiquitously expressed VDR is much higher compared to 25(OH)D. VDR activation finally leads to the regulation of hundreds of genes by binding to so called vitamin D responsive elements (VDRE) on the DNA . Degradation of 1,25(OH)2D and 25(OH)D is initiated by 24-hydroxylation and subsequent metabolism to less active vitamin D metabolites. There is almost universal agreement that serum levels of 25(OH)D are the best parameter to assess the vitamin D status, but there is no consensus what is an adequate and what is an inadequate vitamin D status [1, 3, 4, 23, 37, 38] . The rationale for vitamin D cut-off levels was initially based on the fact that in states of low 25(OH)D levels, disturbances in calcium and bone metabolism emerge. Whereas many vitamin D experts consider 25(OH)D levels ≥75 nmol/L (divide by 2.496 to convert nmol/L to ng/mL) as sufficient others classify 25(OH)D levels of ≥50 nmol/L as a sufficient vitamin D status [1, 3, 4, 23, 37, 38] . Hence, the general consensus seems to be that 25(OH)D levels <50 nmol/L (divide by 2.496 to convert nmol/L to ng/mL) are below the normal range for 25(OH)D. In this context, large epidemiological studies in general populations showed that almost every second to every third individual suffers from 25(OH)D levels below this threshold of 50 nmol/L. This underscores the significance of vitamin D deficiency as a public health problem [1, 13, 39] .