The 'cocaine blues' and other problems in epidemiologic studies of vitamin D and cancer.

Compelling evidence from epidemiologic and laboratory studies and from clinical trials indicates that vitamin D metabolites decrease the risk of several common cancers. To set the stage for future progress in the area of vitamin D and cancer risk, it is imperative to clear the stage of encumbrances that have retarded progress. This introduction highlights six key areas that have been sources of confusion: 1) the difference between vitamins and hormones, 2) the meaning of the term vitamin “deficiency”, 3) the need for an explicit understanding of which form of vitamin D pertains to what hypothesis, 4) the timing of vitamin D exposure with respect to cancer risk, 5) the mechanism(s) of action of vitamin D metabolites, and 6) the difference between disease prevention and therapy with vitamin D metabolites. Many of these topics have been the subjects of recent reviews. Vitamins are essential substances required for normal functioning that the body cannot manufacture and therefore must consume; an example is vitamin C, a deficiency of which causes the disease scurvy. Conversely, hormones are essential substances that are synthesized in one part of the body, transferred via the blood, and exert effects at distant sites. 1,25(OH)2D is the result of synthesis in the skin that begins when ultraviolet radiation converts 7-dehydrocholesterol to vitamin D3 (cholecalciferol). Vitamin D3 subsequently undergoes two hydroxylations, the first in the liver, forming 25-(OH)vitamin D, which has little biologic activity, and the second in the kidney and in extra-renal sites, forming 1,25(OH)2D, the active vitamin D metabolite. 1,25(OH)2D exerts numerous effects in the body, classically on bone and mineral homeostasis, and therefore it is a hormone. Conversely, because cholecalciferol is synthesized in the skin following ultraviolet radiation, it is not a vitamin. Arguably, “vitamin D” could be considered a vitamin among African Americans. This is because persons with darkly pigmented skin make much less vitamin D from the same amount of solar exposure as persons with lightly pigmented skin since melanin, the major pigment in human skin, absorbs ultraviolet light and inhibits vitamin D synthesis. Consequently, the prevalence of hypovitaminosis D in the United States is much higher among African Americans than among Caucasian Americans. Profound vitamin D deficiency during childhood causes the bone-deforming disease rickets. Consequently, vitamin D deficiency historically was defined in terms of bone; the absence of rickets in childhood or the absence of osteomalacia in adulthood was considered to represent vitamin D sufficiency. However, the recognition that many “non-classical” organs, such as the prostate gland, respond to and even synthesize the vitamin D hormone, mandates that the concept of vitamin D deficiency be revised. In short, vitamin D “deficiency” is most meaningful with respect to a specific end point: the amount of vitamin D sufficient to maintain a normal skeleton is likely insufficient to maintain the differentiated phenotype of prostate and other cells. Scientific acceptance of the hypothesis that vitamin D metabolites inhibit the development of cancer faced several obstacles. Chief among these was the difficulty in understanding how differences in serum levels of vitamin D metabolites could influence the behavior of “nonclassical” cells, even though such cells possessed receptors for 1,25(OH)2D (vitamin D receptors, or VDR). This is because, unlike serum levels of 25(OH)D, which are known to decrease with distance from the equator and are lower among persons with dark pigmentation, serum levels of 1,25(OH)2D are tightly regulated and do not vary with geographic latitude or race. This difficulty was effectively removed by the demonstration that normal prostate cells (and, by inference, other “non-classical” cells) possess the enzyme 25-(OH)vitamin D-1hydroxylase and, like the kidney, synthesize 1,25(OH)2D from circulating levels of 25(OH)D. The autocrine synthesis of 1,25(OH)2D was subsequently shown for many other organs, such as the colon, breast, and pancreas, and is Dr. Schwartz is with the Departments of Cancer Biology and Epidemiology and Prevention, Wake Forest University, Winston-Salem, North Carolina, USA. Please address all correspondence to: Dr. Gary Schwartz, Departments of Cancer Biology and Epidemiology and Prevention, Wake Forest University, Medical Center Blvd., Winston-Salem, NC 27157, USA; Phone: 336-716-7446; Fax: 336-716-5687; E-mail: [email protected]. doi: 10.1301/nr.2007.aug.S75–S76