Strengths and Limitations of Current Epidemiologic Studies: Vitamin D as a Modifier of Colon and Prostate Cancer Risk

Abstract

The existence of an association between latitude and cancer mortality rate had been known since the 1930s, but it was in 1980 that Garland et al. first hypothesized that the potential benefit of sun exposure was due to vitamin D. Initially, the hypothesis focused on colon cancer, but it was later extended to other types of cancer including prostate cancer. Many cell types, normal as well as neoplastic, are now known to express vitamin D receptors and 1-hydroxylase, which can convert 25(OH)-vitamin D, abbreviated 25(OH)D, to 1,25(OH)2 vitamin D, abbreviated 1,25(OH)2D, which is the natural ligand of the vitamin D receptor. Activation of the vitamin D receptor induces or inhibits transcription of a number of genes that influence proliferation, invasiveness, angiogenesis, metastatic potential, differentiation, and apoptosis. The autocrine or perhaps paracrine influences of 1,25(OH)2D acting through these genes could potentially help to retard cancer causation or progression in some tissues. The hypothesis that high circulating levels of 25(OH)D or the two sources of vitamin D, sun exposure and intake, are associated with lower risk of cancer has been examined in epidemiologic studies. A brief summary of results for colorectal and prostate cancer is provided here. The association between latitude as a surrogate of sun exposure and vitamin D level, first observed in the United States, has now been confirmed in diverse populations such as in Japan. The circulating 25(OH)D level accounts for all sources of vitamin D, plus conversion of vitamin D into 25(OH)D. Circulating 25(OH)D has a relatively long half-life of about 2 to 3 weeks and thus can provide a fairly stable indicator of long-term vitamin D status. Studies that have examined 25(OH)D levels prospectively in relation to risk of colorectal cancer have generally supported an inverse association. In a recent meta-analysis of these studies, individuals with serum 25(OH)D levels equal to or greater than 33 ng/mL (82 nmol/L) had a 50% lower incidence of colorectal cancer than those with relatively low levels ( 12 ng/mL or 30 nmol/L). The combined number of colorectal cancer cases in the meta-analysis was 535, and the results were highly statistically significant. Studies that have examined vitamin D intake in relation to colorectal cancer or adenoma risk have also tended to show an inverse association, which, as expected, is smaller in magnitude than that estimated from studies considering 25(OH)D levels. The studies of vitamin D intake tend to support a protective effect of 25(OH)D, but a limitation of these studies is that a confounding effect of calcium intake could not be definitively excluded because of the fairly high correlation between calcium and vitamin D intakes in the United States (where milk is fortified with vitamin D). The results for prostate cancer are less clear than those for colorectal cancer. Interestingly, studies that have examined sun exposure have generally been supportive of an inverse association. For example, in a case-control study of prostate cancer conducted in the United Kingdom, where vitamin D deficiency is relatively common, regular foreign holidays in sunny climates, a higher sunbathing score, and higher exposure to solar radiation were strongly associated with a reduced risk. Because of the retrospective and subjective assessment of exposure, recall bias is a possibility in this study. However, studies using surrogate measures of recent past or long-term sun exposure, such as prior non-melanoma skin cancer or reflectometry, have also suggested that exposure to sun is associated with a lower risk of prostate cancer. In contrast to the studies that have examined sun exposure surrogates, studies that have examined circulating 1,25(OH)2D or 25(OH)D levels have yielded inconsistent results. Further, although dietary studies have been limited in number, they tend not to support an association between vitamin D intake and lower prostate cancer risk. Dr. Giovannucci is with the Departments of Nutrition and Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA, and the Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston. Please address all correspondence to: Dr. Edward Giovannucci, Department of Nutrition, Harvard School of Public Health, 665 Huntington Ave., Boston, MA 02115, USA; Phone: 617-432-4648; Fax: 617-4322435; E-mail: egiovann@hsph.harvard.edu. doi: 10.1301/nr.2007.aug.S77–S79