Large Populations and Rigorous Methods Needed for Vitamin D Research

Division of Cancer Epidemiology And Genetics

Vitamin D has been thought to potentially affect a person’s cancer risk, and in laboratory studies, researchers have shown that it can decrease cancer cell growth, promote cancer cell death, and affect other cellular processes. However, studies examining the relationship between blood levels of vitamin D in humans and the risk of many types of cancer have proved largely inconclusive. “At this stage, we don’t really have a clear idea if vitamin D is protective against cancer, and if so against which cancers, or at what stage in a cancer’s development,” explained D. Michal Freedman, J.D., M.P.H., Ph.D., an epidemiologist in NCI’s Division of Cancer Epidemiology and Genetics.

Dr. Freedman and her colleagues from NCI and the Centers for Disease Control and Prevention (CDC) recently analyzed blood samples taken from over 16,000 participants in the Third National Health and Nutrition Examination Survey* (NHANES III) between 1988 and 1994. At the time of the researchers’ analysis, the survey participants had been followed through 2000.

During that period, 536 NHANES III participants died of cancer. Although the researchers did not find any association between blood levels of vitamin D and total cancer mortality, they did find a reduced risk of death from colorectal cancer in participants with higher levels of circulating vitamin D. NCI and CDC investigators have continued to follow the NHANES III participants and are using additional data collected on cancer mortality. “With larger numbers, and more cancer deaths, we have opportunities to look at risks by sex, within racial and ethnic groups, and for less common cancer sites,” she explained.

Another aspect of Dr. Freedman’s work is focusing on how the methods of measuring vitamin D levels may affect the results of epidemiologic studies. In recent research**, she and her colleagues examined whether a delay between blood collection and processing could alter the end results for vitamin D measurement. Fortunately, they found that this was not the case and that variations in the collection process did not affect the measurements.

Currently, they are examining the degree of correlation between levels of vitamin D in blood samples taken during one season with those found in samples collected during other seasons (i.e., winter versus summer)***. “There’s an assumption that a single measurement is a good surrogate for usual or long-term circulating levels of vitamin D, and a single measurement is what’s been used in most studies,” explained Dr. Freedman. “We want to examine this assumption, because our study could shed light on the strength of the evidence provided by the single-sample studies.”