A paper Tuesday in JAMA reports long-term follow-up from the SELECT trial. The SELECT trial (Selenium and Vitamin E Cancer Prevention Trial) was a trial in which over 35,000 men were randomized to selenium, vitamin E, both or neither in order to determine if either supplement could reduce the incidence of prostate cancer. The initial report showed no benefit, but now with further follow-up, the investigators report a 17 percent increase in the incidence of prostate cancer for the group that received vitamin E.
To put this a bit in context, it is useful to recognize that no one has a clue as to the biological mechanism for this, if it is true. Furthermore, it is useful to look back at a landmark study, the ATBC study (Alpha-Tocopherol, Beta-Carotene), which randomized heavy smokers in a similar fashion to either vitamin E (alpha-tocopherol) or carotene to test whether either vitamin could prevent lung cancer. In that trial (published in 1994) there was a similar uproar when an unexpected increase in lung cancer death occurred in the group that received beta carotene, with no effect by vitamin E. Interestingly, as a secondary analysis, there was a marked reduction in prostate cancer incidence among those who received vitamin E in that study, which was one of the reasons for the current SELECT trial, to confirm the beneficial effects of vitamin E on prostate cancer.
A broader perspective may be necessary as well to appreciate the vitamin supplement issue. Most of our interest or belief in vitamins stems from observational studies in which people are asked about their intake of diet (like fruits and vegetables) or of vitamin supplements, and then their past or subsequent development of cancer is correlated statistically with their answers. Such studies are fraught with all sorts of potential biases and errors, and one must have a jaded eye when weighing their results. For example, someone who eats a healthful diet may also engage in many other healthy lifestyle activities, and thus it makes it difficult to separate out the real beneficial factor if cancer incidence is low. As a result, a randomized trial, such as the SELECT trial or the ATBC study, is essential to be certain about the truth of an association.
As a result, despite their difficulty and the cost required to undertake them, quite a number of such randomized trials have been conducted for a variety of supplements and a number of cancers. In brief, one would have to conclude that almost none have shown a significant benefit for any cancer. In other words, once a person is not vitamin-deficient, there is really very little evidence to support that vitamin or mineral supplementation beyond that has any dramatic beneficial impact on cancer risk. If anything, as the studies are starting to suggest, it may be more harmful than beneficial.
The question I would ask the reader is, why are we surprised? Whatever suggested to us that high or mega doses of natural products would make us healthier or prevent cancer or other illnesses? For most chemicals and metabolites, the body is extraordinarily well-regulated. In fact, most of the hormones, physiology, and bodily functions are precisely designed to keep our critical blood levels and body functions within very narrow bounds of "normality." Our temperature does not stray very far from 98.6 degrees, nor our serum albumin from 4.1, nor our blood glucose from 100 if we are healthy. Thus, the concept that, if we are healthy, bombarding our body with high doses of mineral or vitamins should be beneficial does not sound on the face like a logical hypothesis.
Sounds to me like the science is starting to bear that out.