The largely unregulated "nutritional supplements" industry, which is expected to reach $11 billion dollars in sales in the US this year, has an enormous stake in any research related to the effects of vitamins on health. This huge industry has, understandably, seized upon every scientific study showing potential health benefits associated with vitamins and other dietary supplements, while ignoring numerous studies that have failed to identify any such benefits. Caught in the middle between the "nutritional supplements" industry (an oxymoronic term, if there ever was one) and a large body of contradictory scientific research, the American consumer is basically left to figure out the indications for taking these supplements on his or her own.

Most of the studies looking at the disease prevention effects of vitamins and other dietary supplements rely upon either dietary surveys or randomized trials in which volunteers take either supplement pills or placebo pills. The dietary survey method is more common than the randomized, prospective, double-blinded method of study, and for several reasons. The dietary survey method is a simple and inexpensive means of studying the impact of dietary factor on public health, and enables researchers to efficiently study large numbers of people. Unfortunately, studies that rely upon dietary surveys are less powerful than prospective randomized studies, as dietary surveys rely upon the memories of study participants, and so the retrospective data derived from this type of study is far more subjective than prospectively acquired data. An additional problem with the dietary survey model of research is that confounding factors among the study participants might not be uncovered by the research team managing the study, resulting in skewed data that does not actually explain the effects of diet, or dietary supplements, on the disease processes being evaluated. For example, if a group of study participants is divided into sub-groups defined by the level of their self-reported intake of, say, vitamin C, the research team will then statistically analyze the incidence of certain diseases in this group of volunteers as a function of their vitamin C intake. But if the research team fails to recognize, for example, that the study participants with the highest self-reported intake of vitamin C also exercise more regularly than the people who consume the lowest levels of vitamin C, an attempt to correlate a reduction in cardiovascular disease incidence with high levels of dietary vitamin C might be completely erroneous. In this example, differences in exercise habits between volunteers with the highest levels of vitamin C in their diets and those with the lowest levels would be a confounding factor. In my example, I have used a glaringly obvious potential confounding factor, exercise levels, to illustrate my point. However, the interactions of various dietary factors, genetic predispositions, lifestyle behaviors, the side effects of medications, gender, age, environmental factors, emotional and mental health issues, and numerous other simple and complex factors (and all of them potentially interacting with each other at multiple different levels), make it extremely difficult for researchers to identify and correct for every conceivable confounding factor. On the other hand, matching study participants as carefully as possible prior to beginning a new study, and then giving them either a nutritional supplement or a placebo pill (and not allowing either the research team or the study volunteers to know which pill they are actually receiving), bypasses many of the potential weaknesses of retrospective dietary survey studies.

When I review studies that look at the potential health impact of a specific dietary supplement, I am looking for simplicity, power, and accuracy. The more focused the study is on a specific endpoint (for example, what is the effect of garlic supplementation, within a controlled prospective study, on blood cholesterol levels?), the more likely that the study's conclusions are going to be accurate. A new study in the Journal of the National Cancer Institute has piqued my interest, particularly in view of the very contradictory scientific evidence available that vitamin E might reduce the risk of certain cancers. Up front, one potential limitation of this study is that it uses a mouse model instead of a human model, and we know that what works in a mouse doesn't always work in a human being. Having made this disclaimer, this study, at a minimum, demonstrates a biochemical mechanism whereby a specific form of vitamin E might actually reduce the risk of tumor formation and progression.

In this study, an established and validated test for detecting DNA mutations was used to assess the impact of two different chemical forms of vitamin E on the presence of two tumor-associated DNA mutations in mice with experimentally induced tumors. Two different groups of tumor-bearing mice were fed diets high in either alpha-tocopherol or gamma-tocopherol, two different forms of vitamin E that are found in natural sources of this fat soluble vitamin. A third group of mice received normal mice chow, without vitamin E supplementation. The authors found that the tumors in the mice fed high levels of alpha-tocopherol, but not gamma tocopherol, underwent a significant reduction in the expression of two mutated genes associated with genetic instability in their tumors. Interestingly, the 28 mice that were fed 50 mg or less per kg of body weight of alpha-tocopherol per day all experienced a significant reduction in the function of one of the two mutated tumor-associated genes that were studied. Among the 18 mice that received 100 mg/kg body weight of alpha-tocopherol per day, 7 (39%) had evidence of suppressed function of this same tumor-associated mutant gene (these results, in both groups of mice, were based upon comparison with the third group of mice who did not receive any dietary supplementation with any form of vitamin E). Although the small numbers of mice in each group may limit conclusions regarding a dose-response effect, these results at least suggest that there may be an optimal dose of supplemental alpha-tocopherol necessary to shutdown the function of, at least, this one tumor-associated mutant gene (at least in mice…).

This is a very intriguing study, and demonstrates a specific biochemical mechanism whereby alpha-tocopherol is able to reduce, or down-regulate, the function of mutant genes associated with genetic instability in tumors. Of course, this same mechanism of alpha-tocopherol anti-tumor function must be demonstrated in humans before the results of this study can be generalized to human beings. However, this elegant and focused research study provides powerful evidence that at least one form of the vitamin E that we consume from natural dietary sources can, at least in mice, reverse the function of tumor-associated mutant genes.

BRIEFLY…

Cancer Journal: Previous studies have shown contradictory effects of the popular cholesterol-reducing statin drugs on the risk of developing cancer. In an age-matched study of 975 women with a history of breast cancer and 1,007 women without breast cancer, current or prior use of statin drugs was not associated with an increased risk of developing breast cancer. Among women who used statin drugs for more than 5 years, there was actually a 30% reduction in the relative risk of developing breast cancer when compared to women who had never taken statin drugs.

Cancer Journal: A total of 23,618 postmenopausal Danish women were followed for an average of 4.8 years, and the incidence of breast cancer in this group of women was correlated with hormone replacement therapy (HRT) use. Overall, the women who reported prior or current HRT use experienced a 222% increase in the relative risk of developing breast cancer when compared to women who reported no current or prior HRT use. The women who reported continuous long-term HRT use, not surprisingly, had the highest risk of developing breast cancer, and breast tumors sensitive to estrogen stimulation in particular. More evidence that chronic HRT is associated with a significantly increased risk of breast cancer….

Journal of Urology: The relationship, if any, between attention deficit/hyperactivity disorder (ADHD) and nighttime bed-wetting (enuresis) has not previously been well studied. A total of 120 children, age 6 to 12 years, with a history of enuresis were evaluated. Among these 120 children, 15% were diagnosed with the full ADHD syndrome, while 22.5% met the diagnostic criteria for the ADHD inattentive subtype (i.e., attention deficit disorder, but without hyperactivity). This is much higher than the generally accepted incidence of ADHD in the general population, which is estimated to be 3 to 5%. Moreover, the study determined that the older children (those 9 to 12 years of age) with enuresis in this study were more likely than the younger children to have ADHD.