Selenium and the Limits of Lumping All Antioxidants Together

Dear Editor, outcomes-so the label “antioxidant” spans agents that harm, agents that do nothing, and agents that help, depending on the endpoint. Reviews and meta-analyses frequently pool diverse compounds under the single heading of “antioxidants,” as if membership in that category predicted a common biological effect. I write to argue, using selenium as a case in point, that this lumping obscures more than it reveals. Selenium illustrates the point sharply. According to PubMed, although the epidemiological data are described as somewhat controversial, numerous studies suggest selenium compounds can negatively affect cancer progression, and in vitro and animal work has linked selenium supplementation to reduced cell migration, invasion, angiogenic factors, and metastasis in several cancer types, prompting proposals that selenium may act as an anti-metastatic as well as a preventive agent. This is a very different profile from that of high-dose Vitamin E. The divergence within the category is substantial. According to PubMed, while high-dose vitamin E has been associated with increased all-cause mortality, and a large meta-analysis found that beta-carotene supplementation increased mortality and stroke risk, the same analysis reported that omega-3 fatty acids, folic acid, and coenzyme Q10 conferred benefits on specific