Antioxidants and cancer

Cancer and Society Quackery Antioxidants and cancer For the Oxford University review see Br J Cancer 2011; 104: 6–11 John Heseltine/Science Photo Library For more on the ViMiS database see http://journals.cambridge. org/action/displayAbstract?from Page=online&aid=8093819 One of biology’s paradoxes is that we require oxygen for life, but it can be a highly reactive molecule that can damage cells through the production of reactive oxygen species (ROS). Plants and animals thus keep stores of many types of antioxidants, such as glutathione, vitamins C and E, and specific enzymes, which have generally been thought to either prevent ROS being made or remove them before they can cause harm. However, because ROS also have useful functions, including natural host defence against cancer cells, we should know which antioxidants might be useful, how much one should take, or whether they might cause harm. Many so-called superfoods originate from Chinese and Ayurvedic systems of medicine and have been used medicinally for thousands of years. Although their benefits are not evidence-based in the western sense, such long-term use lends credence to the idea of superfoods for those fashionable and health-conscious individuals searching for an elixir of life. Since research in the 1970s suggested vitamin C was an effective cancer therapeutic, superfoods that include antioxidants have spawned an entire industry aimed at mopping up or squelching these supposedly harmful ROS. This trend has been met with an Antioxidant-containing superfoods have spawned an entire industry 996 apathetic response from the scientific community. After all, how much harm can antioxidants do? Human studies examining the oftenrepeated inverse connection between fresh produce and cancer have led many to recommend an intake of five servings of fruit or vegetables per day, but even this benign idea is questionable— findings from a review done at Oxford University showed no firm benefits. Genetic variation means that no two fruits or vegetables contain the same amount of antioxidants and our own genes and gut flora will in turn affect our ability to absorb them. A solution, then, has been to administer known amounts of a variety of these compounds as a supplement. However, once again, not only have many trials shown equivocal results, but the ATBC, CARET, and SELECT studies were all stopped early because of the emergence of data implicating high concentrations of vitamins A, C, and E, and β-carotene as potentially promoting carcinogenesis. The development of the Vitamin and Mineral Supplement (ViMiS) database at Cambridge University, which includes 2066 supplements and 16 586 ingredients, is thus likely to be very useful in the identification of effective antioxidant preventatives and the establishment of their potential toxic effects or benefits. A reasonable inference is that cancer and its risk is regulated by a balance between specific toxins and the proteins and processes that control their concentrations. Thus, under normal physiological circumstances, ROS concentrations are controlled by a feedback loop in which the inputs include cellular stressors and the output includes activation of pathways inside cells, which in turn controls the activity of genes that produce proteins implicated in detoxification and elimination of ROS through direct binding and mopping mechanisms and actions. Research published in the past few weeks shows that cancercausing genes themselves activate an antioxidant pathway via a range of mechanisms, and that production of ROS is actually part of an anticancer host defence. It’s a balance between having some ROS to ward off cancer cells but not too many to cause damage to normal tissues. Some of the newest, most exciting bench-to-bedside targeted anticancer therapeutics specifically inhibit the activity of genes that act to reduce ROS concentrations, so by keeping these switched off, ROS concentrations remain high. This can subvert cancer cells’ ability to survive when nutrients and oxygen are scarce, especially when the cancer is just starting to develop. In the same way drug studies are now being personalised, and directed to the right cancer in the right person at the right time, studies are needed to establish whether antioxidants can change cancer risk or the outcome of cancer, depending on a person’s genomic or indeed tumorigenic fingerprint. Until these data are mature, aiming to kill cancer cells by oxidative stress while concomitantly reducing ROS is a flawed and populist approach, and that’s what the evidence suggests thus far. The specific antioxidants present in dietary or supplemental form that might be beneficial or harmful is unclear, so although an increase in fresh fruit, vitamin, or supplement consumption seems to do no harm, we still await supportive evidence. To take antioxidants out of context might be more dangerous than we realise. Intelligence is recognising that a tomato is a fruit. Wisdom is knowing not to put one in a fruit salad. *Justin Stebbing, Colin A Hart Imperial College, London, UK (JS) and University of Leicester, UK (CAH) j.stebbing@imperial.ac.uk www.thelancet.com/oncology Vol 12 October 2011