Although many new targeted drugs have made it as far as the clinic and hold promise for the treatment of people already diagnosed with cancer, preventing the development of cancer makes both economic and clinical sense. Early detection, where possible, is certainly a valid approach, but many chemopreventative agents have not proven their worth. Why is this? Epidemiological evidence has guided us towards quitting smoking, exercising, losing weight and eating 5 portions of fruit and vegetables a day in order to reduce our chances of developing cancer. But epidemiology alone is not always a good enough guide to judge the use of chemopreventative agents, including those that are contained in our daily fruit and vegetable intake. This, and related aspects, are reviewed on page 867, by Eva Szabo, who asks: where next for cancer chemoprevention trials?

Although some trials have shown the efficacy of chemopreventative agents in high-risk individuals (for example, the use of tamoxifen to prevent breast cancer), many others have shown no benefit and a few have even identified harm (the acceleration of lung cancer incidence in smokers that took β-carotene). Other agents seem to be effective, but their side effects look likely to restrict their use in all but the highest risk patients. We need, argues Szabo, a much clearer biological and clinical insight into how a potential chemopreventative agent works before we attempt to evaluate its effect on suppressing tumour development.

Will we ever find chemopreventative agents that are available to all, irrespective of cancer risk? No seems the most likely answer. Any side effects will always be a barrier to use in an otherwise healthy individual, but with careful planning we should be able to identify chemicals that reduce the risk of cancer development in patients with precancerous lesions.