Controlled trials and chemoprevention of cancer

Epidemiological and laboratory studies provide preliminary evidence that a compound may prevent certain types of clinical cancer. The final proof for practical application demands two controlled trials with similar, decisive results. Controlled chemoprevention trials on clinical cancer are large, time-consuming and expensive, whereas studies on cancer surrogates are smaller but less reliable. Rational trial design often lacks sufficient information about the sensitive period and the time from that point to clinically delectable cancer. The correct dose of chemopreventive agent and an expected preventive fraction of cancer are also often based on informed guesswork. Long trials call for special arrangements to guarantee the staying will of the participants and key research personnel. Although large chemoprevention trials are currently being carried out without any certainty of successful outcome, the situation is not so different from the early days of chemoprevention trials for cardiovascular diseases. Cancer trials will be conducted based on the 'learning-by-doing' approach, and in the more distant future based on research designed to provide information for trial needs.