Long-term use of non-steroidal anti-inflammatory drugs (NSAIDs) can be regarded as an effective approach for cancer chemoprevention, as demonstrated by a bulk of clinical and experimental evidence. However, the clinical use of these drugs as chemopreventive agents is limited by many open questions about the optimal drug, dose, duration of therapy and knowledge about the mechanism(s) by which these drugs act. In particular, the recent data on cardiovascular toxicity of coxibs has posed some limitations on the use of NSAIDs for cancer chemoprevention in the general population. The situation is different in certain genetically susceptible subgroups, such as in individuals with genetic mutations associated with hereditary nonpolyposis colon cancer (HNPCC) or familiar adenomatous polyps (FAP) in whom lifetime risk increases up to 70-90% and in whom the benefit of a chemopreventive drug might justify its use even in the presence of adverse effects.