Mutations in the p53 tumour suppressor gene lead to uncontrolled cell division and are found in over 50% of all human tumours, including 60% of lung cancers. Research published in 1996 by Denissenko and colleagues demonstrated patterned in-vitro mutagenic effects on p53 of benzo[a]pyrene, a carcinogen present in tobacco smoke. We investigated the tobacco industry's response to p53 research linking smoking to cancer. We searched online tobacco document archives, including the Legacy Tobacco Documents Library and Tobacco Documents Online, and archives maintained by tobacco companies such as Philip Morris and R J Reynolds. Documents were also obtained from the British American Tobacco Company depository in Guildford, UK. Informal correspondence was carried out with scientists, lawyers, and tobacco control experts in the USA and Europe. We found that executives and scientists at the highest levels of the tobacco industry anticipated and carefully monitored p53 research. The tobacco industry's own scientists conducted research which appeared to cast doubt on the link between smoking and p53 mutations. Researchers and a journal editor with tobacco industry ties participated in the publication of this research in a peer-reviewed journal without clear disclosure of their tobacco industry links. Tobacco industry responses to research linking smoking to carcinogenic p53 mutations mirror prior industry efforts to challenge the science linking smoking and lung cancer. The extent of tobacco industry involvement in p53 research and the potential conflict of interest discussed here demonstrate the need for consistent standards for the disclosure and evaluation of such potential conflicts in biomedical research.