BACKGROUND: Adjuvant radiation therapy for breast cancer has been related to excess esophageal cancer risk, but population-based data are scanty. PATIENTS AND METHODS: We considered esophageal cancer risk among 11 130 breast cancer patients diagnosed between 1974 and 2002 in the Swiss cantons of Vaud and Neuchâtel, and followed-up to the end of 2002, for a total of 75 900 women-years at risk. RESULTS: Overall, 18 cases were observed compared with 8.9 expected, corresponding to a standardised incidence ratio (SIR) of 2.0 [95% confidence interval (CI) 1.2-3.2]. The SIR was 1.6 in the first 10 years after diagnosis and 3.3 for >/=10 years after diagnosis, 2.3 for cases diagnosed between 1974 and 1988 and 1.5 for those diagnosed after 1988, 2.3 (based on 15 cases) for squamous cell cancer and 1.3 (based on three cases) for adenocarcinomas, and 2.9 for the upper third, 2.3 for the middle third and 1.9 for the lower third of the esophagus. CONCLUSIONS: These data confirm an excess esophageal cancer risk following treatment for breast cancer which could not be explained by confounding of tobacco or alcohol alone. The excess risk tended to decrease for cases diagnosed after 1988, leaving open the issue of the risk of modern radiotherapy for breast cancer on esophageal cancer.