To study the late health effects of nasopharyngeal radium irradiation, a nonconcurrent prospective study was conducted in The Netherlands. Exposed subjects (n = 2510) were treated unilaterally with nasopharyngeal radium irradiation at five ear, nose, and throat clinics between 1945 and 1965 and followed up through February 1, 1985. Unexposed subjects (n = 2199) were matched to treated patients according to clinic, year of birth, and first consultation within 5 years. No overall significant increase in cancer mortality was found. A statistically significant difference was noted only for cumulative all-site cancer incidence, with the exposed having twice as many verified cancers as the nonexposed. This excess risk was caused mainly by head and neck tumors and other tumors (lung, digestive tract, and urogenital system). The relative risk for these specific sites, however, was not significantly different from 1.00. These findings corroborate the hypothesis that there is some risk of tumor induction involved with nasopharyngeal radium irradiation at the dose level applied to this population. Because of ongoing interest in this subject, I plan to extend the follow-up of the cohort through December 31, 1995, and I am studying the possibility of expanding the study population.