Concern has been expressed that exposure to fertility drugs might be associated with a risk of ovarian cancer. We have examined the incidence of breast and ovarian cancer in a cohort of 10,358 women referred for in-vitro fertilisation (IVF) treatment in Victoria, Australia, between 1978 and 1992. The "exposed" group (n = 5564) had had ovarian stimulation to induce multiple folliculogenesis and the "unexposed" group (n = 4794) had been referred for IVF but were untreated or had had "natural cycle" treatment without ovarian stimulation. Duration of follow-up ranged from 1 to 15 years. Cases of cancer were determined by record linkage with data from population-based cancer registries. 34 cases of invasive breast cancer and 6 of invasive ovarian cancer were observed. A comparison with the expected numbers, derived by applying age-standardised general population rates to the cohort gave standardised incidence ratios (SIR) for breast cancer of 0.89 (95% CI 0.55-1.46) in the exposed group and 0.98 (0.62-1.56) in the unexposed group, and for ovarian cancer SIRs were 1.70 (0.55-5.27) and 1.62 (0.52-5.02), respectively. Rates of all cancers were not significantly different from general population rates. The relative risk (RR) of cancer, adjusted for age and infertility type, was, in the treated group compared with the untreated group, 1.11 (95% CI 0.56-2.20) for breast cancer and 1.45 (0.28-7.55) for ovarian cancer. The risk of body of uterus cancer was increased in the exposed and unexposed groups combined (SIR 2.84 [1.18-6.81]). Women with unexplained infertility, independent of IVF exposure, had significantly increased risks of ovarian cancer (RR = 19.19 [2.23-165.0]) and body of uterus cancer (RR = 6.34 [1.06-38.0]) compared with women with known causes of infertility. This relatively short-term follow-up suggests that ovarian stimulation with IVF is not associated with an increased risk of breast cancer. Although there was no significantly increased risk of ovarian cancer after ovarian stimulation with IVF the small number of cases limits the conclusions that can be drawn. Longer-term follow-up of large cohorts of women who have been in IVF programmes will be necessary.