Fertility Treatment Increases Risk of Ovarian Cancer

Flora E. van Leeuwen, MD

Subfertile women who undergo ovarian stimulation for in vitro fertilization (IVF) have an increased risk of ovarian malignancies, according to the results of a large nationwide study by a Dutch team.

Flora E. van Leeuwen, MD, with the Netherlands Cancer Institute in Amsterdam, and colleagues examined data on 19,146 subfertile women who underwent at least one IVF ovarian stimulation treatment between 1980 and 1985 and 6006 subfertile women who did not undergo IVF.

Currently, 1.2% to 2.3% of children born in Western countries are conceived using assisted reproductive technologies. In recent years, multiple studies have examined the relationship between ovarian stimulation with IVF treatment and ovarian cancer risk. However, results are unreliable because of short follow-up, low statistical power, and a failure to control for important risk factors such as subfertility and parity.

For their study, the investigators obtained data on reproductive factors from the study participants and detailed information on subfertility cause and treatment from medical records, and determined the incidence of ovarian malignancies through 2007 through linkage with disease registries. The median duration of follow-up was 14.7 years.

Compared with the general population rates, there was a significantly increased risk for borderline ovarian tumors in the IVF group (standardized incidence ratio [SIR], 1.93; 95% CI, 1.31-2.73).

The overall SIR for invasive ovarian cancer was not elevated, but increased with longer follow-up after first IVF (P =.02). After 15 or more years, the SIR for invasive ovarian cancer in the IVF group was 3.54 (95% CI, 1.62-6.72; P for trend =.02).

After controlling for age, parity, and the cause of subfertility, the risks of borderline ovarian tumors and of all ovarian malignancies combined in the IVF cohort (hazard ratio [HR], 4.23; 95% CI, 1.25-14.33) were significantly raised compared with risks in the subfertile cohort (HR, 2.14; 95% CI, 1.07-4.25).

Van Leeuwen and associates emphasized that while their results “give reason for some concern,” they are tempered by the fact that they are “based on rather small numbers.” Also, no dose-response relationship was found, and the increased risk for invasive ovarian cancer was not statistically significant in multivariate analyses.

Accordingly, they said that larger prospective cohort studies of IVF-treated women are needed with longer follow-up and a subfertile comparator group not treated with IVF to lend more insight into the relationship between ovarian stimulation for IVF treatment and ovarian malignancies. Dose-response analyses with more statistical power would also be helpful.