Cancer Risk After the Use of Ovulation-Stimulating Drugs

Abstract

This large retrospective study investigating the effect of ovulation-stimulating drugs on the subsequent risk of ovarian cancer included infertility patients from 5 large reproductive endocrinology centers in the years 1965 to 1988. Information was gathered from extensive review of medical records, follow-up data, patient questionnaires mailed in 1998 (10 years after the last year of inclusion), and cancer registries or similar sources such as the National Death Index. Clomiphene citrate (3277 patients) and human gonadotropins, including Pergonal (Serano, Rockland, MA), Humegon (Organon, West Orange, NJ), or Metrodin (Serono) (866 patients), were used for ovulation stimulation. The diagnoses of cancer were verified in discharge summaries and operative and pathology reports. The total number of infertility patients was 12,193. Figure 1 shows the procedure followed for data management. Over 80% of the patients had at least 15 years of follow up (median follow-up period 18.8 months). There were 45 cases of ovarian cancer diagnosed in the total number of infertility patients. Table 1 shows the risk of ovarian cancer in the study patients, with and without exposure to ovulation-inducing drugs, compared with the general population. The risk for infertility patients was 1.98 (95% confidence interval [CI], 1.4–2.6). There were no significant increases in the risk of ovarian cancer among patients who received either clomiphene or gonadotropins. There were also no significant differences in risk between patients who took clomiphene and those who took gonadotropins. Eleven women died from ovarian cancer (standardized mortality ratio [SMR], 1.94; 95% CI, 0.3–4.2). The rate of mortality did not differ between those who received ovulation-inducing drugs and those who did not. Extensive analysis of possible risk factors for ovarian cancer, including demographic data, causes of infertility, medical history, and histologic findings, with adjustment for infertility drug use, found only 1 significant predictive variable. Zero gravidity at the first clinic visit was the only factor with a significant influence on the development of ovarian cancer (42.3% of those without ovarian cancer compared with 57.8% of those who developed ovarian cancer; P <0.05). Use of clomiphene or gonadotrophins did not influence the development of ovarian cancer. There was also no effect seen when considering either total dosage or exposure to both clomiphene and gonadotropins. Except for zero gravidity, no subset of patients was seen that had an increased risk of ovarian cancer compared with the total study group.Fig. 1: Field and analytic status of eligible study subjects, women evaluated for infertility, 1965–1988. Reproduced with permission. Obstet Gynecol 2004;103:1194–1203. Copyright © 2004, Lippincott Williams & Wilkins. All rights reserved.TABLE 1: Standardized incidence ratios comparing ovarian cancer among infertile patients with the general population, overall and stratified by usage of clomiphene and gonadotropins