SAN ANTONIO — Ovarian stimulation for in vitro fertilization was linked to an increased risk of ovarian cancer 15 years later in a large cohort study that followed thousands of women in the Netherlands.
Compared with controls who had fertility problems but did not undergo in vitro fertilization (IVF), women who underwent IVF were more than four times as likely to develop “borderline” tumors and 1.5 times more likely to develop invasive ovarian cancer. Overall, IVF conferred a relative risk of 2.05 for all ovarian malignancies.
The “borderline” tumors, also known as low-malignant-potential tumors, tended to occur earlier than the invasive ovarian cancers—for which an increase in incidence did not become apparent until 15 years after treatment, Dr. Curt W. Burger reported at the Society of Gynecologic Oncologists' annual meeting.
Whether borderline tumors eventually become invasive is subject to debate, noted Dr. Burger, a gynecologist at Erasmus University Medical Center in Rotterdam, the Netherlands.
“The clinical implications are modest,” he said, estimating the cumulative individual risk of developing an ovarian tumor before age 55 years as 0.45% for the general population and 0.71% for women who have undergone IVF.
Dr. Wendy R. Brewster of the University of North Carolina, Chapel Hill, called the results “quite troubling” in a discussion of the study.
Both Dr. Brewster and Dr. Burger reviewed a long line of studies that failed to prove increased incidence of ovarian cancer after ovarian stimulation. Among the earlier reports were two by Dr. Burger, based on shorter follow-up.
All 12 IVF centers in the Netherlands participated in the study. The initial cohort comprised 18,970 women who received IVF treatment between 1983 and 1995, and a control group of 7,536 subfertile women who sought help but were not treated with IVF.
About two-thirds of the women—67% of the total population and 74% of the IVF group—responded to questionnaires on reproductive risk factors between 1997 and 1999. The investigators reviewed their medical records and, with written permission, followed their cancer diagnoses through linkage with the Netherlands Cancer Registry through 2007.
At a median follow-up of 14.7 years, he reported 61 ovarian cancers were observed in the IVF group and 16 in the control group versus expectations of 38.4 and 15.6, respectively, in those populations. The standardized incidence ratio (SIR) for the IVF group was 1.59.
In the IVF group the SIR for invasive cancers peaked in the first year, probably because of screening after IVF, and at or after 15 years (3.94 and 3.22, respectively). Borderline tumors also showed a peak the first year, and most were found within 10 years of treatment; between years 5 and 9 after treatment, 12 were found (SIR 2.18).
All told, 55 ovarian cancers (SIR 1.49) were found after the first year in the IVF group: 28 invasive cancers (SIR 1.30) and 27 borderline tumors (SIR 1.76).
Dr. Glenn L. Schattman, chairperson of the Practice Committee of the Society for Assisted Reproductive Technology, affiliate of the American Society for Reproductive Medicine, called the study interesting but noted that it 'does not take into account whether the IVF patients were successful in achieving a pregnancy or what their previous pregnancy histories and ovarian cancer risk factors were. It also does not give the dosages of the stimulant drugs they took. It was a retrospective study, and such studies have limitations.”