Risk of endometrial cancer in women treated with ovary-stimulating drugs for subfertility

Abstract

Background: Medical treatment for subfertility principally involves the use of ovary-stimulating agents, including selective oestrogen receptor modulators (SERMs), such as clomiphene citrate, gonadotropins, gonadotropin-releasing hormone (GnRH) agonists and antagonists, as well as human chorionic gonadotropin. Ovary-stimulating drugs may act directly or indirectly upon the endometrium (lining of the womb). Nulliparity and some causes of subfertility are recognized as risk factors for endometrial cancer. Objectives: To evaluate the association between the use of ovary-stimulating drugs for the treatment of subfertility and the risk of endometrial cancer. Search methods: A search was performed in CENTRAL, MEDLINE (Ovid) and Embase (Ovid) databases up to July 2016, using a predefined search algorithm. A search in OpenGrey, ProQuest, ClinicalTrials.gov, ZETOC and reports of major conferences was also performed. We did not impose language and publication status restrictions. Selection criteria: Cohort and case-control studies reporting on the association between endometrial cancer and exposure to ovary-stimulating drugs for subfertility in adult women were deemed eligible. Data collection and analysis: Study characteristics and findings were extracted by review authors independently working in pairs. Inconsistency between studies was quantified by estimating I2. Random-effects (RE) models were used to calculate pooled effect estimates. Separate analyses were performed, comparing treated subfertile women versus general population and/or unexposed subfertile women, to address the superimposition of subfertility as an independent risk factor for endometrial cancer. Main results: Nineteen studies were eligible for inclusion (1,937,880 participants). Overall, the quality of evidence was very low, due to serious risk of bias and indirectness (non-randomised studies (NRS), which was reflected on the GRADE assessment. Six eligible studies, including subfertile women, without a general population control group, found that exposure to any ovary-stimulating drug was not associated with an increased risk of endometrial cancer (RR 0.96, 95% CI 0.67 to 1.37; 156,774 participants; very low quality evidence). Fifteen eligible studies, using a general population as the control group, found an increased risk after exposure to any ovary-stimulating drug (RR 1.75, 95% CI 1.18 to 2.61; 1,762,829 participants; very low quality evidence). Five eligible studies, confined to subfertile women (92,849 participants), reported on exposure to clomiphene citrate; the pooled studies indicated a positive association ( RR 1.32; 95% CI 1.01 to 1.71; 88,618 participants; very low quality evidence), although only at high dosage (RR 1.69, 95% CI 1.07 to 2.68; two studies; 12,073 participants) and at a high number of cycles (RR 1.69, 95% CI 1.16 to 2.47; three studies; 13,757 participants). Four studies found an increased risk of endometrial cancer in subfertile women who required clomiphene citrate compared to a general population control group (RR 1.87, 95% CI 1.00 to 3.48; four studies, 19,614 participants; very low quality evidence). These data do not tell us whether the association is due to the underlying conditions requiring clomiphene or the treatment itself. Using unexposed subfertile women as controls, exposure to gonadotropins was associated with an increased risk of endometrial cancer (RR 1.55, 95% CI 1.03 to 2.34; four studies; 17,769 participants; very low quality evidence). The respective analysis of two studies (1595 participants) versus the general population found no difference in risk (RR 2.12, 95% CI 0.79 to 5.64: very low quality evidence). Exposure to a combination of clomiphene citrate and gonadotropins, compared to unexposed subfertile women, produced no difference in risk of endometrial cancer (RR 1.18, 95% CI 0.57 to 2.44; two studies; 6345 participants; very low quality evidence). However, when compared to the general population, an increased risk was found , suggesting that the key factor might be subfertility, rather than treatment (RR 2.99, 95% CI 1.53 to 5.86; three studies; 7789 participants; very low quality evidence). Authors' conclusions: The synthesis of the currently available evidence does not allow us to draw robust conclusions, due to the very low quality of evidence. It seems that exposure to clomiphene citrate as an ovary-stimulating drug in subfertile women is associated with increased risk of endometrial cancer, especially at doses greater than 2000 mg and high (more than 7) number of cycles. This may largely be due to underlying risk factors in women who need treatment with clomiphene citrate, such as polycystic ovary syndrome, rather than exposure to the drug itself. The evidence regarding exposure to gonadotropins was inconclusive.