In vitro fertilization is associated with an increased risk of borderline ovarian tumours

Abstract

Objectives

To compare the risk of borderline ovarian tumours in women having in vitro fertilization (IVF) with women diagnosed with infertility but not having IVF.

Methods

This was a whole-population cohort study of women aged 20–44 years seeking hospital infertility treatment or investigation in Western Australia in 1982–2002. Using Cox regression, we examined the effects of IVF treatment and potential confounders on the rate of borderline ovarian tumours. Potential confounders included parity, age, calendar year, socio-economic status, infertility diagnoses including pelvic inflammatory disorders and endometriosis and surgical procedures including hysterectomy and tubal ligation.

Results

Women undergoing IVF had an increased rate of borderline ovarian tumours with a hazard ratio (HR) of 2.46 (95% confidence interval [CI] 1.20–5.04). Unlike invasive epithelial ovarian cancer, neither birth (HR 0.89; 95% CI 0.43–1.88) nor hysterectomy (1.02; 0.24–4.37) nor sterilization (1.48; 0.63–3.48) appeared protective and the rate was not increased in women with a diagnosis of endometriosis (HR 0.31; 95% CI 0.04–2.29).

Conclusions

Women undergoing IVF treatment are at increased risk of being diagnosed with borderline ovarian tumours. Risk factors for borderline ovarian tumours appear different from those for invasive ovarian cancer.

Introduction

Borderline epithelial ovarian tumours are a heterogeneous group of neoplasms, first described in 1929 [1]. For some time they were believed to be precursors of invasive epithelial ovarian cancer, but they have been gradually recognised as a separate entity and were classified as such by the International Federation of Gynecology and Obstetrics (FIGO) in 1970 [2]. Unlike invasive epithelial ovarian cancer, borderline epithelial ovarian tumours, also known as tumours of low malignant potential, have an indolent disposition, do not destructively invade the underlying ovarian stroma [3], are more likely to be diagnosed in women of reproductive age [4], and have a favourable prognosis, with more than 95% of women surviving five years beyond diagnosis [5]. They represent around 15% of all ovarian neoplasms [6].

Some authors maintain that risk factors for borderline ovarian tumours are the same as those for invasive epithelial ovarian cancer, but the evidence is scant and contradictory. For example, giving birth has been found to be protective in some studies [7], [8], possibly protective in another [9] and not protective in others [10], [11], [12]. Oral contraceptive use appears in some studies [9], [12] to confer protection, but not in others [7], [11]. Harris et al. [8] found a protective effect of sterilization; Mosgaard et al. [10] found no effect.

With regard to treatment with fertility drugs, most studies have found an increased risk of borderline ovarian tumours after fertility drug treatment [8], [12], [13], [14], [15], although some have not [10], [11]. The only study that focussed specifically on in vitro fertilization (IVF) found that IVF treatment was associated with a four-fold increase in the risk of borderline ovarian tumours [16].

Given the important role that IVF plays in the current management of infertility, we believe that the relationship between IVF and borderline ovarian tumours deserves further investigation. The aim of the present study was to examine the association between IVF treatment and risk of borderline ovarian tumours in a cohort of women seeking treatment for infertility, considering also the confounding effects of known or potential ovarian cancer risk factors including parity, infertility diagnosis, sterilization, hysterectomy and socio-economic status.