Optimisation of GnRH antagonist use in ART

Abstract

This thesis focuses on the optimisation of controlled ovarian stimulation for IVF using exogenous FSH and GnRH antagonist co-treatment, by studying the timing of the initiation of GnRH antagonist co-medication and the role of ovarian reserve markers in optimising ovarian response and reproductive outcome. In this thesis we describe the results of a multicentre randomised controlled trial (RCT) which studied the effect of early (cycle day (CD) 2) versus late (CD 6) GnRH antagonist initiation on the stimulation phase endocrine profile as well as the impact on live birth rate per started cycle (LBR) and the cumulative live birth rate (CLBR). Early initiation of GnRH antagonist treatment resulted in a more stable endocrine profile, with more physiological levels of estradiol and LH during the follicular phase. However, live birth rates did not differ significantly between the two groups, although a non-significant trend towards a higher LBR and CLBR was observed in the late start group compared with the early start group. Thus, despite a more stable endocrine profile early initiation of GnRH antagonist does not appear to improve clinical outcomes of IVF compared with standard midfollicular initiation. Therefore the current late start GnRH antagonist protocol remains the best protocol at present and further adjustments in the timing of GnRH antagonist co-treatment are deemed useless.

We further assessed the impact of elevated early follicular progesterone levels in GnRH antagonist co-treated cycles on ongoing pregnancy rate by using prospective data from the RCT in combination with a systematic review and meta-analysis. It was demonstrated that elevated early follicular progesterone levels have a negative impact on the chance of achieving a pregnancy. In view of the low incidence of elevated early follicular progesterone levels and the absence of an effective strategy, routine screening is, however, not recommended.

It may be more useful to focus on the individualisation of GnRH antagonist co-treated cycles by using ovarian reserve tests such as AMH. This thesis further describes the clinical value of AMH and other patient characteristics for the prediction of high or low ovarian response in controlled ovarian stimulation using GnRH antagonist co-treatment. It was demonstrated that AMH had a better accuracy for the prediction of both high and low response as compared to age and BMI. Furthermore, AMH had a better accuracy for the prediction of high response than for low response and showed similar accuracy as reported in previous studies on GnRH agonists. Finally, it was demonstrated that although AMH was associated with pregnancy, its predictive accuracy was very modest even when assessing the cumulative pregnancy rate.Thus, AMH alone or in combination with age, is unlikely to alter clinical decisions based on the chance of an ongoing pregnancy or live birth after IVF treatment and should therefore preferably only be used for ovarian response prediction.