Association between the number of oocytes retrieved and cumulative live birth rate in IVF treatment: menopur in GnRH antagonist single embryo transfer - high responder (MEGASET-HR) trial outcomes

Abstract

Although retrieval of high numbers of oocytes (>15) is linked with increased OHSS incidence, agonist trigger and freeze-only approaches nearly eliminate the severe form of the syndrome. Physicians therefore strategically pursue high oocyte yields and the prospect of high cumulative live birth rates, translated as increased efficiency at lower cost to patients. Although published data consistently shows no benefit of high response in fresh transfer cycles, the impact on cumulative live birth rates remains unclear. We evaluated this question on data obtained from the recently concluded randomized controlled MEGASET-HR trial. Analysis of the association between pregnancy outcomes and oocyte number resulting from stimulation with highly-purified human menopausal gonadotropin (HP-hMG) or recombinant follicle stimulating hormone (rFSH) in fresh and frozen cycles. This multicenter, randomized, assessor-blind, non-inferiority study included ovulatory women aged 21-35 years with body mass index 18-30 kg/m2 and serum AMH ≥ 5 ng/mL who were randomized 1:1 to a 150 IU starting dose of HP-hMG (n=311) or rFSH (n=309). Human chorionic gonadotropin was used to trigger oocyte maturation. Oocytes were fertilized by intracytoplasmic sperm injection and fresh single blastocyst transfer was performed; resultant ongoing pregnancy (detectable fetal heartbeat 8-9 weeks after fresh transfer) was the primary endpoint. Those with excessive ovarian response were required to have all embryos frozen after agonist trigger. Live birth outcomes for all fresh and frozen transfers occurring within 6 months of randomization were collected. Data presented are mean ± standard deviation unless otherwise indicated. Patient demographics were similar between HP-hMG and rFSH arms, respectively: age (30.0±3.1, 30.4±3.0 years), BMI (24.4±3.3, 24.3±3.4 kg/m2), AMH (7.8±3.6, 7.5±2.4 ng/mL), antral follicle count (30.5±15.5, 31.0±12.2). The average number of oocytes per patient in the rFSH arm (22.2±11.54) was approximately 7 higher than in the hMG arm (15.1±10.12), a striking difference in ovarian response that was however, accompanied by statistically significant increases in rates of OHSS (21.4% vs 9.7%; p<0.05) and pregnancy loss (25.5% vs 14.5%; p<0.05). Despite increased numbers of transfers in rFSH treated patients, the cumulative live birth rate was similar between groups (51.5% vs. 50.6%, p>0.05). Consistent with published data, fresh transfer ongoing pregnancy rates were lower in patients with ≥15 compared to ≤14 oocytes retrieved for both rFSH and HP-hMG. Interestingly, cumulative live birth rates also peaked and plateaued beyond 11-14 retrieved oocytes. Supernumerary oocytes did not confer a cumulative live birth rate benefit to patients predicted to be high-responders; moreover, they may have contributed to the increase in observed adverse events. Retrieval of high numbers of oocytes (>15) had a negative impact on the ongoing pregnancy rate in fresh transfer cycles with only a marginal benefit to cumulative live birth rates. Taken together the data suggest that optimal fresh and cumulative efficacy and safety outcomes can be achieved through retrieval of 8-14 oocytes through proper selection of gonadotropin type and dose.