Medroxyprogesterone acetate versus ganirelix in oocyte donation: a randomized controlled trial

STUDY QUESTION: Is oral medroxiprogesterone acetate (MPA) non-inferior compared to ganirelix with respect to the number of mature oocytes (MII) retrieved at ovum pick-up (OPU) in oocyte donation cycles? SUMMARY ANSWER: MPA is comparable to ganirelix in terms of number of MII retrieved at OPU in oocyte donation cycles. WHAT IS KNOWN ALREADY: Oral treatment with MPA inhibits the pituitary LH surge during ovarian stimulation in infertile patients. Because of its negative effect on the endometrium, MPA suppression is combined with freeze-all. Published reports indicate that both the number of MII retrieved and pregnancy rates from these oocytes are comparable to short protocol of GnRH agonists during IVF cycles with freeze-all. MPA might allow for more comfortable and cost-effective ovarian stimulation. STUDY DESIGN, SIZE, DURATION: Randomized clinical trial, open-label, single center, to assess the non-inferiority of MPA (10 mg/day) versus ganirelix (0.25 mg/day) from Day 7, in ovarian stimulation cycles triggered with triptoreline acetate. Trigger criterion was >= 3 follicles of diameter >18 mm. PARTICIPANTS/MATERIALS, SETTING, METHODS: Overall, 252 oocyte donors were selected (eligible), 216 were randomized and 173 reached OPU: 86 under MPA and 87 under ganirelix. The main outcome was the number of MII retrieved at OPU. Secondary outcomes were embryological laboratory outcomes and reproductive outcomes in recipients. The study was powered to test that the lower limit of the 95% confidence interval of the difference in retrieved MII between groups will be above the non-inferiority limit of -3. Differences were tested using a two-sided Student's t-test or a Pearson's Chi(2) test, as appropriate. MAIN RESULTS AND THE ROLE OF CHANCE: All participants were in their first cycle of oocyte donation. On average, donors were 24 (SD 4.5) years old and with a BMI of 23 (SD 2.9) kg/m(2). Duration of stimulation was similar in both groups (11.2 days), as well as the total gonadotropin dose up to trigger (2162 IU in MPA and 2163 IU in ganirelix). The number of MII retrieved was no different: 15.1 (SD 8.3) with MPA and 14.6 (SD 7.0), 95% CI of the difference -2.78, -1.83 excluding the pre-defined non-inferiority limit (-3). Recipients and embryo transfer (ET) characteristics were also similar between groups. The average age of recipients was 42 (SD 4.8) years and the BMI was 24 (SD 4.4) kg/m(2). The mean number of MII assigned to each recipients was 6.7 (SD 1.2) in MPA and 6.6 (SD 1.2) in ganirelix (P = 0.58). MII were fertilized with partner sperm in 84% cycles overall and fertilization rate was 76% in MPA versus 74% in ganirelix (P = 0.34). Overall, there was 54% of double ET and 46% of single ET, with 40% of ETs were performed in D5. In spite of similar recipients and cycle characteristics, reproductive outcomes were unexpectedly lower with MPA. Biochemical pregnancy rate was 44 versus 57% (P = 0.023); clinical pregnancy rate 31 versus 46% (P = 0.006); ongoing pregnancy rate 27 versus 40%, (P = 0.015) and live birth rate 22 versus 31%, (P = 0.10). LIMITATIONS, REASONS FOR CAUTION: Although oocyte recipient and ET characteristics are similar among groups, this RCT has been designed under a hypothesis of non-inferiority in the number of MII obtained and recipients were not randomized; therefore, the reproductive outcomes in recipients should be evaluated with extreme caution. WIDER IMPLICATION OF THE FINDINGS: Ovarian stimulation using MPA for prevention of LH surge yields comparable number of MII oocytes compared to ganirelix in oocyte donation cycles. The unexpected finding in reproductive outcomes should be further investigated.