A randomised trial (MEGASET) comparing highly purified menotropin and recombinant FSH in a GnRH antagonist cycle with single blastocyst transfer.

Introduction: There are several RCTs comparing highly purified menotropin
(HP-hMG) with recombinant FSH (rFSH) in the long GnRH agonist cycle,
but there are limited data in GnRH antagonist cycles. A multicentre trial was
conducted to compare the ongoing pregnancy rates with HP-hMG vs rFSH in
GnRH antagonist cycles. The trial design aimed at minimising all major sources
of methodological variation and included single blastocyst transfer in all
patients.
Material and Methods: MEGASET (MENOPUR in GnRH Antagonist Cycles
with Single Embryo Transfer) was a randomised, assessor-blind, multicentre,multinational, non-inferiority trial. Infertile women aged 21-34 years and
eligible for ICSI due to unexplained or mild male factor infertility were randomised.
Controlled ovarian stimulation was started on day 2-3 of the cycle
with HP-hMG (MENOPUR 1200 IU, Ferring Pharmaceuticals) (N = 374) or
rFSH (PUREGON 900 IU, MSD/Schering-Plough) (N = 375). The gonadotropin
dose was fixed at 150 IU SC daily for the first five days. From day 6
of stimulation and onwards, gonadotropin dose adjustments of 75 IU were allowed,
and daily GnRH antagonist (ORGALUTRAN, ganirelix acetate 250
?g, MSD/Schering-Plough) administration was initiated. Recombinant hCG
(OVITRELLE, Serono) 250 ?g SC was administered as soon as ? 3 follicles
of ? 17mm were observed. All oocytes were inseminated using ICSI. Single
blastocyst transfer was done on day 5 after oocyte retrieval. Luteal support was
provided with vaginal progesterone 600 mg daily (UTROGESTAN, Seid) from
the day after oocyte retrieval until assessment of ?hCG. The primary endpoint
was ongoing pregnancy defined as presence of a viable fetus 10-11 weeks after
blastocyst transfer. The objective was to establish non-inferiority with a limit
of -10% for both per-protocol (PP) and intention-to-treat (ITT) populations.
Results: The two treatment groups were comparable with respect to causes of
infertility and demographic and pre-stimulation parameters. Oocyte retrieval
was performed in 97% and transfer in 83% of the patients in both groups. The
ongoing pregnancy rate was 30% with HP-hMG and 27% with rFSH for the
PP-population (95% CI of difference: -3.8; 9.8) and 29% and 27% respectively
for the ITT-population (95% CI of difference: -4.2; 8.6). Thus, non-inferiority
of HP-hMG to rFSH with respect to ongoing pregnancy rate was demonstrated.
Ongoing pregnancy rate per blastocyst transfer was 36% for HP-hMG and 32%
for rFSH in the PP-population and 35% for HP-hMG and 32% for rFSH in the
ITT-population. Serum estradiol (p <0.01) and progesterone (p <0.05) were
significantly higher with rFSH on day 6 of stimulation. At the end of stimulation,
estradiol (p <0.001) was significantly higher with HP-hMG. The proportion
of patients who on day 6 of stimulation maintained the starting dose of 150
IU was similar (67% for HP-hMG and 73% for rFSH). The number of oocytes
per retrieval was significantly (p <0.001) higher in the rFSH group than in the
HP-hMG group: 10.7 ± 5.8 and 9.1 ± 5.2, respectively. Blastocyst stage was
reached by 52% of the embryos on day 5 in the HP-hMG group and 49% in
the rFSH group. Endometrial thickness and echogenic pattern were comparable
between groups. Moderate/severe early ovarian hyperstimulation syndrome
(OHSS) occurred at a rate of 1.6% in both groups. The percentage of patients
with interventions because of excessive ovarian response, causing cycle cancellations
and/or other actions to prevent or treat OHSS (albumin administration
and paracentesis) was significantly (p <0.05) higher in the rFSH group (6.4%)
compared to the HP-hMG group (2.9%).
Conclusions: Controlled ovarian stimulation with HP-hMG in a GnRH antagonist
cycle provides ongoing pregnancy rates that are at least similar to that
achieved with rFSH. rFSH yields more oocytes than HP-hMG, but results in
more interventions due to excessive responses.