Is a freeze-all policy the optimal solution to circumvent the effect of late follicular elevated progesterone? A multicentric matched-control retrospective study analysing cumulative live birth rate in 942 non-elective freeze-all cycles.

STUDY QUESTION: Is late follicular elevated progesterone (LFEP) in the fresh cycle hindering cumulative live birth rates (CLBRs) when a freeze only strategy is applied? SUMMARY ANSWER: LFEP in the fresh cycle does not affect the CLBR of the frozen transfers in a freeze only approach, nor the embryo freezing rate. WHAT IS KNOWN ALREADY: Ovarian stimulation promotes the production of progesterone (P) which has been demonstrated to have a deleterious effect on IVF outcomes. While there is robust evidence that this elevation produces impaired endometrial receptivity, the impact on embryo quality remains a matter of debate. In particular, previous studies have shown that LFEP is associated with a hindered CLBR. However, most clinical insight on the effect of progesterone on embryo quality in terms of CLBRs have focused on embryo transfers performed after the fresh transfer, thus excluding the first embryo of the cohort. To be really informative on the possible detrimental effects of LFEP, evidence should be derived from freeze-all cycles where no fresh embryo transfer is performed in the presence of progesterone elevation, and the entire cohort of embryos is cryopreserved. STUDY DESIGN, SIZE, DURATION: This was a matched case-control, multicentre (three centres), retrospective analysis including all GnRH antagonist ICSI cycles in which a freeze all (FA) policy of embryos on day 3/5/6 of embryonic development was applied between 2012 and 2018. A total of 942 patients (471 cases with elevated P and 471 matched controls with normal P values) were included in the analysis. Each patient was included only once. PARTICIPANTS/MATERIALS, SETTING, METHODS: The sample was divided according to the following P levels on the day of ovulation triggering: <1.50 ng/ml and ≥1.50 ng/ml. The matching of the controls was performed according to age (±1 year) and number of oocytes retrieved (±10%). The main outcome was CLBR defined as a live-born delivery after 24 weeks of gestation. MAIN RESULTS AND THE ROLE OF CHANCE: The baseline characteristics of the two groups were similar. Estradiol levels on the day of trigger were significantly higher in the elevated P group. There was no significant difference in terms of fertilisation rate between the two groups. The elevated P group had significantly more cleavage stage frozen embryos compared to the normal P group while the total number of cryopreserved blastocyst stage embryos was the same. The CLBR did not differ between the two study groups (29.3% and 28.2% in the normal versus LFEP respectively, P = 0.773), also following confounder adjustment using multivariable GEE regression analysis (accounting for age at oocyte retrieval, total dose of FSH, progesterone levels on the day of ovulation trigger, day of freezing, at least one top-quality embryo transferred and number of previous IVF cycles, as the independent variables). LIMITATIONS, REASONS FOR CAUTION: This is a multicentre observational study based on a retrospective data analysis. Better extrapolation of the results could be validated by performing a prospective analysis. WIDER IMPLICATIONS OF THE FINDINGS: This is the first study demonstrating that LFEP in the fresh cycle does not hinder CLBR of the subsequent frozen cycles in a FA approach. Thus, a FA strategy circumvents the issue of elevated P in the late follicular phase. STUDY FUNDING/COMPETING INTEREST(S): No funding was received for this study. Throughout the study period and manuscript preparation, authors were supported by departmental funds from: Centre for Reproductive Medicine, Brussels, Belgium; Infertility Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy; Centro Scienze Natalità, San Raffaele Scientific Institute, Milan, Italy; and IVI-RMA, Lisbon, Portugal. E.S. has competing interests with Ferring, Merck-Serono, Theramex and Gedeon-Richter outside the submitted work. E.P. reports grants from Ferring, grants and personal fees from Merck-Serono, grants and personal fees from MSD and grants from IBSA outside the submitted work. All the other authors have no conflicts of interest to declare. TRIAL REGISTRATION NUMBER: N/A.

Do we trust scientific evidence? A multicentre retrospective analysis of first IVF/ICSI cycles before and after the OPTIMIST trial.

STUDY QUESTION: Has the practice of individualizing the recombinant-FSH starting dose been superseded after the largest randomized controlled trial (RCT) in assisted reproduction technology (ART), the OPTIMIST trial? SUMMARY ANSWER: The OPTIMIST trial has influenced our ART daily practice to a limited degree, but adherence is still generally poor. WHAT IS KNOWN ALREADY: Although the 'one size fits all' approach has been discouraged for decades by most authors, the OPTIMIST study group demonstrated in a large prospective RCT that, in general, dosage individualization does not improve the prospects for live birth, although it may decrease ovarian hyperstimulation syndrome (OHSS) risk in expected high responders. STUDY DESIGN, SIZE, DURATION: Retrospective analysis of all first in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) cycles from 1st January 2017 to 31st December 2018, before and after the OPTIMIST publication on November 2017. PARTICIPANTS/MATERIALS, SETTING, METHODS: Two thousand six hundred and seventy-seven patients, between 18 and 42 years old, undergoing their first IVF-ICSI cycle in seven Italian fertility centres, were included. Patients were allocated to three groups according to their ovarian reserve markers: predicted poor ovarian responders (POR), predicted normo-responders (NR) and expected hyper-responders (HRs). MAIN RESULTS AND THE ROLE OF CHANCE: Between 2017 and 2018, there was an overall increase in prescription of the standard 150 IU dose proposed by the OPTIMIST trial and a reduction in the use of a starting dose >300 IU. After subgroup analysis, the decrease in doses >300 IU remained significant in the POR and NR sub-groups. LIMITATIONS, REASONS FOR CAUTION: The retrospective nature of the study. Physicians need time to adapt to new scientific evidence and a comparison between 2017 and 2019 may have found a greater impact of the Optimist trial, although other changes over the longer time span might have increased confounding. We cannot be sure that the observed changes can be attributed to knowledge of the OPTIMIST trial. WIDER IMPLICATIONS OF THE FINDINGS: Clinicians may be slow to adopt recommendations based on RCTs; more attention should be given to how these are disseminated and promoted. STUDY FUNDING/COMPETING INTEREST(S): No external funding was used for this study. E.P. reports grants and personal fees from MSD, grants from Ferring, from IBSA, grants and personal fees from Merck, grants from TEVA, grants from Gedeon Richter, outside the submitted work. E.S. reports grants from Ferring, grants and personal fees from Merck-Serono, grants and personal fees from Theramex, outside the submitted work. All other authors do not have conflicts of interest to declare. TRIAL REGISTRATION NUMBER: Not applicable.

Top quality blastocyst formation rates in relation to progesterone levels on the day of oocyte maturation in GnRH antagonist IVF/ICSI cycles.

Cycles with progesterone elevation during controlled ovarian stimulation (COS) for IVF/ICSI are commonly managed with a "freeze-all" strategy, due to a well-recognized detrimental effect of high progesterone levels on endometrial receptivity. However, also a detrimental effect of elevated progesterone on day-3 embryo quality has recently been found with regards to top quality embryo formation rate. Because blastocyst culture and cryopreservation are largely adopted, we deemed relevant to determine whether this detrimental effect is also seen on blastocyst quality on day 5-6. This issue was investigated through a large two-center retrospective study including 986 GnRH antagonist IVF/ICSI cycles and using top quality blastocyst formation rate as the main outcome. Results showed that on multivariate analysis sperm motility (p<0.01) and progesterone levels at ovulation triggering (p = 0.01) were the only two variables that significantly predicted top quality blastocyst formation rate after adjusting for relevant factors including female age, BMI, basal AMH and total dose of FSH used for COS. More specifically, progesterone levels at induction showed an inverse relation with top quality blastocyst formation (correlation coefficient B = -1.08, 95% CI -1.9 to -0.02) and ROC curve analysis identified P level >1.49 ng/ml as the best cut-off for identification of patients at risk for the absence of top quality blastocysts (AUC 0.55, p<0.01). Our study is the first to investigate the top quality blastocyst formation rate in relation to progesterone levels in IVF/ICSI cycles, showing that increasing progesterone is associated with lower rates of top quality blastocyst. Hence, the advantages of prolonging COS to maximize the number of collected oocytes might eventually be hindered by a decrease in top quality blastocysts available for transfer, if increasing progesterone levels are observed. This observation extends the results of two recent studies focused on day-3 embryos and deserves further research.