Monday was ovarian stimulation day. Three trials presented in two free communication sessions and one satellite symposium, and all before lunch. Anyone arriving on time at Room 101 for the “Optimising ovarian stimulation” session – a room built for more than 500 people – could not get in, let alone find a seat. So the disappointed were marched en masse to Room 5C, which also filled quickly to its capacity of 450 seats. No wonder ESHRE’s SIG Reproductive Endocrinology is planning new stimulation guidelines.
First off was the ESPART trial, described by presenter Peter Humaidan of Aarhus University Hospital, Denmark, as the “largest RCT in poor ovarian response”, and another last word on LH supplementation in patients meeting at least two of the Bologna criteria (age >40, a previous cycle with ≤3 oocytes retrieved with conventional stimulation and an AMH level 0.12–1.3 ng/mL).
Following down-regulation, 462 women were randomised to FSH/LH and 477 to FSH alone, with a primary endpoint set at the number of oocytes retrieved. Results showed no significant difference between the groups for this endpoint, nor for ongoing pregnancy and live birth rate. Asked if this meant the end of LH supplementation in stimulation in POR patients, Humaidan, recalling Churchill and thinking ahead to IVF subgroups and suboptimal patients, said no, not the beginning of the end but the end of the beginning.
The issue of the day, however, was not LH but the question of individualised or standard dosing in patients whose predicted response was classified according to a several biomarkers, including AMH, bodyweight and AFC. The Optimist trial, a Dutch multicentre study of more than 1500 subjects split into two RCTs, found little reason for individualised dosing in its cumulative LBRs over an 18-month study period. Prediction, however, was based only on AFC, but outcome was the same in both trials – that individualised dosing based on AFC does not improve cumulative LBR in either predicted poor or high responders. However, in predicted hyper-responders there were fewer cancellations in the individualised group and “slightly more” oocytes collected. In this same group mild OHSS occurred in 3% of those with individualised dosing and in 12% of the standard group. Incidence of severe OHSS was comparable (1.2% vs 1.1%).
With poor response a major theme of this congress – and particularly the frustration of responding successfully to it – safety through the prediction of over-response is an increasingly proposed justification for AMH testing. The ESTHER trial, presented at a sponsored symposium of Ferring, also pitched an individualised dosing regimen against a standard dose to minimise risk of poor and excessive response. Unlike the Optimist trial, ESTHER predicted response according to AMH and bodyweight, but made its assessments in a first treatment cycle. The individualised regimen was based on a validated model which, said Professor Scott Nelson in a press meeting, aims “to get every possible egg”. And so it proved, with more cycles reaching egg retrieval target in the individualised group than in the standard. There were also fewer interventions for OHSS in this group, again emphasising the safety potential of response prediction. However, in terms of outcome, ongoing pregnancy and implantation rates were the same in both groups.
And as for the poor responding patient, there was still no good news – at least from supplementary therapies. Rob Norman from Australia found no benefit of adjunctive growth hormone in an underpowered RCT, while Stratis Kolibianakis reported similarly disappointing findings for DHEA from a meta-analysis.
Simon Brown, Editor of Focus on Reproduction