P-485 Performance of ERA test after PGT-A in RIF patients defined according to ESHRE 2023 good practice recommendations: a retrospective case-control study

Abstract

Abstract Study question Is Endometrial-Receptivity-Array (ERA) test clinically-useful among Repeated-Implantation-Failure (RIF) women defined according to ESHRE 2023 good-practice-recommendations (GPR) and undergoing single euploid blastocyst transfer? Summary answer ERA-test showed a higher “non-receptive endometrium” rate among RIF. Nevertheless, the live-birth-rate (LBR) per first euploid transfer was independent from progesterone timing according to ERA-test. What is known already ERA-test evaluates the expression of 238 genes from an endometrial biopsy. It aims at predicting the window of implantation and adjust progesterone timing accordingly. The data are controversial due to poor study design and/or limited sample size. Few studies investigated ERA-test in a setting with euploid transfers and/or among RIF patients. Moreover, RIF definition is largely inconsistent leading to a significant bias in the interpretation of the existing literature and treatment. ESHRE 2023 GPR revised RIF definition. Here, we leveraged this updated definition and PGT-A to assess ERA-test diagnostic/therapeutic performance in these women. Study design, size, duration Retrospective case-control study during 2013-2021 including 2211 patients with at least one euploid blastocyst after PGT-A. Hysteroscopy was conducted in case of suspect uterine pathologies, biopsy in case of suspect inflammation, and therapy in case of endometritis. 265 patients (12%) were classified RIF according to the ESHRE GPR (= no successful implantation despite an estimated cumulative chance ≥60%). 48 RIF patients requested ERA-test before undergoing vitrified-warmed euploid single blastocyst transfer, while 217 did not. Participants/materials, setting, methods qPCR/NGS-based PGT-A was conducted to assess non-mosaic aneuploidies. Normal thyroid function, vitamin-D levels, uterine cavity, and absence of sactosalpynx were confirmed. Whenever the ERA test suggested altered receptivity, progesterone timing was modified accordingly. The primary outcome was LBR per first euploid transfer in the non-ERA versus ERA groups, with a sub-analysis in ERA receptive versus non-receptive. Outcomes were adjusted for confounders assessed via logistic regressions. Main results and the role of chance RIF patients requesting ERA-test were older (39.1±4.5 versus 36.7±3.5 years, p < 0.01), and had already undergone 3.7±2.0 transfers resulted in implantation failures (1.5±1.3 of euploid blastocysts) versus 2.6±1.7 transfers (1.0±0.8 of which of euploid blastocysts) in the control (p < 0.01). Nevertheless, all these variables were not associated with the primary outcome under investigation. 44 non-RIF patients requested ERA-test in the same study period. Notably, the rate of pre-/post-receptive endometria was 39.6% (N = 19/48) among RIF and 13.6% (N = 6/44) among non-RIF women (p < 0.01). In fact, maternal age was not associated with “non-receptive” responses after ERA-test, the number of previous implantation failures showed an Odds-Ratio 1.29, 95%CI 1.03-1.62, p = 0.03. Blastocyst quality and day of full-blastulation were similar in ERA and non-ERA patients, as well as in ERA receptive and non-receptive patients. Positive pregnancy test, biochemical-pregnancy-loss, and miscarriage rates were all similar among non-ERA and ERA, and among ERA receptive and non-receptive patients. Indeed, also the LBR was similar among non-ERA and ERA (N = 98/220, 44.5% versus N = 17/48, 35.4%, p = 0.26; Odds-Ratio adjusted for blastocyst quality and day of full-blastulation: 0.76, 95%CI:0.39.1.48, p = 0.42) and among ERA receptive and non-receptive patients (N = 10/29, 34.5% versus N = 7/19, 36.8%, p = 0.99; Odds-Ratio adjusted for blastocyst quality and day of full-blastulation: 0.99, 95%CI:0.3-3.4, p = 0.99). Limitations, reasons for caution Our standard clinical workflow does not entail ERA-test, but the couple may request it. The study is retrospective with limited sample size. Patients requesting ERA-test were older with more previous failures. Non-ERA and ERA receptive patients underwent either Hormone-Replacement-Therapy or Modified-Natural-Cycle. Cost-effectiveness analyses are missing. Wider implications of the findings A promising association was reported between the novel definition of RIF and “non-receptive” responses, suggesting that ERA-test might unveil endometrial receptivity issues. Nevertheless, positive, and negative predictive values upon implantation are largely missing and its clinical utility as a therapeutic tool in RIF women remains questionable. Trial registration number none