Live birth rates are unrelated to sex-steroid levels on ET day in a dydrogesterone-based ‘programmed-ovulatory FET’ protocol: a multi-centric prospective cohort study

STUDY QUESTION What are the effects of three-times-a-day 10 mg oral dydrogesterone (DYD), initiated in the late follicular phase of natural menstrual cycles to induce endometrial receptivity for frozen-thawed embryo transfer (FET), on progesterone levels, indicative of ovulation on the day of FET, and how are levels of DYD, 20α-dihydrodydrogesterone (DHD), progesterone (P) and estradiol (E2) on the day of FET associated with clinical outcomes? SUMMARY ANSWER Late follicular phase initiation of oral 30 mg DYD is compatible with progesterone levels indicative of ovulation in 98% of cases, and DYD, DHD, P and E2 blood levels on the day of FET do not show a consistent relationship with live birth achievement, whether evaluated in isolation or interaction. WHAT IS KNOWN ALREADY Hormone replacement treatment (HRT) regimens for FET have come under scrutiny due to: (i) the risk of insufficient progesterone exposure with conventional dosing schemes, and (ii) maternal and fetal risks associated with the iatrogenic absence of a corpus luteum. Oral DYD 10 mg three-times-a-day (tid) is considered unlikely to interfere with ovulation or corpus luteum formation and does not exhibit cross-reactivity with progesterone in ELISA. Therefore, it can be used to induce endometrial receptivity (ie, to schedule the timing of FET in a natural cycle) and provide luteal phase support (LPS) while allowing ovulation to occur independently of the implantation window and enabling the monitoring of endogenous progesterone serum levels. STUDY DESIGN, SIZE, DURATION Nested within a multi-centric, prospective, clinical cohort study (NCT03507673), 559 normally cycling women from the routine care population who underwent FET in a spontaneous menstrual cycle (12/2021-8/2023) had DYD, DHD, P and E2 levels on day of FET measured by high performance liquid chromatography/tandem mass spectroscopy (HPLC/MS/MS) or Roche Elecsys®-Immunoassay (ELISA). PARTICIPANTS/MATERIALS, SETTING, METHODS Starting from cycle day 10, women underwent endocrine (E2, LH and P) and sonographic monitoring until the following criteria were ideally met: a leading follicle ≥16 mm, endometrial thickness ≥ 6 mm, and E2 ≥ 180 pg/ml. At this point, physicians could initiate oral DYD 10 mg (tid) or, if there was no luteinising hormone (LH) rise in serum, postpone initiation within 2 days of monitoring to align with patient and/or centre preferences (e.g., avoiding weekends). FET for day 2 to day 5 embryos was performed on day 3 to 6 of DYD intake. MAIN RESULTS AND THE ROLE OF CHANCE DYD and DHD plasma levels on day 3 to 6 of intake replicate previous findings. In FET with day 4 or 5 embryos, serum P levels indicated ovulation in 98.1% (>1.5 ng/ml) and 95.4% (>3.0 ng/ml) of cases and levels were higher (ratio of geometric means 2.88, 95% CI 2.48 to 3.35) than in FET with day 2 or 3 embryos. Progesterone levels on day of FET were unrelated (ratio of geometric means 1.13, 95% CI 0.95 to 1.34) to follicular size (≤16mm vs. >16mm) and only slightly higher (ratio of geometric means 1.36, 95% CI 1.19 to 1.56) in cases of serum LH level ≥12.6 IU/L versus <12.6 IU/L on day of last monitoring (ie before DYD initiation). When analysed stratified for FET timing (day 2/3 versus 4/5 embryos), risk differences (RD) for live birth in subjects were only marginally and not statistically significant different in patients with low (≤25th percentile) versus normal-high (>25th percentile) hormone levels (DYD RD -5.3%, 95% CI -14.0 to 3.2, p = 0.227; DHD RD -4.0%, 95% CI -14.6 to 6.0, p = 0.428; Progesterone RD 2.9%, 95% CI -7.0 to 12.3, p = 0.597; E2 RD -3.3%, 95% CI -12.2 to 5.3, p = 0.487). These findings were confirmed across the whole range of hormone values by logistic regression, and no interaction effects of the evaluated hormones on live birth rates were found at the unadjusted significance level 0.05. LIMITATIONS, REASONS FOR CAUTION All outcomes were assessed only in subjects undergoing FET, not on intention-to-treat. Oral DYD 10 mg (tid), similar to other progestins given to induce endometrial receptivity or for LPS, will likely still interfere with the LH surge, ovulation and luteal phase characteristics, warranting further investigations using a true natural cycle as the reference. WIDER IMPLICATIONS OF THE FINDINGS We propose the term ‘Programmed-Ovulatory (PO-) FET’ for this protocol. The PO-FET protocol enables scheduling the window of implantation, allows ovulation and corpus luteum formation, provides double gestagenic support, and may eliminate the need for control measurements of sex steroids on the day of FET. It allows monitoring of corpus luteum activity during the luteal phase and early pregnancy, is injection-free, and has low drug costs. This protocol needs to be tested in RCTs against other ovulatory FET protocol options, such as the ‘natural cycle (NC) FET’ or ‘managed NC FET’. STUDY FUNDING/COMPETING INTEREST(S) The trial was funded through university resources, excluding the expenses related to plasma and serum sample handling, storage and shipment, as well as the HPLC/MS/MS and ELISA analyses of DYD, DHD, E2 and progesterone. These specific tasks were financially supported by Abbott Products Operations AG (Allschwil, Switzerland); payment was made to Department of Gynecological Endocrinology and Reproductive Medicine, University Hospital of Schleswig-Holstein, Campus Lübeck. Abbott Products Operations AG had no impact on the study protocol, conduct, data analysis or data interpretation.