The pregnancy outcomes among women receiving individualized algorithm dosing with follitropin delta: a systematic review of randomized controlled trials

Purpose To investigate whether the ovarian stimulation with follitropin delta in an individualized algorithm-based manner is inferior to recombinant human-follicle stimulating’s follitropin alfa or follitropin beta conventional dosing regarding a series of established primary endpoints. Methods We conducted a registered systematic review (CRD42024512792) on PubMed-MEDLINE, Web of Science™, Cochrane Database of Systematic Reviews, and Scopus. Our search was designed to cover all relevant literature, particularly randomized controlled trials. We critically and comparatively analyzed the outcomes for each primary endpoint based on the intervention, reflected by the positive βhCG test, clinical pregnancy, vital pregnancy, ongoing pregnancy, live birth, live birth at 4 weeks, and multiple pregnancies. Results Six randomized controlled trials were included in the quality assessment as priority manuscripts, revealing an 83.3% low risk of bias. Follitropin delta led to non-significant differences in each parameter of interest from positive βhCG test (691; 53.44% vs. 602; 46.55%), ongoing pregnancies (603; 53.79% vs. 518; 46.20%), clinical and vital pregnancies (1,073; 52.80% vs. 959; 47.19%), to live birth and at 4 weeks (595; 54.14% vs. 504; 45.85%) with only 2 losses, and even multiple pregnancies (8; 66.66% vs. 4; 33.33%). However, follitropin delta was well-tolerated among hypo- and hyper-responders without significant risk of ovarian hyperstimulation syndrome and/or preventive interventions in contrast with follitropin alfa or follitropin beta. Conclusion The personalized individualized-based algorithm dosing with follitropin delta is non-inferior to conventional follitropin alfa or follitropin beta. It is as effective in promoting a similar response in women without significant comparable adverse effects. Supplementary Information The online version contains supplementary material available at 10.1007/s10815-024-03146-1.


Introduction
The pioneering work on gonadotropins [1,2] corroborated with the joint applications in translational medicine [3] paved the way towards synthesizing novel recombinant human follicle-stimulating hormones (r-hFSHs) of high purity through different biological processes [4].These preparations [5] advanced as essential parts for ovarian stimulation and response of in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) within current assisted reproductive technologies (ART) protocols [3].
Given the inter-individual heterogeneity and variability across ethnic populations [6][7][8][9], the need for predictive factors adapted for patient stratification in acquiring high-quality embryos for transfer was widely accepted.Thus, it arose the idea of shifting from standardization to individualization to alleviate the risks of cycle cancelation, poor ovarian stimulation, and ovarian hyperstimulation syndrome (OHSS) [10][11][12] and culminated in overcoming the initial limitations of implementation in clinical practice [10,[12][13][14].
While conventional r-FSHs uses international reference standard (IU) calculation with Steelman-Pohley bioassay, follitropin delta is dosed by mass (µg) because of the specific bioactivity, with an established dosing algorithm following pharmaco-kinetics/dynamic modeling exercise [15].In light with the European Society of Human Reproduction and Embryology (ESHRE) guidelines recommendations [14], the approved dosing algorithm comprises both the body weight and anti-Müllerian hormone (AMH) level that impacts the drug exposure distribution volume and ovarian response [18][19][20][21].
Although randomized controlled trials represent the best approach to compare interventional treatments, they are limited by generalizability and strict inclusion criteria of patients under specific conditions [22].Considering this argument, our purpose in this primary systematic review is to shed light on the non-inferiority potential by comparing the pregnancy outcomes represented by the established primary endpoints (positive βhCG test, clinical pregnancy, vital pregnancy, ongoing pregnancy, live birth, live birth at 4 weeks, and multiple pregnancies) among women undergoing conventional ovarian stimulation with follitropin alfa/ follitropin beta compared with the individualized dosing algorithm of follitropin delta.

Methodology and registration
This protocol was designed to adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [23] and registered in the International Prospective Register of Systematic Reviews database (PROSPERO) (CRD42024512792).

Ethical approval
This systematic review did not required Institutional Review Board (IRB) consent or evaluation from another panel of expertise, as research data were extracted from published studies.

Database search
Inquiries in distinct bibliographic academic databases that include PubMed-MEDLINE -United States National Library of Medicine (NLM, 1996), Web of Science™ (WOS) (Clarivate Analytics, 1997), Cochrane Database of Systematic Reviews (CDSR) (Cochrane Library, 1993), and Scopus (Elsevier, 2004) [24] were performed to identify, rank, and analyze potential suitable studies using MeSH (Medical Subject Headings) terms.The searches were restricted between December 12, 2016, and January 21, 2024.The date of inception represents the interval from authorization by the European Medicines Agency (EMA) and the current state of knowledge.We used synonyms, from the marketing name "Rekovelle" to "Follitropin Delta" and "FE 999049" to even abbreviations of the two projects entitled "The Evidencebased Stimulation Trial with Human rFSH in Europe and Rest of World" ("ESTHER-1") and "ESTHER-2.""Follitropin Delta" and "FE 999049″ are listed as [Supplementary Concept] and not Major Topics [Majr] that were introduced on June 11, 2017, respectively, July 8, 2016, according to the NLM official website.Both can be identified with the following credentials: MeSH Unique ID and Registry Number: C000620228-076WHW89TW and C000608977.

Strategy and strings
We used a dedicated terminology for Search #1 that relies on sole vocabulary components "Follitropin Delta" OR "FE 999049" OR "Rekovelle" followed by "ESTHER-1" OR "ESTHER-2" and complex clusters using Boolean operators ("AND" or "OR") for Search #2.The complete string sets for each database are available in Supplementary File 1.

Study selection
References retrieved were imported to Mendeley -Reference Management Software (v.1.19.8)(Elsevier, 2013) and de-duplicated by accessing the "check for duplicates" function followed by a second manual screening for accuracy.O.-D.I. and T.A. assessed the titles ± abstracts of each record for relevance and tangency with the scope, while the list of articles considered was subsequently scanned by assessing the entire content.Divergent opinions were resolved by common consent between each author.A tabular form for the retrieved records can be consulted in Supplementary File 2.

Objectives
The main resolution of this manuscript consists in answering the question of whether the third r-hFSH is superior to follitropin alfa/beta for one or multiple interest primary endpoints: positive βhCG test, clinical pregnancy, vital pregnancy, ongoing pregnancy, live birth, live birth at 4 weeks, and multiple pregnancies.We designed a Patient (P), Intervention (I), Comparator (C), and Outcome (O) (PICO) structure to develop the main research question and the criteria for inclusion and exclusion.The adopted PICO format is presented in Supplementary File 3.

Data extraction
Series of evidence in a tabular format using Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA, USA) for sorting and coding were independently extracted through a standardized form developed to characterize included studies by B.D., O.-D.I., and R.M. that describe methodological data: author's first name, year of publication, journal, country or countries, participant's age, study design and population, and outcome measures reported as number and percentage (%).

Quality assessment
B.D. and O.-D.I. independently evaluated the quality of each included study using the Revised Cochrane risk-of-bias tool for randomized trials (RoB 2) [25] via https:// sites.google.com/ site/ risko fbias tool/ (accessed on 28 January 2024) by providing guidance and different packages depending on the type of study.This tool provides a framework that classifies biases into five different domains: (1) bias arising from the randomization process; (2) bias due to deviations from intended interventions; (3) bias due to missing outcome data; (4) bias in measurement of the outcome; (5) bias in selection of the reported result.Answering to series of prompt questions varies from "yes," "probably yes," "probably no," "no," or "no information," and calculating the overall risk may be categorized as "low risk of bias," "some concerns," and "high risk of bias."For transparency of the RoB 2 evaluation, we employed the Risk-of-bias VISualization (ROBVIS) [26] in the present systematic review.

Inclusion/exclusion criteria
Whether the manuscripts were structured respecting the IMRaD format, they had to be written in English, report original data, and be published in a peer-reviewed journal.Therefore, primary research from RCTs was considered if available as full-length articles.

Studies reviewed
From 55 papers identified between December 12, 2016, and January 21, 2024, including duplicates and studies that had no relevance to the primary research question, 14 were subsequently removed as being duplicates after review.Of the remaining 41 records subjected to full-text assessment, an additional 32 manuscripts were excluded during the first phase and other 3 during the second step, as detailed in the PRISMA flowchart in Fig. 1.The complete list with the individual and overall number of publications per year and lists of records can be consulted in the Supplementary File 4. Therefore, we retained six exclusive and multinational RCTs [27][28][29][30][31][32] conducted primarily in Asia (n = 4) and Europe (n = 2), n = 1 in 2017, n = 3 in 2021, and n = 2 in 2022.

Risk of bias assessment
There has been an overall low risk (green color) of bias with an 83.3% quality on the assignment to the intention-to-treat (ITT), with only 16.7% indicating some concerns (yellow color) on one study [31] in D1 attributable to the "Bias arising from the randomization process" due to scarcity of baseline data to compare the differences between the groups and enrollment of underweight, overweight, and obese participants (Fig. 2).

Positive βhCG
Positive serum βhCG at 13-15 days after blastocyst transfer was established in five [27-29, 31, 32] out of six [30] eligible RCTs and totals 1293 confirmed tests.Precisely, 691 (53.44%) women experienced OS with individualized follitropin delta in doses ranging from 1 µg to 12 µg/d guided by the AMH concentration.On the other hand, 602 (46.55%) resulted from the administration of the other two conventional r-FSHs, from which 584 (97%) of follitropin alfa and 18 (2.99%) of follitropin beta.However, the study of Sánchez et al. [31] which relied on adding choriogonadotropin beta (CG beta) and observed a negative response in contrast with the placebo (odds ratio -OR vs. placebo; 0.53; confidence interval -CI 95% (0.30; 0.93); P = 0.0264).Although Ishihara et al. [30] conducted a similar study with the same primary endpoints [28], they did not report the associated data.In their prior experience, they have recorded elevated levels of βhCG among 68 participants (34/34)

Vital and clinical pregnancy
In the context of vital and clinical pregnancies regarded as at least one intrauterine gestational sac with a fetal heartbeat 5-6 weeks after transfer, three RCTs [27,28,32] have

Multiple pregnancies
Andersen et al. [27] are the sole investigators who monitored the likelihood of multiple pregnancies and observed that 12 were obtained following ovarian stimulation: 8 (66.66%) in the follitropin alfa and half in the follitropin delta (33.33%).

Discussion
In this first systematic review, we aimed to summarize evidence on the non-inferiority potential of individualized algorithm-based follitropin delta administration in terms of pregnancy outcomes compared with conventional dosing.It has been estimated that a daily dose of 10.0 µg follitropin delta corresponds to 150 IU/d [33] by possessing well tolerability up to 12-24 µg in Chinese women [34] with respect to the relatively same percentages on pregnancy outcomes [35,36].
Considering the low immunogenicity among women undergoing multiple stimulation cycles [35], the post hoc analyses emphasize a similar number of oocytes yielded and retrieved [33,35,37], irrespective of ovarian reserve, which interestingly was subsequently contradicted on further examinations [36].Regardless of the reported mean retrieval of > 10 oocytes [38][39][40] except in one instance [41] and that > 40% achieved the established optimal range between 8 and 14 oocytes [38,40,41], the pregnancy
Of note is that Asian women exhibit a higher risk of OHSS owing to the ethnic-related differences in weight than European females [36,37,43], but in a predictable dose-dependent and dose-exposure proportionality [44].This observation refutes additional evidence on reduced risk of moderate/ severe OHSS and preventive interventions in participants that followed ovarian stimulation with follitropin delta [43,45].OHSS is a common iatrogenic complication reported in clinical trials.Most cases ranged from mild to moderate OHSS [38,39,46] or isolated severe [38] and circumstances of any grade OHSS where the incidence was higher than in the ESTHER-1, but with fewer moderate or severe cases [47].
Cumulative data highlight the safety profile of follitropin delta irrespective of the ovarian reserve as was tested in categories of patients regarded as normal, low, and high responders [36,37] indicative by the AMH value [43] even though it appears to differ in a dose-response manner depending on the AMH via a r-hFSH and endocrine parameters-follicular development interconnection [19].However, its variability across stages of the menstrual cycle through multiple measurements suggests a limited impact considering the number of oocytes ± 1 when AMH < 15 pmol/L attained and dose adjustments of ± 1.5 μg when AMH ≥ 15 pmol/L [48].It is worth noting that previous studies have found that commercial AMH assays may exhibit an intercycle variation exceeding 20% and reach 163% [49][50][51], which was associated with a lack of standardized AMH tests [52][53][54].This has first limited result comparisons, but this concern was resolved since current assays started to display concordance for gonadotrophin prescribing [55].
The MARCS trial results regarding a higher mean number of oocytes retrieved and good quality blastocysts [47] are conflicting with the reports of another study in which the authors argue the presence of a lower proportion of day 3 good and intermediate blastocysts [56].Moreover, PROFILE [39] and DELTA [38] trials, in parallel with that who preceded [42] or compared the results obtained in the ESTHER programs [40,41,46], have reached congruent conclusions on primary and secondary outcomes.Even if ET was fresh or frozen during a woman's first stimulation cycle, the rate of major congenital disorders remained lower throughout the first 4 weeks after birth [42].
More recent studies from members of the ESTHER group [57] or conducted on specific populations from Europe [58] and North America [59,60] that might imply the concomitant administration of long gonadotropin-releasing hormone (GnRH) agonist [61] added further proof to the applicability into the "real-world" and extrapolation in clinical practice.Follitropin delta combined with menotropin [61] led to changes in endocrine and reproductive profile [57], while alone had no effect [58], and without notable clinical outcomes [57,61].It appears that even a combined approach necessitates OHSS preventive measures [61], besides being constantly reported and varying from early to moderate, even late and graded from moderate to severe [57].
Fertility nurses rated the GONAL-f pen injector to be less prone to handling errors than other injectors [62] as there was an isolated situation in which the women wrongfully exposed themselves for short term to 72 µg follitropin delta for 3 consecutive days which interestingly had no major adverse events (AEs) [63].

Strengths and limitations of the study
This is the first systematic review conducted on this topic, and the quality of the RCTs included indicates a low risk of bias.However, it is important to note that there is a relatively low number of studies overall, and all of these have been predominantly conducted by the same team members.

Conclusions
Based on all the aspects covered in this systematic review, it can be concluded that follitropin delta provides a more consistent response than follitropin alfa or beta, thus being non-inferior.Follitropin delta proved to be a reliable r-hFSH dedicated to ovarian stimulation.This manuscript consolidates the actual spectrum of knowledge and may assist clinicians and researchers in future studies to translate this in clinical practice.

Fig. 1
Fig. 1 PRISMA flow diagram of the systematic review

Fig. 2
Fig. 2 Overall assessment quality of the RTCs based on the ROB-2 tool

Table 1
Methodological data of the studies included in this systematic review

Table 2
Results of the established primary endpoints