Toni Michael 1, Vinayak Smith 2,3, Beverley Vollenhoven 1,3,4,5, Sophie Stocker 1,2,6
1 School Of Pharmacy, The University Of Sydney (Sydney, Australia), 2 Virtus Health Pty Ltd (Sydney, Australia), 3 Department of Obstetrics and Gynaecology, Monash University (Melbourne, Australia), 4 Monash IVF (Melbourne, Australia), 5 Monash Health (Melbourne, Australia), 6 Department of Clinical Pharmacology and Toxicology, St. Vincent’s Hospital (Sydney, Australia)
Introduction/Objectives: Controlled ovarian hyperstimulation (COH) in women undergoing in vitro fertilisation (IVF) involves daily administration of recombinant follicle-stimulating hormone (FSH; follitropin alfa) to stimulate follicular maturation. As follicles grow, they initially secrete inhibin-B prior to a rise in oestradiol [1], making inhibin-B a sensitive dynamic biomarker of follicular recruitment and early ovarian response. Substantial inter individual variability exists in FSH pharmacokinetics, inhibin-B dynamics, and the final number of oocytes retrieved [2]. Current dosing strategies rely heavily on empirical protocols and limited early-cycle hormonal feedback.
Population pharmacokinetic/pharmacodynamic (popPK/PD) modelling represents an opportunity to mechanistically quantify sources of variability in FSH and inhibin-B, integrate clinically measured biomarkers, and identify covariates that may support individualised FSH dosing to optimise oocyte retrieval. This study aimed to (i) update and extend a previously published popPK/PD model of FSH and inhibin-B, (ii) evaluate the influence of covariates, and (iii) assess model performance using real world IVF clinical data from Australian patients to inform individualised FSH dosing practices.
Methods: Data were obtained from 81 Australian women undergoing IVF. Repeated FSH (PK) and inhibin-B (PD) measurements were collected at baseline and approximately 12 hours after last dose (Ctmid) at standard clinic visits. A published [3] one compartment PK model with first order absorption and linear elimination, linked to a PD turnover model for inhibin-B production stimulated by FSH, served as the a priori model. Endogenous FSH was converted from μg/L to IU/L to reflect the Australian data. The a priori model was evaluated using the dataset and demonstrated consistent structural performance. The parameter estimates of the a priori model were re-estimated using the dataset, followed by covariate evaluation. Covariates evaluated included baseline age, weight, anti-Müllerian Hormone (AMH), COH cycle number, oestradiol, and luteinizing hormone (LH). Covariate evaluation used conditional sampling use for stepwise approach based on correlation tests [4] for automated correlation driven selection, assessing for physiological plausibility and improvements in model fit. Model performance was assessed using standard diagnostic plots and visual predictive checks (VPC; 90% prediction interval). Parameter precision was assessed via relative standard errors. Models were estimated using non-linear mixed-effects modelling through the stochastic approximation expectation-maximization algorithm using Monolix (2024R1).
Results: Of 81 patients, all contributed ≥2 FSH concentrations and 28 contributed ≥2 inhibin-B concentrations to the PK and PD models, respectively. The typical patient was 36 years old (SD 4.7), weighing 67 kg (SD 11.6), was 1.64 m tall (SD 0.6), and taking 250 IU/d (range 125-450 IU/d) of follitropin alfa. The updated model improved fit relative to the a priori model (BICc = 2196.77 vs. 2240.15; −2LL = 2066.64 vs. 2127.4).
The final PK parameter and covariate relationships are as follows:
CL/F=0.53 *(Weight/70)^0.92 L/h
Baseline [FSH]=0.62 * e^(-0.0012 * baseline [oestradiol]) IU/L
The final PD parameter and covariate relationships are as follows:
Inhibin-B Kout=1.01 * e^(0.026 * baseline [LH]) h^-1
Baseline [Inhibin-B]=21.28 * e^(0.049 * baseline [AMH]) pg/mL
These population parameter estimates were similar to the a priori model. VPCs indicated adequate description of both PK and PD data despite sparse clinical sampling schedules typical of routine IVF practice.
Conclusions: The integration of covariates improved the predictive performance of the popPK/PD model of FSH and inhibin-B. Individualised dosing of FSH should consider the patients weight, baseline oestradiol, LH, and AMH concentration. All identified covariate effects are biologically plausible. The scaling impact of weight on clearance is well-established, and as oestradiol increases FSH concentrations decreases because oestradiol suppresses FSH production [5]. As LH increases, inhibin-B elimination increases because LH regulates the secretion of inhibin-B [5]. As AMH increases baseline inhibin-B also increases, which corresponds with AMH representing the number of small follicles present in the ovaries [6]. Inclusion of these ovarian response biomarkers as time-varying covariates using in-cycle data would be useful to inform FSH dose adjustments. Overall, the updated model supports future development of model informed precision dosing tools for FSH, particularly for early identification and management of poor responders. Despite sparse data, the model robustly characterised patient trajectories, demonstrating feasibility for real world clinical application. Planned in silico simulations will evaluate personalised dosing strategies for predicted poor responders and assess potential improvements in COH outcomes.
References:
[1] Eldar-Geva T et al. Relationship between serum inhibin A and B and ovarian follicle development after a daily fixed dose administration of recombinant follicle-stimulating hormone. J Clin Endocrinol Metab. 2000 Feb;85(2):607-13.
[2] Bahadur G et al. Correlation of IVF outcomes and number of oocytes retrieved: a UK retrospective longitudinal observational study of 172 341 non-donor cycles. BMJ Open. 2023 Jan 2;13(1):e064711.
[3] Ebid AHIM et al. Population PK-PD-PD Modeling of Recombinant Follicle Stimulating Hormone in In Vitro Fertilization/Intracytoplasmic Sperm Injection: Implications on Dosing and Timing of Gonadotrophin Therapy. J Clin Pharmacol. 2021 May;61(5):700-713.
[4] Ayral G et al. A novel method based on unbiased correlations tests for covariate selection in nonlinear mixed effects models: The COSSAC approach. CPT Pharmacometrics Syst Pharmacol. 2021 Apr;10(4):318-329.
[5] Robertson DM et al. Interrelationships between ovarian and pituitary hormones in ovulatory menstrual cycles across reproductive age. J Clin Endocrinol Metab. 2009 Jan;94(1):138-144.
[6] Robertson DM et al. Interactions between serum FSH, inhibin-B and antral follicle count in the decline of serum AMH during the menstrual cycle in late reproductive age. Endocrinol Diab Metab. 2021Aug;4(2):e00172.
Reference: PAGE 34 (2026) Abstr 12083 [www.page-meeting.org/?abstract=12083]
Poster: Clinical Applications