Merigolix QSP Model for Endometriosis Treatment

Overview

A comprehensive Quantitative Systems Pharmacology (QSP) model for merigolix, a GnRH antagonist under development for endometriosis treatment. This model integrates pharmacokinetics, hypothalamic-pituitary-gonadal (HPG) axis dynamics, and clinical endpoints to support dose selection and clinical trial design.

Model Structure

Pharmacokinetic Model

• Model Type: Two-compartment with first-order absorption and lag time
• Key Parameters:
  • Absorption rate constant (Ka): 3.39 h⁻¹
  • Absorption lag time (ALAG1): 0.302 h
  • Apparent clearance (CL/F): 549 L/h
  • Central volume (Vc/F): 1690 L
  • Intercompartmental clearance (Q/F): 158 L/h
  • Peripheral volume (Vp/F): 643 L

Dose-Dependent Bioavailability

Dose (mg)	Relative Bioavailability
20	0.532
40	0.75
80	1.0 (reference)
120	1.0
160	1.0
240	1.5
320	2.02

Pharmacodynamic Model - Estradiol Suppression

• Mechanism: Indirect response model with competitive GnRH receptor antagonism
• Key Parameters:
  • Maximum inhibition (Imax): 0.95
  • Half-maximal inhibitory concentration (IC50): 2.5 ng/mL
  • Baseline estradiol (BASE_E2): 80.0 pg/mL
  • Negative feedback rate constant (Ktol): 0.0165
  • Hill coefficient (gamma): 1.0

HPG Axis Dynamics

LH Dynamics

• Baseline LH: 5.0 mIU/mL
• LH Surge: Sharp spike at ovulation (Day 14)
  • Gaussian width: 6 hours (for spike-like appearance)
  • Surge factor: 8.0
  • Triggered by E2 > 150 pg/mL
• Drug Effect:
  • LH surge abolished when GnRH_effect < 0.5
  • Enhanced suppression via GnRH_effect^1.5

FSH Dynamics

• Baseline FSH: 6.0 mIU/mL
• Early Follicular Boost: 1.5x elevation (Day 1-5)
• Ovulatory Surge: Smaller than LH (factor: 1.5)
• Drug Effect: Enhanced suppression via GnRH_effect^1.5

Bone Mineral Density (BMD) Model

• Type: Threshold model
• E2 Threshold: 40 pg/mL
  • E2 ≥ 40 pg/mL: No BMD change (equilibrium)
  • E2 < 40 pg/mL: BMD loss proportional to E2 suppression
• Maximum BMD Loss Rate: 0.0000065 (calibrated for ~2.4% loss over 168 days at full suppression)
• Clinical Validation: Model reproduces clinical BMD changes:
  • 0 mg: Stable (0% change)
  • 80-240 mg: -0.65% to -2.4% over 24 weeks

Pain Score Model

• Baseline pain: 7 (NRS 0-10)
• Pain turnover rate: 0.008 h⁻¹
• EC50 for E2 effect on pain: 25 pg/mL
• Minimum achievable pain: 1.5

Hot Flash Model

• EC50 for E2 effect on hot flashes: 15 pg/mL
• Maximum hot flashes: 5 per day

Validation Data Sources

Clinical Trials

1. Merigolix Phase 2a (NCT05138562)

   • E2 suppression data
   • Pain score reduction (NRS 0-10)
   • Doses: 80, 120, 160, 240 mg QD

2. SPIRIT 1/2 Trials (Relugolix - structural analog)

   • BMD changes
   • Hot flash frequency
   • Extended treatment data

3. LIBERTY 1/2 Trials (Elagolix - GnRH antagonist class)

   • Additional validation for hormone dynamics
   • Long-term safety endpoints

Simulation Parameters

Treatment Duration

• Pre-treatment observation: 28 days
• Treatment period: 168 days (24 weeks)
• Post-treatment recovery: 112 days (16 weeks)
• Total simulation: 308 days (44 weeks)

Doses Simulated

• 0 mg (placebo)
• 80 mg QD
• 120 mg QD
• 160 mg QD
• 240 mg QD

Inter-Individual Variability (IIV)

Parameter	CV (%)
CL/F	53.9
Vc/F	60.8
Ka	86.5
ALAG1	23.7
IC50	184.1
BASE_E2	47.3

Files

R Scripts

• merigolix-qsp.R: Main QSP model and simulation
• validation_data_and_plots.R: Validation against clinical data
• app.R: Shiny dashboard for interactive exploration

Output Files

• *_pop.csv/png: Population simulation results (n=5 per dose)
• Files without suffix: Single patient simulation

Figures

• figure_hormone_profiles_pop.png: E2, LH, FSH dynamics
• figure_clinical_outcomes_pop.png: Pain, Lesion, Hot Flash, BMD
• figure_validation_combined.png: Model validation vs clinical data
• figure_exposure_response_pop.png: E-R relationships
• figure_imaging_parameters_pop.png: Uterine/ovarian volume, endometrial thickness
• figure_laboratory_parameters_pop.png: Lipids, liver enzymes

Usage

Run Full Simulation

source("merigolix-qsp.R")

Run Validation Plots

source("validation_data_and_plots.R")

Launch Shiny Dashboard

shiny::runApp("app.R")

Key Model Features

1. Menstrual Cycle Integration: 28-day cycle with follicular/luteal phases
2. Drug-Induced Anovulation: LH surge abolished during treatment
3. Threshold-Based BMD: Clinically realistic bone safety profile
4. Multi-Endpoint Modeling: Efficacy (pain, lesion) and safety (BMD, hot flash) endpoints
5. Population Variability: Built-in IIV for PK and PD parameters

Target Product Profile

Efficacy Target

• E2 suppression to 20-40 pg/mL range ("partial suppression")
• Pain score reduction to ≤3 (NRS)
• Preservation of bone-protective E2 levels

Safety Constraints

• BMD loss < 3% over treatment period
• Minimal vasomotor symptoms (hot flashes)

References

1. Merigolix Phase 2a Study Protocol (NCT05138562)
2. Relugolix SPIRIT trials (Taylor et al., NEJM 2021)
3. Elagolix LIBERTY trials (Taylor et al., NEJM 2017)

Authors

Research Team, 2025

License

For research purposes only.