Posters

Authors

Zhe Huang¹, Guibi Yu ¹, Skanda Narayanaswamy¹, Mats. O. Karlsson¹

Affiliations

1 Uppsala University (Uppsala, Sweden)

Objectives

Causality and homogeneity in exposure–response (E-R) analyses can be assessed by the partitioned effects (PE) model, based on an instrumental variable (IV) framework [1]. Homogeneity being a single E-R relation regardless of reason for variation in exposure. In the PE model, randomized dose serves as the IV for exposure, enabling causal interpretation under the key exclusion restriction assumption that dose affects response only through exposure.

In this moxonidine case study, we applied the PE model to decompose the E–R relationship into dose, covariate (COV), and random effects (RE) components, quantifying their contributions to response variability. To assess the validity of the exclusion restriction assumption and add mechanistic understanding, we conducted mediation analyses.

Methods

The moxonidine dataset originated from a Phase II study in patients with NYHA class II–III congestive heart failure who received placebo (n = 23) or twice-daily moxonidine at three randomized dose levels (0.1, 0.2, or 0.3 mg; n = 74). Repeated PK samples and PD measurements, including plasma noradrenaline (NA) and blood pressure (BP), were obtained over 8 hours after the first dose and steady-state doses. The published PKPD model [2,3] served as the reference for PE model comparison, with NA and BP reductions described by inhibitory Emax functions.

A PE model with three components—dose effect, covariate effect (COV), and random effects (RE)—was developed for the PK–BP relationship. To quantify contributions to BP, the pure typical concentration (PTC) represented the effect attributable to dose, the typical concentration (TC) minus PTC represented the effect attributable to COV, and the individual concentration (C) minus TC represented the effect attributable to RE.

To investigate the D–E–R relationship, a mediation framework was developed incorporating (i) a KPD model to characterize the direct dose effect on BP and (ii) a PK–BP model to describe the exposure-mediated indirect effect. An additional structure assuming NA as a mediator in the PK–BP pathway (PK-NA-BP) was also evaluated for mechanistic understanding and indirect support for the exclusion restriction assumption.

To quantify the proportion of the total effect mediated by exposure or NA, counterfactual decomposition [4] was employed to estimate the mediation proportion (MP). Model comparisons used likelihood ratio tests (LRTs) and Bayesian Information Criterion (BIC).

Results

The fit of the PE model was not significantly better than that of the PKPD model (p>0.05). Hence there is no evidence of deviation from homogeneity in the PKPD model. PD parameters were similar between the PE models causal (i.e., IV) relation (Slope 0.10±0.05; Emax 0.34±0.11) and the PKPD model (0.11±0.04; 0.33±0.10). The PD parameters of the COV and RE components of the PE model were similar, but estimated with higher imprecision as only 5.8% and 10.9% of the PK variability are attributed to these components.

In the D–E–R mediation analysis, the mediation model achieved the lowest BIC, outperforming the indirect-only (ΔBIC = +14) and direct-only models (ΔBIC = +30). The median MP was 98.9% (IQR: 98.5–99.3%), indicating near-complete mediation through exposure, although three individuals showed low MP values and were better described by the KPD model.

In the PK-NA-BP analysis, the median MP was 7% (IQR:3.2-20.5%) and 63 of 74 subjects had MP < 50%, indicating that most of the effect was driven directly by exposure. The direct-only model provided the lowest BIC, the indirect-only model was similar (ΔBIC = +0.47), the mediation model showed the highest BIC (ΔBIC = +7.5).

Conclusion

The performed analysis addressed standard assumptions made, but typically not assessed, in the course of an ER analysis. The results supported the original PKPD model as a valid representation also of the causal relation.

Acknowledgement

The work was performed with resources from the project INVENTS, which has received funding from the European Union’s Horizon Europe Research and Innovation programme under grant agreement 101136365.

References

1. 1. Karlsson MO,BrundavanamD. Addressing Causality and Homogeneity Assumptions in Exposure-Response Analyses. Clin Pharmacol Ther. 2026;119:703–12. https://doi.org/10.1002/cpt.70132
   2. Karlsson MO, Jonsson EN, Wiltse CG, Wade JR. Assumption testing in population pharmacokinetic models: illustrated with an analysis ofmoxonidinedata from congestive heart failure patients. J Pharmacokinet Biopharm. 1998;26:207–46. https://doi.org/10.1023/a:1020561807903
   3. Brynne L, McNay JL, Schaefer HG, Swedberg K, Wiltse CG, Karlsson MO. Pharmacodynamic models for the cardiovascular effects ofmoxonidinein patients with congestive heart failure. Br J Clin Pharmacol. 2001;51:35–43. https://doi.org/10.1046/j.1365-2125.2001.01320.x
   4. GouloozeSC, Marostica E, Snelder N. Tutorial on Causal Mediation Analysis for Pharmacometricians. CPT Pharmacomet Syst Pharmacol. 2026;15:e70193. https://doi.org/10.1002/psp4.70193

Authors

Meemansa Sood¹, Karen Sinclair¹, Gregory Pinault¹, Marina Savelieva¹

Affiliations

1 Novartis Pharma AG (Basel, Switzerland)

Objectives

Drug development in rare diseases and paediatrics is challenging due to limited disease understanding, small sample sizes, single-arm trials and recruitment difficulties, resulting in scarce clinical data that makes robust estimation of efficacy, safety or disease progression from a single study difficult. While modelling and simulation tools such as PKPD and disease progression modelling or Virtual Patient Cohorts can address these evidence gaps, regulatory acceptance remains limited due to uncertainty around data quality and model qualification.
The Horizon EU-funded INVENTS Consortium aims to overcome these challenges by improving model-based extrapolation capability and enhancing regulatory acceptance of approaches such as Virtual Patient Cohorts in orphan and paediatric diseases. This effort relies on access to good-quality, securely anonymised clinical data. To support this, Novartis shared large, high-quality anonymised individual-level data (including PK, biomarkers and clinical endpoints) along with disease area expertise and population PK/PKPD models for two compounds: Fingolimod and Secukinumab. Pooled anonymised datasets were created to support development, testing and validation of new modelling approaches while maintaining high information content. This work describes the anonymisation and data-sharing process, primarily focusing on Fingolimod, providing a foundation for advancing implementation and acceptance of new methodologies with regulatory bodies.

Methods

All data were processed within EU GDPR-compliant secure environments and deployed on controlled high-performance computing platforms accessible to consortium partners. The process ensured datasets met regulatory requirements while retaining sufficient utility for robust modelling, simulation and virtual patient generation. Tailored anonymisation workflows balanced privacy protection with preservation of key clinical, biomarker and PKPD variables.
For Fingolimod, four Phase III clinical trials in patients with relapsing-remitting multiple sclerosis (RRMS)—three in adults and one in paediatrics—were pooled. Variables that were feasibly recognisable, replicable and distinguishable were labelled as identifiers: direct identifiers (items that alone allow identification, e.g. subject ID) or indirect identifiers (items that can identify participants when combined, e.g. race). Variables underwent transformations including randomisation of subject identifiers, suppression of free text, PHUSE-compliant date shifting, generalisation of race, suppression of demographics such as height and ethnicity, and generalisation of MS duration into three-year intervals. Both direct and indirect identifiers were transformed iteratively to reduce re-identification risk while balancing data utility.

Results

The implemented framework enabled consortium members to securely access individual patient-level data from 12 Phase III RCTs containing over 4,000 adult and paediatric patients treated with secukinumab for plaque-type psoriasis, psoriatic arthritis and juvenile idiopathic arthritis, and four Phase III trials for fingolimod comprising over 3,500 adults and over 200 paediatric patients with RRMS.

Certain changes occurred during anonymisation while preserving analytical utility. In one adult RRMS study, mean MS duration changed from 8.18 years (SD 6.6) pre-anonymisation to 6.79 years (SD 6.58) post-anonymisation, with minimal impact on median values. Correlation structures critical for PKPD and disease progression modelling were largely retained; for example, the correlation between MS duration and age decreased only modestly (45% to 41%). In the paediatric study, broader shifts in MS duration were observed reflecting the smaller sample size, but core covariate relationships remained stable (correlation with age shifted from −60% to −70%). Importantly, modelling and simulation results derived from anonymised datasets were consistent with analyses from original data, demonstrating that utility for model-based decision-making was preserved.

Conclusion

This work demonstrates that well-designed, risk-based anonymisation enables advanced quantitative sciences respectful of privacy regulations, and highlights the importance of strong coordination, robust governance and technical infrastructure. It provides a transferable model for future public–private data-sharing initiatives in clinical research. The INVENTS project has implemented a collaborative framework between industrial and academic stakeholders, enabling development of novel clinical trial methodologies in rare and paediatric diseases. The anonymised data was further used to perform credibility evaluation of the PKPD model to support future generation of virtual patient cohorts for paediatric RRMS.

References

1. 1. https://invents-he.eu/
   2. International Council for Harmonisation (ICH). E11A: Pediatric Extrapolation. 2024
   3. U.S. Food and Drug Administration (FDA). E11A Pediatric Extrapolation: Guidance for Industry. 2024.
   4. Musuamba, Flora T., et al. “Scientific and regulatory evaluation of mechanistic in silico drug and disease models in drug development: Building model credibility.” CPT: pharmacometrics & systems pharmacology 10.8 (2021): 804-825.
   5. PAGE 34 (2026)Abstr12296 [www.page-meeting.org/?abstract=12296]

Authors

Giulia Monchietto¹, Sarah Zohar¹, Karen Sinclair²

Affiliations

1 Inserm, Inria, Université Paris-Cité (Paris, France)

2 Novartis Pharma AG (Basel, Switzerland)

Introduction

Drug development is challenging in rare and pediatric diseases, due to recruitment limitations and ethical constraints. Generation of virtual patients (VPs) is an attractive approach in a setting where patients do not exist at numbers required to conduct powered clinical trials. Model-informed drug development (MIDD [1]), facilitates the process of VP generation using computational models integrating multi-source data, however, a specific framework is yet to be developed which considers:

• how much knowledge, and of which type, is required,
• which methods are suitable in different situations,
• how reliable are the generated VPs,

A 2025 systematic review [2] highlighted this gap, underlining that only 18/76 in-silico trial publications addressed rare or pediatric populations. Our work follows six-steps to combine multi-source knowledge, available methodology and clinical studies to facilitate reliable VP cohort generation. Specifically, we focus on generating VPs through stochastic simulations of concentration-time profiles via population pharmacokinetic (popPK) models to demonstrate the purpose of this workflow.

Methods

A framework to generate VPs has been initialised, implementing 6 steps. Steps 1-4 are the focus of this work, while steps 5-6 will be developed in future work.

1. 1. Objective statement: define the specific question of interest for themodelingactivity, and where it fits in the VP generation framework. 
   2. Knowledge assessment: categorize available information into relevant domains to eventually understand the impact of each knowledge domain on VP generation:

• drug knowledge (e.g. PK/PD models),
• disease knowledge (endpoints, progression models),
• knowledge coming from clinical trials (in one or more indications),
• population knowledge (clinical and demographic attributes).

1. 3. Methodological selection and procedure: identify methods and procedures applicable to various scenarios mimicking different amounts of available knowledge, and link them within an overarching framework to provide guidance on which approaches to follow for VP generation under different scenarios.

Initially we focus on extrapolation of PK to a new indication under different scenarios, considering many approaches such as external validation, integrating information from distinct indications, and model ensembling methods. [3-9]

1. 4. Model evaluation: assess candidate models to evaluate robustness and associated prediction capability

For extrapolation of PK to a new indication, classic approaches of prediction-based diagnostics and simulation-based tools (e.g. prediction errors and visual predictive checks) are utilised to ensure accurate reproducibility of the central tendency and variability of the target populations. Published guidelines on external evaluation and on model evaluation diagnostics are leveraged [10-11].

5/6. Uncertainty quantification and VP generation: the associated uncertainty quantification of the modeling approach on the generated VPs will be a focus of future work.

Results

We present a simple case study to illustrate the implementation of the framework, whose objective is to augment cohorts of patients with psoriatic arthritis (PsA), treated with Secukinumab, using a population pharmacokinetic model (step 1). Available knowledge comes from an established popPK model for Secukinumab built on data from patients with plaque psoriasis (PsO) [12], however the clearance and absorption rate parameters are scaled to mimic a scenario where the target and reference population have differing PK profiles. Knowledge from clinical trials is fixed to N=50 PsA patients (target population)(step 2). Direct extrapolation from the scaled model to the PsA population was performed (step 3), but evaluation (step 4) revealed that the model failed to meet predefined population-based predictive criteria. Consequently, we iterate through step 3 to assess alternative approaches (investigation of sampling scheme optimisation, model re-fitting on the target population, model re-fitting on pooled indications, etc.), After re-evaluating each approach in Step 3, model evaluation (step 4) is further repeated to identify robust approaches for VP generation. VP generation may then performed with confidence, but we emphasize that the uncertainty assessment to be performed in step 5 will provide the full measure of model risk, allowing regulators to understand the total level of confidence in the evidence generated to support drug development.

Conclusions

Results show an example of applying a framework designed to provide a structured approach to VP data generation under varying amounts of knowledge volume and types. Future work will expand this framework to accommodate additional knowledge scenarios with more complex models and techniques, including VP generation in pediatric populations, and integrating pharmacodynamic modeling.

References

1. 1. International Council for Harmonisation (ICH). General Principles for Model-Informed Drug Development M15.Final version 2026
   2. Chen, B., Schneider, L.C., Röver, C. et al. In Silico Clinical Trials in Drug Development: A Systematic Review. Ther Innov Regul Sci (2025). https://doi.org/10.1007/s43441-025-00893-w
   3. Nyberg J et al. PopED: an extended, parallelized, nonlinear mixed effects models optimal design tool. Comput Methods Programs Biomed. 2012;108(2):789–805.
   4. Gisleskog O, Karlsson MO, B SL. Use of Prior Information to Stabilize a Population Data Analysis. J. Pharmacokinet. Pharmacodyn. 2002;29(5-6):473–505.
   5. Sahasrabudhe SA, Bonate PL. Pharmacokinetic comparability between two populations using nonlinear mixed effect models: a Monte Carlo study. J. Pharmacokinet. Pharmacodyn. 2023;50:189–201.
   6. Uster DW, et al. A Model Averaging/Selection Approach Improves the Predictive Performance of Model-Informed Precision Dosing: Vancomycin as a Case Study. Clin Pharm Therap. 2021;109(1):175–-183.
   7. Agema BC et al. Selecting the Best Pharmacokinetic Models for a Priori Model‑Informed Precision Dosing with Model Ensembling. Clin. Pharmacokinet. 2024;63:1449–1461.
   8. Chan A et al. Synthetic Model Combination: A new machine-learning method for pharmacometric model ensembling. Pharmacometrics Syst Pharmacol. 2023;12(7):953–962.
   9. Chan Kwong AH et al. Prior information for population pharmacokinetic and pharmacokinetic/pharmacodynamic analysis: overview and guidance with a focus on the NONMEM PRIOR subroutine. J. Pharmacokinet. Pharmacodyn. 2020;47:431–446.
   10. Nguyen, T.H.T., et al. and for the Model Evaluation Group of the International Society of Pharmacometrics (ISoP) Best Practice Committee (2017), Model Evaluation of Continuous Data Pharmacometric Models: Metrics and Graphics. CPT Pharmacometrics Syst. Pharmacol., 6: 87-109
   11. Comets E et al. Computing normalised prediction distribution errors to evaluate nonlinear mixed-effect models: the npde add-on package for R. Comput Methods Programs Biomed. 2008 May;90(2):154-66
   12. Bruin G et al. Comparison of Pharmacokinetics, Safety and Tolerability of Secukinumab Administered Subcutaneously Using Different Delivery Systems in Healthy Volunteers and in Psoriasis Patients. Clin Pharm Therap. 2017;57(7):876–885.

Authors

Chengli Yu ¹, Moreno Ursino ², Sylvie Retout ^3,6, Marylore Chenel ⁴, Emmanuelle Comets^1,5, Yumi Cleary ⁶

Affiliations

1 Université Paris Cité and Université Sorbonne Paris Nord, IAME, Inserm (Paris, France)

2 Université Paris Cité, HeKA, Inserm (Paris, France)

3 Institut Roche (Boulogne-Billancourt, France)

4 Pharmetheus (Uppsala, Sweden)

5 Univ Rennes, Inserm, EHESP, Irset (Rennes, France)

6 Pharmaceutical Sciences, Roche Pharma Research and Early Development, Roche Innovation Center (Basel, Switzerland)

Introduction

The recent decade has seen considerable growth of paediatric physiologically-based pharmacokinetic (PBPK) applications both in academia and industry, however, the number of regulatory submissions that includes paediatric PBPK modelling remains relatively small (1). Among the current use of PBPK in marketing authorization applications, most of the cases were not considered qualified for the intended use (2). The recent ICH M15 guideline (3) provides a framework for Model-Informed Drug Development (MIDD) evidence assessment which can be applied to examine the reliability of a PBPK model.
Objectives: We searched and reviewed European Public Assessment Reports (EPARs) including paediatric PBPK modelling to summarise the context of use (CoU) and modelling approaches. We then retrospectively assessed their regulatory influence in decision making within the ICH M15 framework, considering model influence and consequence of wrong decision.

Methods

We used the open database platform (https://paediatricdata.eu/) to retrieve all EPARs that included paediatric PBPK modelling between 2015 and 2024. A structural review was conducted to collect core elements of: 1) Product information, 2) General model attributes, 3) Context of use, 4) Modelling work flow and evaluation, 5) Regulatory review outcomes. Model influence and consequence of wrong decision were assessed as low/medium/high based on available information from EPARs. When available, comments on reasons for acceptance or rejection were noted. Each EPAR was read by two readers and discrepancies between the assessments were resolved by consensus. Descriptive data analysis was performed in R.

Results

A total of 25 distinct cases were included for the final EPARs review. Among the cases, oncology was the most common therapeutic area (N = 9), then central nervous system (CNS) disease (N = 3) and cardiovascular disease (N = 3). 14 products had both adult and paediatric oral formulations, with bioequivalence demonstrated in 9 cases.

Most paediatric PBPK analyses (N = 23) were intended to support paediatric dose extrapolation from adult exposure, 6 of which involved more than one CoU. Fifteen extrapolations mainly targeted patients aged 2-8 years based on sparse paediatric PK data, while 10 spanned broader age range around 0-8 years without paediatric PK data. Remaining objectives included prediction of drug-drug interactions (DDI, N = 4), food effect (N = 3), and organ impairment to inform label in lieu of clinical study (N = 1).

Most reported paediatric PBPK models were scaled from a previously developed adult PBPK models. CYP3A-mediated metabolism and renal elimination were frequent elimination processes. Over the 18 cases with reported data, a median of 48 (13 – 881) paediatric patients were included in the evaluation datasets. Details of model evaluation were mostly unreported, and exposure comparison between prediction and observation (eg, AUC0-24h, Cmax, Css) was common among reported cases. Six cases compared paediatric PBPK results with population PK results when paediatric data was available, and 14 cases mentioned uncertainty evaluation but further information was mostly unavailable.

According to the regulator’s decisions, most paediatric PBPK approaches were considered premature, but model adequacy was explicitly acknowledged in only half of the cases. EMA accepted 14 cases, 6 of them were applied to select dose for paediatric studies, 4 for dose extrapolation with limited paediatric data, and 3 for formulation extrapolation. Only 2 cases were approved for dose extrapolation without observations. Rejected cases were mostly intended for dose extrapolation (N = 8), including even 4 cases with PK data from targeted age groups. The remaining rejected cases applied for a waiver concerning DDI studies (N = 3) and paediatric study dose selection (N = 2). The main reason stated for rejection was lack of data for evaluation. A trend between model influence and the regulatory decision was noted. More approved cases were associated with low intended model influence (low/medium/high: 6/5/3 cases) and majority of rejected cases were high influence applications (2/4/5).

Conclusion

PBPK modelling shows promise to support paediatric clinical development, however, regulatory acceptance remains to be cautious. In our survey, approval of paediatric PBPK modelling was more likely when the model was intended for study dose selection, adequately evaluated, had low influence on the decision and involved low risk.

References

1. 1. Johnson, T. N., Small, B. G. & Rowland Yeo, K. Increasing application of pediatric physiologically based pharmacokinetic models across academic and industry organizations. CPT Pharmacometrics Syst Pharmacol 11, 373–383 (2022).
   2. Mehrotra N, BhattaramA, Earp JC, Florian J, Krudys K, Lee JE, et al. Role of Quantitative Clinical Pharmacology in Pediatric Approval and Labeling. Drug Metab Dispos. 2016 Jul;44(7):924-33.
   3. Cole S, Hay JL, Luzon E, Nordmark A, Rusten IS. European regulatory perspective on pediatric physiologically based pharmacokinetic models. Int J Pharmacokinet. 2017;2(2):113-24
   4. Burckart GJ, Seo S, Pawlyk AC, McCune SK, Yao LP, Giacoia GP, Wang Y, Zineh I. Scientific and Regulatory Considerations for an Ontogeny Knowledge Base for Pediatric Clinical Pharmacology. Clin Pharmacol Ther. 2020 Apr;107(4):707-709.
   5. ICHM15 Guideline on general principles for model-informed drug development_Step 4

Authors

Tarini Singh ¹, Meemansa Sood ², Letao Li ¹, Sarah Zohar ¹, Moreno Ursino¹, Marina Savelieva ²

Affiliations

1 Inserm, Université Paris Cité, Inria, HeKA, F-75015 Paris, France

2 Novartis Pharma AG, Basel, Switzerland

Objectives

INVENTS project (https://invents-he.eu/) aims to assess and improve modelling and simulation approaches and their regulatory acceptance to expedite drug development in rare diseases including paediatrics. As part of this initiative, Novartis Pharma AG provided access to anonymised data from four Phase III clinical trials of Fingolimod in adult and paediatric patients with Relapsing-Remitting Multiple Sclerosis (RRMS). Paediatric Multiple Sclerosis (MS) is rare and represents only 3-5% of all MS cases worldwide, predominantly of the RRMS subtype [1]. The rarity of the disease and its paediatric manifestation make drug development challenging. This work focuses on assessing whether population PK and sequential population PKPD models developed using anonymised adult data in combination with external evidence can be refined and qualified to capture the dynamics of paediatric patients.

Methods

The methodology strategy was aligned with the ICH M15 guideline [2] and the framework proposed by Musuamba et al [3]. The anonymised dataset comprised 1856 adults (median age 38 years, 71.7% female, 93.1% Caucasian) and 106 paediatrics (median age 16 years, 66.03% female, 93.40% Caucasian). The patients were administered a once-daily dose of Fingolimod ranging from 0.25 to 1.25 mg. Anonymisation resulted in global suppression of BMI, ethnicity, height and time of first visit, while race was generalised to three categories and MS duration was generalised to 3-year intervals. To evaluate the impact of data anonymisation, all the results were tested for their consistency with previous PK analyses [4, 5]. Emphasis was placed on ensuring similar estimates for parameters and covariate effects and their statistical significance. A linear model for steady-state PK concentrations [4] as well as non-linear mixed effects (NLME) 2-compartment population PK model with first-order absorption and linear elimination from the prior published model in adult healthy volunteers [5] was tested using adult data only. Based on this, the choice was made to adapt the NLME model with time lag, absorption, clearance, volume(s) of distribution and inter-compartmental clearance to enable full longitudinal analysis. The predicted concentrations from the PK model were then used to evaluate the time course of absolute lymphocyte counts (ALC), a key biomarker for RRMS. Model adequacy was assessed using goodness-of-fit plots, relative standard errors (RSE), and shrinkage. Covariate and error model selection followed the Stochastic Approximation for Model Building Algorithm [6]. In order to verify the suitability of the adult model built for paediatric extrapolation, the PK parameters derived from the adult cohort were used to simulate drug concentrations for 1000 paediatric individuals (using the observed paediatric demographic data). Furthermore, the parameter estimates from the adult PD model was used to simulate the lymphocyte response for the (simulated) 1000 individuals. The model was then refitted to the observed paediatric data, with the quality of fit evaluated using goodness of fit plots.

Results

As a result of credibility assessments, the 2-compartment population PK model and a direct inhibitory Emax population PKPD lymphocyte model were finally selected. The anonymised dataset yielded PK and PKPD predictions consistent with results from previously developed models for adults and paediatrics prior to anonymisation. Covariate for the PK model included adjustment of Clearance for race, age and body weight at baseline. PKPD lymphocyte model comprised Imax adjusted for age, prior MS treatment and sex; IC50 for prior MS treatment, and baseline lymphocyte count for age and body weight. The PK and sequential PKPD model yielded RSE <30% for all parameters and covariate effects. Simulation-based verification for both PK and PD showed that most of the observed paediatric data was within 5-95% prediction intervals, supporting the suitability of the model use for extrapolation.

Conclusions

We presented here first steps in credibility evaluation of PK and PKPD models for paediatric patients with MS. Future work will focus on refining extrapolation and extending the analysis to additional efficacy endpoints to assess their suitability to generate virtual control arms for prospective RRMS trials. Incorporating established model‑credibility frameworks and harmonised MIDD principles ensures that the modelling approach is scientifically robust and aligned with evolving regulatory expectations.

References

1. 1. Dahlke F, Arnold DL, Aarden P,GanjgahiH, Häring DA, Čuklina J, Nichols TE, Gardiner S, Bermel R, Wiendl H. Characterisation of MS phenotypes across the age span using a novel data set integrating 34 clinical trials (NO.MS cohort): Age is a key contributor to presentation. Mult Scler.2021;27(13):2062‑2076. doi:10.1177/1352458520988637.
   2. International Council for Harmonisation (ICH). ICH M15: General principles on model‑informed drug development. Draft Guideline. ICH; 2024.
   3. Musuamba FT,SkottheimRusten I, Bursi R, Emili L, Wangorsch G, Karlsson KE, et al. Scientific and regulatory evaluation of mechanistic in silico drug and disease models in drug development: Building model credibility. CPT Pharmacometrics Syst Pharmacol. 2021;10(9):997‑1010.
   4. Population pharmacokinetic and exposure-lymphocyte count analysis of FTY720 (Fingolimod/Gilenya) in pediatric patients with multiple sclerosis, Mita M Thapar, Colm Farrell, Gordon Graham, Olivier Petricoul, https://www.page-meeting.org/wp-content/uploads/pdf_abstracts/8203-Poster_FTY720_v3_17May2018.pdf
   5. Wu K, Mercier F, David OJ,SchmouderRL, Looby M. Population pharmacokinetics of fingolimod phosphate in healthy participants. J Clin Pharmacol. 2012;52(7):1054‑1068. doi:10.1177/0091270011409229.
   6. Prague M, Lavielle M. SAMBA: A novel method for fast automatic model building in nonlinear mixed‑effects models. CPT Pharmacometrics Syst Pharmacol. 2022;11(2):161‑172. doi:10.1002/psp4.12742.