Arnaud Nativel1, Hippolyte Darré1, Perrine Masson1, Loic Ethève1, Nicolas Ratto1, Adèle L'Hostis1, Claudio Monteiro1
1Nova In Silico
Introduction: Pemetrexed is an essential antifolate agent frequently prescribed alongside platinum compounds to treat metastatic non-squamous NSCLC. Similarly to natural occurring folate, pemetrexed can undergo an unique metabolism called folypolyglutamination that facilitates its intracellular accumulation, notably in tumor tissue. In metastatic disease, characterizing tissue-specific exposure—including in brain, bone, liver, and lung metastases—is important for understanding efficacy and toxicity over multiple cycles. Objectives: We sought to develop a PBPK model that simulates pemetrexed’s disposition in primary and metastatic NSCLC lesions, validate it against clinical pharmacokinetic datasets, and integrate it into a quantitative systems pharmacology NSCLC model in Jinko to enable prospective three-year simulations of combination regimens reflective of a phase III trial setting. Methods: A full-body PBPK structure was employed, including dedicated compartments for bone, brain, liver, and secondary lung lesions. Pemetrexed polyglutamination and its effect on pharmacokinetic and efficacy were considered in every compartment deemed relevant. Model parameters were calibrated using clinical data on pemetrexed’s plasma profiles over multiple cycles (500 mg/m² every 21 days) [1]. After validation, the PBPK model was combined with the ISELA-V2 disease model on the Jinko platform, which is uniquely capable of running long-term virtual trials that reflect dynamic tumor evolution and multi-drug strategies. Results: Simulations showed strong agreement with observed plasma concentrations: less than 10% deviation for Cmax (Observed: 131 ug/ml – CV 29%, Simulated reference patient: 137 ug/ml ) and AUC ( Observed: 188 ug.h/ml – CV 27%, Simulated: 172 ug.h/ml) following intravenous 10-minutes infusion of 500 mg/m² pemetrexed . Tissue-level predictions revealed comparable distributions in metastatic compartments, aligning with reported efficacy data in advanced NSCLC. When tested under an extended FLAURA2-like protocol, the integrated QSP–PBPK platform accurately reproduced pemetrexed exposures and related outcomes (progression-free survival) over the 36 months of the clinical trial. Conclusion: This PBPK model accounts for pemetrexed’s tissue-specific partitioning in the context of advanced NSCLC with multiple metastatic sites. Its deployment within the Jinko system demonstrates the feasibility of simulating multi-year phase III trials and highlights the potential for mechanistic in silico approaches to optimize therapeutic regimens and guide personalized treatment strategies in metastatic lung cancer.
[1] Alain C. Mita et al. (2006, February). Phase I and Pharmacokinetic Study of Pemetrexed Administered Every 3 Weeks to Advanced Cancer Patients With Normal and Impaired Renal Function. Journal of Clinical Oncology, 10.1200/JCO.2004.00.9720
Reference: PAGE 33 (2025) Abstr 11392 [www.page-meeting.org/?abstract=11392]
Poster: Drug/Disease Modelling - Oncology