Pascale Rietveld 1, Stijn Koolen 1, Chen Ning 2, Ron Mathijssen 1, Birgit Koch 1, Pieter Annaert 2, Sebastiaan Sassen 1
1 Erasmus MC (Rotterdam, The Netherlands), 2 KU Leuven (Leuven, Belgium)
Objectives
Intraperitoneal (IP) irinotecan is currently being investigated for treatment of peritoneal metastases, to overcome limited IP drug availability associated with systemic administration. However, quantitative understanding of drug transport across the peritoneal-plasma barrier and resulting IP exposure remains limited. Irinotecan is an anti-cancer drug that is converted to the highly active metabolite SN-38, which is subsequently inactivated by UGT1A1-mediated glucuronidation into SN-38G[1]. Reduced UGT1A1 activity results in increased systemic SN-38 exposure and toxicity risk[2, 3]. Physiologically-based pharmacokinetic (PBPK) modeling provides a mechanistic framework to investigate pharmacokinetics under alternative administration routes and genetic variability[4]. To date, no PBPK model describing IP drug administration of irinotecan exists. Therefore, this study aimed to develop a PBPK model for IP irinotecan and evaluate the effect of reduced UGT1A1 activity on systemic SN-38 exposure by use of simulations.
Methods
A whole-body PBPK model was developed in PK-Sim v.12 (Open Systems Pharmacology Suite) using clinical pharmacokinetic data from 53 patients with colorectal and gastric peritoneal metastases receiving 75 mg IP irinotecan in the INTERACT, INTERACT-gastric and INTERACT-II studies[5]. Plasma concentrations of irinotecan, SN-38, and SN-38G and IP concentrations of SN-38 and SN-38G were used for model calibration and evaluation. Because PK-Sim lacks a peritoneal compartment, the interstitial space of the small intestine was used as a surrogate for dosing in the IP compartment. The effective permeability, surface area, and interstitial volume fraction parameters were optimized to reproduce delayed drug exchange across the peritoneal-plasma barrier. Tissue distribution was predicted using PK-Sim standard partition coefficient and permeability algorithms under perfusion-limited assumptions. Metabolic pathways were implemented mechanistically: irinotecan conversion into SN-38 via Carboxylesterase (CES) 1 and CES2, and glucuronidation of SN-38 into SN-38G via UGT1A1. In vitro values for maximum rates per mg protein in tissues were translated to enzyme-specific rates using quantitative proteomics-based enzyme abundances for CES1, CES2 and UGT1A1. Local IP metabolism was added by fixing experimentally measured CES enzyme concentrations to the proxy interstitial compartment. Model performance was evaluated using observed versus predicted concentration-time profiles, exposure metrics, and absolute average fold error (AAFE). UGT1A1 poor metabolizer scenarios were simulated by scaling glucuronidation capacity (Vmax) to 20-50% of wild-type activity.
Results
The PBPK model adequately described systemic and IP pharmacokinetics of irinotecan and its metabolites, with predicted exposures within two-fold of observed data (AAFE = 1.28). The model described delayed systemic absorption following IP administration and sustained IP exposure. Reduced UGT1A1 activity resulted in increased systemic SN-38 AUC (87.9 ng*h/mL) compared with normal metabolizers (59.1 ng*h/mL). Simultaneously, SN-38G exposure decreased and the SN-38G/SN-38 AUC ratio declined substantially, confirming model sensitivity to altered glucuronidation capacity.
Conclusions
We developed the first mechanistic model describing IP administration in humans. A surrogate interstitial compartment enabled simulation of peritoneal barrier effects within the whole-body PBPK environment. Simulations demonstrate that reduced UGT1A1 activity increases systemic SN-38 exposure after IP administration. This framework provides a generalizable strategy for integrating non-standard administration routes into PBPK platforms, and it provides a step towards development of physiologically-based peritoneal models for translational optimization of IP chemotherapy.
References:
[1] de Man FM, Goey AKL, van Schaik RHN, Mathijssen RHJ, Bins S. Individualization of Irinotecan Treatment: A Review of Pharmacokinetics, Pharmacodynamics, and Pharmacogenetics. Clin Pharmacokinet. 2018;57(10):1229-54.
[2] Atasilp C, Biswas M, Jinda P, Nuntharadthanaphong N, Rachanakul J, Hongkaew Y, et al. Association of UGT1A1*6, UGT1A1*28, or ABCC2 c.3972C>T genetic polymorphisms with irinotecan-induced toxicity in Asian cancer patients: Meta-analysis. Clin Transl Sci. 2022;15(7):1613-33.
[3] Hulshof EC, de With M, de Man FM, Creemers GJ, Deiman B, Swen JJ, et al. UGT1A1 genotype-guided dosing of irinotecan: A prospective safety and cost analysis in poor metaboliser patients. Eur J Cancer. 2022;162:148-57.
[4] Kuepfer L, Niederalt C, Wendl T, Schlender JF, Willmann S, Lippert J, et al. Applied Concepts in PBPK Modeling: How to Build a PBPK/PD Model. CPT Pharmacometrics Syst Pharmacol. 2016;5(10):516-31.
[5] van Eerden RAG, de Boer NL, van Kooten JP, Bakkers C, Dietz MV, Creemers GM, et al. Phase I study of intraperitoneal irinotecan combined with palliative systemic chemotherapy in patients with colorectal peritoneal metastases. Br J Surg. 2023;110(11):1502-10.
Reference: PAGE 34 (2026) Abstr 11951 [www.page-meeting.org/?abstract=11951]
Poster: Drug/Disease Modelling - Absorption & PBPK