(S-086) QSP I-O & ADCs: Integration of an antibody-drug conjugates mechanistic model into an immuno‐oncology quantitative systems pharmacology platform
Associate Director, Quantitative Systems Pharmacology Daiichi Sankyo Inc., United States
Disclosure(s):
Yafei Wang: No financial relationships to disclose
Objectives: Antibody-drug conjugates (ADCs) synergize targeted chemotherapy with monoclonal antibody precision, while immuno-oncology (I-O) therapies harness the immune system to combat cancer. Combining these modalities holds promise for overcoming resistance and enhancing therapeutic efficacy in oncology [1]. However, optimizing such combinations requires a robust framework to predict pharmacokinetic-pharmacodynamic (PK-PD) interactions. In this work, we integrate a mechanistic ADC model into an in-house I-O quantitative systems pharmacology (QSP) platform, enabling the design of virtual clinical trials to explore ADC/I-O combination therapies.
Methods: A mechanistic ADC module was developed, incorporating physiologically based pharmacokinetic models for ADC components (antibody and payload), and soluble receptors across plasma, peripheral tissues, and the tumor microenvironment (TME). The module simulates antibody-receptor monovalent and bivalent binding, internalization, intracellular payload cleavage, and cancer cell death in the TME. Two drug-to-antibody ratio (DAR) models were implemented: (i) an individual DAR model, tracking payload deconjugation kinetics across ADC variants [2], and (ii) an analytically derived average DAR model to simplify computational complexity. Both models were embedded into an in-house I-O QSP platform, which models the cancer-immunity cycle and the TME cancer immunity subcycle [3], and immune checkpoint interactions. This integration allows simulation of ADC monotherapy and combination regimens with I-O agents (e.g., checkpoint inhibitors).
Results: The integrated individual DAR model accurately represented the physiological distribution of various ADC states, soluble receptors, and payloads within plasma and the TME, both extracellularly and intracellularly. Comparative analysis revealed that the average DAR model approximated key outcomes of the individual DAR model—such as plasma payload levels, intracellular payload concentrations, and tumor shrinkage—with high fidelity under physiological conditions. This parity suggests the average DAR model can accelerate QSP workflows without compromising accuracy. Furthermore, the platform considered immune modulation in the TME during ADC/I-O combination therapy, highlighting synergies between payload-induced immunogenic cell death and immune activation.
Conclusions: By merging ADC mechanistic modeling with I-O QSP, this work establishes a versatile platform for virtual clinical trial design. Future studies will explore extracellular payload release, immunomodulatory effects of ADCs [4], and patient-specific virtual cohort calibration for indications such as breast or lung cancer. The platform’s capability to simulate combination therapies could help streamline early clinical development, reduce trial costs, and identify optimal dosing regimens.
Citations: [1] Y., P., et al., Cell Death Dis 15, 433 (2024). DOI: 10.1038/s41419-024-06837-w [2] C., Y., et al., AAPS J 19, 1002 (2017). DOI: 10.1208/s12248-017-0100-x [3] M., I., et al., Immunity 56, 2188 (2023). DOI: 10.1016/j.immuni.2023.09.011 [4] T., L., et al., Nat Commun 16, 3167 (2025). DOI: 10.1038/s41467-025-58266-8
