A Mechanism-Based PK/PD Model for Hematological Toxicities Induced by Antibody-Drug Conjugates

Antibody-drug conjugates (ADCs) are complex drug platforms composed of monoclonal antibodies (mAbs) conjugated to potent cytotoxic drugs (payloads) via chemical linkers, enabling selective payload delivery to neoplastic cells, resulting in improved efficacy and reduced toxicity. Brentuximab vedotin (Adcetris®, SGN-35) and adotrastuzumab emtansine (Kadcyla®, T-DM1) are the two FDA-approved and commercially available ADCs, and both drugs exhibit ADC-related thrombocytopenia and neutropenia. A pharmacokinetic/pharmacodynamic (PK/PD) model for ADCs was developed to identify the analyte from each ADC that is most associated with the observed hematopoietic toxicities and to determine the role of the apparent in vivo payload release rate on the severity of thrombocytopenia and neutropenia. Murine xenograft experiments and data from literature were combined, and the PK of both ADCs and their analytes were described with two-compartment models, with linear elimination and first-order payload release rate constants (k rel). ADC-associated hematotoxicities were captured with a previously published PD model for myelosuppression driven by various analytes. ADC half-lives were about 5 days, and k rel values were 0.46 (T-DM1) and 0.12 h−1 (SGN-35). The lifespans of platelets following T-DM1 and neutrophils following SGN-35 were 3.73 and 4.72 days. Comparison of alternate model structures suggested that mechanisms of myelosuppression are payload-driven for SGN-35 and ADC pinocytosis-dependent for T-DM1. Model simulations suggested that a 4-fold increase (T-DM1) and 70% decrease (SGN-35) in k rel would improve hematotoxicity to grade 1. The proposed model successfully captured the PK and associated myelosuppression of both ADCs and might serve as a general PK/PD platform for assessing hematological toxicities to ADCs.