IV-22 Siddharth Sukumaran

Development of a Mechanism-based pharmacokinetic model for MMAE conjugated ADCs

Siddharth Sukumaran, Crystal Zhang, Douglas Leipold, Keyang Xu, Kapil Gadkar, Marija Milojic-Blair, Bonnee Rubinfeld, Paul Fielder, Kedan Lin* and Saroja Ramanujan*

Genentech INC

Objectives: Antibody drug conjugates (ADC) are often produced and administered as a mixture of conjugated antibodies with different drug to antibody ratios (DAR) resulting in complex heterogeneous disposition kinetics. Results from in-vivo pre-clinical studies suggest that ADCs with different DAR show differential total antibody clearance. The purpose of this study is to develop a mathematical model that can describe and predict the complex PK behavior of MMAE conjugated ADCs by incorporating known mechanisms including DAR dependent antibody clearance and drug deconjugation. Pharmacokinetic disposition of ADCs in cynomolgous monkeys was used for model development.

Methods: Model development, data fitting and simulations were performed in MathWorks SimBiology software. PK data for total antibody and antibody conjugated MMAE (AcMMAE) concentrations were obtained after single intravenous administration of 0.3 mg/kg and 1 mg/kg Anti-Steap1 (or Napi2b)-Vc-MMAE (average DAR~3.5) and 1 mg/kg unconjugated Anti-Steap1 (or Napi2b) antibody in cynomolgous monkeys. Total antibody was measured by ELISA and AcMMAE by LC-MS. The structure of the mathematical model was developed based on the known mechanisms and observed impact of DAR on ADC pharmacokinetics. Specifically, models accounting for DAR dependent clearance and drug deconjugation were assessed by simultaneous fitting of different PK analytes.

Results: The model includes 10 distinct species (DAR0-8 ADC and unconjugated antibody), each represented by a two compartment model (central and peripheral) with common distribution parameters. Optimal fits were obtained, with deconjugation rates linearly dependent on DAR and proteolytic clearance increasing exponentially with DAR. The final model described PK profiles (total antibody and/or AcMMAE concentrations) of both Anti-Steap1 and Anti-Napi2b ADCs along with their unconjugated antibody counterparts in cyno. Final estimated parameters related to DAR dependent clearance and de-conjugation were comparable between Anti-Steap1 and Anti-Napi2b suggesting a common mechanism irrespective of the target.

Conclusions: The integrated PK model was developed based on the known mechanisms of ADC disposition kinetics and the model serves as a platform for describing complex PK behavior of MMAE conjugated ADCs in cynomolgous monkeys with potential translational application for human PK prediction.

Reference: PAGE 23 () Abstr 3073 [www.page-meeting.org/?abstract=3073]

Poster: Drug/Disease modeling - Oncology