(M-027) Model-based meta-analysis of safety for monomethyl auristatin E-conjugated antibody drug conjugates in cancer patients

Mari Shiomi, Dr. – Principal Scientist, MSD K.K.; Ichiro Ieiri, Dr. – Dean, Professor, School of Pharmacy in Fukuoka, International University of Health and Welfare; Takeshi Hirota, Dr. – Associate Professor, Department of Pharmacy, Kyushu University Hospital, Fukuoka, Japan

Objectives: MMAE is one of the most-commonly used payloads for ADCs in clinical development as well as in the clinic for cancer treatment. Due to its potent cytotoxicity by preventing tubulin polymerization, which causes arrest of cell cycle and apoptosis, MMAE-conjugated ADCs often requires safety management for adverse events (AEs) during treatment, such as bone marrow toxicity, peripheral neuropathy, or skin reactions, etc. The purpose of this MBMA was to characterize the safety profiles across MMAE-conjugated ADCs in a quantitative manner and to assess the relationships between MMAE exposure and AE frequences.

Methods: A comprehensive literature search was conducted in the public databases for the approved MMAE-conjugated ADCs, including brentuximab vedotin, disitamab vedotin, enfortumab vedotin, tisotumab vedotin and polatuzumab vedotin, administered alone or in combination to cancer patients. The pharmacokinetic data for unconjugated-MMAE in the blood, the frequency of AEs of interest (anemia, neutropenia, hyperglycemia, peripheral neuropathy, skin reactions and alopecia) and the study-level baseline characteristics were retrieved. A meta-regression model was applied to derive a functional form of AE rate dependence on MMAE exposure to quantitatively estimate parameters that would characterize dependence [1, 2]. The following model was used to describe the data with maximum likelihood: logit(PrAE) = β0 + β1*Cmmae, where logit(PrAE) refers to the logit-transformed probability of a given AE and Cmmae corresponds to the average concentration of MMAE normalized over the standard dosing cycle per ADC. After the base model for exposure dependence of AE rates was established, trial-level patient baseline characteristics were included to explore potential covariates. Meta-regression model development and evaluation were performed using R software version 4.3.0.

Results: A total of 59 articles were identified, covering 5,842 patients in 155 dosing cohorts treated with ADC therapies. For anemia, neutropenia and peripheral neuropathy, no exposure dependence was found; the overall AE rates were, 24.6%, 23.4% and 31.5%, respectively. The developed models indicated that the all-grade AE rates for hyperglycemia, skin reactions and alopecia were driven by MMAE: the regression coefficient (β1) describing the payload exposure was 0.371 (95%CI: 0.0636-0.679), 0.493 (0.195-0.790), 0.806 (0.544-1.07), respectively.

Conclusion: The MBMA results provided a quantitative overview of safety profiles across MMAE-conjugated ADCs associated with MMAE exposure. This framework also helps to better understand and manage the safety when MMAE-conjugated ADCs are administered to patients in the clinic.

Citations: [1] Shulgin B et al. Oncoimmunology. 2020; 9(1): 1748982.
[2] Zhang R et al. CPT Pharmacometrics Syst Pharmacol. 2022; 11(8):1135-1146.