(T-029) A Model-Based Meta-Analysis Framework to Predict and Manage ADC-Related Grade 3+ Neutropenia Across Clinical Studies

Zeel Shah – Quantitative Pharmacology and Data Analytics (QPDA) – BMS; Jian Zhou – Quantitative Pharmacology and Data Analytics (QPDA) – BMS; Anna Kondic – Quantitative Pharmacology and Data Analytics (QPDA) – BMS; Chuanpu Hu – Quantitative Pharmacology and Data Analytics (QPDA) – BMS; Gaohua Lu – Quantitative Pharmacology and Data Analytics (QPDA) – BMS; Yili Qian – Quantitative Pharmacology and Data Analytics (QPDA) – BMS; Ada Zhuang – Clinical Pharmacology (CP) – BMS

Enter your core abstract text here using the suggested layout. Abstracts are text only: No figures/tables. Do not paste author or affiliation information in the body of your abstract. Max character count not including spaces for abstract body is 2800.:

Objectives:
Antibody-drug conjugates (ADCs) are an important class of targeted cancer therapies, but their clinical use is often limited by toxicities such as neutropenia. This study aimed to characterize the safety profile of ADCs in solid tumors using a model-based meta-analysis (MBMA) approach, focusing on grade 3+ neutropenia as the primary endpoint. By integrating trial data across ADC programs, this analysis provides insights into grade 3+ neutropenia risk, identifies contributing factors, and supports study design and dose optimization. The model simulations can also be utilized to assess how internal ADC assets compare with marketed ADCs in the absence of head-to-head trials. MBMA addresses data heterogeneity challenges, evaluates dose-response relationships and covariate effects to enable reliable cross-study comparisons.

Methods:
A literature database of ADC clinical trials in solid tumors was utilized, comprising 55 randomized studies. For grade 3+ neutropenia, 27 trials (35 treatment arms) representing 6403 patients were included. A binomial regression model estimated the probability of grade 3+ neutropenia and accounted for between-trial and between-arm variability. It also explored covariates such as payload type, linker type, antibody type, tumor indication, trial phase, patient demographics and dosing regimen. Furthermore, an effort is underway to augment the literature-based dataset using a PBPK modeling tool to characterize the PK of ADC and payload for FDA-approved ADCs.

Results:
The preliminary MBMA model successfully predicted the probability of grade 3+ neutropenia in alignment with observed trial data. Grade 3+ neutropenia remains as one of the major safety concerns for several ADCs in solid tumors. Key findings indicated that combination with Taxanes was associated with increased probability of grade 3+ neutropenia. The analysis enabled ranking of comparators, supporting benchmarking of ADC treatment strategies.

Conclusion:
This study demonstrated MBMA’s utility for characterizing ADC-induced neutropenia in solid tumors. By identifying key covariates and enabling cross-comparator analysis, MBMA serves as a powerful tool for safety profiling and internal asset benchmarking. When paired with PBPK modeling tool, MBMA can help identify and interpret key covariates, enabling better-informed decisions for ADC clinical development.

Citations: No citations

Keywords: Model-Based Meta-Analysis, ADC, Grade 3+ Neutropenia