Scholarlink

<p>CD3-bispecific antibodies (CD3-BsAbs) represent an emerging modality with promising anticancer potential. Despite increasing regulatory approvals, the development of CD3-BsAbs remains challenging. CD3-BsAb candidates are routinely assessed and compared via <i>in vitro</i> workflows. However, protocol heterogeneity across experimental laboratories constrains cross-study potency comparisons. To address this, we developed an <i>in vitro</i> Quantitative System Pharmacology (QSP) model that mechanistically characterizes key processes underlying CD3-BsAb activity. The aim was to establish a framework adaptable to diverse <i>in vitro</i> conditions. The current framework comprises (a) single-cell trimer formation sub-model, (b) trimer-mediated T-cell activation and differentiation sub-model, (c) effector T-cell mediated tumor cell killing sub-model. We evaluated the framework using DuoBody-CD3x5T4 (CD3 equilibrium dissociation constant (K<sub>D</sub>) = 683&#xa0;nM) data from 14 solid tumor cell lines spanning 5T4 expression of 9,447–61,686 molecules/cell and drug concentrations of 1.76E-05–42.8 nM. For a subset of cell lines, we also included additional data comparing DuoBody-CD3x5T4 with bsIgG1-CD3x5T4 (CD3 K<sub>D</sub> = 16&#xa0;nM) and assessing effector-to-target (E:T) ratios of 1:1–8:1. All <i>in vitro</i> data were pooled into a single modeling dataset. A joint fit of T-cell activation and tumor cell cytotoxicity across the interconnected sub-models accurately captured the data and demonstrated mechanistic consistency. The model yielded mechanistically meaningful parameters, such as the per-T cell trimer count required to achieve half-maximal T-cell activation (EC50_act, estimated to be 2.12–4.6 trimers/T cell). The model's mechanistic structure and versatility suggest its potential to serve as a platform to predict drug effects across diverse assay conditions, quantify assay-dependent effects, and guide candidate selection.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An In Vitro Quantitative Systems Pharmacology Platform for Characterizing CD3-Bispecific Antibody-Mediated T-Cell Activation and Tumor Cell Cytotoxicity

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月28日

查看全文
  • Xuanzhen Yuan,
  • Craig Thalhauser,
  • Nasrin Afzal,
  • Kristel Kemper,
  • Guohua An,
  • Tommy Li

摘要

CD3-bispecific antibodies (CD3-BsAbs) represent an emerging modality with promising anticancer potential. Despite increasing regulatory approvals, the development of CD3-BsAbs remains challenging. CD3-BsAb candidates are routinely assessed and compared via in vitro workflows. However, protocol heterogeneity across experimental laboratories constrains cross-study potency comparisons. To address this, we developed an in vitro Quantitative System Pharmacology (QSP) model that mechanistically characterizes key processes underlying CD3-BsAb activity. The aim was to establish a framework adaptable to diverse in vitro conditions. The current framework comprises (a) single-cell trimer formation sub-model, (b) trimer-mediated T-cell activation and differentiation sub-model, (c) effector T-cell mediated tumor cell killing sub-model. We evaluated the framework using DuoBody-CD3x5T4 (CD3 equilibrium dissociation constant (KD) = 683 nM) data from 14 solid tumor cell lines spanning 5T4 expression of 9,447–61,686 molecules/cell and drug concentrations of 1.76E-05–42.8 nM. For a subset of cell lines, we also included additional data comparing DuoBody-CD3x5T4 with bsIgG1-CD3x5T4 (CD3 KD = 16 nM) and assessing effector-to-target (E:T) ratios of 1:1–8:1. All in vitro data were pooled into a single modeling dataset. A joint fit of T-cell activation and tumor cell cytotoxicity across the interconnected sub-models accurately captured the data and demonstrated mechanistic consistency. The model yielded mechanistically meaningful parameters, such as the per-T cell trimer count required to achieve half-maximal T-cell activation (EC50_act, estimated to be 2.12–4.6 trimers/T cell). The model's mechanistic structure and versatility suggest its potential to serve as a platform to predict drug effects across diverse assay conditions, quantify assay-dependent effects, and guide candidate selection.

Graphical Abstract