Genentech Research and Early Development (gRED) Computational Sciences, Genentech, Building 45, 620 E Grand Ave, South San Francisco, California, 94080, USA.
Independent Consultant, South San Francisco, California, USA.
AAPS J. 2024 Jul 15;26(4):84. doi: 10.1208/s12248-024-00954-2.
Measurement of anti-drug antibodies (ADA) to assess the incidence of ADA in a clinical trial is a critical step in immunogenicity assessment during the development of a protein therapeutic. We developed novel graphical approaches to illustrate clinical trial ADA data for the PD-L1 inhibitor atezolizumab (Tecentriq) that included a systematic analysis of the impact of the timing of ADA sampling and ADA assay drug tolerance on reported ADA incidence. We found that approaches used across the industry for ADA incidence analysis provide a limited view of immunogenicity in oncology studies, where ADA detection may be confounded by both drug dosage and patient attrition. Moreover, these approaches can miss important temporal information about the immune response. Our results demonstrated that the methodology of ADA assessment for the atezolizumab program was specifically designed to capture most ADA responses to ensure accurate reporting of ADA incidence. We further showed that the use of sparse sampling and/or ADA test methods with insufficient drug tolerance may result in a significant underreporting of ADA incidence. We conclude that the comparison of ADA incidence between different drugs can be highly misleading and that a test method with appropriate sensitivity in the presence of the drug and a clinical sampling scheme that is aligned with ADA responses to a drug is required to accurately report ADA incidence.
测定抗药物抗体（ADA）以评估临床试验中的 ADA 发生率，是在蛋白治疗药物开发过程中进行免疫原性评估的关键步骤。我们开发了新颖的图形方法来展示 PD-L1 抑制剂阿特珠单抗（Tecentriq）的临床试验 ADA 数据，其中包括对 ADA 采样时间和 ADA 检测药物耐量对报告 ADA 发生率的影响进行系统分析。我们发现，行业内用于 ADA 发生率分析的方法提供了一种有限的观点，即免疫原性在肿瘤学研究中可能会受到药物剂量和患者流失的双重影响。此外，这些方法可能会错过有关免疫反应的重要时间信息。我们的研究结果表明，阿特珠单抗项目的 ADA 评估方法专门设计用于捕获大多数 ADA 反应，以确保准确报告 ADA 发生率。我们进一步表明，稀疏采样和/或 ADA 检测方法缺乏药物耐量可能导致 ADA 发生率的显著低估。我们得出结论，不同药物之间 ADA 发生率的比较可能具有很大的误导性，并且需要一种在药物存在时具有适当灵敏度的检测方法和与药物 ADA 反应一致的临床采样方案，以准确报告 ADA 发生率。
