Abstract：
Regulatory guidance for global drug development relies on animal studies to evaluate safety risks for humans, including risk of reproductive toxicity. Weight-of-evidence approaches (WoE) are increasingly becoming acceptable to evaluate risk. A WoE for developmental risk of monoclonal antibodies (mAbs) was evaluated for its ability to retrospectively characterize risk and to determine the need for further in vivo testing based on the remaining uncertainty. Reproductive toxicity studies of 65 mAbs were reviewed and compared to the WoE. Developmental toxicities were absent in 52/65 (80%) mAbs. Lack of toxicity was correctly predicted in 29/52 (56%) cases. False positive and equivocal predictions were made in 9/52 (17%) and 14/52 (27%) cases. For 3/65 (5%) mAbs, the findings were equivocal. Of mAbs with developmental toxicity findings (10/65, 15%), the WoE correctly anticipated pharmacology based reproductive toxicity without any false negative predictions in 9/10 (90%) cases, and in the remaining case (1/10, 10%) an in vivo study was recommended due to equivocal WoE outcome. Therefore, this WoE approach could characterize presence and absence of developmental risk without animal studies. The current WoE could have reduced the need for developmental toxicity studies by 42% without loss of important patient information in the label.
摘要：
全球药物开发的监管指导依赖于动物研究来评估人类的安全风险。包括生殖毒性的风险。证据权重方法(WoE)越来越成为评估风险的可接受方法。对单克隆抗体（mAb）的发育风险的WoE进行了回顾性表征风险的能力评估，并根据剩余的不确定性确定是否需要进一步进行体内测试。对65mAb的生殖毒性研究进行了回顾，并与WoE进行了比较。52/65（80％）mAb中不存在发育毒性。在29/52（56％）病例中正确预测了缺乏毒性。在9/52（17％）和14/52（27％）病例中进行了假阳性和模棱两可的预测。对于3/65（5%）单克隆抗体，调查结果模棱两可。在具有发育毒性的单克隆抗体中（10/65，15%），WoE正确预测了基于药理学的生殖毒性，在9/10(90%)病例中没有任何假阴性预测，在其余病例中（1/10，10%），由于WoE结果不明确，建议进行体内研究。因此,这种WoE方法可以在没有动物研究的情况下描述发育风险的存在和不存在。当前的WoE可以将发育毒性研究的需求减少42％，而不会丢失标签中的重要患者信息。
