器官芯片等非动物测试方法正加速替代传统动物实验

非动物替代方法（NAMs）在药物开发和毒理学测试中的应用正在加快推进。2022年末通过的FDA现代化法案2.0明确授权在新药临床前研究中使用这些替代技术1，随后美国监管部门进一步明确了政策方向：2025年FDA承诺将动物研究变为"例外而非常规"，2026年9月FDA颁布新规则，正式用"非临床测试"替代"动物测试"的表述1。美国国家卫生研究院也在2025年宣布，申请动物模型研究经费的研究人员必须同时纳入非动物研究1。欧委会、英国、经合组织等也相继宣布了逐步淘汰动物测试的计划1。

这一转变的科学基础正在加强。现有数据显示，92%进入美国临床试验的药物最终未能获批上市，部分原因在于动物实验未能准确预测人体反应1。相比之下，新兴技术展现出更高的预测准确性：Emulate公司开发的肝脏芯片能正确识别约87.5%在动物试验中安全通过但对人肝脏有毒性的药物1；由牛津大学和Janssen制药团队合作开发的心脏细胞计算模拟在预测心律不齐方面的准确率为89%，高于传统动物研究的75%1。

尽管技术进步明显，但验证、标准化、文化转变和人才培训仍是实现广泛采用的关键挑战1。历史经验表明这一过程需要时间——1990年代开发的单核细胞激活试验直到2010年才被欧洲药典正式接受作为兔热原试验的替代1。目前，约30%的动物实验用于早期药物开发和监管测试，其余70%用于基础生物学研究1。

Non-animal methods (NAMs) such as organ-on-chip technology are increasingly replacing traditional animal testing in drug development and toxicology. The FDA Modernization Act 2.0, passed in late 2022, explicitly authorized the use of NAMs to substitute for animal experiments in preclinical drug research 1. Building on this momentum, the FDA committed in 2025 to making animal research an "exception rather than the norm" and introduced new rules in September 2026 that replace the term "animal testing" with "non-clinical testing" 1.

The shift reflects growing recognition that animal models often fail to predict human outcomes. Among drugs entering U.S. clinical trials, 92% ultimately fail to gain approval, partly because animal testing does not accurately forecast human responses 1. In contrast, emerging technologies demonstrate superior predictive power. Emulate's liver-on-chip correctly identified approximately 87.5% of drugs that passed animal trials but proved toxic to human livers 1, while a computational heart cell model developed by Oxford University and Janssen Pharmaceuticals achieved 89% accuracy in predicting arrhythmias—outperforming the 75% accuracy of animal research 1.

Regulatory bodies worldwide are accelerating this transition. The U.S. National Institutes of Health announced in 2025 that researchers seeking funding for animal model studies must also incorporate non-animal research 1. The European Commission, United Kingdom, and Organisation for Economic Co-operation and Development have similarly announced plans to phase out animal testing 1. However, widespread adoption faces significant barriers. Validation, standardization, cultural shifts, and workforce training remain critical challenges, and historical precedent suggests the process can be lengthy—a monocyte activation test developed in the 1990s took until 2010 to gain acceptance by the European Pharmacopoeia as a replacement for the rabbit pyrogen test 1. Currently, approximately 30% of animal experiments support early-stage drug development and regulatory testing, with the remaining 70% dedicated to basic biological research 1.