The Efficacy of Trastuzumab in Animal Models of Breast Cancer: A Systematic Review and Meta-Analysis

Background
Breast cancer is the most frequent cancers and is the second leading cause of cancer death among women. Trastuzumab is an effective treatment, the first monoclonal antibody directed against the human epidermal growth factor receptor 2 (HER2). To inform the development of other effective treatments we report summary estimates of efficacy of trastuzumab on survival and tumour volume in animal models of breast cancer.
Methods
We searched PubMed and EMBASE systematically to identify publications testing trastuzumab in animal models of breast cancer. Data describing tumour volume, median survival and animal features were extracted and we assessed quality using a 12-item checklist. We analysed the impact of study design and quality and evidence for publication bias.
Results
We included data from 83 studies reporting 169 experiments using 2076 mice. Trastuzumab treatment caused a substantial reduction in tumour growth, with tumours in treated animals growing to 32.6% of the volume of tumours in control animals (95%CI 27.8%-38.2%). Median survival was prolonged by a factor of 1.45 (1.30–1.62). Many study design and quality features accounted for between-study heterogeneity and we found evidence suggesting publication bias.
Conclusion
We have found trastuzumab to be effective in animal breast cancer models across a range of experimental circumstances. However the presence of publication bias and a low prevalence of measures to reduce bias provide a focus for future improvements in preclinical breast cancer research.

背景
乳腺癌是女性群体中最常见的恶性肿瘤，亦是女性癌症相关死亡的第二大诱因。曲妥珠单抗（trastuzumab）是一款有效的治疗药物，也是首款靶向人表皮生长因子受体2（human epidermal growth factor receptor 2, HER2）的单克隆抗体。为助力其他有效治疗手段的开发，本研究报告了曲妥珠单抗在乳腺癌动物模型中对生存期与肿瘤体积的疗效汇总评估结果。

方法
研究人员系统检索了PubMed与EMBASE数据库，以筛选在乳腺癌动物模型中测试曲妥珠单抗的相关文献。提取了涵盖肿瘤体积、中位生存期及实验动物特征的相关数据，并采用12条目清单对纳入研究的质量进行评估。同时分析了研究设计与研究质量对研究结果的影响，以及发表偏倚相关证据。

结果
本研究最终纳入83项研究的数据，涵盖169项实验，共涉及2076只实验小鼠。曲妥珠单抗治疗可显著抑制肿瘤生长：给药组动物的肿瘤最终体积仅为对照组动物的32.6%（95%置信区间：27.8%~38.2%）。受试动物的中位生存期较对照组延长1.45倍（95%置信区间：1.30~1.62）。多项研究设计与质量特征可解释研究间的异质性，且本研究发现存在发表偏倚的相关证据。

结论
本研究证实，在多种实验条件下，曲妥珠单抗对乳腺癌动物模型均具有治疗有效性。但当前研究中存在发表偏倚，且减少偏倚的防控措施普及率较低，这为未来临床前乳腺癌研究的改进指明了方向。