研究来那替尼治疗乳腺癌引起肝毒性和胃肠道不良反应风险。

检索Cochrane、PubMed、Web of Science、中国知网、维普及万方等相关数据库，检索时间为从建库以来至2018年3月1日，收集关于来那替尼治疗乳腺癌且描述了肝毒性[丙氨酸氨基转移酶(ALT)或天门冬氨酸氨基转移酶(AST)升高]和胃肠道不良反应(包括腹泻、呕吐或恶心)等不良事件的随机对照临床研究。不良反应按严重程度分为1～5级，其中所有级别包括1～5级；严重级别包括3级及以上。由2名研究人员独立完成文献筛选、数据提取，运用Cochrane手册推荐的风险评价工具对文献质量进行评价。采用相对危险度(relative risk, RR)和95%置信区间(confidence interval, CI)作为效应量；采用Q检验分析数据的异质性，并根据异质性结果选择固定效应模型或随机效应模型。

本研究共纳入4项RCT研究，共3 745例患者。其中2项研究为来那替尼单用，另外2项研究为来那替尼联用紫杉醇。来那替尼组的所有级别ALT与AST升高发生率与对照组比较，差异无统计学意义(ALT：RR＝1.37，95%CI：0.50～3.76，P＝0.54; AST: RR＝1.19，95%CI：0.36～3.92，P＝0.78)，但严重级别的ALT升高发生率来那替尼组高于对照组(RR＝3.63，95%CI：1.58～8.36，P<0.01)，来那替尼组的严重级别AST升高发生率与对照组比较，差异无统计学意义(RR＝2.08，95%CI：0.80～5.43，P＝0.13)。关于胃肠道不良反应，所有级别与严重级别的腹泻发生率在来那替尼组均高于对照组(所有级别：RR＝2.06，95%CI：1.38～3.08，P<0.01；严重级别：RR＝8.77, 95%CI：2.91～26.40，P<0.01)；所有级别与严重级别的呕吐发生率在来那替尼组均高于对照组(所有级别：RR＝2.13，95%CI：1.43～3.18, P<0.01；严重级别：RR＝6.22，95%CI：3.16～12.27, P<0.01)。所有级别和严重级别的恶心发生率，来那替尼组与对照组比较，差异无统计学意义(所有级别：RR＝1.28, 95%CI：0.81～2.04，P＝0.29;严重级别：RR＝3.10, 95%CI: 0.87～11.00，P＝0.08)。

来那替尼用于治疗乳腺癌后，严重级别的ALT升高、腹泻和呕吐的发生率增加，AST升高和恶心的发生率没有明显变化，因此，需要密切监测患者的肝功能和胃肠道功能。

To investigate the risk of hepatic toxicity and gastrointestinal events in breast cancer patients treated with neratinib.

The Cochrane, PubMed, Web of Science, CNKI, VIP and Wanfang databases were searched for randomized controlled trials in breast cancer patients treated with neratinib from the earliest data available to March 1, 2018. The enrolled studies described hepatic toxicity such as elevated alanine aminotransferase(ALT) or elevated aspartate aminotransferase(AST) or gastrointestinal events induding diarrhea, vomiting and nausea. All adverse events were divided into 5 grades(1-5). The adverse events at all levels included all 5 grades and high-level adverse events refered to the events at grade 3 and above. Two researchers independently completed literature retrieval and data extraction, and evaluated the quality of the included studies using the recommended bias tools of the Cochrane handbook. The relative risk (RR) and 95% confidence interval (CI) were used as effect quantities and Q test was used to analyze the heterogeneity of the data. A fixed effect model or a random effect model was chosen according to the results of the heterogeneity.

Four studies were included, among which neratinib monotherapy was used in two studies and neratinib combined with paclitaxel in other two studies. The results showed that there was no significant difference in the incidence of ALT and AST increase at all levels between neratinib group and control group (ALT: RR=1.37, 95%CI: 0.50-3.76, P=0.54; AST: RR=1.19, 95%CI: 0.36-3.92, P=0.78). The incidence of high-level elevation of ALT was significantly higher in neratinib group than in control group (RR=3.63, 95%CI: 1.58-8.36, P<0.01), while no significant difference was found in the incidence of high-level elevation of AST (RR=2.08, 95%CI: 0.80-5.43, P=0.13). As to gastrointestinal events, the incidences of nausea at all levels and high level were significantly higher in neratinib group than in control group(at all levels: RR=2.06, 95%CI: 1.38-3.08, P<0.01; high-level: RR=8.77, 95%CI: 2.91-26.40, P<0.01); the incidences of vomiting at all levels and high level were significantly higher in neratinib group than in control group(at all levels: RR=2.13, 95%CI: 1.43-3.18, P<0.01; high-level: RR=6.22, 95%CI: 3.16-12.27, P<0.01). The incidence of nausea at all levels and high level presented no significant difference between neratinib group and control group (at all levels: RR=1.28, 95%CI: 0.81-2.04, P=0.29; high-level: RR=3.10, 95%CI: 0.87-11.00, P=0.08).

The application of neratinib in breast cancer patients will increase the incidences of high-level ALT increase, diarrhea and vomiting, but no significant changes are observed in AST increase and nausea. Therefore, the liver function and gastrointestinal function should be closely monitored if neratinib is used against breast cancer.