PD-1抑制剂致NSCLC患者肝损伤发生的影响因素分析及风险预测模型建立

Analysis of influencing factors and establishment of risk prediction model of liver injury in patients with non-small cell lung cancer caused by PD-1 inhibitorLi Youhong, WU Fan, Sun Zhouliang, Chen Qiulin (;School of Pharmacy, Fujian Medical University, Fuzhou 350108, China；The First Affiliated Hospital of Xiamen University, Xiamen 361003, China)

关键词:  PD-1抑制剂  非小细胞肺癌  风险预测模型  肝脏损伤

Key words:  PD-1 inhibitors  Non-small cell lung cancer  Risk prediction model  Liver injury

摘要:

目的 分析PD-1抑制剂致非小细胞肺癌(NSCLC )患者肝损伤发生的影响因素,构建风险预测模型并进行列线图构建。方法 回顾性分析厦门市某医院自2019年至2023年12月收治的使用PD-1抑制剂进行治疗的NSCLC患者共249例。以是否发生肝损伤将患者分为肝损组和非肝损组。分析患者临床资料,采用Logistic回归分析探讨肝损伤发生影响因素,再根据影响因素进行列线图绘制,模型检验。结果 249例患者中,共有60名(26.5%)患者发生不同程度的肝损伤,189名(73.5%)患者未发生肝损伤。年龄、用药阶段、大细胞癌类型、饮酒史、WBC/10₉^.L_-1^{是肝损伤发生的}影响因素。列线图模型验证C-index为0.746(95%CI：0.600-0.836)；模型验证kappa为0.160；决策曲线显示阈值概率在5%-73%范围内模型净获益值较高。结论 年龄、用药阶段、大细胞癌、饮酒史、WBC为PD-1抑制剂致NSCLC患者肝损伤的影响因素,本模型构建效果较好。

Abstract:

Objective To analyze the influencing factors of liver injury in patients with non-small cell lung cancer (NSCLC) induced by PD-1 inhibitors, and to construct a risk prediction model and line map. Methods: A total of 249 NSCLC patients treated with PD-1 inhibitors in a hospital in Xiamen from 2019 to December 2023 were retrospectively analyzed. The patients were divided into liver injury group and non-liver injury group according to the presence or absence of liver injury. The clinical data of the patients were analyzed, Logistic regression analysis was used to explore the influencing factors of liver injury, and then the nomogram was drawn according to the influencing factors, and the model was tested. Results Among the 249 patients, 60 (26.5%) patients had different degrees of liver injury, and 189 (73.5%) patients did not have liver injury. Age, medication stage, large cell carcinoma type, drinking history, WBC/109.L-1 were the influencing factors of liver injury. The C-index of the nomogram model was 0.746 (95%CI: 0.600-0.836). The kappa value of model validation was 0.160. The decision curve showed that the net benefit value of the model was higher when the threshold probability was in the range of 5%-73%. Conclusions: Age, stage of treatment, large cell carcinoma, drinking history, and WBC are the influencing factors of liver injury in patients with NSCLC induced by PD-1 inhibitor, and the model has a good construction effect.