预测中晚期(BCLC B/C期)肝细胞癌肝切除术后生存与复发的列线图开发与验证
本研究基于中国3家医院2010至2020年间接受治愈性肝切除术的中晚期肝细胞癌患者,开发并外部验证了预测生存与复发的列线图。列线图在训练队列中预测生存和复发的C指数分别为0.777(95%CI 0.759-0.800)和0.719(95%CI 0.697-0.742),优于4种常规分期系统(P < .001),并在两个外部验证队列中得到良好验证。
要点
- 列线图预测生存的C指数为0.777(95%CI 0.759-0.800),预测复发的C指数为0.719(95%CI 0.697-0.742)。
- 列线图预测性能优于4种常规肝细胞癌分期系统(P < .001),并能准确区分低、中、高风险亚组。
- 列线图结果在两个外部验证队列中得到良好验证,或有助于个体化手术决策及辅助治疗策略制定。
核心结论
- 开发并验证了预测中晚期肝细胞癌患者生存与复发的列线图,该结果与巴塞罗那临床肝癌分期系统的手术指南相矛盾。
- 列线图或可帮助临床医生制定个体化手术决策、评估长期预后并制定新辅助/辅助抗复发治疗策略。
研究设计与样本
- 多中心队列研究,分析2010至2020年间中国3家医院接受治愈性肝切除术的中晚期肝细胞癌患者。
- 东方肝胆外科医院队列作为训练队列构建列线图,吉林市第一医院和福建孟超肝胆医院队列作为外部验证队列。
- 通过单变量和多变量Cox回归分析确定生存和复发的独立术前预测因素,使用一致性指数和校准曲线衡量预测准确性。
- 共1328例患者符合纳入标准。
主要发现
- 预测生存和复发的列线图分别使用10个和6个独立变量构建。
- 训练队列中列线图预测生存的C指数为0.777(95%CI 0.759-0.800),预测复发的C指数为0.719(95%CI 0.697-0.742)。
- 列线图预测性能优于4种常规肝细胞癌分期系统(P < .001)。
- 列线图可将风险准确分层为低、中、高亚组,结果在两个外部验证队列中得到良好验证。
临床怎么用
- 作者认为列线图或可帮助临床医生制定个体化手术决策、评估长期预后并制定新辅助/辅助抗复发治疗策略。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Despite the Barcelona Clinic Liver Cancer system discouraging hepatectomy for intermediate/advanced hepatocellular carcinoma, the procedure is still performed worldwide, particularly in Asia. This study aimed to develop and validate nomograms for predicting survival and recurrence for these patients. METHODS: We analyzed patients who underwent curative-intent hepatectomy for intermediate/advanced hepatocellular carcinoma between 2010 and 2020 across 3 Chinese hospitals. The Eastern Hepatobiliary Surgery Hospital cohort was used as the training cohort for the nomogram construction, and the Jilin First Hospital and Fujian Mengchao Hepatobiliary Hospital cohorts served as the external validation cohorts. Independent preoperative predictors for survival and recurrence were identified through univariable and multivariable Cox regression analyses. Predictive accuracy was measured using the concordance index and calibration curves. The predictive performance between nomograms and conventional hepatocellular carcinoma staging systems was compared. RESULTS: A total of 1,328 patients met the inclusion criteria. The nomograms for predicting survival and recurrence were developed using 10 and 6 independent variables, respectively. Nomograms' concordance indices in the training cohort were 0.777 (95% confidence interval 0.759-0.800) and 0.719 (95% confidence interval 0.697-0.742) for survival and recurrence, outperforming 4 conventional staging systems (P < .001). Nomograms accurately stratified risk into low, intermediate, and high subgroups. These results were validated well by 2 external validation cohorts. CONCLUSION: We developed and validated nomograms predicting survival and recurrence for patients with intermediate/advanced hepatocellular carcinoma, contradicting Barcelona Clinic Liver Cancer surgical guidelines. These nomograms may facilitate clinicians to formulate personalized surgical decisions, estimate long-term prognosis, and strategize neoadjuvant/adjuvant anti-recurrence therapy.
引用
- Diao YK, Sun L, Wang MD, et al. Development and validation of nomograms to predict survival and recurrence after hepatectomy for intermediate/advanced (BCLC stage B/C) hepatocellular carcinoma. Surgery. 2024. PMID: 38734502. doi:10.1016/j.surg.2024.03.028
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。