Department of Medicine, Taipei Veterans General Hospital, Taipei: No 201, Sec. 2, Shipai Rd, Taipei, 112, Taiwan.
Faculty of Medicine, National Yang-Ming University School of Medicine, Taipei: No 155, Sec. 2, Linong St, Taipei, 112, Taiwan.
Sci Rep. 2017 Aug 11;7(1):7914. doi: 10.1038/s41598-017-08382-3.
The nomogram of the Barcelona Clinic Liver Cancer (BCLC) has accurate outcome prediction. This study aims to propose a treatment-integrated nomogram derived from BCLC for patients with hepatocellular carcinoma (HCC). A total of 3,371 patients were randomly grouped into derivation (n = 2,247) and validation (n = 1,124) sets. Multivariate Cox proportional hazards model was used to generate the nomogram from tumor burden, cirrhosis, performance status (PS) and primary anti-cancer treatments. Concordance indices and calibration plots were used to evaluate the performance of nomogram. The derivation and validation sets had the same concordance index of 0.774 (95% confidence intervals: 0.717-0.826 and 0.656-0.874, respectively). In calibration plots, survival distributions predicted by the nomogram and observed by the Kaplan-Meier method were similar at 3- and 5-year for patients from derivation and validation sets. Validation group patients divided into 10 subgroups by the original and new treatment-integrated BCLC nomogram were used to evaluate the prognostic performance of integrating primary anti-cancer treatments. Compared to the nomogram of original BCLC system, the treatment-integrated nomogram of BCLC system had larger linear trend and likelihood ratio X. In conclusion, based on the results of concordance index tests, integrating primary anti-cancer treatments into the BCLC system provides similar discriminatory ability.
巴塞罗那临床肝癌(BCLC)的列线图具有准确的预后预测能力。本研究旨在为肝癌(HCC)患者提出一种基于 BCLC 的综合治疗列线图。共纳入 3371 名患者,随机分为推导组(n=2247)和验证组(n=1124)。使用多变量 Cox 比例风险模型从肿瘤负担、肝硬化、体力状况(PS)和原发性抗癌治疗中生成列线图。一致性指数和校准图用于评估列线图的性能。推导组和验证组的一致性指数相同,为 0.774(95%置信区间:0.717-0.826 和 0.656-0.874)。在校准图中,推导组和验证组的列线图预测的生存分布与 Kaplan-Meier 方法观察到的生存分布在 3 年和 5 年时相似。验证组患者按原始和新的综合 BCLC 列线图分为 10 个亚组,用于评估综合原发性抗癌治疗的预后性能。与原始 BCLC 系统的列线图相比,BCLC 系统的综合治疗列线图具有更大的线性趋势和似然比 X。总之,基于一致性指数检验的结果,将原发性抗癌治疗纳入 BCLC 系统提供了相似的区分能力。