基于列线图构建和验证肝细胞癌伴门静脉癌栓患者肝切除术后的预后评估模型

目的 分析影响肝细胞癌伴门静脉癌栓（PVTT-HCC）患者肝切除术后预后的影响因素，并基于列线图模型构建和验证预后评估模型。方法 本研究为回顾性队列研究，选择2008年1月—2017年11月在本院行肝切除术的PVTT-HCC患者为研究对象，随访截至2021年1月。主要预测结局为1、3、5年总生存率。按照7∶3的比例将患者随机分为训练集和验证集，在训练集中采用Cox比例风险回归分析影响预后的影响，并基于影响因素构建列线图模型。同时在训练集和验证集中采用C-index评价模型的区分度，一致性曲线评估模型的校准度。结果 共231例患者符合纳入排除标准纳入分析，其中训练集162例，验证集69例。Cox比例风险回归模型显示，AFP≥400 μg/L、AST≥40 U/L、ALP≥80 U/L、肿瘤个数>1个及肿瘤包膜不完整是影响预后的危险因素。在训练集中，列线图模型预测1、3、5年总生存率的C-index分别为0.826（95%CI: 0.791~0.861）、0.818（95%CI：0.782~0.854）、0.781（95%CI：0.742~0.820），在验证集中分别为0.814（95%CI：0.777~0.851）、0.798（95%CI：0.758~0.837）、0.769（95%CI：0.728~0.810）。校正曲线显示列线图模型在训练集和验证集均有较好的校准度。结论 本研究构建的列线图模型可准确预测PVTT-HCC患者的预后。

Construction and verification of the prognosis evaluation model for patients in hepatocellular carcinoma with portal vein tumor thrombus after hepatectomy based on the nomogram

Objective To analyze the factors affecting the prognosis of patients with hepatocellular carcinoma with portal vein tumor thrombosis (PVTT-HCC) after hepatectomy, and to construct and verify the prognostic evaluation model based on the nomogram model. Methods This retrospective cohort study selected patients with PVTT-HCC who underwent hepatectomy in Cangzhou Central Hospital from January 2008 to November 2017 as the research subjects, and the follow-up was as of January 2021. The main predicted outcome was the 1-, 3-, and 5-year overall survival rates. The patients were randomly divided into training cohort and validation cohort by a ratio of 7∶3. The Cox proportional hazard regression analysis was used to analyze the impact of prognosis in the training cohort, and a nomogram model was constructed based on the influencing factors. Meanwhile, C-index was used to evaluate the distinction of the model in both the training cohort and the validation cohort, and the consistency curve was used to evaluate the calibration of the model. Results A total of 231 patients met the inclusion and exclusion criteria, including 162 cases of the training cohort and 69 cases of the validation cohort. The Cox proportional hazard regression model showed that AFP≥400 μg/L, AST≥40 U/L, ALP≥80 U/L, number of tumors >1 and tumor envelope incompleteness were risk factors affecting the prognosis. The C-index of the nomogram model predicted that the 1-, 3-, and 5-year overall survival rates were 0.826 (95%CI: 0.791-0.861), 0.818 (95%CI: 0.782-0.854), and 0.781 (95%CI: 0.742-0.820), respectively, in the training cohort, and 0.814 (95%CI: 0.777-0.851), 0.798 (95%CI: 0.758-0.837), and 0.769 (95%CI: 0.728-0.810), respectively, in the validation cohort. The calibration curves showed that the nomogram model had a good calibration degree in both the training cohort and the validation cohort. Conclusions The nomogram model constructed in this study can accurately predict the prognosis of patients with PVTT-HCC.