Xiu-Ping Zhang1, Zhen-Hua Chen1, Teng-Fei Zhou2, Le-Qun Li3, Min-Shan Chen4, Tian-Fu Wen5, Jie Shi1, Wei-Xing Guo1, Meng-Chao Wu1, Wan Yee Lau6, Shu-Qun Cheng7. 1. Department of Hepatic Surgery VI, Eastern Hepatobiliary Surgery Hospital, Second Military Medical University Shanghai, China. 2. The No.313 Hospital of PLA, Huludao, Liaoning, China. 3. Department of Hepatobiliary Surgery, Affiliated Tumour Hospital of Guangxi Medical University, Nanning, China. 4. Department of Hepatobiliary Oncology of Sun Yat-sen University Cancer Center, Guangzhou, China. 5. Department of Liver Surgery and Liver Transplantation Center, West China Hospital, Sichuan University, Chengdu, Sichuan Province, 610041, China. 6. Department of Hepatic Surgery VI, Eastern Hepatobiliary Surgery Hospital, Second Military Medical University Shanghai, China; Faculty of Medicine, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, China. 7. Department of Hepatic Surgery VI, Eastern Hepatobiliary Surgery Hospital, Second Military Medical University Shanghai, China. Electronic address: chengshuqun@aliyun.com.
Abstract
BACKGROUND: Portal vein tumour thrombus (PVTT) is a significant poor prognostic factor for hepatocellular carcinoma (HCC). Patients with PVTT limited to a first-order branch or above of the main portal vein (MPV) could benefit from R0 liver resection (LR). A nomogram is needed to predict early postoperative recurrence (ER) in HCC patients with PVTT and to guide selection of these patients for adjuvant therapy to reduce postoperative recurrence risks. METHODS: HCC patients with PVTT limited to a first-order branch or above of the MPV after R0 LR as an initial therapy were included. A nomogram using data from a retrospective training cohort was developed with the Cox regression model. The model was tested in a prospective internal validation cohort and three external validation cohorts. RESULTS: Of 979 patients, 657 developed postoperative ER (67.1%). ER occurred in 165 of 264 patients (62.5%) in the training cohort, 146 of 218 patients (70.0%) in the internal validation cohort, and 204 of 284 patients (71.8%), 77 of 113 patients (68.1%), and 65 of 100 patients (65%) in the three external validation cohorts, respectively. The nomogram included the following variables: hepatitis B surface antigen (HBsAg), PVTT, HBV DNA, satellite nodules, α-fetoprotein, and tumour diameter. The ROC were 0.836, 0.763, 0.802, 0.837, and 0.846 in predicting ER in the five respective cohorts. CONCLUSION: A nomogram was developed and validated to predict postoperative ER in patients with HCC with PVTT after R0 LR. This nomogram could select appropriate patients with high ER risks for postoperative adjuvant therapy.
BACKGROUND: Portal vein tumour thrombus (PVTT) is a significant poor prognostic factor for hepatocellular carcinoma (HCC). Patients with PVTT limited to a first-order branch or above of the main portal vein (MPV) could benefit from R0 liver resection (LR). A nomogram is needed to predict early postoperative recurrence (ER) in HCC patients with PVTT and to guide selection of these patients for adjuvant therapy to reduce postoperative recurrence risks. METHODS: HCC patients with PVTT limited to a first-order branch or above of the MPV after R0 LR as an initial therapy were included. A nomogram using data from a retrospective training cohort was developed with the Cox regression model. The model was tested in a prospective internal validation cohort and three external validation cohorts. RESULTS: Of 979 patients, 657 developed postoperative ER (67.1%). ER occurred in 165 of 264 patients (62.5%) in the training cohort, 146 of 218 patients (70.0%) in the internal validation cohort, and 204 of 284 patients (71.8%), 77 of 113 patients (68.1%), and 65 of 100 patients (65%) in the three external validation cohorts, respectively. The nomogram included the following variables: hepatitis B surface antigen (HBsAg), PVTT, HBV DNA, satellite nodules, α-fetoprotein, and tumour diameter. The ROC were 0.836, 0.763, 0.802, 0.837, and 0.846 in predicting ER in the five respective cohorts. CONCLUSION: A nomogram was developed and validated to predict postoperative ER in patients with HCC with PVTT after R0 LR. This nomogram could select appropriate patients with high ER risks for postoperative adjuvant therapy.