20 - Incidence and Factors Predictive of Recurrent Rupture in People with Rupture of Unresectable Hepatocellular Carcinoma

Sunday, April 12, 2026

4:12 PM - 4:21 PM ET

Presenting Author(s)

Tianfan Pan, MD (he/him/his)

Department of Interventional Radiology
Jiangyin People's Hospital, China (People's Republic)

Author/Co-author(s)

YW

Yong Wang, PhD

Center of Interventional Radiology and Vascular Surgery
Zhongda Hospital, China (People's Republic)
BS

Bai Sun, MD

Department of Oncology
The First People’s Hospital of Changzhou, China (People's Republic)
YG

Yang Gao, MD

Department of Radiology
The Affiliated Cancer Hospital of Nanjing Medical University, China (People's Republic)
XX

Xinjian Xu, MD

Department of Interventional Radiology
Jiangyin People’s Hospital, China (People's Republic)
XH

Xiangzhong Huang, MD

Department of Interventional Radiology
Jiangyin People’s Hospital, China (People's Republic)
JL

Jian Lu, PhD

Center of Interventional Radiology and Vascular Surgery
Zhongda Hospital, China (People's Republic)
JG

Jinhe Guo, PhD

Center of Interventional Radiology and Vascular Surgery
Zhongda Hospital, China (People's Republic)

Purpose: Spontaneous rupture hemorrhage of hepatocellular carcinoma (HCC) has a low incidence and a high risk of death. Unresectable rupture hepatocellular (rHCC) carcinoma has a high early hemostasis rate after transcatheter arterial embolization (TAE). However, there is still a risk of recurrent rupture (RR) during subsequent treatment. There is a lack of studies on the Incidence and factors predictive of recurrent rupture in these patients. Our first aim was to describe the incidence of RR in people with unresectable rHCC. The second aim was to identify RR risk factors and afterwards verify them in a validation cohort.

Materials and Methods: This is a multicenter, retrospective observational study evaluating risk factors for RR in 94 people with unresectable rHCC from Jan 2010 to Dec 2023 in four medical centers. In the training cohort, the potential value of parameters to predict RR was analyzed by univariate and multivariate Cox analysis. Findings were validated in 79 individuals with unresectable rHCC.

Results: A total of 173 patients were included in this study. Of the 94 participants in training cohort, 17 (18.1%) presented recurrent rupture during follow-up (cumulative incidence: 6.6, 13.2, 16.3, and 24.6% at 3, 6, 12, and 24 months, respectively; the mean and median time: 9.6 and 7.0 months, respectively). In addition, MELD score > 15, tumor size > 10cm and received targeted therapy were the independent factors predicting RR (hazard ratio 8.40, 95% CI 2.25-34.10, p= 0.002; hazard ratio 3.36, 95% CI 1.05-10.74, p= 0.041; hazard ratio 5.72, 95% CI 1.78-18.43, p= 0.003, respectively). Of the 79 participants in validation cohort, 14 individuals (17.7%) presented RR during follow-up (cumulative incidence: 6.9, 15.6, 18.2, and 32.8% at 3, 6, 12, and 24 months, respectively; the mean and median time: 10.9 and 7.0 months, respectively). The predictive value of these factors was showed good stability and prediction accuracy when assessed externally in validation cohort (C-index: 0.872, 95% CI: 0.773-0.972).

Conclusion: People with unresectable rHCC are at risk of recurrent rupture. Factors of MELD score > 15, tumor size > 10cm and received targeted therapy are independent risk factors for recurrent rupture.