Back to all papers

AI in predicting the macrotrabecular-massive subtype of HCC and informing treatment selection: a multi-center and prospective validation study.

July 11, 2026pubmed logopapers

Authors

Wei R,Jiang H,Zuo M,He X,Cao F,Song B,Li S,Li W,Liu W,Li C,Li X,Han J,Fu Y,Yan D,He W,Duan F,Zhao X,An C

Affiliations (20)

  • Department of Gastrointestinal Surgery, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China.
  • Department of Radiology, Functional and Molecular Imaging Key Laboratory of Sichuan Province, West China Hospital, Sichuan University, Chengdu, China.
  • Department of Minimally Invasive Interventional Therapy, Sun Yat-sen University Cancer Center, Guangzhou, China.
  • State Key Laboratory of Oncology in South China, Guangzhou, China.
  • Guangdong Provincial Clinical Research Center for Cancer, Guangzhou, China.
  • Shcool of Electronic Information, Nortwest University, Xi'an, China.
  • Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou, China.
  • Department of Interventional Therapy, Guangdong Provincial Hospital of Chinese Medicine and Guangdong Provincial Academy of Chinese Medical Sciences, Guangdong, China.
  • Department of Interventional Radiology and Vascular Surgery, The First Affiliated Hospital of Jinan University, Guangzhou, China.
  • Department of Oncology, Fifth Medical Center of Chinese PLA General Hospital, Beijing, China.
  • Interventional Radiology Department, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Tai'an, China.
  • Department of Interventional Therapy, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
  • Department of Oncology, Beijing Luhe Hospital Affiliated to Capital Medical University, Beijing, China.
  • Department of Gastrointestinal Surgery, Xiang'an Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, China. [email protected].
  • Department of Oncology, Fifth Medical Center of Chinese PLA General Hospital, Beijing, China. [email protected].
  • Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, China. [email protected].
  • Department of Minimally Invasive Interventional Therapy, Sun Yat-sen University Cancer Center, Guangzhou, China. [email protected].
  • State Key Laboratory of Oncology in South China, Guangzhou, China. [email protected].
  • Guangdong Provincial Clinical Research Center for Cancer, Guangzhou, China. [email protected].
  • Department of Oncology, Fifth Medical Center of Chinese PLA General Hospital, Beijing, China. [email protected].

Abstract

The macrotrabecular-massive (MTM) subtype of hepatocellular carcinoma (HCC) is associated with aggressive biology and poor prognosis. We aimed to develop a CT-based artificial intelligence model (DeepCT-MTM) for the noninvasive prediction of MTM-HCC and investigate its prognostic utilities as well as biological underpinnings. A total of 3118 patients with HCC were included from 20 tertiary-care hospitals. DeepCT-MTM was developed and validated among 832 patients with early-stage HCC undergoing resection (the resection set) and extrapolated to 2286 patients (including 480 prospectively-collected ones) with intermediate/advanced-stage HCC receiving IATs. DeepCT-MTM's predictive performance for MTM-HCC was evaluated using the area under the receiver operating characteristic curve (AUC), and its prognostic values were investigated for progression-free survival (PFS) and overall survival (OS). In the external test cohort of the resection set, DeepCT-MTM predicted MTM-HCC with an AUC of 0.845. The DeepCT-MTM-predicted high-risk group had worse PFS and OS across all IAT sets (all P < 0.05).. DeepCT-MTM is effective for noninvasively predicting MTM-HCC and may help selecting patients who benefit from a combination of IAT with immunotherapy and anti-angiogenic therapy. However, prospective validations are warranted for these hypothesis-generating findings.

Topics

Journal Article

Ready to Sharpen Your Edge?

Subscribe to join 11k+ peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.