Login / Signup

Preoperative predictors of non-transplantable recurrence after resection for early-stage hepatocellular carcinoma: application in an East Asian cohort.

Wei-Feng LiYi-Hao YenYueh-Wei LiuChih-Chi WangChee-Chien YongChih-Che Lin
Published in: Updates in surgery (2022)
A French study found that three preoperative factors (i.e. alpha-fetoprotein (AFP) > 100 ng/ml, image-diagnosed tumor number > 1, and cirrhosis) could predict non-transplantable recurrence (NTR) after liver resection (LR) for early-stage hepatocellular carcinoma (HCC). We aimed to evaluate whether this model could be applicable in an East Asian cohort from a country in which the majority of patients undergo living donor liver transplantation (LT). This retrospective study enrolled consecutive patients who underwent LR for transplantable HCC between 2011 and 2018 in our institution. The occurrence of NTR after LR was analyzed in a competing risks analysis, with death and transplantable recurrence as competing events. A total of 309 patients were included. The five-year overall survival and recurrence-free survival were 79.0% and 51.4%, respectively (median follow-up: 32.0 months). Recurrence was noted in 94 (30.4%) patients. NTR was noted in 35 (11.3%) patients. Univariate analysis showed that cirrhosis (sub-distribution hazard ratio (SHR) = 2.301, 95% CI = 1.046-5.065; p = 0.038) and image-diagnosed tumor number > 1 (SHR = 2.32; 95% CI = 1.11-4.86; p = 0.026) were associated with NTR, whereas AFP > 100 ng/ml (SHR = 1.56; 95% CI = 0.59-4.10; p = 0.37) was not associated with NTR. In the presence of 0, 1, and 2 factors (i.e. cirrhosis or image-diagnosed tumor number > 1), the NTR rates were 7.2%, 10.8%, and 29.0%, respectively. The results showed that the French model was applicable to our cohort. In the presence of two factors (i.e. cirrhosis and image-diagnosed tumor number > 1), risks and benefits of upfront LT could be discussed with the patient and the living donor.
Keyphrases
  • early stage
  • end stage renal disease
  • ejection fraction
  • newly diagnosed
  • free survival
  • peritoneal dialysis
  • deep learning
  • radiation therapy
  • machine learning
  • risk assessment
  • patients undergoing
  • data analysis