TY  - JOUR
AU  - Rabasco Meneghetti, Asier
AU  - Campani, Claudia
AU  - Roux, Charles
AU  - Carrero, Zunamys Itzel
AU  - Popica, Dan-Adrian
AU  - Amaddeo, Giuliana
AU  - Lequoy, Marie
AU  - Hollande, Clémence
AU  - Mouri, Sarah
AU  - Wagner, Mathilde
AU  - Plaforet, Vincent
AU  - Sidali, Sabrina
AU  - Ronot, Maxime
AU  - Rudler, Marika
AU  - Luciani, Alain
AU  - Sutter, Olivier
AU  - Spitzer, Eleonore
AU  - Regnault, Hélène
AU  - El Mouhadi, Sanaâ
AU  - Ozenne, Violaine
AU  - Ganne-Carrié, Nathalie
AU  - Bouattour, Mohamed
AU  - Nault, Jean Charles
AU  - Thabut, Dominique
AU  - Kather, Jakob Nikolas
AU  - Allaire, Manon
TI  - Detection of Esophageal Varices and Prediction of Hepatic Decompensation in Unresectable Hepatocellular Carcinoma using AI: AI Detection of Varices and Decompensation.
JO  - Journal of hepatology
VL  - nn
SN  - 0168-8278
CY  - Amsterdam [u.a.]
PB  - Elsevier Science
M1  - DKFZ-2026-00359
SP  - nn
PY  - 2026
N1  - #DKTKZFB9# / #NCTZFB9# / epub
AB  - In hepatocellular carcinoma (HCC) with cirrhosis, portal hypertension (PHT) worsens outcomes. Esophagogastroduodenoscopy (EGD), current screen for esophageal varices (EVs), is invasive and delays therapy. We aimed to develop and externally validate non-invasive models to detect EV and predict hepatic decompensation (bleeding, ascites or hepatic encephalopathy), a major cause of HCC mortality, using routine contrast-enhanced CT (CECT) and clinical data.This multicenter retrospective study included 489 patients with unresectable HCC treated with Atezolizumab-Bevacizumab (AtezoBev) from five French centers, split into a development (n=279) and external validation (n=210) cohorts. Arterial-phase CECTs were processed through a Deep Learning pipeline using a foundation model (HepatoSageCT). Logistic and Cox models generated clinical models and combined models integrating the HepatoSageCT scores with key clinical variables for EVs and hepatic decompensation. Performance was assessed using AUCROC, sensitivity, specificity, concordance index and cause-specific hazard ratio.Portosystemic shunts (PSS) at imaging identified EVs with AUCROC of 0.78, increasing to 0.84 when combined with HepatoSageCT. A decision algorithm incorporating PSS and HepatoSageCT missed 4.2
KW  - deep learning (Other)
KW  - hepatic decompensation (Other)
KW  - hepatocellular carcinoma (Other)
KW  - portal hypertension (Other)
KW  - portosystemic shunt (Other)
KW  - varices (Other)
LB  - PUB:(DE-HGF)16
C6  - pmid:41679555
DO  - DOI:10.1016/j.jhep.2026.01.021
UR  - https://inrepo02.dkfz.de/record/309874
ER  -