Yangyue Zhang, Hye Won Lee, Huapeng Lin, Emmanuel Tsochatzis, Salvatore Petta, Elisabetta Bugianesi, Masato Yoneda, MING-HUA ZHENG, Hannes Hagström, Jérôme Boursier, José Luís Calleja, George Boon-Bee Goh, W.K. Chan, Rocio Gallego-Durán, Arun J. Sanyal, Victor de Lédinghen, Philip N. Newsome, Jian-Gao Fan, Michelle Lai, Charles Fournier, Grace Lai-Hung Wong, Grazia Pennisi, Angelo Armandi, Atsushi Nakajima, Wen-Yue Liu, Ying Shang, Marc de Saint-Loup, Elba Llop, Kevin Teh, Carmen Lara-Romero, Amon Asgharpour, Sara Mahgoub, Mandy Sau-Wai Chan, Clémence M. Canivet, Manuel Romero-Gomez, Seung Up Kim, Vincent Wai-Sun Wong, Laurent Castéra, Terry Cheuk-Fung Yip
BACKGROUND AND AIMS: Liver stiffness measurement (LSM) by vibration-controlled transient elastography correlates well with liver-related events (LREs), but previous head-to-head comparisons with liver histology were limited by small sample size. This study aimed to compare the prognostic performance of LSM and histology for predicting LREs in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). APPROACH AND RESULTS: We analyzed data from 3532 metabolic dysfunction-associated steatotic liver disease patients (mean age 51.9 y, 57.3% male) in the multicenter vibration-controlled transient elastography-prognosis cohort, all having undergone both LSM and liver biopsy. The primary outcome was LREs, defined as hepatic decompensation, liver transplantation, or liver-related death. Secondary outcomes included HCC and decompensation analyzed separately. Median baseline LSM was 8.8 kPa; 33.5% had F3-F4 fibrosis. Over a median follow-up of 56.6 months, 126 patients (3.6%) developed LREs (123 decompensation). LSM and histology demonstrated similar prognostic performance for LREs, with comparable 5-year AUROC values (0.870 vs. 0.869), integrated AUROCs (0.878 vs. 0.852), and integrated precision-recall curves (0.137 vs. 0.0.68). The 5-year integrated Brier scores were also similar (1.389% vs. 1.391%), and the integrated discrimination improvement index showed no significant difference. Similar results were found across all the outcomes, time points, and sensitivity analyses. CONCLUSIONS: In this large metabolic dysfunction-associated steatotic liver disease cohort, LSM by vibration-controlled transient elastography showed comparable prognostic accuracy to histology. As a noninvasive tool, LSM may serve as an alternative surrogate endpoint in clinical trials.