Data from A Machine Learning Algorithm Facilitates Prognosis Prediction and Treatment Selection for Barcelona Clinic Liver Cancer Stage C Hepatocellular Carcinoma

• DOI: 10.1158/1078-0432.c.7309446
• Publication Date: 2024-07-01T07:23:57Z
• Authors: Ji W. Han; Soon K. Lee; Jung H. Kwon; Soon W. Nam; Hyun Yang; Si H. Bae; Ji H. Kim; Heechul Nam; Chang W. Kim; Hae L. Lee; Hee Y. Kim; Sung W. Lee; Ahlim Lee; U I. Chang; Do S. Song; Seok-Hwan Kim; Myeong J. Song; Pil S. Sung; Jong Y. Choi; Seung K. Yoon; Jeong W. Jang
• Keywords: Liver Cancer

<div>AbstractPurpose:<p>Given its heterogeneity and diverse clinical outcomes, precise subclassification of Barcelona Clinic Liver Cancer stage C (BCLC-C) hepatocellular carcinoma (HCC) is required for appropriately determining patient prognosis selecting treatment.</p>Experimental Design:<p>We recruited 2,626 patients with BCLC-C HCC from multiple centers, comprising training/test (<i>n</i> = 1,693) validation cohorts 933). The XGBoost model was chosen maximum performance among the machine learning (ML) models. Patients were categorized into low-, intermediate-, high-, very high-risk subgroups based on estimated prognosis, this named CLAssification via Machine (CLAM-C).</p>Results:<p>The areas under receiver operating characteristic curve CLAM-C predicting 6-, 12-, 24-month survival 0.800, 0.831, 0.715, respectively—significantly higher than those conventional models, which consistent in cohort. four had significantly different median overall survivals, difference maintained various treatment modalities. Immune-checkpoint inhibitors transarterial therapies associated better tyrosine kinase (TKI) low- intermediate-risk subgroups. In cases first-line systemic therapy, identified atezolizumab–bevacizumab as best particularly group. later-line nivolumab TKIs low-to-intermediate-risk subgroup, whereas high- to subgroup.</p>Conclusions:<p>ML modeling effectively subclassified HCC, potentially aiding allocation. Our study underscores potential utilization ML terms prognostication allocation HCC.</p></div>