Objective To identify the risk factors for early death in metastatic hepatocellular carcinoma (mHCC) patients and construct a clinical prediction nomogram for mHCC patients based on clinical information and survival data of mHCC patients from the Surveillance, Epidemiology, and End Results (SEER) database.

Methods 1 691 eligible mHCC patients were downloaded and selected from the SEER database. Using R language, a nomogram was constructed based on the risk factors in multifactorial logistic regression. The predictive performance of the nomogram was evaluated through receiver operating characteristic curve (ROC) curve, calibration curve, and clinical decision curve.

Results The risk factors for early death in mHCC patients included: AFP (P<0.01), T stage (P<0.05), N stage (P<0.01), radiotherapy (P<0.01), chemotherapy (P<0.001), surgery (P<0.01), and lung metastasis (P<0.001). A nomogram prediction model was constructed based on these seven variables. Validation results showed that the model achieved area under curve (AUC) values of 0.819 and 0.788 in the training and validation sets, respectively, indicating good discriminative ability. The calibration curve demonstrated a high consistency between the predictions and actual observations. Decision curve analysis further confirmed its significant clinical net benefit.

Conclusions The nomogram can calculate the probability of early death in mHCC patients. It assists clinical doctors in identifying patients with high-risk mHCC for drawing up personalized clinical decisions.