Development of a Predictive Model for Early Death in Metastatic Hepatocellular Carcinoma Based on the SEER Database
| Author | Weiguo Liu | en |
| Author | Moli Chen | en |
| Author | Xue Wei | en |
| Author | Peng Wang | en |
| Accessioned Date | 2026-10-27T03:50:41Z | |
| Issued Date | 2026-12-31 | en |
| Abstract | Background: Hepatocellular carcinoma (HCC) often has an insidious onset and high metastatic potential. Early death, defined as death within 3 months of initial diagnosis, is common in metastatic HCC (mHCC). However, few studies have used current staging systems to predict this outcome. Objectives: This study aimed to develop and internally validate a predictive model for early death in mHCC (American Joint Committee on Cancer [AJCC], 8th edition, M1 stage IVB) using the Surveillance, Epidemiology, and End Results (SEER) database and to construct a nomogram for individual risk estimation. Methods: Data from patients diagnosed with metastatic HCC between 2018 and 2023 were extracted from the SEER database (version 9.0.43.0). A total of 5,877 patients were randomly divided into training (n = 4,114) and validation (n = 1,763) cohorts at a ratio of 7:3. Univariable and multivariable logistic regression analyses were performed to identify predictors. A nomogram was subsequently constructed, and its performance was evaluated using the concordance index (C-index), calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA). Results: The early death rate was 62.5% in the training cohort and 63.4% in the validation cohort. The final multivariable model retained nine predictors: age, tumor size, alpha-fetoprotein (AFP), bone metastasis, brain metastasis, lung metastasis, marital status, radiotherapy, and chemotherapy. The nomogram achieved a C-index of 0.740 in the training cohort and 0.746 in the validation cohort, outperforming the conventional tumor-node-metastasis (TNM) staging system (C-index: 0.531). Calibration curves demonstrated close agreement between predicted and observed outcomes, and decision curve analysis (DCA) indicated net clinical benefit. A reduced model excluding radiotherapy and chemotherapy had a validation C-index of 0.605. Conclusions: This SEER-based nomogram integrates nine readily available clinical variables to predict early death in patients with mHCC and demonstrates robust discrimination, calibration, and clinical utility. It provides an intuitive quantitative tool to identify high-risk patients and guide individualized treatment and follow-up strategies. | en |
| DOI | https://doi.org/10.5812/hepatmon-171940 | en |
| URI | https://brieflands.com/journals/hepatmon/articles/171940 | en |
| Keyword | Hepatocellular Carcinoma | en |
| Keyword | Metastatic | en |
| Keyword | Early Death | en |
| Keyword | SEER Database | en |
| Keyword | Nomogram | en |
| Keyword | Prediction Model | en |
| Publisher | Brieflands | en |
| Title | Development of a Predictive Model for Early Death in Metastatic Hepatocellular Carcinoma Based on the SEER Database | en |
| Type | Research Article | en |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- hepatmon-26-1-171940-publish-pdf.pdf
- Size:
- 982.83 KB
- Format:
- Adobe Portable Document Format
- Description:
- Article/s PDF