Abstract
Distant-metastatic hepatocellular carcinoma (DM-HCC) represents the terminal stage of liver cancer, characterized by the spread of cancer cells from the primary tumor site in the liver to distant organs, such as the lungs, bones, or lymph nodes. This study aims to develop a predictive model for assessing cancer-specific survival (CSS) in patients with DM-HCC. We extracted data from the surveillance, epidemiology, and end results (SEER) database for 651 patients initially diagnosed with DM-HCC between 2010 and 2017. The patients were randomly allocated to the training and validation cohorts at a 7:3 ratio. Cox regression analysis was used to identify predictors of CSS. The nomogram was validated using concordance index (C-index), receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis. Multivariate Cox regression analysis showed that alpha-fetoprotein (AFP) expression, grade, T stage, lung metastasis, surgery, radiation, and chemotherapy were independent predictors of CSS. A predictive model was established based on these factors, and a nomogram method was used for visualization. The C-index of the nomogram were 0.721 and 0.726 in the training and validation sets, respectively. In the training set, the time-dependent area under the curve (AUC) values were 0.796, 0.797, and 0.766 at 4, 8, and 12 months, respectively. In the validation set, time-dependent AUC values were 0.767, 0.771, and 0.834 at 4, 8, and 12 months, respectively. Calibration curves indicated good consistency between the actual observed values and the nomogram predictions for CSS, while decision curve analysis demonstrated that the nomogram has good clinical application value. Simultaneously, a risk classification system was established, perfectly categorizing DM-HCC patients into 2 risk groups: low- and high-risk groups. We developed a nomogram with high accuracy for predicting the prognosis of patients with DM-HCC, which is important for enhancing the survival and prognosis of this patient population.