Influencing Factors and Nomogram Model Construction of Cancer-related Fatigue in Primary Liver Cancer Patients Post-TACE Treatment

Acad Radiol. 2025 Sep 29:S1076-6332(25)00713-5. doi: 10.1016/j.acra.2025.07.043. Online ahead of print.

ABSTRACT

RATIONALE AND OBJECTIVES: Cancer-related fatigue (CRF) is a common and persistent symptom in primary liver cancer patients undergoing transarterial chemoembolization (TACE), significantly impacting their quality of life and treatment adherence. Despite its multifactorial origins, including physiological, psychological, and treatment-related factors, effective tools for predicting CRF risk are lacking. This study aims to identify key predictors of post-TACE CRF and develop a nomogram to quantify individual risk.

METHODS: This study included 202 primary liver cancer patients treated with TACE between 2020 and 2023. The dataset was randomly divided into a training set (70%, n = 141) and a validation set (30%, n = 61). Collected variables encompassed demographic data (age, gender, BMI), disease characteristics (tumor stage, size, liver function), treatment details (TACE sessions, chemotherapeutics, embolic agents), and post-procedural psychological assessments (anxiety, depression scores). Significant predictors of CRF were identified using univariate Logistic regression on the training set and incorporated into a multivariate Logistic model to construct a nomogram. The model’s performance was evaluated via Harrell’s C-index, AUC, and calibration curves on both the training and validation sets. Internal validation was further conducted using Bootstrap resampling.

RESULTS: Tumor size (OR = 1.65, 95% CI: 1.32-2.15, P < 0.001), number of TACE sessions (OR = 1.82, 95% CI: 1.45-2.37, P < 0.001), anxiety score (OR = 1.45, 95% CI: 1.20-1.75, P < 0.001), and Child-Pugh class (OR = 2.35, 95% CI: 1.75-3.19, P < 0.001) were independent predictors of CRF. The nomogram demonstrated a C-index of 0.85 (95% CI: 0.80-0.90) and an AUC of 0.88 (95% CI: 0.83-0.93) in the training set, with a C-index of 0.83 (95% CI: 0.77-0.89) and an AUC of 0.86 (95% CI: 0.81-0.91) in the validation set. Calibration curves indicated good agreement between predicted and observed CRF outcomes, with Brier Scores of 0.1252 and 0.1292 in the training and validation sets, respectively.

CONCLUSION: This study developed a robust nomogram for predicting CRF in primary liver cancer patients undergoing TACE, providing a valuable tool for personalized risk assessment and proactive management strategies.

PMID:41027782 | DOI:10.1016/j.acra.2025.07.043