We use cookies necessary to giving you a better online experience. By using our website you consent to all cookies in accordance with our Privacy Policy.
Yes, I accept
Menu
Search Opener
A contrast enhanced CT histogram driven nomogram for predicting post radiotherapy liver regeneration

 

A contrast‑enhanced CT histogram‑driven nomogram for predicting post‑radiotherapy liver regeneration in hepatocellular carcinoma


Liu B, Xu Y, Zhang X, Qi W, Ren B, Kong W, Zhou W.


Radiat Oncol. 2025 Oct 29;20(1):159. doi: 10.1186/s13014-025-02734-4. PMID: 41163014; PMCID: PMC12574058

 

Abstract:
Background: Postoperative radiotherapy (RT) is increasingly used in hepatocellular carcinoma (HCC) patients with high-risk pathological features to reduce recurrence. However, RT may compromise liver regeneration (LR), particularly in patients with underlying hepatic dysfunction. Accurate evaluation of LR potential is therefore critical for individualized treatment planning.

Methods: In this retrospective, multicenter study, 315 HCC patients were enrolled—139 in the training cohort and 85 and 91 in two external validation cohorts. All patients underwent postoperative intensity-modulated radiotherapy (IMRT), and pre-RT portal-venous phase contrast-enhanced computed tomography (CE-CT) scans were analyzed to extract whole-liver histogram features. Patients with a ≥ 10% increase in remnant liver volume within 1 year were classified as effective liver regeneration (ELR), and the remainder as impaired liver regeneration (ILR). Independent predictors of ELR were identified using logistic regression, and a nomogram was developed and evaluated with ROC, calibration, and decision-curve analyses.

Results: Among the 315 patients, 187 (59.4%) achieved ELR and 128 (40.6%) showed ILR. Multivariate analysis identified kurtosis, mean liver dose (Liver-Dmean), standardized remnant liver volume (SRLV), and fibrosis stage as independent predictors of ELR. The nomogram demonstrated robust performance with AUCs of 0.866 (95% CI: 0.804–0.928) in the training cohort and 0.794 (95% CI: 0.698–0.889) and 0.733 (95% CI: 0.617–0.848) in the validation cohorts.

Conclusion: We established and externally validated a CE-CT histogram-based nomogram that accurately predicts LR following adjuvant IMRT in high-risk HCC patients. This model offers a clinically applicable tool to optimize postoperative management by balancing oncologic control with preservation of hepatic function.