Establishment and validation of a prognostic pomogram in unresectable hepatocellular carcinoma treated with intensity modulated radiotherapy: a real world study

Background To establish a prognostic model to predict the overall survival (OS) in patients with unresectable hepatocellular carcinoma (HCC) treated with intensity modulated radiotherapy (IMRT). Methods The unresectable HCC patients treated with IMRT were retrospectively analyzed and randomized into development cohort (n = 237) and validation cohort (n = 103) in a 7:3 ratio. We developed a prognosis model with the multivariate Cox regression analysis in the development cohort to derive the predictive nomogram, which was then validated in the validation cohort. Model performance was evaluated by the c-index, the area under curve(AUC) and the calibration plot. Results A total of 340 patients were enrolled. Tumor numbers > 3 (HR = 1.69, 95% CI = 1.21–2.37), AFP ≥ 400 ng/ml (HR = 1.52, 95% CI = 1.10–2.10), PLT < 100 × 10^9(HR = 1.7495% CI = 1.11–2.73), ALP > 150U/L (HR = 1.65, 95% CI = 1.15–2.37) and prior surgery (HR = 0.63, 95% CI = 0.43–0.93) were independent prognostic factors. The nomogram based on independent factors was constructed. The c-index for OS prediction was 0.658 (95% CI, 0.647–0.804) and 0.683 (95% CI, 0.580–0.785) in the development and validation cohort, respectively. The nomogram demonstrated good discriminative ability with AUC rates of 0.726, 0.739 and 0.753 at 1-year, 2-year and 3-year models in the development cohort, and 0.715, 0.756 and 0.780 in the validation cohort, respectively. Additionally, good prognostic discrimination of the nomogram is also reflected in stratifying patients into two subgroups with distinct prognosis. Conclusions We constructed a prognostic nomogram for predicting the survival of patients with unresectable HCC treated with IMRT.