Abstract
460
Objectives: The aim of this study was to identify a radiomics signature and generate a predictive radiomics scoring system for progression-free survival (PFS) and overall survival (OS) in patients treated with transarterial radioembolization using Yttrium-90 (90Y-TARE) for unresectable hepatocellular carcinoma (uHCC).
Methods: We retrospectively analyzed pretreatment 18F-FDG PET/CT images of 47 patients (median age, 68; range 23.5 to 86.5 years) undergoing 90Y-TARE for uHCC (31 resin, 16 glass). For each patient, based on radiomic signature from whole liver semi automatic segmentation, PFS- and OS predictive PET-radiomics scores (pPET-RadScore) were obtained using the least absolute shrinkage and selection operator (LASSO) Cox regression model. Using X-tile software, the optimal threshold of pPET-RadScore for OS and PFS served as cutoff to separate high and low-risk patients. Their survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test. Furthermore, the prognostic value of PFS- and OS pPET-RadScores, Barcelona Clinic Liver Cancer (BCLC) staging system and serum alpha-fetoprotein level was analyzed to predict PFS and OS in a multivariate analysis.
Results: Out of a total 108 radiomics features, 69 were highly correlated and excluded from the analysis. The pPET-RadScores generated from the non null coefficients of the LASSO regression from the 39 remaining radiomics features were significantly correlated with survival for i) PFS (median of 10.2 mo [95% confidence interval CI: 7.3-13.1 mo] in low-risk group [PFS pPET-RadScore < 0.13] vs. 6.1 mo [95% CI: 3.1-9.2 mo] in high-risk group [PFS pPET-RadScore > 0.13]; P = 0.007) and ii) OS (median of 30.9 mo [95% CI: 0.8-61 mo] in low-risk group [OS pPET-RadScore < -0.07] vs. 8.9 mo [95% CI: 6.7-11.1 mo] in high-risk group [OS pPET-RadScore > -0.07]; P = 0.015). The multivariate analysis confirmed PFS pPET-RadScore (hazard ratio (HR): 30.3, 95% CI: 2.89-317; P = 0.004) and OS pPET-RadScore (HR: 15.4, 95% CI: 2.97-79.7; P = 0.001) as independent negative predictors.
Conclusion: Based on pretherapy 18F-FDG PET/CT, we have identified a radiomics signature and generated a predictive radiomics scoring system as independent negative predictor for PFS and OS in patients treated with 90Y-TARE for uHCC. Research Support: None