Background: Although transarterial chemoembolization(TACE) is the first-line therapy for intermediatestage hepatocellular carcinoma(HCC), it is not suitable for all patients. This study aimed to determine how to selec...Background: Although transarterial chemoembolization(TACE) is the first-line therapy for intermediatestage hepatocellular carcinoma(HCC), it is not suitable for all patients. This study aimed to determine how to select patients who are not suitable for TACE as the first treatment choice. Methods: A total of 243 intermediate-stage HCC patients treated with TACE at three centers were retrospectively enrolled, of which 171 were used for model training and 72 for testing. Radiomics features were screened using the Spearman correlation analysis and the least absolute shrinkage and selection operator(LASSO) algorithm. Subsequently, a radiomics model was established using extreme gradient boosting(XGBoost) with 5-fold cross-validation. The Shapley additive explanations(SHAP) method was used to visualize the radiomics model. A clinical model was constructed using univariate and multivariate logistic regression. The combined model comprising the radiomics signature and clinical factors was then established. This model’s performance was evaluated by discrimination, calibration, and clinical application. Generalization ability was evaluated by the testing cohort. Finally, the model was used to analyze overall and progression-free survival of different groups. Results: A third of the patients(81/243) were unsuitable for TACE treatment. The combined model had a high degree of accuracy as it identified TACE-unsuitable cases, at a sensitivity, specificity, and area under the receiver operating characteristic curve(AUC) of 0.759, 0.885, 0.906 [95% confidence interval(CI): 0.859-0.953] in the training cohort and 0.826, 0.776, and 0.894(95% CI: 0.815-0.972) in the testing cohort, respectively. Conclusions: The high degree of accuracy of our clinical-radiomics model makes it clinically useful in identifying intermediate-stage HCC patients who are unsuitable for TACE treatment.展开更多
Background:Radiofrequency ablation(RFA)is one of the effective therapeutic modalities in patients with hepatocellular carcinoma(HCC).However,there is no proper method to evaluate the HCC response to RFA.This study aim...Background:Radiofrequency ablation(RFA)is one of the effective therapeutic modalities in patients with hepatocellular carcinoma(HCC).However,there is no proper method to evaluate the HCC response to RFA.This study aimed to establish and validate a clinical prediction model based on dual-energy com-puted tomography(DECT)quantitative-imaging parameters,clinical variables,and CT texture parameters.Methods:We enrolled 63 patients with small HCC.Two to four weeks after RFA,we performed DECT scanning to obtain DECT-quantitative parameters and to record the patients’clinical baseline variables.DECT images were manually segmented,and 56 CT texture features were extracted.We used LASSO al-gorithm for feature selection and data dimensionality reduction;logistic regression analysis was used to build a clinical model with clinical variables and DECT-quantitative parameters;we then added texture features to build a clinical-texture model based on clinical model.Results:A total of six optimal CT texture analysis(CTTA)features were selected,which were statis-tically different between patients with or without tumor progression(P<0.05).When clinical vari-ables and DECT-quantitative parameters were included,the clinical models showed that albumin-bilirubin grade(ALBI)[odds ratio(OR)=2.77,95%confidence interval(CI):1.35-6.65,P=0.010],λAP(40-100 keV)(OR=3.21,95%CI:3.16-5.65,P=0.045)and IC AP(OR=1.25,95%CI:1.01-1.62,P=0.028)were asso-ciated with tumor progression,while the clinical-texture models showed that ALBI(OR=2.40,95%CI:1.19-5.68,P=0.024),λAP(40-100 keV)(OR=1.43,95%CI:1.10-2.07,P=0.019),and CTTA-score(OR=2.98,95%CI:1.68-6.66,P=0.001)were independent risk factors for tumor progression.The clinical model,clinical-texture model,and CTTA-score all performed well in predicting tumor progression within 12 months after RFA(AUC=0.917,0.962,and 0.906,respectively),and the C-indexes of the clinical and clinical-texture models were 0.917 and 0.957,respectively.Conclusions:DECT-quantitative parameters,CTTA,and clinical variables were helpful in predicting HCC progression after RFA.The constructed clinical prediction model can provide early warning of potential tumor progression risk for patients after RFA.展开更多
基金supported in part by grants from the National Key Research and Development Program (2019YFC0118100 and 2017YFC0110903)the National Natural Science Foundation of China (12026602 and 81802649)+2 种基金the Guangdong Key Area Re-search and Development Program (2020B010165004)the Shen-zhen Key Basic Science Program (JCYJ20180507182437217)the Shenzhen Key Laboratory Program (ZDSYS201707271637577)
文摘Background: Although transarterial chemoembolization(TACE) is the first-line therapy for intermediatestage hepatocellular carcinoma(HCC), it is not suitable for all patients. This study aimed to determine how to select patients who are not suitable for TACE as the first treatment choice. Methods: A total of 243 intermediate-stage HCC patients treated with TACE at three centers were retrospectively enrolled, of which 171 were used for model training and 72 for testing. Radiomics features were screened using the Spearman correlation analysis and the least absolute shrinkage and selection operator(LASSO) algorithm. Subsequently, a radiomics model was established using extreme gradient boosting(XGBoost) with 5-fold cross-validation. The Shapley additive explanations(SHAP) method was used to visualize the radiomics model. A clinical model was constructed using univariate and multivariate logistic regression. The combined model comprising the radiomics signature and clinical factors was then established. This model’s performance was evaluated by discrimination, calibration, and clinical application. Generalization ability was evaluated by the testing cohort. Finally, the model was used to analyze overall and progression-free survival of different groups. Results: A third of the patients(81/243) were unsuitable for TACE treatment. The combined model had a high degree of accuracy as it identified TACE-unsuitable cases, at a sensitivity, specificity, and area under the receiver operating characteristic curve(AUC) of 0.759, 0.885, 0.906 [95% confidence interval(CI): 0.859-0.953] in the training cohort and 0.826, 0.776, and 0.894(95% CI: 0.815-0.972) in the testing cohort, respectively. Conclusions: The high degree of accuracy of our clinical-radiomics model makes it clinically useful in identifying intermediate-stage HCC patients who are unsuitable for TACE treatment.
基金the National Key Research and Development Program of China(2019YFC0118100)the National Natural Science Foundation of China(81671760,81873910 and 62171167)the Natural Science Foundation of Shang-hai(19ZR1457800).
文摘Background:Radiofrequency ablation(RFA)is one of the effective therapeutic modalities in patients with hepatocellular carcinoma(HCC).However,there is no proper method to evaluate the HCC response to RFA.This study aimed to establish and validate a clinical prediction model based on dual-energy com-puted tomography(DECT)quantitative-imaging parameters,clinical variables,and CT texture parameters.Methods:We enrolled 63 patients with small HCC.Two to four weeks after RFA,we performed DECT scanning to obtain DECT-quantitative parameters and to record the patients’clinical baseline variables.DECT images were manually segmented,and 56 CT texture features were extracted.We used LASSO al-gorithm for feature selection and data dimensionality reduction;logistic regression analysis was used to build a clinical model with clinical variables and DECT-quantitative parameters;we then added texture features to build a clinical-texture model based on clinical model.Results:A total of six optimal CT texture analysis(CTTA)features were selected,which were statis-tically different between patients with or without tumor progression(P<0.05).When clinical vari-ables and DECT-quantitative parameters were included,the clinical models showed that albumin-bilirubin grade(ALBI)[odds ratio(OR)=2.77,95%confidence interval(CI):1.35-6.65,P=0.010],λAP(40-100 keV)(OR=3.21,95%CI:3.16-5.65,P=0.045)and IC AP(OR=1.25,95%CI:1.01-1.62,P=0.028)were asso-ciated with tumor progression,while the clinical-texture models showed that ALBI(OR=2.40,95%CI:1.19-5.68,P=0.024),λAP(40-100 keV)(OR=1.43,95%CI:1.10-2.07,P=0.019),and CTTA-score(OR=2.98,95%CI:1.68-6.66,P=0.001)were independent risk factors for tumor progression.The clinical model,clinical-texture model,and CTTA-score all performed well in predicting tumor progression within 12 months after RFA(AUC=0.917,0.962,and 0.906,respectively),and the C-indexes of the clinical and clinical-texture models were 0.917 and 0.957,respectively.Conclusions:DECT-quantitative parameters,CTTA,and clinical variables were helpful in predicting HCC progression after RFA.The constructed clinical prediction model can provide early warning of potential tumor progression risk for patients after RFA.
