| Part 1 MRI-based radiomics value for predicting the survival of patients with locally advanced cervical squamous cell cancer treated by concurrent chemoradiotherapyPurpose:To investigate the value of magnetic resonance imaging(MRI)-based radiomics in predicting the survival of patients with locally advanced cervical squamous cell cancer(LACSC)treated with concurrent chemoradiotherapy(CCRT).Material and methods:Between January 2014 and December 2018,a total of 185 patients(training group:n=128;testing group:n=57)with LACSC treated with CCRT were retrospectively enrolled in this study.A total of 400 radiomics features were extracted from T2-weighted imaging,apparent diffusion coefficient map,arterial-and delayed-phase contrast-enhanced MRI.Univariate Cox regression and least absolute shrinkage and selection operator Cox regression was applied to select radiomics features and clinical characteristics that could independently predict progression-free survival(PFS)and overall survival(OS),to establish the radiomics score,clinical predicting model and combined predicting model.Nomograms and calibration curves were then generated.The predictive capability of the prediction model was evaluated using Harrell’s C-index.Survival curves were generated using the Kaplan-Meier method,and compared with log-rank test.Results:The radiomics score achieved significantly better predictive performance for the estimation of PFS(C-index,0.764 for training and 0.762 for testing)and OS(C-index,0.793 for training and 0.750 for testing),compared with the 2018 International Federation of Gynecology and Obstetrics(FIGO)staging system(C-index for PFS,0.657 for training and 0.677 for testing;C-index for OS,0.665 for training and 0.633 for testing)and clinical predicting model(C-index for PFS,0.731 for training and 0.725 for testing;C-index for OS,0.708 for training and 0.693 for testing).The clinical predicting model for PFS estimation was constructed with T stage,tumor maximum-diameter,and lymph node metastasis(LNM)position,the clinical predicting model for OS estimation was constructed with T stage and LNM position.The combined predicting model constructed with T stage,LNM position,and radiomics score achieved the best performance for the estimation of PFS(C-index,0.792 for training and 0.809 for testing)and OS(C-index,0.822 for training and 0.785 for testing),which were significantly higher than those of the radiomics score.Conclusion:The MRI-based radiomics score could provide effective information in predicting the PFS and OS in patients with LACSC treated with CCRT.The combined model,including MRI-based radiomics score and clinical characteristics,showed the best prediction performance.Part 2 The value of texture analysis of ADC in predicting the survival of patients with stage ⅢCr cervical squamous cell cancer treated by concurrent chemoradiotherapyPurpose:To investigate the value of texture analysis of apparent diffusion coefficient(ADC)in predicting the survival of patients with 2018 International Federation of Gynecology and Obstetrics(FIGO)stage ⅢCr cervical squamous cell cancer(CSCC)treated with concurrent chemoradiotherapy(CCRT).Material and methods:Between January 2014 and December 2018,a total of 91 patients with stage ⅢCr CSCC treated by CCRT were retrospectively enrolled in this study.Clinical variables include age,body mass index,serum levels of squamous cell carcinoma antigen,tumor grade,FIGO 2018 staging system,T stage,tumor maximum size,and the number of lymph node metastasis(LNM).Twenty-one first-order texture features were extracted from pretreatment ADC maps.Univariate Cox hazard regression analyses was performed to evaluate the value of texture features in predicting progression-free survival(PFS)and overall survival(OS).Texture features with a P value<0.2 in the univariate analysis were selected.Meanwhile,the correlation between the selected first-order features was calculated,and features with a coefficient r ≥ 0.8 were removed accordingly,remaining features were then incorporated into the multivariate Cox hazard regression analyses.Using univariate and multivariate Cox hazard regression analyses to construct clinical predicting model and combined predicting model.Nomograms and calibration curves were then generated.The predictive capability of the prediction model was evaluated using Harrell’s C-index.Survival curves were generated using the Kaplan-Meier method,and compared with log-rank test.Results:The C-index values of the 2018 FIGO staging system for PFS and OS were 0.629 and 0.630,respectively.The clinical predicting model for PFS estimation includes T stage and the number of LNM,while the clinical predicting model for OS estimation includes T stage,the number of LNM,and tumor grade,and the C-index values were respectively 0.674 and 0.724,which were significantly higher than those of the 2018 FIGO staging system(P<0.05).In texture features,mean absolute deviation(MAD)is an independent predictor for PFS,MAD and energy were independent predictors for OS.The combined predicting model was constructed by clinical variables and texture features,the C-index values for PFS and OS were respectively 0.750 and 0.832,which were significantly higher than clinical predicting model and 2018 FIGO staging system(P<0.05).Conclusion:The texture analysis of the ADC maps is helpful to improve the accuracy of prognosis prediction for patients with stage ⅢCr CSCC treated by CCRT.The texture analysis of the ADC maps could be used along with clinical prognostic biomarkers to predict PFS and OS in patients with stage ⅢCr CSCC treated by CCRT.Part 3 MRI-based radiomics for pretreatment prediction of response to concurrent chemoradiotherapy in locally advanced cervical squamous cell cancerPurpose:To investigate the value of magnetic resonance imaging(MRI)-based radiomics in predicting the treatment response to concurrent chemoradiotherapy(CCRT)in patients with locally advanced cervical squamous cell cancer(LACSC).Material and methods:Between January 2014 and December 2019,a total of 198 patients(training:n=138;testing:n=60)with LACSC treated with CCRT were retrospectively enrolled in this study.Responses were evaluated by MRI and clinical data performed at one month after completion of CCRT according to RECIST standards,and patients were divided into the residual group and nonresidual group.Clinical variables include age,body mass index,serum levels of squamous cell carcinoma antige,tumor grade,2018 International Federation of Gynecology and Obstetrics(FIGO)stage,T stage,tumor maximum size,and the lymph node metastasis(LNM).A total of 400 radiomics features were extracted from T2-weighted imaging,apparent diffusion coefficient map,arterial-and delayed-phase contrast-enhanced MRI.The radiomics score was constructed with a feature selection strategy.Logistic regression analysis was used for multivariate analysis of clinical variables,and the clinical predicting model and combined predicting model were constructed.Nomograms and calibration curves were then generated.The performance of all models was assessed using receiver operating characteristic curves(ROC).Delong test was used to compare the difference of area under the ROC curve(AUC).Results:Among the clinical variables,tumor grade and FIGO stage were independent risk factors,and the AUCs of the clinical predicting model were 0.741 and 0.749 in the training and testing groups.The radiomics score,consisting of 4 radiomics features,showed good predictive performance with an AUC of 0.819 in the training group and 0.776 in the testing group,which were higher than the clinical predicting model,but the difference was not statistically significant.The combined predicting model constructed with tumor grade,FIGO stage,and radiomics score achieved the best performance,with an AUC of 0.857 in the training group and 0.842 in the testing group,which were significantly higher than the clinical predicting model.Conclusion:MRI-based radiomics features could be used as a noninvasive biomarker to improve the ability to predict the treatment response to CCRT in patients with LACSC. |
PDF Full Text Request