Machine learning-based radiomics for predicting outcomes in cervical cancer patients undergoing concurrent chemoradiotherapy.

PURPOSES: To investigate the value of machine learning-based radiomics for predicting disease-free survival (DFS) and overall survival (OS) undergoing concurrent chemoradiotherapy (CCRT) for patients with locally advanced cervical cancer (LACC). MATERIALS AND METHODS: In this multicentre study, 700 patients with IB2-IVA cervical cancer who underwent CCRT with ongoing follow-up were retrospectively analyzed. Three-dimensional radiomics features of primary lesions and its surrounding 5 mm region in T2WI sequences were collected. Six machine learning methods were used to construct the optimal radiomics model for accurate prediction of DFS and OS after CCRT in LACC patients. Eventually, TCGA and GEO databases were used to explore the mechanisms of radiomics in predicting the progression and survival of cervical cancer. This study adhered CLEAR for reporting and its quality was assessed using RQS and METRICS. RESULTS: In the prediction of DFS, the RSF model combined tumor and peritumor radiomics demonstrated the best predictive efficacy, with the AUC for predicting 1-year, 3-year, and 5-year DFS in the training, validation, and test sets of 0.986, 0.989, 0.990, and 0.884, 0.838, 0.823, and 0.829, 0.809, 0.841, respectively. In the prediction of OS, the GBM model best performer, with AUC of 0.999, 0.995, 0.978, and 0.981, 0.975, 0.837, and 0.904, 0.860, 0.905. Differential genes in TCGA and GEO suggest that the prediction of radiomics model may be associated with KDELR2 and HK2. CONCLUSION: Machine learning-based radiomics models help to predict DFS and OS after CCRT in LACC patients, and the combination of tumor and peritumor information has higher predictive efficacy, which can provide a reliable basis for therapeutic decision-making in cervical cancer patients.

Prediction of survival of patients with cervical cancer after concurrent chemoradiotherapy based on clinical and imaging parameters / 中华放射肿瘤学杂志

Objective:To investigate the value of nomograms based on clinical parameters, apparent diffusion coefficient (ADC) and MRI-derived radiomics in predicting survival of patients with locally advanced cervical cancer (LACC) after concurrent chemoradiotherapy (CCRT).Methods:Clinical data of 423 patients with IB-IVA cervical cancer treated with CCRT at Anhui Provincial Hospital Affiliated to Anhui Medical University from March 2014 to March 2020 were retrospectively analyzed and randomly divided into the training and validation groups at a ratio of 2∶1 using the simple randomization method. The values of ADC min, ADC mean, ADC max and 3D texture parameters of diffusion weighted imaging (DWI), T 2WI, T 2WI-fat suppression of pre-treatment primary lesions in all patients were measured. The least absolute shrinkage and selection operator (LASSO) algorithm and logistic regression analysis were used to screen the texture features and calculate radiomics score (Rad-score). Cox regression analysis was employed to construct nomogram models for predicting overall survival (OS) and cancer-specific survival (CS) of patients with LACC after CCRT, which were subject to internal and external validation. Results:Squamous cell carcinoma antigen (SCC-Ag), external beam radiotherapy dose, ADCmin and Rad-score were the independent prognostic factors for OS and CS of LACC patients after CCRT and constituted predictive models for OS and CS. The area under the receiver operating characteristic (ROC) curve (AUC) of two models in predicting 1-year, 3-year, 5-year OS and CS was 0.906, 0.917, 0.916 and 0.911, 0.918, 0.920, with internally validated consistency indexes (C-indexes) of 0.897 and 0.900. Then, models were brought into the validation group for external validation with AUC of 0.986, 0.942, 0.932 and 0.986, 0.933, 0.926 in predicting 1-year, 3-year, 5-year OS and CS.Conclusion:The nomograms based on clinical parameters, ADC values and MRI-derived radiomics are of high clinical value in predicting OS and CS of patients with LACC after CCRT, which can be used as prognostic markers for patients with cervical cancer to certain extent.

Nomogram based on IVIM-DWI and radiomics in predicting recurrence after concurrent chemoradiotherapy for patients with cervical cancer / 中华放射肿瘤学杂志

Objective:To investigate the value of nomogram based on intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) and MRI-derived radiomics for predicting recurrence after concurrent chemoradiotherapy (CCRT) in patients with locally advanced cervical cancer (LACC).Methods:Clinical data of 111 patients with ⅠB-ⅣA cervical cancer who underwent CCRT at Anhui Provincial Hospital from December 2014 to December 2019 and were continuously followed up were retrospectively analyzed. Pre-treatment IVIM-DWI parameters (ADC, D, D * and f) and pre- and post-treatment 3D texture parameters (from axial T 2WI) of the primary lesions were measured. Least absolute shrinkage and selection operator (LASSO) algorithm and multivariate logistic regression analysis were used to filter texture features and calculate radiomics score (Rad-score). A Cox regression model was used to analyze independent risk factors for recurrence after CCRT in patients with LACC and construct a nomogram. Results:External beam radiotherapy dose, f value , pre-treatment Rad-score and post-treatment Rad-score ( HR=0.204, 3.253, 2.544, 7.576) were the independent prognostic factors for recurrence after CCRT in cervical cancer patients and jointly formed the nomogram. The area under curve (AUC) of the nomogram for predicting 1-, 3- and 5-year disease-free survival (DFS) was 0.895, 0.888 and 0.916, with internal validation C-indexes of 0.859, 0.903 and 0.867, respectively. The decision curves analysis showed that the nomogram has a higher net clinical benefit compared to other models, and the clinical impact curves further visualized its predictive accuracy. Conclusions:The nomogam based on IVIM-DWI and radiomics has high clinical value in predicting recurrence after CCRT in patients with LACC, providing reference for prognostic assessment and individualized treatment of cervical cancer patients.