Background Prospective evaluation of the results of volumetric modulated arc therapy (VMAT) for sinonasal cancer compared to 3D conformal radiation therapy (3DCRT). Materials and Methods We prospectively evaluated 34 patients (pts) treated with postoperative VMAT with simultaneous integrated boost (SIB; RapidArc) from 2011 to 2015. These pts were retrospectively compared with 24 pts treated with 3DCRT from 2003 to 2011. The two sets were not significantly different on sex, mean age, tumor site, stage, histology. Efficacy and toxicity were evaluated. Results Median follow‐up was 45 months (range: 6‐143 months). Three‐year overall survival was 85.2% in VMAT‐SIB versus 65.2% in 3DCRT (P = .02). Three‐year local control was 81.2% in VMAT‐SIB versus 62.5% in 3DCRT (P = .04). There was a reduction of acute (<0.09) and late (0.03) ocular toxicity of grade ≥ 2 for pts with VMAT‐SIB. Conclusion VMAT significantly improved local control and overall survival in sinonasal cancer with lower rate of toxicity.

In this study, a radiomics analysis was conducted to provide insights into the differentiation of radionecrosis and tumor progression in multiparametric MRI in the context of a multicentric clinical trial. First, the sensitivity of radiomic features to the unwanted variability caused by different protocol settings was assessed for each modality. Then, the ability of image normalization and ComBat-based harmonization to reduce the scanner-related variability was evaluated. Finally, the performances of several radiomic models dedicated to the classification of MRI examinations were measured. Our results showed that using radiomic models trained on harmonized data achieved better predictive performance for the investigated clinical outcome (balanced accuracy of 0.61 with the model based on raw data and 0.72 with ComBat harmonization). A comparison of several models based on information extracted from different MR modalities showed that the best classification accuracy was achieved with a model based on MR perfusion features in conjunction with clinical observation (balanced accuracy of 0.76 using LASSO feature selection and a Random Forest classifier). Although multimodality did not provide additional benefit in predictive power, the model based on T1-weighted MRI before injection provided an accuracy close to the performance achieved with perfusion.

No abstract