Background  Despites advances in radiation technology, skull base osteoradionecrosis (ORN) continues to be a rare, devastating, and hard to treat complication of radiotherapy. We present three cases of anterior skull base ORN in a cohort treated with intensity-modulated radiation therapy (IMRT). Case Series  Three patients developed anterior skull base ORN after receiving at least one round of IMRT. ORN was diagnosed through either nasal endoscopy or imaging findings. The first was a 59-year-old woman with a sinonasal squamous cell carcinoma. Her chemoradiation history was notable for reirradiation and a high dose of radiation (143.3 Gy). The second was a 55-year-old man with recurrent nasopharyngeal carcinoma, whose history was notable for a high dose of radiation (∼140 Gy) and for being reirradiated. The final patient was a 37-year-old woman with an unremarkable history who received radiotherapy (65.0 Gy) for an esthesioneuroblastoma. One patient was asymptomatic and did not receive ORN-specific therapy. The other two were treated with a combination of medical and surgical intervention with successful short-term outcomes (no evidence of infection). Conclusion  Anterior skull base ORN can be treated through conservative and surgical means to achieve successful short-term outcomes. Further investigation of long-term outcomes is warranted.

Purpose

Most radiomic studies use the features extracted from the manually drawn tumor contours for classification or survival prediction. However, large interobserver segmentation variations lead to inconsistent features and hence introduce more challenges in constructing robust prediction models. Here, we proposed an automatic workflow for glioblastoma (GBM) survival prediction based on multimodal magnetic resonance (MR) images.

Methods and materials

Two hundred eighty-five patients with glioma (210 GBM, 75 low-grade glioma) were included. One hundred sixty-three of the patients with GBM had overall survival data. Every patient had 4 preoperative MR images and manually drawn tumor contours. A 3-dimensional convolutional neural network, VGG-Seg, was trained and validated using 122 patients with glioma for automatic GBM segmentation. The trained VGG-Seg was applied to the remaining 163 patients with GBM to generate their autosegmented tumor contours. The handcrafted and deep learning (DL)-based radiomic features were extracted from the autosegmented contours using explicitly designed algorithms and a pretrained convolutional neural network, respectively. One hundred sixty-three patients with GBM were randomly split into training (n = 122) and testing (n = 41) sets for survival analysis. Cox regression models were trained to construct the handcrafted and DL-based signatures. The prognostic powers of the 2 signatures were evaluated and compared.

Results

The VGG-Seg achieved a mean Dice coefficient of 0.86 across 163 patients with GBM for GBM segmentation. The handcrafted signature achieved a C-index of 0.64 (95% confidence interval, 0.55-0.73), whereas the DL-based signature achieved a C-index of 0.67 (95% confidence interval, 0.57-0.77). Unlike the handcrafted signature, the DL-based signature successfully stratified testing patients into 2 prognostically distinct groups.

Conclusions

The VGG-Seg generated accurate GBM contours from 4 MR images. The DL-based signature achieved a numerically higher C-index than the handcrafted signature and significant patient stratification. The proposed automatic workflow demonstrated the potential of improving patient stratification and survival prediction in patients with GBM.