Volumetric Response and Survival of Patients With Bulky IDH-Mutated Grade 3 Glioma Managed With FET-FDG-Guided Integrated Boost IMRT.

AIMS: Despite relatively favourable outcomes associated with IDH-mutant grade 3 gliomas, many patients present with diffuse non-enhancing disease involving multiple brain regions, prompting concern over both durable disease control and the morbidity associated with large volume radiation therapy. This study audits volumetric response, survival and functional outcomes in this 'large volume' subgroup that undergoes intensity modulated radiation therapy (IMRT). MATERIALS AND METHODS: From a prospective database of 187 patients with IDH-mutant grade 3 gliomas managed with IMRT between 2008 and 2020, recorded PTV was divided into quartiles. The top quartile, termed the 'large volume cohort' (LVC), was identified. IMRT involved FET-FDG guided integrated boost (59.4/54Gy in 33 fractions). Manual volumetric segmentation of baseline, four months and 13 months post-IMRT tumour were performed for T1, T2 and T1gd MRI sequences. The primary endpoint was volumetric reduction on the T1 and T2 sequences at 13 months and analysed with relapse-free survival (RFS) and overall survival (OS). Morbidity endpoints were assessed at year four post-IMRT and included performance status (ECOG PS) and employment outcomes. RESULTS: The fourth quartile (LVC) identified 44 patients for whom volumetric analysis was available. The LVC had median PTV of 320cm3 compared to 186.2cm3 for the total group. Anaplastic astrocytoma and oligodendroglioma were equally distributed and tumour sites were frontal (54%), temporal (18%) and parietal lobes (16%). Median follow-up for survivors was 71.5 months. Projected 10-year RFS and OS in LVC was 40% and 62%, compared to 53% and 62% respectively in the overall cohort. The RFS (p = 0.06) and OS (p = 0.65) of the LVC was not significantly different to other PTV quartiles; however the impact of PTV volume reached significance when analysed as a continuous variable (RFS p < 0.01; OS p = 0.02). Median T1 volumes were 26.1cm3, 8.0cm3 and 5.3cm3 at months +0, +3 and +12, respectively. The corresponding T2 volumes were 120.8cm3, 29.1cm3 and 26.3cm3. The median T1 and T2 volume reductions were 77% (q1-3: 57-92%) and 78% (q1-3: 60-85%) at 13 months post-IMRT. Initial T2 volume was associated with worse RFS (p = 0.04) but not OS (p = 0.96). There was no association between median T2 volume reduction and RFS (p = 0.77). For patients assessable at year 4 post-IMRT, no late CTCAE Grade 3/4 toxicity events were recognised. 92% of patients were ECOG PS 0-1, 45% were employed at prior capacity and 28% were working with impairment. CONCLUSION: Patients with large volume IDH-mutant Grade 3 glioma demonstrated significant tumour reduction post-IMRT, and good long-term outcomes with respect to survival and functional status. Although larger IMRT volumes were associated with poorer RFS, this was also associated with the initial volume of non-enhancing tumour.

Genetic Discrimination of Grade 3 and Grade 4 Gliomas by Artificial Neural Network.

Gliomas, including anaplastic gliomas (AG; grade 3) and glioblastomas (GBM; grade 4), are malignant brain tumors associated with poor prognosis and low survival rates. Current classification systems based on histopathology have limitations due to intratumoral heterogeneity. The treatment and prognosis are distinctly different between grade 3 and grade 4 gliomas patients. Therefore, there is a need for molecular markers to differentiate these tumors accurately. In this study, we aimed to identify a gene expression signature using an artificial neural network (ANN) in application to microarray and serial analysis of gene expression (SAGE) data for grade 3 (AG) and grade 4 (GBM) gliomas discrimination. We acquired gene expression data from publicly available datasets on glial tumors of grades 3 and 4-a total of 93 grade 3 gliomas and 224 grade 4 gliomas. To select genes for classification, we implemented an artificial neural network-based method using a combination of self-organized maps (SOM) and perceptron. In general, we implemented a multi-stage procedure that involved multiple runs of a genetic algorithm to identify genes that provided optimal clusterization on the SOM. We performed this procedure multiple times, resulting in different sets of genes each time. Eventually, we selected several genes that appeared most frequently in the reduced sets and performed classification using them. Our analysis identified a set of seven genes (BCAS4, GLUD2, KCNJ10, KCND2, AKR7A2, FOLR1, and KIAA0319). The classification accuracy using this gene set was 87.5%. These findings suggest the potential of this gene set as a molecular marker for distinguishing grade 3 (AG) from grade 4 (GBM) gliomas.

A model combined machine learning with imaging omics characteristics in differentiating anaplastic glioma from glioblastoma / 中华神经医学杂志

Objective:To construct and validate a prediction model combined machine learning with imaging omics characteristics in differentiating anaplastic glioma from glioblastoma.Methods:Imaging data of 241 patients with anaplastic glioma or glioblastoma, confirmed by pathology in our hospital from August 2005 to August 2012, were retrospectively collected. These patients were divided into a training group ( n=140) and a verification group ( n=101) according to random number table method. MRIcron software was used to delineate tumor boundaries of patients from the training group on preoperative T1 enhanced MR imaging. The regions of interest (ROIs) were outlined on preoperative T1 enhanced MR imaging, and the radiomic features were extracted from ROIs by Matlab software. Least absolute shrinkage and selection operator (LASSO) regression model was used to screen the features, and then, the selected features were used to construct the prediction model by support vector machine (SVM) classifier. The area under the curve (AUC) of receiver operating characteristic (ROC) curve was used to evaluate the predictive efficacy of the model. Results:In these 241 patients, 101 were with anaplastic glioma and 140 were with glioblastoma confirmed by pathology. In the training group and validation group, there was statistical difference in age between patients with anaplastic glioma and glioblastoma ( P<0.05); there was no significant difference in gender distribution, tumor location, and percentages of tumor necrosis or edema between patients with anaplastic glioma and glioblastoma ( P>0.05). Totally, 431 radiomic features were extracted; 11 radiomic features were screened by LASSO regression model and the prediction model was established. The AUC of ROC curve was 0.942 and 0.875, respectively, in the training group and validation group. Conclusion:The prediction model combined machine learning and imaging omics characteristics can effectively discriminate anaplastic glioma from glioblastoma.