Search

Background Short-course partial brain radiotherapy ± chemotherapy for older patients with GBM extends survival but there is no validated evidence for prediction of individual risk of acute radiotherapy-related side effects. Methods This prospective multicentre observational trial recruited patients with newly diagnosed GBM aged ≥65 planned for cranial radiotherapy. Baseline MRI scans were analyzed for markers of brain resilience including relative total brain volume (ratio of cerebrospinal fluid (CSF) volume to total intracranial volume (TIV)) and their relationship to change in quality of life (QoL). Results 126 patients enrolled: mean age 72 years (range 65-83). 77% had debulking surgery. 79% received radiotherapy with concurrent TMZ, and 21% received palliative radiotherapy alone. The median OS was 10.7 months. After accounting for age, sex, treatment, and baseline MoCA score, there was a relationship between baseline CSF:TIV and change in QoL score at 8 weeks post treatment. For each unit point of increase in CSF:TIV, there was a corresponding decrease in QoL score of 1.72 (95% CI −3.24 to −0.19 P  = .027). 35 participants were too unwell to complete questionnaires or had died by the 8 week follow-up visit. In this subgroup, post hoc logistic regression showed baseline CSF:TIV was related to the risk of non-attendance (OR 1.35, 95% CI 1.01 to 1.80, P  = .042). Cox regression models showed baseline CSF:TIV was associated with worsened OS (HR 1.41, 95% CI 1.19 to 1.66, P  < .001). Conclusions This study provides evidence to support the use of an imaging biomarker to help assess the risk:benefit ratio for radiotherapy. [ABSTRACT FROM AUTHOR]

Background: The aim was to predict survival of glioblastoma at 8 months after radiotherapy (a period allowing for completing a typical course of adjuvant temozolomide), by applying deep learning to the first brain MRI after radiotherapy completion., Methods: Retrospective and prospective data were collected from 206 consecutive glioblastoma, isocitrate dehydrogenase -wildtype patients diagnosed between March 2014 and February 2022 across 11 UK centers. Models were trained on 158 retrospective patients from 3 centers. Holdout test sets were retrospective (n = 19; internal validation), and prospective (n = 29; external validation from 8 distinct centers). Neural network branches for T2-weighted and contrast-enhanced T1-weighted inputs were concatenated to predict survival. A nonimaging branch (demographics/MGMT/treatment data) was also combined with the imaging model. We investigated the influence of individual MR sequences; nonimaging features; and weighted dense blocks pretrained for abnormality detection., Results: The imaging model outperformed the nonimaging model in all test sets (area under the receiver-operating characteristic curve, AUC P = .038) and performed similarly to a combined imaging/nonimaging model (P > .05). Imaging, nonimaging, and combined models applied to amalgamated test sets gave AUCs of 0.93, 0.79, and 0.91. Initializing the imaging model with pretrained weights from 10 000s of brain MRIs improved performance considerably (amalgamated test sets without pretraining 0.64; P = .003)., Conclusions: A deep learning model using MRI images after radiotherapy reliably and accurately determined survival of glioblastoma. The model serves as a prognostic biomarker identifying patients who will not survive beyond a typical course of adjuvant temozolomide, thereby stratifying patients into those who might require early second-line or clinical trial treatment., (© The Author(s) 2024. Published by Oxford University Press on behalf of the Society for Neuro-Oncology.)