Neurocognitive and Resting-State Functional Magnetic Resonance Imaging Changes in Patients With Diffuse Gliomas After Chemoradiation Therapy

Purpose: This prospective observational study employed resting-state functional magnetic resonance imaging (rs-fMRI) to investigate network-level disturbances associated with neurocognitive function (NCF) changes in patients with gliomas following partial-brain radiation therapy (RT). Methods and Materials: Adult postoperative patients with either isocitrate dehydrogenase (IDH)–wild-type or IDH-mutant gliomas underwent computerized NCF testing and rs-fMRI at baseline and 6 months post-RT. rs-fMRI data were assessed using seed-based functional connectivity (FC). NCF changes were quantified by the percent change in age-normalized composite scores from baseline (ΔNCF_comp). Connectivity regression analysis assessed the association between network FC changes and NCF changes, using a split-sample approach with a 26-patient training set and a 6-patient validation set, iterated 200 times. Permutation tests evaluated the significance of network selection. Results: Between September 2020 and December 2023, 43 patients were enrolled, with 32 completing both baseline and follow-up evaluations. The mean ΔNCF_comp was 2.9% (SD, 13.7%), with 38% experiencing a decline. Patients with IDH-mutant glioma had similar NCF changes compared with those with IDH–wild-type glioma. Intrahemispheric FC was similar between ipsilateral and contralateral hemispheres for 91% of patients at baseline, and 69% had similar intrahemispheric FC change posttreatment. FC changes accounted for a moderate fraction of variance in NCF changes (mean R², 0.301; SD, 0.249), with intranetwork FC of the parietal memory network (PMN-PMN, P = .001) and internetwork FC between the PMN and the visual network (PMN-VN, P = .002) as the most significant factors. Similar findings were obtained by sensitivity analyses using only the FC data from the hemisphere contralateral to the tumor. Conclusions: Post-RT rs-fMRI changes significantly reflected NCF decline, highlighting rs-fMRI as a promising imaging biomarker for neurocognitive decline after RT.