Nihar Kathrani1, Richa Singh Chauhan2, Abhishek Kotwal2, Karthik Kulanthaivelu2, Maya Dattatraya Bhat3,4, Jitender Saini2, Chandrajit Prasad2, Dhritiman Chakrabarti5, Vani Santosh6, Alok Mohan Uppar7, Dwarakanath Srinivas7. 1. Interventional Radiology, Paras Hospital, Gurugram, Haryana, India. 2. Department of Neuroimaging and Interventional Radiology, National Institute of Mental Health and Neurosciences (NIMHANS), Hosur Road, Near Bangalore Milk Dairy, Karnataka, Bengaluru, 5600029, India. 3. Department of Neuroimaging and Interventional Radiology, National Institute of Mental Health and Neurosciences (NIMHANS), Hosur Road, Near Bangalore Milk Dairy, Karnataka, Bengaluru, 5600029, India. mayabhat05@yahoo.co.in. 4. Faculty Quarters, NIMHANS, Hosur Road, Near Bangalore Milk Dairy, Karnataka, Bengaluru, 5600029, India. mayabhat05@yahoo.co.in. 5. Department of Neuroanaesthesia and NeuroCritical Care, National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru, India. 6. Department of Neuropathology, National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru, India. 7. Department of Neurosurgery, National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru, India.
Abstract
PURPOSE: H3K27M-mutant diffuse midline gliomas (M-DMGs) exhibit a clinically aggressive course. We studied diffusion-weighted imaging (DWI) and perfusion (PWI) MRI features of DMG with the hypothesis that DWI-PWI metrics can serve as biomarkers for the prediction of the H3K27M mutation status in DMGs. METHODS: A retrospective review of the institutional database (imaging and histopathology) of patients with DMG (July 2016 to July 2020) was performed. Tumoral apparent diffusion coefficient (ADC) and peritumoral ADC (PT ADC) values and their normalized values (nADC and nPT ADC) were computed. Perfusion data were analyzed with manual arterial input function (AIF) and leakage correction (LC) Boxerman-Weiskoff models. Normalized maximum relative CBV (rCBV) was evaluated. Intergroup analysis of the imaging variables was done between M-DMGs and wild-type (WT-DMGs) groups. RESULTS: Ninety-four cases (M-DMGs-n = 48 (51%) and WT-DMGs-n = 46(49%)) were included. Significantly lower PT ADC (mutant-1.1 ± 0.33, WT-1.23 ± 0.34; P = 0.033) and nPT ADC (mutant-1.64 ± 0.48, WT-1.83 ± 0.54; P = 0.040) were noted in the M-DMGs. The rCBV (mutant-25.17 ± 27.76, WT-13.73 ± 14.83; P = 0.018) and nrCBV (mutant-3.44 ± 2.16, WT-2.39 ± 1.25; P = 0.049) were significantly higher in the M-DMGs group. Among thalamic DMGs, the min ADC, PT ADC, and nADC and nPT ADC were lower in M-DMGs while nrCBV (corrected and uncorrected) was significantly higher. Receiver operator characteristic curve analysis demonstrated that PT ADC (cut-off-1.245), nPT ADC (cut-off-1.853), and nrCBV (cut-off-1.83) were significant independent predictors of H3K27M mutational status in DMGs. CONCLUSION: DWI and PWI features hold value in preoperative prediction of H3K27M-mutation status in DMGs.
PURPOSE: H3K27M-mutant diffuse midline gliomas (M-DMGs) exhibit a clinically aggressive course. We studied diffusion-weighted imaging (DWI) and perfusion (PWI) MRI features of DMG with the hypothesis that DWI-PWI metrics can serve as biomarkers for the prediction of the H3K27M mutation status in DMGs. METHODS: A retrospective review of the institutional database (imaging and histopathology) of patients with DMG (July 2016 to July 2020) was performed. Tumoral apparent diffusion coefficient (ADC) and peritumoral ADC (PT ADC) values and their normalized values (nADC and nPT ADC) were computed. Perfusion data were analyzed with manual arterial input function (AIF) and leakage correction (LC) Boxerman-Weiskoff models. Normalized maximum relative CBV (rCBV) was evaluated. Intergroup analysis of the imaging variables was done between M-DMGs and wild-type (WT-DMGs) groups. RESULTS: Ninety-four cases (M-DMGs-n = 48 (51%) and WT-DMGs-n = 46(49%)) were included. Significantly lower PT ADC (mutant-1.1 ± 0.33, WT-1.23 ± 0.34; P = 0.033) and nPT ADC (mutant-1.64 ± 0.48, WT-1.83 ± 0.54; P = 0.040) were noted in the M-DMGs. The rCBV (mutant-25.17 ± 27.76, WT-13.73 ± 14.83; P = 0.018) and nrCBV (mutant-3.44 ± 2.16, WT-2.39 ± 1.25; P = 0.049) were significantly higher in the M-DMGs group. Among thalamic DMGs, the min ADC, PT ADC, and nADC and nPT ADC were lower in M-DMGs while nrCBV (corrected and uncorrected) was significantly higher. Receiver operator characteristic curve analysis demonstrated that PT ADC (cut-off-1.245), nPT ADC (cut-off-1.853), and nrCBV (cut-off-1.83) were significant independent predictors of H3K27M mutational status in DMGs. CONCLUSION: DWI and PWI features hold value in preoperative prediction of H3K27M-mutation status in DMGs.
Authors: David N Louis; Caterina Giannini; David Capper; Werner Paulus; Dominique Figarella-Branger; M Beatriz Lopes; Tracy T Batchelor; J Gregory Cairncross; Martin van den Bent; Wolfgang Wick; Pieter Wesseling Journal: Acta Neuropathol Date: 2018-03-01 Impact factor: 17.088