Gliomatosis cerebri (GC) growth pattern: A single-center analysis of clinical, histological, and molecular characteristics of GC and non-GC glioblastoma

TY - JOUR

T1 - Gliomatosis cerebri (GC) growth pattern: A single-center analysis of clinical, histological, and molecular characteristics of GC and non-GC glioblastoma

AU - Divé, Iris

AU - Weber, Katharina J

AU - Hartung, Tabea I

AU - Steidl, Eike

AU - Wagner, Marlies

AU - Hattingen, Elke

AU - Franz, Kea

AU - Fokas, Emmanouil

AU - Ronellenfitsch, Michael W

AU - Herrlinger, Ulrich

AU - Harter, Patrick N

AU - Steinbach, Joachim P

N1 - © The Author(s) 2023. Published by Oxford University Press, the Society for Neuro-Oncology and the European Association of Neuro-Oncology.

PY - 2023

Y1 - 2023

N2 - BACKGROUND: The biological understanding of glioblastoma (GB) with gliomatosis cerebri (GC) pattern is poor due to the absence of GC-specific studies. Here, we aimed to identify molecular or clinical parameters that drive GC growth.METHODS: From our methylome database of IDH (isocitrate dehydrogenase)-wildtype GB, we identified 158 non-GC and 65 GC cases. GC cases were subdivided into diffuse-infiltrative (subtype 1), multifocal (subtype 2), or tumors with 1 solid mass (subtype 3). We compared clinical, histological, and molecular parameters and conducted a reference-free tumor deconvolution of DNA methylation data based on latent methylation components (LMC).RESULTS: GC subtype 1 less frequently showed contrast-enhancing tumors, and more frequently lacked morphological GB criteria despite displaying GB DNA methylation profile. However, the tumor deconvolution did not deliver a specific LMC cluster for either of the GC subtypes. Employing the reference-based analysis MethylCIBERSORT, we did not identify significant differences in tumor cell composition. The majority of both GC and non-GC patients received radiochemotherapy as first-line treatment, but there was a major imbalance for resection. The entire GC cohort had significantly shorter overall survival (OS) and time to treatment failure (TTF) than the non-GC cohort. However, when filtering for cases in which only stereotactic biopsy was performed, the comparison of OS and TTF lost statistical significance.CONCLUSIONS: Our study offers clinically relevant information by demonstrating a similar outcome for GB with GC growth pattern in the surgically matched analysis. The limited number of cases in the GC subgroups encourages the validation of our DNA methylation analysis in larger cohorts.

AB - BACKGROUND: The biological understanding of glioblastoma (GB) with gliomatosis cerebri (GC) pattern is poor due to the absence of GC-specific studies. Here, we aimed to identify molecular or clinical parameters that drive GC growth.METHODS: From our methylome database of IDH (isocitrate dehydrogenase)-wildtype GB, we identified 158 non-GC and 65 GC cases. GC cases were subdivided into diffuse-infiltrative (subtype 1), multifocal (subtype 2), or tumors with 1 solid mass (subtype 3). We compared clinical, histological, and molecular parameters and conducted a reference-free tumor deconvolution of DNA methylation data based on latent methylation components (LMC).RESULTS: GC subtype 1 less frequently showed contrast-enhancing tumors, and more frequently lacked morphological GB criteria despite displaying GB DNA methylation profile. However, the tumor deconvolution did not deliver a specific LMC cluster for either of the GC subtypes. Employing the reference-based analysis MethylCIBERSORT, we did not identify significant differences in tumor cell composition. The majority of both GC and non-GC patients received radiochemotherapy as first-line treatment, but there was a major imbalance for resection. The entire GC cohort had significantly shorter overall survival (OS) and time to treatment failure (TTF) than the non-GC cohort. However, when filtering for cases in which only stereotactic biopsy was performed, the comparison of OS and TTF lost statistical significance.CONCLUSIONS: Our study offers clinically relevant information by demonstrating a similar outcome for GB with GC growth pattern in the surgically matched analysis. The limited number of cases in the GC subgroups encourages the validation of our DNA methylation analysis in larger cohorts.

U2 - 10.1093/noajnl/vdad131

DO - 10.1093/noajnl/vdad131

M3 - SCORING: Journal article

C2 - 38024242

VL - 5

SP - vdad131

JO - NEURO-ONCOL ADV

JF - NEURO-ONCOL ADV

SN - 2632-2498

IS - 1

ER -