Surgery for Diffuse WHO Grade II Gliomas: Volumetric Analysis of a Multicenter Retrospective Cohort From the German Study Group for Intraoperative Magnetic Resonance Imaging.
Adolescent
Adult
Aged
Brain Neoplasms
/ mortality
Cohort Studies
Female
Glioma
/ mortality
Humans
Kaplan-Meier Estimate
Magnetic Resonance Imaging
/ methods
Male
Margins of Excision
Middle Aged
Neurosurgical Procedures
/ methods
Progression-Free Survival
Proportional Hazards Models
Retrospective Studies
Young Adult
Extent of resection
Gross total resection
Intraoperative MRI
Low-grade glioma
Molecular markers
Overall survival
Progression-free survival
Volumetric analysis
Journal
Neurosurgery
ISSN: 1524-4040
Titre abrégé: Neurosurgery
Pays: United States
ID NLM: 7802914
Informations de publication
Date de publication:
01 01 2020
01 01 2020
Historique:
received:
24
12
2018
accepted:
18
07
2019
pubmed:
2
10
2019
medline:
25
8
2020
entrez:
2
10
2019
Statut:
ppublish
Résumé
In diffuse WHO grade II gliomas (LGG), the extent of resection (EOR) required to achieve significant survival benefits remains elusive. To evaluate the association of residual volume (RV) and EOR with progression-free survival (PFS) or overall survival (OS) in LGG in a retrospective, multicenter series by the German study group of intraoperative MRI (GeSGIM). Consecutive cases were retrospectively assessed from 5 centers. Tumors were volumetrically quantified before and after surgery, and clinical data were analyzed, including IDH mutations and neurologic deficits. Kaplan-Meier estimates, accelerated failure time models (AFT), and multivariate Cox regression models were calculated to identify determinants of survival. A total of 140 cases were analyzed. Gross total resection (GTR) was associated with significantly longer PFS compared to any incomplete resection (P = .009). A significant survival disadvantage was evident even for small (>0-5 ml) residuals and increased for moderate (>5-20 ml) and large remnants (>20 ml) P = .001). Accordingly, PFS increased continuously for 20% incremental steps of EOR (P < .001). AFT models supported the notion of a continuous association of RV and EOR with PFS. Multivariate Cox regression models confirmed RV (P = .01) and EOR (P = .005) as continuous prognosticators of PFS. Univariate analysis showed significant associations of RV and EOR with OS. Our data support the hypothesis of a continuous relationship of RV and EOR with survival for LGG with superiority seen for GTR. Hence, GTR should be achieved whenever safely feasible, and resections should be maximized whenever tumor has to be left behind to spare function.
Sections du résumé
BACKGROUND
In diffuse WHO grade II gliomas (LGG), the extent of resection (EOR) required to achieve significant survival benefits remains elusive.
OBJECTIVE
To evaluate the association of residual volume (RV) and EOR with progression-free survival (PFS) or overall survival (OS) in LGG in a retrospective, multicenter series by the German study group of intraoperative MRI (GeSGIM).
METHODS
Consecutive cases were retrospectively assessed from 5 centers. Tumors were volumetrically quantified before and after surgery, and clinical data were analyzed, including IDH mutations and neurologic deficits. Kaplan-Meier estimates, accelerated failure time models (AFT), and multivariate Cox regression models were calculated to identify determinants of survival.
RESULTS
A total of 140 cases were analyzed. Gross total resection (GTR) was associated with significantly longer PFS compared to any incomplete resection (P = .009). A significant survival disadvantage was evident even for small (>0-5 ml) residuals and increased for moderate (>5-20 ml) and large remnants (>20 ml) P = .001). Accordingly, PFS increased continuously for 20% incremental steps of EOR (P < .001). AFT models supported the notion of a continuous association of RV and EOR with PFS. Multivariate Cox regression models confirmed RV (P = .01) and EOR (P = .005) as continuous prognosticators of PFS. Univariate analysis showed significant associations of RV and EOR with OS.
CONCLUSION
Our data support the hypothesis of a continuous relationship of RV and EOR with survival for LGG with superiority seen for GTR. Hence, GTR should be achieved whenever safely feasible, and resections should be maximized whenever tumor has to be left behind to spare function.
Identifiants
pubmed: 31574147
pii: 5580032
doi: 10.1093/neuros/nyz397
doi:
Types de publication
Journal Article
Multicenter Study
Langues
eng
Sous-ensembles de citation
IM
Pagination
E64-E74Informations de copyright
Copyright © 2019 by the Congress of Neurological Surgeons.