Prediction of isocitrate dehydrogenase genotype in brain gliomas with MRI: Single-Shell versus Multishell Diffusion Models

Matteo Figini, Marco Riva, Mark Graham, Gian Marco Castelli, Bethania Fernandes, Marco Grimaldi, Giuseppe Baselli, Federico Pessina, Lorenzo Bello, Hui Zhang, Alberto Bizzi

Research output: Contribution to journalArticle

Abstract

Purpose: The primary aim of this prospective observational study was to assess whether diffusion MRI metrics correlate with isocitrate dehydrogenase (IDH) status in grade II and III gliomas. A secondary aim was to investigate whether multishell acquisitions with advanced models such as neurite orientation dispersion and density imaging (NODDI) and diffusion kurtosis imaging offer greater diagnostic accuracy than diffusion-tensor imaging (DTI). Materials and Methods: Diffusion MRI (b = 700 and 2000 sec/mm2) was performed preoperatively in 192 consecutive participants (113 male and 79 female participants; mean age, 46.18 years; age range, 14–77 years) with grade II (n = 62), grade III (n = 58), or grade IV (n = 72) gliomas. DTI, diffusion kurtosis imaging, and NODDI metrics were measured in regions with or without hyperintensity on diffusion MR images and compared among groups defined according to IDH genotype, 1p/19q codeletion status, and tumor grade by using Mann-Whitney tests. Results: In grade II and III IDH wild-type gliomas, the maximum fractional anisotropy, kurtosis anisotropy, and restriction fraction were significantly higher and the minimum mean diffusivity was significantly lower than in IDH-mutant gliomas (P = .011, P = .002, P = .044, and P = .027, respectively); areas under the receiver operating characteristic curve ranged from 0.72 to 0.76. In IDH wild-type gliomas, no difference among grades II, III, and IV was found. In IDH-mutant gliomas, no difference between those with and those without 1p/19q loss was found. Conclusion: Diffusion MRI metrics showed correlation with isocitrate dehydrogenase status in grade II and III gliomas. Advanced diffusion MRI models did not add diagnostic accuracy, supporting the inclusion of a single-shell diffusion-tensor imaging acquisition in brain tumor imaging protocols.
Original languageEnglish
Pages (from-to)788-796
Number of pages9
JournalRadiology
Volume289
Issue number3
DOIs
Publication statusPublished - Dec 1 2018

Fingerprint

Isocitrate Dehydrogenase
Glioma
Genotype
Diffusion Magnetic Resonance Imaging
Brain
Diffusion Tensor Imaging
Anisotropy
Neurites
Neuroimaging
ROC Curve
Brain Neoplasms
Observational Studies
Prospective Studies

Cite this

Prediction of isocitrate dehydrogenase genotype in brain gliomas with MRI : Single-Shell versus Multishell Diffusion Models. / Figini, Matteo; Riva, Marco; Graham, Mark; Castelli, Gian Marco; Fernandes, Bethania; Grimaldi, Marco; Baselli, Giuseppe; Pessina, Federico; Bello, Lorenzo; Zhang, Hui; Bizzi, Alberto.

In: Radiology, Vol. 289, No. 3, 01.12.2018, p. 788-796.

Research output: Contribution to journalArticle

@article{611eec9ed53f4f27ba06ff174651a784,
title = "Prediction of isocitrate dehydrogenase genotype in brain gliomas with MRI: Single-Shell versus Multishell Diffusion Models",
abstract = "Purpose: The primary aim of this prospective observational study was to assess whether diffusion MRI metrics correlate with isocitrate dehydrogenase (IDH) status in grade II and III gliomas. A secondary aim was to investigate whether multishell acquisitions with advanced models such as neurite orientation dispersion and density imaging (NODDI) and diffusion kurtosis imaging offer greater diagnostic accuracy than diffusion-tensor imaging (DTI). Materials and Methods: Diffusion MRI (b = 700 and 2000 sec/mm2) was performed preoperatively in 192 consecutive participants (113 male and 79 female participants; mean age, 46.18 years; age range, 14–77 years) with grade II (n = 62), grade III (n = 58), or grade IV (n = 72) gliomas. DTI, diffusion kurtosis imaging, and NODDI metrics were measured in regions with or without hyperintensity on diffusion MR images and compared among groups defined according to IDH genotype, 1p/19q codeletion status, and tumor grade by using Mann-Whitney tests. Results: In grade II and III IDH wild-type gliomas, the maximum fractional anisotropy, kurtosis anisotropy, and restriction fraction were significantly higher and the minimum mean diffusivity was significantly lower than in IDH-mutant gliomas (P = .011, P = .002, P = .044, and P = .027, respectively); areas under the receiver operating characteristic curve ranged from 0.72 to 0.76. In IDH wild-type gliomas, no difference among grades II, III, and IV was found. In IDH-mutant gliomas, no difference between those with and those without 1p/19q loss was found. Conclusion: Diffusion MRI metrics showed correlation with isocitrate dehydrogenase status in grade II and III gliomas. Advanced diffusion MRI models did not add diagnostic accuracy, supporting the inclusion of a single-shell diffusion-tensor imaging acquisition in brain tumor imaging protocols.",
author = "Matteo Figini and Marco Riva and Mark Graham and Castelli, {Gian Marco} and Bethania Fernandes and Marco Grimaldi and Giuseppe Baselli and Federico Pessina and Lorenzo Bello and Hui Zhang and Alberto Bizzi",
year = "2018",
month = "12",
day = "1",
doi = "10.1148/radiol.2018180054",
language = "English",
volume = "289",
pages = "788--796",
journal = "Radiology",
issn = "0033-8419",
publisher = "Radiological Society of North America Inc.",
number = "3",

}

TY - JOUR

T1 - Prediction of isocitrate dehydrogenase genotype in brain gliomas with MRI

T2 - Single-Shell versus Multishell Diffusion Models

AU - Figini, Matteo

AU - Riva, Marco

AU - Graham, Mark

AU - Castelli, Gian Marco

AU - Fernandes, Bethania

AU - Grimaldi, Marco

AU - Baselli, Giuseppe

AU - Pessina, Federico

AU - Bello, Lorenzo

AU - Zhang, Hui

AU - Bizzi, Alberto

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Purpose: The primary aim of this prospective observational study was to assess whether diffusion MRI metrics correlate with isocitrate dehydrogenase (IDH) status in grade II and III gliomas. A secondary aim was to investigate whether multishell acquisitions with advanced models such as neurite orientation dispersion and density imaging (NODDI) and diffusion kurtosis imaging offer greater diagnostic accuracy than diffusion-tensor imaging (DTI). Materials and Methods: Diffusion MRI (b = 700 and 2000 sec/mm2) was performed preoperatively in 192 consecutive participants (113 male and 79 female participants; mean age, 46.18 years; age range, 14–77 years) with grade II (n = 62), grade III (n = 58), or grade IV (n = 72) gliomas. DTI, diffusion kurtosis imaging, and NODDI metrics were measured in regions with or without hyperintensity on diffusion MR images and compared among groups defined according to IDH genotype, 1p/19q codeletion status, and tumor grade by using Mann-Whitney tests. Results: In grade II and III IDH wild-type gliomas, the maximum fractional anisotropy, kurtosis anisotropy, and restriction fraction were significantly higher and the minimum mean diffusivity was significantly lower than in IDH-mutant gliomas (P = .011, P = .002, P = .044, and P = .027, respectively); areas under the receiver operating characteristic curve ranged from 0.72 to 0.76. In IDH wild-type gliomas, no difference among grades II, III, and IV was found. In IDH-mutant gliomas, no difference between those with and those without 1p/19q loss was found. Conclusion: Diffusion MRI metrics showed correlation with isocitrate dehydrogenase status in grade II and III gliomas. Advanced diffusion MRI models did not add diagnostic accuracy, supporting the inclusion of a single-shell diffusion-tensor imaging acquisition in brain tumor imaging protocols.

AB - Purpose: The primary aim of this prospective observational study was to assess whether diffusion MRI metrics correlate with isocitrate dehydrogenase (IDH) status in grade II and III gliomas. A secondary aim was to investigate whether multishell acquisitions with advanced models such as neurite orientation dispersion and density imaging (NODDI) and diffusion kurtosis imaging offer greater diagnostic accuracy than diffusion-tensor imaging (DTI). Materials and Methods: Diffusion MRI (b = 700 and 2000 sec/mm2) was performed preoperatively in 192 consecutive participants (113 male and 79 female participants; mean age, 46.18 years; age range, 14–77 years) with grade II (n = 62), grade III (n = 58), or grade IV (n = 72) gliomas. DTI, diffusion kurtosis imaging, and NODDI metrics were measured in regions with or without hyperintensity on diffusion MR images and compared among groups defined according to IDH genotype, 1p/19q codeletion status, and tumor grade by using Mann-Whitney tests. Results: In grade II and III IDH wild-type gliomas, the maximum fractional anisotropy, kurtosis anisotropy, and restriction fraction were significantly higher and the minimum mean diffusivity was significantly lower than in IDH-mutant gliomas (P = .011, P = .002, P = .044, and P = .027, respectively); areas under the receiver operating characteristic curve ranged from 0.72 to 0.76. In IDH wild-type gliomas, no difference among grades II, III, and IV was found. In IDH-mutant gliomas, no difference between those with and those without 1p/19q loss was found. Conclusion: Diffusion MRI metrics showed correlation with isocitrate dehydrogenase status in grade II and III gliomas. Advanced diffusion MRI models did not add diagnostic accuracy, supporting the inclusion of a single-shell diffusion-tensor imaging acquisition in brain tumor imaging protocols.

U2 - 10.1148/radiol.2018180054

DO - 10.1148/radiol.2018180054

M3 - Article

VL - 289

SP - 788

EP - 796

JO - Radiology

JF - Radiology

SN - 0033-8419

IS - 3

ER -