ELECTRONIC SUPPLEMENTARY MATERIAL
Advanced MRI may complement histological diagnosis of lower grade gliomas and help in predicting survival.
Journal of Neuro-Oncology
Cuccarini V, Erbetta A, Farinotti M, Cuppini L, Ghielmetti F, Pollo B, Di Meco F, Grisoli M, Filippini G, Finocchiaro G, Bruzzone MG, and Eoli M.
Corresponding author: Cuccarini Valeria, MD, Unit of Neuro-Radiology, Fondazione IRCCS Istituto Neurologico Carlo Besta, Via Celoria 11, 20133 Milano, Italy, Ph 0039 02 23942449, Fax 0039 02 23942543,
METHODS
Online Resource 1
Conventional Magnetic Resonance Imaging
cMRI was performed with a 1.5 T unit (Siemens, Avanto) using the following protocols: sagittal Spin Echo (SE) T1-weighted imaging (w.i.) with repetition time (TR)/echo time (TE) 500 ms/9 ms and4 mm slice thickness; bi-commissural double SE T2 w.i. and proton density (PD) w.i. with TR/TE 3210 ms/110 ms and TR/TE 3210 ms/22 ms, respectively, and 5 mm slice thickness; coronal Fluid Attenuated Inversion Recovery (FLAIR) with TR/TE 8200 ms/109 ms, Inversion Time (TI) 2500 ms, and 5 mm slice thickness; and bi-commissural SE T1 w.i. with TR/TE 503 ms/9 ms and 6 mm slice thickness. Isotropic post-contrast volumetric T1 w.i. with TR/TE/TI 1640 ms/2 ms/549 ms and1 mmslice thickness was performed after the acquisition of the perfusion MRI data.
The following features were considered for complete cMRI analysis: homogeneity on T1 and T2 images,presence, extension and pattern of contrast enhancement, mass effect, multifocality, leptomeningeal and/or subependimal seeding, and extension to the contralateral hemisphere.
Online Resource 2
Advanced Magnetic Resonance Imaging
aMRI was performed with a 1.5 T unit (Siemens, Avanto).
PWI.Dynamic susceptibility-weighted contrast-enhanced (DSC) PWI was acquired with the following parameters and features: TR/TE 1930 ms/50 ms, bi-commissural, 5 mm with no gap, 50 volumes, Gadovist® administration (0.1 cc/Kg, 5 ml/sec and fixed 3 cc pre-bolus). Maximum tumoural cerebral blood volume (CBV) was obtained by identifying regions of maximum perfusion from colour maps. Three different 24-pixel-sized circular regions of interest (ROI) were placed on the highest colour areas of the solid portions of tumoural lesions, and their average values were recorded. The CBV values were also normalized (rCBV). Each rCBV tumour/normal tissue value was obtained as the ratio between maximum CBV value acquired from a tumour ROI and the CBV value acquired from an identical ROI positioned on the contralateral healthy white matter (CHWM). rCBV max was obtained from the highest average value ROI.
DWI.A single-shot echo-planar sequence that was bi-commissural with TR/TE 3800 ms/74 ms, 3 orthogonal directions, 3.5 mm slice thickness, 35% gap, and b-values of 0-500-1000 s mm-2 was used. Three different 24-pixel-sized circular ROIs were placed on different areas of the tumour on the basis of cMRI and ADC map appearance, and their average values were registered. The ADC values were also normalized (rADC) by dividing tumour ADC values by those obtained from the CHWM.
MRS.Multivoxel 2D Proton MRS with Point Resolved Spectroscopic Imaging (PRESS) was used. The acquisition parameters included TR=1200, TE=135, 32X32 matrix, voxel size 10X10X15 mm, with 6 saturation bands fully enclosing the volume of interest (VOI).The following features were considered: reduction of N-Acetyl Aspartate (NAA) metabolite level, choline peak increase, creatine peak, Cho/Crratio,and the presence of lactate and/or lipids. The two highest Cho/Cr ratios were considered for statistical analysis.
TABLES
Online Resource 3 PWI, MRS, and DWI normalized values (mean±sd in different types of grade II (n=40) gliomas.rCBVmax (PWI) / rADCmin (DWI) / Cho/Crmax (MRS)
FA (n=13) / 1.4±1.1 / 1.9±0.4 / 1.7±0.4
OA (n=22) / 1.5±1.3 / 1.7±0.4 / 1.9±0.6
ODG (n=5) / 2.7±2.9 / 1.5±0.2 / 2.2±1.1
FA: fibrillary astrocytoma; OA: oligoastrocytoma; ODG: oligodendroglioma.
Online Resource 4 Survival data of cases with mismatched radiologic vs histopathologic diagnoses.
CASE / aMRI / Histological Grade (Type) / OS months / PFS months
1 a / HGG / II (ODG) / >65 / 33
2 / HGG / II (FA) / 65 / 10
3 / HGG / II (FA) / 61 / 28
4 a / HGG / II (OA) / >57 / 55
5 b / HGG / II (OA) / >48 / >48
6 c / HGG / II (OA) / 4 / 4
7 b / LGG / III (OA) / >80 / >80
8 a / LGG / III (A) / >71 / 67
9 / LGG / III (OAA) / 66 / 66
10 a / LGG / III (AA) / >60 / 27
11 a / LGG / III (OAA) / >60 / 22
12 a / LGG / III (OAA) / >58 / 30
13 b / LGG / III (OAA) / >57 / >57
14 b / LGG / III (AA) / >57 / >57
15 b / LGG / III (OAA) / >54 / >54
16 a / LGG / IV (GBM) / >49 / 33
17 / LGG / III (AA) / 16 / 5
18 / LGG / III (OAA) / 15 / 8
ODG: oligodendroglioma; FA: fibrillary astrocytoma; OA: oligoastrocytoma; OAA: anaplastic oligoastrocytoma; AA: anaplastic astrocytoma; GBM: glioblastoma.
a alive, progressed disease; b alive, currently free of disease; c partial resection.
Online Resource 5 aMRI cut-off values with respect to histopathology from previous studies that mainly focused on grade II vs grade III gliomas.
Paper first author (year of publication) / No. of Subjects and
Histological Grade / aMRI Parameter (cut-off) / Compared glioma populations
Maia (2005) / 13 II, 7 III / CBV (not specified) / II vs III
Fan (2006) / 14 LLG, 8 HGG
(non-enhancing) / CBV(not specified)
ADC (not specified) / LGG vs HGG
Murakami (2009) / 8 I, 8 II, 12 III, 22 IV / ADC (0.31) / LGG vs HGG
II vs III
Morita (2010) / 6 II, 11 III
(non-enhancing As) / CBV (0.94) / II vs III
Liu (2011) / 37 II, 15 III
(non-enhancing) / CBV(not significant)
ADC (not significant) / II vs III
Haegler (2012) / 23 I- II, 14 III / Association of CE, CBV, CBF
(not specified) / I-II vs III
Hilario (2012) / 32 II, 29 III / ADC (not significant) / II vs III
Liu (2012) / 4 I, 6 II, 8 III, 15 IV / Cho/Cr (2.01) / LGG vs HGG
Khalid (2013) / 46 II, 29 III
(ODG) / ADC (0.925) / II vs III
(ODG)
Roy (2013) / 24 II, 18 III, 20 IV / Cho/Cr (not specified) / LGG vs HGG
Sahin (2013) / 14 II, 6 III
(non-enhancing) / CBV (not significant)
Cho/Cr (1.3) / II vs III
Bradac (2014) / 16 II, 11 III / Cho/Cr (not specified) / II vs III
Cui (2014) / 35 II, 22 III / ADC (not significant) / II vs III
A: astrocytoma; CE: contrast enhancement; HGG: high-grade glioma; LGG: low-grade glioma; ODG: oligodendroglioma; vs: versus.
Online Resource 6 Previous aMRI and outcome analysis studies that mainly focused on LGG.
Study / No. of Subjects and
Histological Grade / aMRI Parameter (cut-off) / Outcome Parameter / Follow up (range)
Chaskis (2006) / 6 I, 9 II, 9 III, 31 IV / CBV (1.7) / OS / 60 m
Law (2008) / 53 II, 84 III, 52 IV / CBV (1.75) / PFS / 11 m (1-119)
Bisdas (2009) / 7 I, 9 II, 5 III, 6 IV / CBV (3.8)a / PFS / 60 m
Caseiras (2010) / 63 II (34 A, 20 ODG, 9 OA) / CBV (1.75) / PFS / 12-29 m
Hattingen (2010) / 23 II, 26 III / rCr (0.93) / PFS / 4-51 m
Hlaihel (2010) / 21 II (ODG) / Cho/Cr (2.4) / PFSb / 24 m
Majchrzak (2012) / 68 II / CBV (1.75) / PFS and OSc / 34 m (21-49)
Sahin (2013) / 14 II, 6 III (non-enhancing) / CBV (1.63) / PFS (trend, p>0.05) / 6-15 m
Cui (2014) / 1 I, 35 II, 22 III, 24 IV / ADC (1.56) / prognosisd / -
Lober (2014) / 20 diffuse pontinee (paediatric) / ADC (1.29) / OS / 7 m (1-36)
A: astrocytoma; m: months; OA: oligoastrocytoma; ODG: oligodendroglioma
aStatistically significant correlation with PFS only excluding OA and ODG; b only five progressions of disease; c only four deaths; d prognostically favourable vs unfavourable gliomas on the basis of correlation with age, KPS, histology, dimension, midline crossing, 1p/19q and IDH status [29]; e only four patients had a histological diagnosis, including one grade II, two grade III and one grade IV.