Characterisation of paediatric brain tumours by their MRS metabolite profiles

Simrandip K. Gill; Heather E. L. Rose; Martin Wilson; Daniel Rodriguez Gutierrez; Lara Worthington; Nigel P. Davies; Lesley MacPherson; Darren R. Hargrave; Dawn E. Saunders; Christopher A. Clark; Geoffery S. Payne; Martin O. Leach; Franklyn A. Howe; Dorothee P. Auer; Tim Jaspan; Paul S. Morgan; Richard G. Grundy; Shivaram Avula; Barry Pizer; Theodoros N. Arvanitis; Andrew C. Peet

doi:10.1002/nbm.5101

Characterisation of paediatric brain tumours by their MRS metabolite profiles

Simrandip K. Gill, Heather E. L. Rose, Martin Wilson, Daniel Rodriguez Gutierrez, Lara Worthington, Nigel P. Davies, Lesley MacPherson, Darren R. Hargrave, Dawn E. Saunders, Christopher A. Clark, Geoffery S. Payne, Martin O. Leach, Franklyn A. Howe, Dorothee P. Auer, Tim Jaspan, Paul S. Morgan, Richard G. Grundy, Shivaram Avula, Barry Pizer, Theodoros N. ArvanitisAndrew C. Peet^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

24 Downloads (Pure)

Abstract

¹H‐magnetic resonance spectroscopy (MRS) has the potential to improve the noninvasive diagnostic accuracy for paediatric brain tumours. However, studies analysing large, comprehensive, multicentre datasets are lacking, hindering translation to widespread clinical practice. Single‐voxel MRS (point‐resolved single‐voxel spectroscopy sequence, 1.5 T: echo time [TE] 23–37 ms/135–144 ms, repetition time [TR] 1500 ms; 3 T: TE 37–41 ms/135–144 ms, TR 2000 ms) was performed from 2003 to 2012 during routine magnetic resonance imaging for a suspected brain tumour on 340 children from five hospitals with 464 spectra being available for analysis and 281 meeting quality control. Mean spectra were generated for 13 tumour types. Mann–Whitney U‐tests and Kruskal–Wallis tests were used to compare mean metabolite concentrations. Receiver operator characteristic curves were used to determine the potential for individual metabolites to discriminate between specific tumour types. Principal component analysis followed by linear discriminant analysis was used to construct a classifier to discriminate the three main central nervous system tumour types in paediatrics. Mean concentrations of metabolites were shown to differ significantly between tumour types. Large variability existed across each tumour type, but individual metabolites were able to aid discrimination between some tumour types of importance. Complete metabolite profiles were found to be strongly characteristic of tumour type and, when combined with the machine learning methods, demonstrated a diagnostic accuracy of 93% for distinguishing between the three main tumour groups (medulloblastoma, pilocytic astrocytoma and ependymoma). The accuracy of this approach was similar even when data of marginal quality were included, greatly reducing the proportion of MRS excluded for poor quality. Children's brain tumours are strongly characterised by MRS metabolite profiles readily acquired during routine clinical practice, and this information can be used to support noninvasive diagnosis. This study provides both key evidence and an important resource for the future use of MRS in the diagnosis of children's brain tumours.

Original language	English
Article number	5101
Number of pages	21
Journal	NMR in biomedicine
Early online date	1 Feb 2024
DOIs	https://doi.org/10.1002/nbm.5101
Publication status	E-pub ahead of print - 1 Feb 2024

Bibliographical note

Funding information:
This work was funded by the CRUK and EPSRC Cancer Imaging Programme at the CCLG, in association with the Medical Research Council (MRC) and Department of Health (England) (C7809/A10342). The study benefitted from the ECMC Paediatric Network funded by Cancer Research UK and NIHR (C8232/A25261). This study has received funding from the CRUK and EPSRC Cancer Imaging Centre in association with the MRC and Department of Health (England) (Grants C1060/A10334 and C1060/A16464) and NHS funding to the NIHR Biomedical Research Centre and the Clinical Research Facility in Imaging at The Institute of Cancer Research and The Royal Marsden Hospital. Birmingham Children's Hospital acknowledges data collection at the NIHR 3T MRI Centre. Professor A. C. Peet acknowledges support from an NIHR Research Professorship (NIHR-RP-R2-12-019). S. K. Gill was partially funded by Action Medical Research and The Brain Tumour Charity (GN2181). H. E. L. Rose was funded by Little Princess Trust in collaboration with the Children's Cancer and Leukaemia Group (CCLG 2019 26) and Children with Cancer (15/188). We also acknowledge funding from Help Happy Help Others Cure and the Children's Research Fund. Martin O. Leach is an NIHR Emeritus Senior Investigator. Dr D. R. Hargrave is supported by the NIHR Biomedical Research Centre at Great Ormond Street Hospital for Children NHS Foundation Trust and University College London. Tim Jaspan and Paul S. Morgan are members of the Roche-sponsored ‘HERBY’ Trial Steering Group, for which their employing institution received financial support. We would like to thank Dr Karen A. Manias for her help in producing Figure 9.

Keywords

1H‐magnetic resonance spectroscopy
paediatric
classification
brain tumours
metabolites

Access to Document

10.1002/nbm.5101Licence: Creative Commons: Attribution (CC BY)

GillS2024CharacterisationFinal published version, 3.83 MBLicence: Creative Commons: Attribution (CC BY)

NIHR Professorship - Supporting Research into New Imaging Methods for Children
Peet, A.
NIHR TRAINEES COORDINATING CENTRE
1/01/13 → 31/03/18
Project: Other Government Departments
Development and Evaluation of MR based Functional Imaging for the Enhanced Management of Childhood Cancer - A Childrens Cancer and Leukaemia Group Initiative
Peet, A. & Arvanitis, T.
Cancer Research UK
1/12/08 → 31/03/15
Project: Research

Cite this

Gill, S. K., Rose, H. E. L., Wilson, M., Rodriguez Gutierrez, D., Worthington, L., Davies, N. P., MacPherson, L., Hargrave, D. R., Saunders, D. E., Clark, C. A., Payne, G. S., Leach, M. O., Howe, F. A., Auer, D. P., Jaspan, T., Morgan, P. S., Grundy, R. G., Avula, S., Pizer, B., ... Peet, A. C. (2024). Characterisation of paediatric brain tumours by their MRS metabolite profiles. NMR in biomedicine, Article 5101. Advance online publication. https://doi.org/10.1002/nbm.5101

@article{241101794fd5450387f058529ba68b1a,

title = "Characterisation of paediatric brain tumours by their MRS metabolite profiles",

abstract = "1H‐magnetic resonance spectroscopy (MRS) has the potential to improve the noninvasive diagnostic accuracy for paediatric brain tumours. However, studies analysing large, comprehensive, multicentre datasets are lacking, hindering translation to widespread clinical practice. Single‐voxel MRS (point‐resolved single‐voxel spectroscopy sequence, 1.5 T: echo time [TE] 23–37 ms/135–144 ms, repetition time [TR] 1500 ms; 3 T: TE 37–41 ms/135–144 ms, TR 2000 ms) was performed from 2003 to 2012 during routine magnetic resonance imaging for a suspected brain tumour on 340 children from five hospitals with 464 spectra being available for analysis and 281 meeting quality control. Mean spectra were generated for 13 tumour types. Mann–Whitney U‐tests and Kruskal–Wallis tests were used to compare mean metabolite concentrations. Receiver operator characteristic curves were used to determine the potential for individual metabolites to discriminate between specific tumour types. Principal component analysis followed by linear discriminant analysis was used to construct a classifier to discriminate the three main central nervous system tumour types in paediatrics. Mean concentrations of metabolites were shown to differ significantly between tumour types. Large variability existed across each tumour type, but individual metabolites were able to aid discrimination between some tumour types of importance. Complete metabolite profiles were found to be strongly characteristic of tumour type and, when combined with the machine learning methods, demonstrated a diagnostic accuracy of 93% for distinguishing between the three main tumour groups (medulloblastoma, pilocytic astrocytoma and ependymoma). The accuracy of this approach was similar even when data of marginal quality were included, greatly reducing the proportion of MRS excluded for poor quality. Children's brain tumours are strongly characterised by MRS metabolite profiles readily acquired during routine clinical practice, and this information can be used to support noninvasive diagnosis. This study provides both key evidence and an important resource for the future use of MRS in the diagnosis of children's brain tumours.",

keywords = "1H‐magnetic resonance spectroscopy, paediatric, classification, brain tumours, metabolites",

author = "Gill, {Simrandip K.} and Rose, {Heather E. L.} and Martin Wilson and {Rodriguez Gutierrez}, Daniel and Lara Worthington and Davies, {Nigel P.} and Lesley MacPherson and Hargrave, {Darren R.} and Saunders, {Dawn E.} and Clark, {Christopher A.} and Payne, {Geoffery S.} and Leach, {Martin O.} and Howe, {Franklyn A.} and Auer, {Dorothee P.} and Tim Jaspan and Morgan, {Paul S.} and Grundy, {Richard G.} and Shivaram Avula and Barry Pizer and Arvanitis, {Theodoros N.} and Peet, {Andrew C.}",

note = "Funding information: This work was funded by the CRUK and EPSRC Cancer Imaging Programme at the CCLG, in association with the Medical Research Council (MRC) and Department of Health (England) (C7809/A10342). The study benefitted from the ECMC Paediatric Network funded by Cancer Research UK and NIHR (C8232/A25261). This study has received funding from the CRUK and EPSRC Cancer Imaging Centre in association with the MRC and Department of Health (England) (Grants C1060/A10334 and C1060/A16464) and NHS funding to the NIHR Biomedical Research Centre and the Clinical Research Facility in Imaging at The Institute of Cancer Research and The Royal Marsden Hospital. Birmingham Children's Hospital acknowledges data collection at the NIHR 3T MRI Centre. Professor A. C. Peet acknowledges support from an NIHR Research Professorship (NIHR-RP-R2-12-019). S. K. Gill was partially funded by Action Medical Research and The Brain Tumour Charity (GN2181). H. E. L. Rose was funded by Little Princess Trust in collaboration with the Children's Cancer and Leukaemia Group (CCLG 2019 26) and Children with Cancer (15/188). We also acknowledge funding from Help Happy Help Others Cure and the Children's Research Fund. Martin O. Leach is an NIHR Emeritus Senior Investigator. Dr D. R. Hargrave is supported by the NIHR Biomedical Research Centre at Great Ormond Street Hospital for Children NHS Foundation Trust and University College London. Tim Jaspan and Paul S. Morgan are members of the Roche-sponsored {\textquoteleft}HERBY{\textquoteright} Trial Steering Group, for which their employing institution received financial support. We would like to thank Dr Karen A. Manias for her help in producing Figure 9.",

year = "2024",

month = feb,

day = "1",

doi = "10.1002/nbm.5101",

language = "English",

journal = "NMR in biomedicine",

issn = "0952-3480",

publisher = "Wiley",

}

Gill, SK, Rose, HEL, Wilson, M, Rodriguez Gutierrez, D, Worthington, L, Davies, NP, MacPherson, L, Hargrave, DR, Saunders, DE, Clark, CA, Payne, GS, Leach, MO, Howe, FA, Auer, DP, Jaspan, T, Morgan, PS, Grundy, RG, Avula, S, Pizer, B, Arvanitis, TN & Peet, AC 2024, 'Characterisation of paediatric brain tumours by their MRS metabolite profiles', NMR in biomedicine. https://doi.org/10.1002/nbm.5101

TY - JOUR

T1 - Characterisation of paediatric brain tumours by their MRS metabolite profiles

AU - Gill, Simrandip K.

AU - Rose, Heather E. L.

AU - Wilson, Martin

AU - Rodriguez Gutierrez, Daniel

AU - Worthington, Lara

AU - Davies, Nigel P.

AU - MacPherson, Lesley

AU - Hargrave, Darren R.

AU - Saunders, Dawn E.

AU - Clark, Christopher A.

AU - Payne, Geoffery S.

AU - Leach, Martin O.

AU - Howe, Franklyn A.

AU - Auer, Dorothee P.

AU - Jaspan, Tim

AU - Morgan, Paul S.

AU - Grundy, Richard G.

AU - Avula, Shivaram

AU - Pizer, Barry

AU - Arvanitis, Theodoros N.

AU - Peet, Andrew C.

N1 - Funding information: This work was funded by the CRUK and EPSRC Cancer Imaging Programme at the CCLG, in association with the Medical Research Council (MRC) and Department of Health (England) (C7809/A10342). The study benefitted from the ECMC Paediatric Network funded by Cancer Research UK and NIHR (C8232/A25261). This study has received funding from the CRUK and EPSRC Cancer Imaging Centre in association with the MRC and Department of Health (England) (Grants C1060/A10334 and C1060/A16464) and NHS funding to the NIHR Biomedical Research Centre and the Clinical Research Facility in Imaging at The Institute of Cancer Research and The Royal Marsden Hospital. Birmingham Children's Hospital acknowledges data collection at the NIHR 3T MRI Centre. Professor A. C. Peet acknowledges support from an NIHR Research Professorship (NIHR-RP-R2-12-019). S. K. Gill was partially funded by Action Medical Research and The Brain Tumour Charity (GN2181). H. E. L. Rose was funded by Little Princess Trust in collaboration with the Children's Cancer and Leukaemia Group (CCLG 2019 26) and Children with Cancer (15/188). We also acknowledge funding from Help Happy Help Others Cure and the Children's Research Fund. Martin O. Leach is an NIHR Emeritus Senior Investigator. Dr D. R. Hargrave is supported by the NIHR Biomedical Research Centre at Great Ormond Street Hospital for Children NHS Foundation Trust and University College London. Tim Jaspan and Paul S. Morgan are members of the Roche-sponsored ‘HERBY’ Trial Steering Group, for which their employing institution received financial support. We would like to thank Dr Karen A. Manias for her help in producing Figure 9.

PY - 2024/2/1

Y1 - 2024/2/1

N2 - 1H‐magnetic resonance spectroscopy (MRS) has the potential to improve the noninvasive diagnostic accuracy for paediatric brain tumours. However, studies analysing large, comprehensive, multicentre datasets are lacking, hindering translation to widespread clinical practice. Single‐voxel MRS (point‐resolved single‐voxel spectroscopy sequence, 1.5 T: echo time [TE] 23–37 ms/135–144 ms, repetition time [TR] 1500 ms; 3 T: TE 37–41 ms/135–144 ms, TR 2000 ms) was performed from 2003 to 2012 during routine magnetic resonance imaging for a suspected brain tumour on 340 children from five hospitals with 464 spectra being available for analysis and 281 meeting quality control. Mean spectra were generated for 13 tumour types. Mann–Whitney U‐tests and Kruskal–Wallis tests were used to compare mean metabolite concentrations. Receiver operator characteristic curves were used to determine the potential for individual metabolites to discriminate between specific tumour types. Principal component analysis followed by linear discriminant analysis was used to construct a classifier to discriminate the three main central nervous system tumour types in paediatrics. Mean concentrations of metabolites were shown to differ significantly between tumour types. Large variability existed across each tumour type, but individual metabolites were able to aid discrimination between some tumour types of importance. Complete metabolite profiles were found to be strongly characteristic of tumour type and, when combined with the machine learning methods, demonstrated a diagnostic accuracy of 93% for distinguishing between the three main tumour groups (medulloblastoma, pilocytic astrocytoma and ependymoma). The accuracy of this approach was similar even when data of marginal quality were included, greatly reducing the proportion of MRS excluded for poor quality. Children's brain tumours are strongly characterised by MRS metabolite profiles readily acquired during routine clinical practice, and this information can be used to support noninvasive diagnosis. This study provides both key evidence and an important resource for the future use of MRS in the diagnosis of children's brain tumours.

AB - 1H‐magnetic resonance spectroscopy (MRS) has the potential to improve the noninvasive diagnostic accuracy for paediatric brain tumours. However, studies analysing large, comprehensive, multicentre datasets are lacking, hindering translation to widespread clinical practice. Single‐voxel MRS (point‐resolved single‐voxel spectroscopy sequence, 1.5 T: echo time [TE] 23–37 ms/135–144 ms, repetition time [TR] 1500 ms; 3 T: TE 37–41 ms/135–144 ms, TR 2000 ms) was performed from 2003 to 2012 during routine magnetic resonance imaging for a suspected brain tumour on 340 children from five hospitals with 464 spectra being available for analysis and 281 meeting quality control. Mean spectra were generated for 13 tumour types. Mann–Whitney U‐tests and Kruskal–Wallis tests were used to compare mean metabolite concentrations. Receiver operator characteristic curves were used to determine the potential for individual metabolites to discriminate between specific tumour types. Principal component analysis followed by linear discriminant analysis was used to construct a classifier to discriminate the three main central nervous system tumour types in paediatrics. Mean concentrations of metabolites were shown to differ significantly between tumour types. Large variability existed across each tumour type, but individual metabolites were able to aid discrimination between some tumour types of importance. Complete metabolite profiles were found to be strongly characteristic of tumour type and, when combined with the machine learning methods, demonstrated a diagnostic accuracy of 93% for distinguishing between the three main tumour groups (medulloblastoma, pilocytic astrocytoma and ependymoma). The accuracy of this approach was similar even when data of marginal quality were included, greatly reducing the proportion of MRS excluded for poor quality. Children's brain tumours are strongly characterised by MRS metabolite profiles readily acquired during routine clinical practice, and this information can be used to support noninvasive diagnosis. This study provides both key evidence and an important resource for the future use of MRS in the diagnosis of children's brain tumours.

KW - 1H‐magnetic resonance spectroscopy

KW - paediatric

KW - classification

KW - brain tumours

KW - metabolites

U2 - 10.1002/nbm.5101

DO - 10.1002/nbm.5101

M3 - Article

C2 - 38303627

SN - 0952-3480

JO - NMR in biomedicine

JF - NMR in biomedicine

M1 - 5101

ER -

Characterisation of paediatric brain tumours by their MRS metabolite profiles

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Projects

NIHR Professorship - Supporting Research into New Imaging Methods for Children

Development and Evaluation of MR based Functional Imaging for the Enhanced Management of Childhood Cancer - A Childrens Cancer and Leukaemia Group Initiative

Cite this