Matthew Grech-Sollars1,2, Katherine Ordidge1,2, Babar Vaqas3, Lesley Honeyfield2, Sameer Khan2, Sophie Camp3, David Towey2, David Peterson3, Federico Roncaroli4, Kevin S O'Neill3, Tara Barwick2,5, and Adam D Waldman1,2
1Division of Brain Sciences, Imperial College London, London, United Kingdom, 2Department of Imaging, Imperial College NHS Healthcare Trust, London, United Kingdom, 3Department of Neurosurgery, Imperial College NHS Healthcare Trust, London, United Kingdom, 4Department of Neuropathology, Imperial College NHS Healthcare Trust, London, United Kingdom, 5Department of Surgery and Cancer, Imperial College London, London, United Kingdom
Synopsis
Choline elevation has been reported as a marker of
aggressive glioma phenotype in numerous in
vivo MRS studies, and more recently 18F-methylcholine-PET has been applied to
glioma characterisation. This study examines the relationship between MRS and
PET choline measures in defined tumour regions, in order to validate these
against tissue biomarkers of choline metabolism and proliferation. Our initial
results raise the possibility that imaging markers of choline metabolism are influenced
by inflammatory and reactive processes for low grade tumours. Background
Choline
containing compounds (cho) are key components of cell membranes; their
metabolism important in membrane turnover, and therefore cellular
proliferation. Cho elevation has been reported as a marker of aggressive glioma
phenotype in numerous in vivo
magnetic resonance spectroscopy (MRS) studies, and more recently 18F-methylcholine-PET
has been applied to glioma characterisation. This preliminary study aims to
examine the relationship between MRS and PET imaging cho measures in defined
tumour regions, and validate these against tissue biomarkers of cho metabolism
and proliferation.
Methods
12 patients
(7 male, 5 female; aged 23-73 years, mean 41 years), with suspected primary
supratentorial glioma suitable for biopsy/resection were recruited to this
pilot study. An MRI protocol, including multivoxel MRS (TE=30ms; TR=1500ms), was
acquired at 3T using a 32 channel head coil; PET/CT was performed with 285MBq
of 18F-methylcholine, with a dedicated 45 minute brain dynamic list mode
acquisition. Stereotactic biopsies were obtained using an ultrasound based neuro-navigational
system from low and high cho areas identified on MRS and PET, and fused with
high resolution structural images (Figure 1).
Results
Preliminary data from patients with low- and mid-grade
tumours suggests spatial concordance between high cho uptake on PET and
elevated MRS cho/cr ratios. In higher grade tumours, areas of high cho PET
uptake correlated with the contrast-enhancing tumour. Higher grade tumours had
increased uptake on PET compared to lower grade tumours and contra-lateral
white matter (Figures 2 and 3). Maximum Cho/Cr ratios on MRS did not show a significant
statistical difference between tumour grades (Figure 3). Tissue from 5 patients (3 WHO grade-II, 2
grade-III) showed high cho areas on imaging correlate with reactive gliosis and
macrophage infiltrates, but not mitotic activity or denser cellularity. In one predominantly grade II astrocytoma,
areas of elevated uptake on PET showed denser cellularity and mitoses, in
keeping with early malignant progression.
Conclusions
Preliminary findings suggest cho PET and MRS may detect
regions of inflammatory infiltrate, rather than showing a simple relationship
with high tumour cellular proliferation.
In one case, elevated cho markers corresponded to early focal malignant
transformation. Whilst these imaging methods show continued promise for glioma
characterisation, further validation against tissue enzymatic and genetic
biomarkers, and clinical outcomes is essential for their rational translation
into wider clinical practice.
Acknowledgements
This
work was funded by the NIHR Imperial Biomedical Research Centre. MGS is partly
funded by The Brain Tumour Charity and Brain Tumour Research Campaign. We would
also like to thank Mr. Daniel Darian for his help in this study. MRI and PET
imaging was carried out at the Imperial College Clinical Imaging Facility.References
No reference found.