Robust and highly selective proton MRSI on a clinical 3 T system using a second order gradient insert, for application in schizophrenia - PROJECT SUMMARY –
The glutamate hypothesis of schizophrenia has generated significant interest in the utility of Magnetic Resonance
Spectroscopy and Spectroscopic Imaging (MRS/MRSI) to study the pathophysiology of schizophrenia in recent
times. The glutamate hypothesis provides a complementary perspective to the well-established dopaminergic
hypothesis, suggesting that dysfunction of the glutamatergic system plays a role in the etiology of schizophrenia.
Previous MRS/MRSI studies conducted have reported alterations of glutamate (Glu) and glutamine (Gln) and
Glx (Glu + Gln) in various brain regions, consistent with neuronal dysfunction in chronic stages of the disease.
The robust and reliable acquisition of MRSI data is however hindered by extracranial lipid contaminants. Most
extracranial lipid suppression methods reported to-date provide inadequate lipid signal suppression, to allow
artifact-free spectroscopic evaluation of cortical grey matter proximal to the skull. As a result, MRSI studies in
schizophrenia conducted to-date at 3 T have two major limitations: 1) exclusion of cortical tissue proximal to
extracranial lipids, and 2) uncertainties associated with Glu/Gln separation, thus predominantly reporting Glx
changes. Because glutamatergic activity is inferred from the Glu/Gln ratio, it is critical to develop accurate and
separate quantitation of Glu and Gln to further the value of testing the glutamate hypothesis.
This work involves the development of a high resolution (~ 5 x 5 x 10 mm3) 3D proton MRSI method using
ECLIPSE, a second order gradient insert, for highly selective elliptical localization over an axial plane. High
selectivity and unparalleled extracranial lipid suppression was previously demonstrated with ECLIPSE, which
allows artifact-free interrogation of cortical tissue adjacent to the skull. The MRSI acquisition will be extended
with an optimized multiple-echo time (TE) acquisition combined with a macromolecule targeted inversion
recovery component to nullify the macromolecular baseline. In combination the proposed MRSI method is
expected to allow reliable quantification of Glu/Gln. Following development of the MRSI sequence, robustness,
reproducibility metrics, and Glu/Gln quantification performance will be evaluated in phantom and on a healthy
cohort of volunteers, followed by tolerance and translatability of the methods evaluated on a heterogenous cohort
of first episode psychosis or patients with recent onset schizophrenia spectrum disorders.
The successful completion of the proposed methods will allow reliable quantitative MRS images of Glu and Gln,
among other commonly detected metabolites, over an axial slab including frontal cortical gray matter proximal
to the skull. Given that the frontal region is the most significantly impacted by schizophrenia, the ability provided
for reliable metabolic images that are sensitive to altered glutamatergic activity is expected to have a high impact
by providing a reliable metabolic imaging tool for characterizing disease severity, heterogeneity, and treatment
effects in patients. The development of reliable Glu/Gln quantification with MRSI at 3 T has further clinical
significance, since the 3 T field strength is the most accessible high field MRI system in the clinical setting.
