Characterization of IDH1-mutational status in gliomas by 31P/H1 MRSI

About 70% of low grade gliomas and secondary glioblastomas harbor a gene
mutation for isocitrate dehydrogenase (IDH1). Because these patients have a
better clinical perspective and IDH1 is a potential therapy target it is
important to identify the mutation. This can be done by analyzing tumor
biopsies from the brain, but as this is a major burden to the patient a
non-invasive biomarker is preferred.

It is known that tumors with the IDH1R132H mutation accumulate the
oncometabolite D-2-hydroxy-glutarate (2HG) to a high level, detectable by
non-invasive 1H MR spectroscopic imaging (MRSI). However, to uncover the signal
of 2HG a special acquisition method is required that under unfavorable
conditions may cause false positives. Hence it is necessary to find other MR
biomarkers to improve the specificity of the non-invasive detection of IDH1
mutations.

In a recent 31P MRSI study of human tumors growing in the mouse brain we
observed that the level of glycerophosphocholine (GPC) is increased and that of
phosphoethanolamine (PE) is decreased in IDHR132H gliomas1. These specific
changes were also observed in extracts of IDH1 mutated cell lines and in tumor
biopsies of patients. The GPC and PE biomarker signals are better resolved and
less affected by field homogeneities than those of 2HG.

To monitor 31P compounds non-invasively in patients we use a dual 1H/31P
detection probe for our clinical MR system with a much better signal-to-noise
than commonly available. With this coil we already successfully obtained 31P
MRS images with GPC and PC signals. The 31P spectra also show signals for
important energy compounds such as phosphocreatine (PCr), ATP and inorganic
phosphate (Pi) and from the resonance of Pi we can derive intracellular tumor
pH. Moreover, in the same examination we obtained 1H MR spectra with signals
that have been found to be of major interest to characterize brain tumors such
as of choline (Cho).

Primary objective: to non-invasively characterize the IDH1-mutational status of
gliomas in patients by measuring the ratio of GPC/PE with 31P MRSI and 2HG
content by 1H MRSI in the same examination. This is expected to make the
identification of the mutation more reliable.

Secondary objective: to include the full information content of the 1H and 31P
MR spectra in the characterization of the tumors. In particular the 31P MR
spectra are of interest as we overcome partial volume effects that hampered
previous 31P MR studies of brain tumors.

The study is observational. Patients will be examined after diagnosis and
before the tumor is (partially) removed.
The measurements take place during the same session in which the neuro
navigation MRI scan will be performed in the context of the treatment protocol.

Patients will be examined on a clinical 3T MR system. They will be in a supine
position in the magnet with the head positioned in a double tuned 1H/31P
birdcage head coil (Rapid Biomedical) equipped with an 8 channel 31P receive
array insert. The measurements take about 60 minutes per patient. Risks of the
examination are equal to the risks of regular clinical MRI exams without the
use of contrast agents.