Metabolic Profiling of Malignant Transformation and IDH-mutation in Diffuse Infiltrating Gliomas

Llewellyn Jalbert¹, Adam Elkhaled¹, Joanna J Phillips², Evan Neill³, Marram P Olson³, Mitchel S Berger⁴, John Kurhanewicz^1,3, Susan M Chang⁴, and Sarah J Nelson^1,3

¹Department of Bioengineering & Therapeutic Sciences, University of California, San Francisco (UCSF), San Francisco, CA, United States, ²Department of Pathology, University of California, San Francisco (UCSF), San Francisco, CA, United States, ³Department of Radiology & Biomedical Imaging, University of California, San Francisco (UCSF), San Francisco, CA, United States, ⁴Department of Neurological Surgery, University of California, San Francisco (UCSF), San Francisco, CA, United States

Synopsis

Patients diagnosed with infiltrating low-grade glioma have a relatively long survival, and a balance is often struck between treating the tumor and impacting quality of life. Aggressive treatments are typically reserved for lesions that have undergoing malignant transformation (MT) to a higher-grade lesion. Mutations in the isocitrate dehydrogenase 1 & 2 oncogenes and production of 2-hydroxyglutarate further characterize these tumors and are associated with improved outcome and treatment sensitivity. In this study, we found distinct metabolic profiles associated with patients' tumors that had undergone MT, as well as contained the IDH-mutated genotype, using proton HR-MAS spectroscopy.

Introduction

Infiltrating gliomas comprise eighty percent of malignant brain tumors and are graded according to criteria set by the World Health Organization (WHO)¹. In contrast to primary Glioblastoma Multiforme (GBM), which has an estimated median overall survival (OS) of 15 months under the current clinical paradigm, patients diagnosed with a WHO Grade II glioma can survive for years, or even decades^2,3. Because of this relatively long survival, a balance is often struck between treating the tumor and impacting quality of life. More aggressive treatments are often reserved for lesions that recur after undergoing malignant transformation (MT) to a higher-grade anaplastic glioma or secondary GBM⁴. Missense mutations in the isocitrate dehydrogenase 1 & 2 (IDH1/2) oncogenes and production of 2-hydroxyglutarate are also present in approximately 70% of these lesions, and are associated with improved outcome and sensitivity to treatment^5,6,7.

Objective

The objective of this study was to characterize the ex vivo metabolic profiles from patients diagnosed with infiltrating glioma. These profiles may enhance the understanding of the metabolic properties of this disease and allow for improved non-invasive monitoring and diagnosis of such patients using in vivo MRSI. Given the key role of 2HG in IDH-mutated lesions, we additionally sought to correlate levels of 2HG with other metabolites and histopathology markers.

Methods

Patient Accrual: One hundred and twenty-six patients with either a prior diagnosis of WHO Grade II glioma or a newly diagnosed, non-enhancing lesion were included in our IRB-approved study. Patients were recruited immediately prior to resection at the time of first diagnosis or at suspected recurrence, when progression to a higher grade is often observed.

In vivo MR Scans: Preoperative MR studies were conducted at either 1.5 or 3 Tesla. In addition to standard T₁ and T₂ weighted anatomical imaging, the scans included 6-directional axial Diffusion Weighted Imaging (DWI) with b=1000s/mm²; lactate-edited 3D MRSI with PRESS volume localization; and dynamic  Perfusion Weighted Imaging (PWI) with a 5ml/s injection of 0.1mmol/kg body weight Gd-DTPA.

Tissue Acquisition: Tissue sample locations  were selected in BrainLab navigation  software based on surgically accessible areas  with low ADC, elevated Choline/N- Acetylaspartate index (CNI), or elevated  PWI peak height and reduced %-recovery. After  surgical excision, tissue samples were  immediately bisected: one half was snap  frozen in liquid nitrogen and stored at -80°C  for ¹H HR-MAS spectroscopy; the other was  formalin-fixed, dehydrated by graded ethanols, and embedded in wax  using standardized techniques for tissue processing and immunohistochemistry evaluation and scoring.

Ex vivo ¹H HR-MAS: Tissue samples were placed in a 35μl zirconium rotor with 3μl 99.9% atom-D deuterium oxide containing 3-trimethylsilyl propionic acid (TSP). Samples were scanned at 11.7 Tesla, 1° C, 2250Hz spin rate in a 4mm gHZ nanoprobe with a Varian INOVA 500 MHz multi-nuclear spectrometer. A 1D Carr-Purcell-Meiboom-Gill (CPMG) was run with TR/TE=4s/144ms, 512 scans, 40,000 acquired points, 90° pulse angle, 20000Hz spectral width, with an acquisition time of 35 minutes. Metabolite levels were evaluated using the High Resolution Quantum Estimation (HR-QUEST) semi-parametric algorithm⁸ and customized brain tumor metabolite basis set. Parameter fits with less than 13% Cramer-Rao error estimates were included for analysis. Mixed-effects modeling and a Kendell correlation test were employed to assess statistical significance (p < 0.05).

Results

A summary of our patient population is presented in Table 1. Fifty-one percent of the population was found to have undergone MT and forty-nine percent were Grade II. The majority (88%) of lesions had IDH-mutations. There was significant elevation of several metabolites from lesions that had undergone MT. These varied based on tumor grade and histological subtype, as presented in Figure 1. IDH-mutant lesions were found to have elevated 2HG levels and decreased glutamate (Glu). Averaged spectra for the distinct histological grades and IDH-genotypes are presented in Figure 2. Spectra of lesions with MT displayed marked elevations of the choline-containing compounds (PC, GPC, and Cho), 2-hydroxyglutate (2HG), and several other metabolites including taurine (Tau), hypotaurine (hTau), glycine (Gly), Glu, glutamine (Gln), glutathione (GSH), alanine (Ala), aspartate (Asp), betaine (Bet), glucose (Glc) and phosphoethanolamine (PE). Measurements of PCr/Cr remained consistent across all grades. A clustered heatmap of the spectral levels across all grades is presented in Figure 3. 2HG levels were found to significantly correlate with mitotic activity by MIB1 staining as well as with the metabolites presented in Table 2.

Conclusions

The ultimate goal of this work is to improve the clinical management of patients with infiltrating glioma. The spectral profiles obtained in this study may aid in developing non-invasive MRSI methods to better diagnose and monitor patients based on underlying tumor metabolism, and further characterize the IDH-mutated molecular subtype.

Acknowledgements

We would like to acknowledge support from the Brain Tumor Research Center at UCSF in collecting and analyzing the tissue samples, as well as from staff in the Margaret Hart Surbeck Laboratory for Advanced Imaging. We would particularly like to express our gratitude to S. Ronen, D. Vigneron, and J. Crane, for their technical assistance and guidance during this project.

References

[1] Riemenschneider MJ, Reifenberger G. Molecular neuropathology of low-grade gliomas and its clinical impact. Adv Tech Stand Neurosurg. 2010; 35:35–64.

[2] Young RM, Jamshidi A, Davis G, Sherman JH. Current trends in the surgical management and treatment of adult glioblastoma. Ann Transl Med. 2015; 3: 121

[3] Claus EB, Walsh KM, Wiencke JK, Molinaro AM, Wiemels JL, Schildkraut JM, Bondy ML, Berger M, Jenkins R, Wrensch M. Survival and low-grade glioma: the emergence of genetic information. Neurosurg Focus. 2015. 38, E6

[4] Grier JT, Batchelor T. Low-grade gliomas in adults. Oncologist. 2006; 11:681–693.

[5] Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009; 360: 765–73.

[6] Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, et al. Cancer- associated IDH1 mutations produce 2-hydroxyglutarate. Nature. 2009; 462:739–744.

[7] Yen KE, Bittinger MA, Su SM, Fantin VR. Cancer-associated IDH mutations: biomarker and therapeutic opportunities. Oncogene. 2010; 29(49):6409-17.

[8] Ratiney H, Albers MJ, Rabeson H, Kurhanewicz J. Semi-parametric time-domain quantification of HR-MAS data from prostate tissue. NMR Biomed. 2010; 23(10): 1146-1157

Figures

Figure 1. Flow diagram of metabolite differences associated with MT and IDH-mutation. Here we present the metabolic differences associated with malignant transformation and IDH-mutation found across all histologies, as well as within individual astrocytoma and oligodendroglioma subtypes.

Figure 2. Averaged HR-MAS spectral profiles from histological grades and IDH-genotypes. After normalizing by tissue weight we produced an averaged spectrum for individual grades and distinct IDH-genotypes. Spectral profiles demonstrated elevations in metabolites associated with MT to Grade III (A) and Grade IV (B), and differences based on IDH-status (C).

Figure 3. Metabolic spectral heatmap across histological grades. The heatmap was generated from the HR-QUEST quantification of individual tissue sample spectra (rows) and organized by metabolite (columns). Data were normalized and hierarchically clustered within each grade providing comprehensive visualization of the dataset and key metabolite differences across histological grades.

Table 1. Glioma patient and tissue sample population by grade, histological subtype, and IDH-mutation status. The patient population comprised astrocytoma, oligodendroglioma, and oligoastrocytoma histological subtypes. The majority of patients (88%) patients harbored IDH-mutant lesions and at the time of recurrence and 64 patients (51%) had undergone MT.

Table 2. Correlation scatterplots of 2HG levels with other metabolites and histopathological measures. We found significant correlation between 2HG levels and mitotic activity (MIB1) as well as several brain tumor metabolites.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)

0228