Hypoxic glucose metabolism in glioblastoma as a potential prognostic factor

  • Takuya Toyonaga
  • Shigeru Yamaguchi
  • Kenji Hirata
  • Kentaro Kobayashi
  • Osamu Manabe
  • Shiro Watanabe
  • Shunsuke Terasaka
  • Hiroyuki Kobayashi
  • Naoya Hattori
  • Tohru Shiga
  • Yuji Kuge
  • Shinya Tanaka
  • Yoichi M. Ito
  • Nagara Tamaki
Original Article

DOI: 10.1007/s00259-016-3541-z

Cite this article as:
Toyonaga, T., Yamaguchi, S., Hirata, K. et al. Eur J Nucl Med Mol Imaging (2017) 44: 611. doi:10.1007/s00259-016-3541-z

Abstract

Purpose

Metabolic activity and hypoxia are both important factors characterizing tumor aggressiveness. Here, we used F-18 fluoromisonidazole (FMISO) and F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) to define metabolically active hypoxic volume, and investigate its clinical significance in relation to progression free survival (PFS) and overall survival (OS) in glioblastoma patients.

Experimental Design

Glioblastoma patients (n = 32) underwent FMISO PET, FDG PET, and magnetic resonance imaging (MRI) before surgical intervention. FDG and FMISO PET images were coregistered with gadolinium-enhanced T1-weighted MR images. Volume of interest (VOI) of gross tumor volume (GTV) was manually created to enclose the entire gadolinium-positive areas. The FMISO tumor-to-normal region ratio (TNR) and FDG TNR were calculated in a voxel-by-voxel manner. For calculating TNR, standardized uptake value (SUV) was divided by averaged SUV of normal references. Contralateral frontal and parietal cortices were used as the reference region for FDG, whereas the cerebellar cortex was used as the reference region for FMISO. FDG-positive was defined as the FDG TNR ≥1.0, and FMISO-positive was defined as FMISO TNR ≥1.3. Hypoxia volume (HV) was defined as the volume of FMISO-positive and metabolic tumor volume in hypoxia (hMTV) was the volume of FMISO/FDG double-positive. The total lesion glycolysis in hypoxia (hTLG) was hMTV × FDG SUVmean. The extent of resection (EOR) involving cytoreduction surgery was volumetric change based on planimetry methods using MRI. These factors were tested for correlation with patient prognosis.

Results

All tumor lesions were FMISO-positive and FDG-positive. Univariate analysis indicated that hMTV, hTLG, and EOR were significantly correlated with PFS (p = 0.007, p = 0.04, and p = 0.01, respectively) and that hMTV, hTLG, and EOR were also significantly correlated with OS (p = 0.0028, p = 0.037, and p = 0.014, respectively). In contrast, none of FDG TNR, FMISO TNR, GTV, HV, patients’ age, or Karnofsky performance scale (KPS) was significantly correlated with PSF or OS. The hMTV and hTLG were found to be independent factors affecting PFS and OS on multivariate analysis.

Conclusions

We introduced hMTV and hTLG using FDG and FMISO PET to define metabolically active hypoxic volume. Univariate and multivariate analyses demonstrated that both hMTV and hTLG are significant predictors for PFS and OS in glioblastoma patients.

Keywords

Anaerobic glycolysis Glioblastoma Fluoromisonidazole Fluorodeoxyglucose Positron emission tomography 

Supplementary material

259_2016_3541_MOESM1_ESM.docx (646 kb)
ESM 1(DOCX 645 kb)

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  • Takuya Toyonaga
    • 1
  • Shigeru Yamaguchi
    • 1
    • 2
  • Kenji Hirata
    • 1
  • Kentaro Kobayashi
    • 1
  • Osamu Manabe
    • 1
  • Shiro Watanabe
    • 1
  • Shunsuke Terasaka
    • 2
  • Hiroyuki Kobayashi
    • 2
  • Naoya Hattori
    • 1
  • Tohru Shiga
    • 1
  • Yuji Kuge
    • 3
  • Shinya Tanaka
    • 4
  • Yoichi M. Ito
    • 5
  • Nagara Tamaki
    • 1
  1. 1.Department of Nuclear MedicineHokkaido University Graduate School of MedicineSapporoJapan
  2. 2.Department of NeurosurgeryHokkaido University Graduate School of MedicineSapporoJapan
  3. 3.Central Institute of Isotope ScienceHokkaido UniversitySapporoJapan
  4. 4.Department of Cancer PathologyHokkaido University Graduate School of MedicineSapporoJapan
  5. 5.Department of BiostatisticsHokkaido University Graduate School of MedicineSapporoJapan

Personalised recommendations