Comparison of FDG and FMISO uptakes and distributions in head and neck squamous cell cancer tumors

Sadek A Nehmeh,Mithat Gönen,John L Humm,Pat Zanzonico,Mohamed B Moussa,Heiko Schöder,Nancy Lee

doi:10.1186/s13550-021-00767-w

Abstract

PurposeGlycolysis is increased by hypoxia, suggesting a possible correlation between the accumulation of 2-[18F]fluoro-2-deoxy-D-glucose (FDG) in malignant tumors and regional hypoxia defined by 1H-1-(3-[18F]fluoro-2-hydroxypropyl)-2-nitroimidazole (FMISO) PET. The aim of this study is to investigate the intra-tumoral spatial distribution and quantitative relationship between FDG and FMISO in a cohort of head and neck squamous cell cancer (HNSCC) patients.MethodsTwenty HNSCC patients with 20 primary tumors and 19 metastatic lymph nodes (LNs) underwent FDG and FMISO PET within 1 week. The metabolic target volume (MTV) was defined on the FDG PET images using a region growing algorithm. The hypoxic volume (HV) was defined by the volume of voxels in an FMISO image within the MTV that satisfy a tumor-to-blood ratio (T/B) greater than 1.2. FDG and FMISO lesions were co-registered, and a voxel-by-voxel correlation between the two datasets was performed. FDG and FMISO TVs’ SUVs were also compared as well as the intra-tumoral homogeneity of the two radiotracers. Separate analysis was performed for the primary tumors and LNs.ResultsTwenty-six percent of the primary tumors and 15% of LNs showed a strong correlation (R > 0.7) between FDG and FMISO intra-tumor distributions when considering the MTV. For the HV, only 19% of primary tumors and 12% of LN were strongly correlated. A weak and moderate correlation existed between the two markers SUVavg, and SUVmax in the case of the primary tumors, respectively. However, this was not the case for the LNs. Good concordances were also observed between the primary tumor’s and LNs HV SUVavgs as well as between the corresponding hypoxic fractions (HF’s).ConclusionsA moderate correlation between FDG and hypoxia radiotracer distribution, as measured by FMISO, seems to exist for primary tumors. However, discordant results were found in the case of LNs. Hypoxia appears to be the dominant driver of high FDG uptake in selected tumors only, and therefore FDG PET images cannot be used as a universal surrogate to identify or predict intra-tumor hypoxia.

Highlights

Tumor hypoxia is an independent prognostic indicator of treatment outcome for several malignancies
Both of these can be driven by hypoxia-inducible factor-1 (HIF-1) transcription, which is activated under tumor hypoxia [9]
The Pearson correlation coefficients (R) between FDG and FMISO intra-tumor distributions, on a voxel-byvoxel basis, for all 20 patients, for both metabolic target volume (MTV) and the hypoxic volume (HV), along with the FDG MTV, maximal tumor SUV, blood SUV, and hypoxic fraction volume (HFV) are summarized in Tables 2 and 3 for the primary tumors and the lymph nodes (LNs), respectively

Summary

Introduction

Tumor hypoxia is an independent prognostic indicator of treatment outcome for several malignancies. Tumor FDG uptake is based on tumor hyperglycolysis (upregulation of glucose transporters (GLUTs) and glycolytic enzymes) [7, 8] Both of these can be driven by hypoxia-inducible factor-1 (HIF-1) transcription, which is activated under tumor hypoxia [9]. This is exacerbated in tumors with low hypoxic fractions due to partial-volume effect. The synthesis of FMISO is currently not widely available, usually requiring the presence of in-house radiopharmacy expertise, which is not available to most nuclear medicine departments

Objectives

Methods

Results

Discussion

Conclusion