Hypoxia and glucose metabolism assessed by FMISO and FDG PET for predicting IDH1 mutation and 1p/19q codeletion status in newly diagnosed malignant gliomas

BACKGROUND: Tumor hypoxia and glycolysis have been recognized as determinant factors characterizing tumor aggressiveness in malignant gliomas. To clarify in vivo hypoxia and glucose metabolism in relation to isocitrate dehydrogenase (IDH) mutation and chromosome 1p and 19q (1p/19q) codeletion status, we retrospectively analyzed hypoxia as assessed by positron emission tomography (PET) with [(18)F]-fluoromisonidazole (FMISO) and glucose metabolism as assessed by PET with [(18)F]-fluoro-2-deoxy-d-glucose (FDG) in newly diagnosed malignant gliomas. METHODS: In total, 87 patients with newly diagnosed supratentorial malignant (WHO grade III and IV) gliomas were enrolled in this study. They underwent PET studies with FMISO and FDG before surgery. The molecular features and histopathological diagnoses based on the 2016 WHO classification were determined using surgical specimens. Maximal tumor-to-normal ratio (TNR) was calculated for FDG PET, and maximal tumor-to-blood SUV ratio (TBR) was calculated for FMISO PET. The PET uptake values in relation to IDH mutation and 1p/19q codeletion status were statistically analyzed. RESULTS: In all tumors and malignant astrocytomas, the median FMISO TBR in IDH-wildtype tumors was significantly higher than that in IDH-mutant tumors (P < 0.001 and P < 0.01, respectively). In receiver operating characteristic (ROC) analysis, the area under the curve showed that the sensitivity for the discrimination was moderate (0.7–0.8) and the specificity was low (0.65–0.68). In the same population, the median FDG TNR in IDH-wildtype tumors tended to be higher than that in IDH-mutant tumors, but the difference was not statistically significant. In WHO grade III anaplastic astrocytomas, there were no significant differences in median FMISO TBR or FDG TNR between IDH-mutant and IDH-wildtype tumors. In IDH-mutant WHO grade III anaplastic gliomas, there were no significant differences in median FMISO TBR or FDG TNR between anaplastic astrocytomas and anaplastic oligodendrogliomas. CONCLUSIONS: Tumor hypoxia as assessed by FMISO PET was informative for prediction of the IDH mutation status in newly diagnosed malignant gliomas. However, the accuracy of the discrimination was not satisfactory for clinical application. On the other hand, glucose metabolism as assessed by FDG PET could not differentiate the IDH-mutant status. Moreover, PET studies using FMISO and FDG could not predict IDH mutation and 1p/19q codeletion status in WHO grade III tumors.