ELEVATED MYOINOSITOL IN DIFFUSE ASTROCYTOMA; A MARKER OF ANAPLASTIC PHENOTYPE, AND NOT ATTRIBUTABLE TO MICROGLIAL INFILTRATION

PURPOSE: Although in vivo MRS has been extensively applied to gliomas, the relevance of myoinositol (Myo) to malignant phenotype and the cell types that contribute signal are poorly understood. Microglial infiltration is associated with Myo elevation in inflammatory and neurodegenerative diseases, and has been implicated in neoangiogenesis in transforming gliomas. The aim of this study was to examine the relationship between Myo and Cho imaging biomarkers and enzymatic tissue markers of Myo synthesis and microglial cell type in grade II and III astrocytomas. SUBJECTS AND METHODS: 28 patients with diffuse astrocytoma (14 diffuse LGG, WHO grade II; 14 anaplastic, WHO grade III; 13 stereotactic biopsies, 15 surgical specimens) underwent multivoxel spectroscopy at 1.5 T or 3T (TE = 30ms, TR = 1500ms, 8x8 grid; Siemens Avanto/Verio, Erlangen, De). Metabolite ratios were estimated using proprietary software. Tissue immunostains with rabbit polyclonal antibodies directed against human inositol monophosphatase (IMPA1; Sigma – dilution 1:400; 25 cases) and Iba1 (Wako – dilution 1:750; 21 cases), were assessed semi-quantitatively. Myo/Cr and Cho/Cr ratios were compared (Kruskal-Wallis) and compared with Iba-1 and IMPA1 levels (Wilcoxon rank-sum), and metabolite ratios with Iba1 (Spearman's rank test). Co-localisation of highest Myo/Cr and Cho/Cr was also measured (Fisher's exact test). RESULTS: Myo/Cr was significantly lower in grade II compared with grade III tumours (p < 0.01). Highest Myo/Cr co-localised with highest Cho/Cr voxels in 72%. IMPA1 scores were significantly higher in grade III compared to grade II lesions (p < 0.01). There was no correlation between semi-quantitative microglial analysis using Iba1 antibody and Myo/Cr or tumour grade. CONCLUSION: Higher Myo in grade III than grade II astrocytomas suggests that it is not a marker of indolent tumour phenotype, as has been previously suggested, and may reflect increased de novo synthesis relating to p53 mutations. Our data do not suggest that microglia account for raised Myo levels in gliomas.