Retrospective review of pediatric patients with supratentorial gliomas or non-neoplastic lesions suspected to be gliomas based on initial MRI. Pts had MRI spectroscopy (MRS) and 18F-DOPA PET within 2 weeks. (Spectroscopy estimates levels of different metabolites, DOPA chosen as uptake of amino acid analogs less in normal brain and uniform in tumor whether enhancing component or not). Twenty-seven patients enrolled aged 4-17 years, 21 with gliomas and 6 with non-neoplastic lesions. Median follow up was 19 months. For PET, sensitivity was 78% (5 false negatives), specificity 83% (1 false positive), accuracy 78%. For MRS, sensitivity was 95% (1 false negative), specificity 83% (1 false positive), accuracy 93%. False negatives for PET and MRS were LGG, and false positive were encephalitis. Statistically significant difference in uptake of DOPA found between HGG and LGG, also seen with choline peaks on MRS, but the difference was lower with MRS. DOPA uptake was also found to be an independent predictor of survival on multivariate analysis. Authors concluded that MRS was better for distinguishing glioma from non-neoplastic lesions as it is more widely available, cheaper and does not involve radiation. DOPA-PET was best for distinguishing low-grade from high-grade lesion.
* In the future, non-invasive imaging techniques such as PET and MR spectroscopy may be able to predict pathology and prognosis of brain tumors, specifically gliomas. Biopsy may still be required though, as molecular information to inform potential targeted therapy is becoming more important.