Imaging Gliomas

DIAGNOSIS The diagnostic evaluation for a patient presenting with recent onset of headaches and/or seizures usually begins with an imaging procedure. The initial procedure is typically MR imaging (MRI) or CT scan. Compared with CT, MRI is more sensitive and provides more anatomic detail. SPECT, FDG-PET and echo planar MRI are additional studies that may have clinical utility. Cerebral angiography is no longer considered a routine examination in the work-up of these patients.

MRI and CT The only test needed to diagnose a brain tumor is cranial MRI. Malignant astrocytomas are usually hypointense on T1-weighted images. These tumors typically enhance heterogeneously and serpiginously following contrast infusion. There may be areas of solid contrast enhancement within a more diffusely serpiginous pattern, or the lesion may show a totally solid pattern of enhancement. Less commonly, malignant astrocytoma enhances in a ring-like pattern that is variable in thickness, with small finger-like projections toward the necrotic center or out from the rim.

The enhancing tumor can be distinguished from the surrounding hypointense signal of edema on T1-weighted sequences. On T2-weighted sequences, the edema is hyperintense and cannot be distinguished from tumor

The outer margin of the T2-weighted abnormality on MRI is the most accurate, although imperfect, measure of the boundary of the malignant astrocytoma. However, isolated tumor cells can extend beyond the edge of the T2-weighted high-signal abnormality into brain that is radiographically normal. There are no typical findings on MRI that are specific for any histologic grade of astrocytoma .

Computed tomography (CT) scans can miss structural lesions, particularly in the posterior fossa, or nonenhancing lesions. As an example, on precontrast CT scans, astrocytomas are often hypodense or isodense compared to normal brain. After contrast infusion, enhancement patterns are similar to those seen with MRI, with contrast enhancement allowing distinction between tumor and surrounding edema. Contrast enhancement on MRI or CT is not specific for tumor and can be due to any process that disrupts the blood-brain barrier. In addition, approximately 30 percent of patients with anaplastic astrocytoma and 4 percent of patients with GBM lack contrast enhancement on CT.

Corticosteroids may alter the imaging picture of a malignant astrocytoma on both CT and MRI and cloud the decision making process. In one report, for example, nine of ten patients had a measurable reduction in the size of the gadolinium-enhancing region or T2-weighted abnormality with corticosteroid treatment. For this reason, it is desirable to delay the use of corticosteroids, if possible, when the diagnosis is in doubt.

SPECT imaging Preoperative thallium-201 single-photon emission computed tomography (SPECT 201-Tl) is useful for distinguishing benign from malignant brain lesions, predicting the histologic grade of brain tumors, and localizing areas of maximum 201-Tl uptake for biopsy . The uptake of 201-Tl is unaffected by steroid treatment

There appears to be a correlation between early and delayed tumor uptake of 201-Tl and the subsequent grade of the surgically resected tumor, allowing a distinction between low and high grade astrocytomas . SPECT 201-Tl has not been as helpful as fluorodexoyglucose (FDG)-PET in distinguishing radiation necrosis from recurrent tumor .

IMT-SPET imaging � Single photon emission tomography (SPET) using iodine-123 alpha methyl-L-tyrosine (IMT) appears to be useful for distinguishing glioma recurrence from benign posttherapeutic change. In one report of 78 patients (68 recurrent gliomas and 10 nontumorous lesions), 62 of the 66 histopathologically confirmed recurrences were identified by IMT-SPET, including all 25 recurrent high grade (grade 4) gliomas, 11 of 12 grade 3 tumors, and 16 of 19 grade 2 recurrences (sensitivity 100, 92, and 84 percent, respectively). All benign lesions were correctly identified, but four histologically confirmed low grade glioma recurrences, and one diffuse glioma was falsely negative.

PET scan Preoperative tumor glucose utilization, as determined by FDG-PET (positron emission tomography), has been evaluated in the preoperative evaluation and posttreatment follow-up of patients with malignant glioma. Glucose utilization in high grade gliomas does not appear to correlate with tumor grade or size, or with patient survival.

FDG-PET has not been directly compared to MRI, or MRI plus PET, for its additive role in tissue diagnosis. PET scans are able to localize the areas of maximum glucose utilization within the tumor, guiding the neurosurgeon to biopsy locations with the most aggressive biologic behavior . An additional use for pretreatment PET scanning may be the ability to map functional areas of the brain, especially in conjunction with functional MRI (see below) prior to surgery or radiation, in order to minimize injury to eloquent areas .

PET scanning may have particular utility in the posttreatment setting:

Following surgical resection, glucose uptake is not normally increased. Patients with hypermetabolic areas on postoperative FDG-PET scans are at risk for early tumor recurrence .

For assessment of the response to therapy, functional imaging modalities such as echo planar MRI or PET imaging are better than MRI or CT. In addition, unlike imaging with MRI or CT, FDG-PET measurement of glucose utilization is not influenced by corticosteroid therapy.

FDG-PET is a helpful noninvasive study for differentiating recurrent tumor from radiation necrosis. Although MRI can provide excellent depiction of soft tissue morphology, it is frequently unable to distinguish edema and treatment-related necrosis from recurrent or residual tumor .

There is one setting, however, in which FDG-PET may be misleading: in patients heavily pretreated with high-dose interstitial or accelerated fraction radiation therapy. In such patients, surgical biopsy or resection is needed for definitive diagnosis.

Echo planar MRI Echo-planar MRI (EPI) is a new technique of functional MRI imaging that provides maps of tumor blood flow, and may permit better resolution of tumor versus surrounding edema at the tumor borders. In addition, imaging after activation of sensory and motor areas by appropriate stimuli may also allow separation of tumor from normal brain preoperatively (functional brain imaging). Ultrafast acquisition of MRI images by EPI is free of motion artifacts, and may be advantageous for patients who are unable to cooperate with standard MRI scanning.

Magnetic resonance spectroscopy MR spectroscopy may improve the differentiation of locally infiltrative tumors from other types of well circumscribed intracranial lesions, including other tumors or metastases In one study of 26 patients with intracranial masses who underwent MRI, proton MR spectroscopy, and stereotactic biopsy, gliomas and lymphomas showed pathologic spectra outside the area of contrast enhancement suggesting infiltrative lesions; in comparison, four nonastrocytic circumscribed tumors (meningioma, pineocytoma, metastases, and germinoma) showed no pathologic spectra outside the region of enhancement . However, diagnostic accuracy was not improved in terms of differentiating the types of infiltrative or circumscribed lesions.