Glioblastoma vs cerebral metastasis
Citation, DOI, disclosures and article data
At the time the article was created Frank Gaillard had no recorded disclosures.View Frank Gaillard's current disclosures
Differentiating a glioblastoma (GBM) from a cerebral metastasis is a frequent challenge, with profound surgical, workup and treatment implications. Unfortunately distinguishing between the two entities is not always straightforward and even experienced neuroradiologists will sometimes struggle. Nonetheless, there are a number of helpful features that will generally make distinguishing between the two relatively straight forward.
In the context of this article, the term glioblastoma is being used to denote both glioblastoma, IDH-wildtype tumors and grade 4 adult-type diffuse astrocytomas, IDH-mutant, which are now considered separate diagnoses.
This article addresses helpful imaging features to aid in distinguishing between these two entities preoperatively. For a general discussion of cerebral metastases, glioblastomas, IDH-wildtype and astrocytomas, IDH-mutant please refer to their respective articles.
You may think this is cheating, but if the patient has a known history of systemic malignancy, especially if it is known to be metastatic then, obviously, an enhancing cerebral lesion is far more likely to be a metastasis. Seeking out this information, either directly from the patient record or looking for clues in other available scans is therefore very much worthwhile. In many instances, if the lesion is not clearly a glioblastoma, systemic imaging can be obtained to look for a primary lesion.
Relatively uncommonly, a patient with a cerebral metastasis will present with no prior history of malignancy and no evidence of a primary lesion on CT and PET-CT, referred to as brain metastases of cancer of unknown primary 11,12.
Conversely, occasionally a patient who has a history of malignancy will have a concurrent unrelated glioblastoma; they are, after all, fairly common tumors in the elderly.
If you can confidently identify solid nodular non-enhancing tumor either separate from or in continuity with the enhancing component, then the diagnosis is almost certainly that of a high-grade diffuse glioma. This is because metastases, unless very small, essentially universally enhance. It is, however, important not to confuse edema for non-enhancing tumor.
Probably the most helpful feature of cerebral metastases, other than known rampant systemic metastatic disease, is their multiplicity, with approximately three-quarters of patients having multiple lesions at presentation 1. However, that means that the remaining 25% of patients present with a single lesion, and in many instances no known systemic illness. To make matters worse it is not uncommon for GBMs to present with multiple enhancing foci.
The best clue in distinguishing between the two is that in the vast majority of cases of GBM with multiple areas of enhancement, they are all embedded within the one area of FLAIR signal abnormality, or linked by abnormal FLAIR signal. This is termed multifocal glioblastoma.
If the enhancing lesions are completely separate, not linked by FLAIR signal abnormality they most likely represent cerebral metastases, as this pattern of enhancement is very uncommon in GBM, and is called multicentric glioblastoma.
Cerebral metastases have a predilection for the grey-white matter junction, and less frequently involve the deeper brain. They tend not to involve the periventricular white matter and rarely extend into the corpus callosum.
In contrast, GBM tends to be centered on the subcortical white matter, with frequent extension and spread along the subependymal tissues. Invasion and expansion of the corpus callosum is also frequently encountered.
Cerebral metastases are typically spheres of foreign tissue, relatively well circumscribed and separate from the adjacent brain parenchyma. They are surrounded by vasogenic edema which can vary greatly in amount, from minimal to vast. Importantly, other than the center of the lesions, which frequently demonstrate necrosis and/or hemorrhage, there are not adjacent areas of non-enhancing tumor. This is particularly helpful in peripheral lesions which involve the cortex. Gliomas often have components that demonstrate cortical expansion without enhancement (non-enhancing "peritumoural" component *).
Gliomas tend to have a more complex shape (less spherical), however even with semi-automated computer assisted interpretation of shape, this difference is difficult to exploit to increase accuracy 10.
Dynamic susceptibility contrast (DSC) MR perfusion can be helpful in distinguishing between cerebral metastases and glioblastoma 2,3. The utility stems from two underlying differences in the vascularity of metastases and primary brain tumors.
Firstly, the vascularity of the enhancing tumor component can be quite different. Metastases have vessels that resemble those of the tissue of origin and as such lack any discernible blood brain barrier. In contrast, glioma vessels, although often abnormal, usually have some remaining blood brain barrier structure 2.
Secondly the vascularity of normal brain surrounding a metastases is histologically normal, and non-enhancing high T2 signal is due to vasogenic edema. In contrast, the non-enhancing region of high T2 signal surrounding a glioma often represents non-enhancing tumor, and thus will also have abnormal vessels 2.
Two metrics that are useful in distinguishing between the two, require assessment of signal intensity-time curves 2,3.
maximal signal intensity drop from the precontrast baseline during the bolus phase of the first pass of gadolinium
correlates with capillary density and rCBV
elevated in peri-enhancing high T2 signal parenchyma in gliomas representing presence of abnormal blood vessels 2
percentage signal intensity recovery
percentage of signal intensity recovered relative to the precontrast baseline at the end of the first pass
correlates with degree of capillary permeability and contrast leakage
higher in enhancing component of tumors, representing more intact BBB in gliomas 2
Arterial spin labeled (ASL) perfusion has also shown increased cerebral blood flow (CBF) in the peri-enhancing high T2 signal brain of gliomas 3.
MR spectroscopy is another potentially useful technique in distinguishing metastases from glioblastomas, although it is in practice by no means perfect as it requires careful placement of voxels relative to the enhancing and non-enhancing components.
The most useful region to interrogate is the non-enhancing high T2 signal region adjacent to the enhancing component. In this region gliomas tend to demonstrate features of tumor infiltration (elevated choline) whereas metastases demonstrate normal or somewhat depressed traces 6-9. Traces obtained from the solid enhancing component are more difficult to interpret and studies have reported variable findings.
non-enhancing "peritumoural" component *: elevation of Cho/Cr and Cho/NAA suggests glioma and reflects tumor infiltration 6,7
Cho/Cr ratio > 1.24: sensitivity 100%, specificity 88.9%, positive (PPV) 80.0%, and negative predictive values (NPV) 100% 7
Cho/NAA ratio > 1.11: sensitivity 100%, specificity 91.1%, PPV 83.3%, and NPV 100% 7
solid enhancing component: elevated in both glioblastoma and metastasis, reflecting high cellular membrane turnover 4
solid enhancing component: if present helpful and indicates glioblastoma 4
although the NAA peak is suppressed in both glioma and metastasis relative to normal brain, a metastasis has no neuronal tissue and therefore should have no NAA, provided the voxel is entirely intralesional so as not to unintentionally include normal adjacent brain
central non-enhancing component: is unhelpful if present as it merely indicates cellular necrosis in both glioblastoma or metastasis
solid enhancing component: high lipid/Cr ratio suggests metastasis 5
* the term "peri-tumoral" is unfortunate as it is now well known that in gliomas tumor cells extend well beyond the enhancing margins, and as such what is being referred to is non-enhancing tumor or "peri-enhancement"
- 1. Schiff D. Single Brain Metastasis. Curr Treat Options Neurol. 2001;3(1):89-99. doi:10.1007/s11940-001-0027-4 - Pubmed
- 2. Cha S, Lupo J, Chen M et al. Differentiation of Glioblastoma Multiforme and Single Brain Metastasis by Peak Height and Percentage of Signal Intensity Recovery Derived from Dynamic Susceptibility-Weighted Contrast-Enhanced Perfusion MR Imaging. AJNR Am J Neuroradiol. 2007;28(6):1078-84. doi:10.3174/ajnr.A0484 - Pubmed
- 3. Essig M, Nguyen T, Shiroishi M et al. Perfusion MRI: The Five Most Frequently Asked Clinical Questions. AJR Am J Roentgenol. 2013;201(3):W495-510. doi:10.2214/AJR.12.9544 - Pubmed
- 4. Ishimaru H, Morikawa M, Iwanaga S, Kaminogo M, Ochi M, Hayashi K. Differentiation Between High-Grade Glioma and Metastatic Brain Tumor Using Single-Voxel Proton MR Spectroscopy. Eur Radiol. 2001;11(9):1784-91. doi:10.1007/s003300000814 - Pubmed
- 5. Bulakbasi N, Kocaoglu M, Ors F, Tayfun C, Uçöz T. Combination of Single-Voxel Proton MR Spectroscopy and Apparent Diffusion Coefficient Calculation in the Evaluation of Common Brain Tumors. AJNR Am J Neuroradiol. 2003;24(2):225-33. PMC7974143 - Pubmed
- 6. Law M, Cha S, Knopp E, Johnson G, Arnett J, Litt A. High-Grade Gliomas and Solitary Metastases: Differentiation by Using Perfusion and Proton Spectroscopic MR Imaging. Radiology. 2002;222(3):715-21. doi:10.1148/radiol.2223010558 - Pubmed
- 7. Server A, Josefsen R, Kulle B et al. Proton Magnetic Resonance Spectroscopy in the Distinction of High-Grade Cerebral Gliomas from Single Metastatic Brain Tumors. Acta Radiol. 2010;51(3):316-25. doi:10.3109/02841850903482901 - Pubmed
- 8. Fink K & Fink J. Imaging of Brain Metastases. Surg Neurol Int. 2013;4(Suppl 4):S209-19. doi:10.4103/2152-7806.111298 - Pubmed
- 9. Delorme S & Weber M. Applications of MRS in the Evaluation of Focal Malignant Brain Lesions. Cancer Imaging. 2006;6(1):95-9. doi:10.1102/1470-7330.2006.0015 - Pubmed
- 10. Blanchet L, Krooshof P, Postma G et al. Discrimination Between Metastasis and Glioblastoma Multiforme Based on Morphometric Analysis of MR Images. AJNR Am J Neuroradiol. 2011;32(1):67-73. doi:10.3174/ajnr.A2269 - Pubmed
- 11. Voorhies R, Sundaresan N, Thaler H. The Single Supratentorial Lesion. An Evaluation of Preoperative Diagnostic Tests. J Neurosurg. 1980;53(3):364-8. doi:10.3171/jns.1980.53.3.0364 - Pubmed
- 12. Balestrino R, Rudà R, Soffietti R. Brain Metastasis from Unknown Primary Tumour: Moving from Old Retrospective Studies to Clinical Trials on Targeted Agents. Cancers (Basel). 2020;12(11):3350. doi:10.3390/cancers12113350 - Pubmed