2020 NPA-03

This case was originally published in 2020. The information provided in this case was accurate and correct at the time of initial program release. Any changes in terminology since the time of initial publication may not be reflected in this case.

The patient is a 67-year-old woman with a history of glioblastoma, IDH wild-type and MGMT-methylated. Following her initial resection, she received standard adjuvant radiation and temozolomide followed by maintenance temozolomide. One year after completion of her radiation therapy, follow-up MRI showed an enhancing mass in the left occipital lobe at the prior surgical resection site (Image A), significantly increased in size from her previous MRI. Clinically she was stable with no new complaints other than her unchanged right homonymous hemianopsia; however, given the increasing size of the lesion on imaging, she underwent repeat resection.

Tissue Site
Brain, left occipital lobe

The whole slide image provided is an H&E-stained section of the left occipital lobe resection specimen.

The best diagnosis in this case is radiation necrosis. Histologic examination reveals large areas of geographic necrosis (Image B) with foci of dystrophic calcifications and fibrinoid necrosis of vessel walls (Image C and Image D). Fibrinous exudates are also seen (Image E). Capillary telangiectatic-like proliferations and areas of hemorrhage are present, as are occasional vessels with hyalinized walls (Image F). Areas of viable parenchyma demonstrate marked gliosis and tissue rarefaction (Image G and Image H). Thrombosis of vessels, mild chronic inflammatory infiltrates, and glia and neurons with atypical pleomorphic nuclei may also be seen.

Standard treatment for glioblastoma includes surgical resection followed by adjuvant concomitant radiation and temozolomide, followed by maintenance temozolomide. Adjuvant radiation is also given to many patients with lower-grade gliomas, as well. A subset of these patients will develop radiation necrosis, which can often mimic tumor recurrence radiographically. Radiation injury can be classified as acute (during radiation treatment), subacute/early delayed (from weeks to a few months after completion of radiation), or late delayed (generally months to years after completion of radiation). Up to 35% to 45% of patients will show radiographic evidence of apparent tumor recurrence in the immediate period after completing concomitant radiation and temozolomide. While approximately half of these patients do have aggressive tumor recurrence, 35% to 50% will actually have subacute/early delayed radiation necrosis, often termed pseudoprogression. It is important to determine whether a patient has tumor progression or radiation necrosis, as this often has therapeutic and/or prognostic implications.

In this case, the patient’s brain MRI approximately one year after completion of radiation therapy showed a 4.3-cm area of heterogeneous enhancement in the left occipital lobe (Image A), significantly increased in size when compared to her prior exam. This increase in enhancement was interpreted as tumor progression, and she therefore underwent a second resection. Standard MRI sequences are unfortunately neither sensitive nor specific for distinguishing tumor progression from radiation necrosis. Radiation necrosis is often said to demonstrate a “soap bubble”- or “swiss cheese”-like enhancement pattern, but it may show rim enhancement indistinguishable from tumor. Additional special imaging sequences including diffusion-weighted imaging, perfusion MRI, and magnetic resonance (MR) spectroscopy may be useful but are still imperfect and are not performed routinely on every patient at every institution. Accurate diagnosis, therefore, still often relies on pathologic examination of the tissue.

Distinguishing recurrent/residual infiltrating tumor cells from reactive astrocytes can be particularly challenging in some cases. This distinction may be complicated by radiation-induced nuclear atypia. While mild hypercellularity and nuclear pleomorphism may be seen in both, mitotic activity is generally restricted to tumor cells and is not a feature of reactive astrocytes. Select ancillary studies may also be beneficial in some cases. In patients with IDH1 R132H-mutant gliomas, immunostaining can be extremely useful in distinguishing tumor cells, which are immunoreactive, from reactive astrocytes, which are nonimmunoreactive. In IDH wild-type tumors, however, this stain is not helpful. Ki67 may be a useful adjunct in some cases, but some studies have shown proliferative rates ranging from 1% to 5% in reactive brain parenchyma, which could overlap with proliferation rates seen in some gliomas. p53 IHC may also be used as part of a panel of stains but should be interpreted cautiously as some studies have shown low immunoreactivity in reactive conditions as well. FISH for EGFR amplification may also be useful in distinguishing tumor from reactive astrocytes in some cases, as nonneoplastic brain does not display EGFR amplification.

Another potential pitfall when examining recurrent gliomas after radiation therapy is the possibility of over-grading a previously lower-grade glioma. Radiation induces necrosis, vascular changes that may superficially mimic endothelial proliferation, and nuclear atypia, all of which can cause confusion when grading recurrent gliomas. As such, it is important that strict histologic criteria for spontaneous tumor necrosis (eg, palisading) and vascular endothelial proliferation (hyperplasia with multiple endothelial cell layers) are utilized.

In practice, the most common histologic finding in previously-treated recurrent glioma specimens is a combination of both radiation necrosis and recurrent/residual tumor. Some studies have shown that the relative proportions of tumor and radiation necrosis are prognostically significant, and, thus, some have advocated that the presence and extent of both tumor and radiation necrosis within the specimen should be reported.

• Radiation necrosis is a common radiographic tumor mimic, and standard MRI sequences are not sensitive or specific in distinguishing tumor progression from radiation necrosis. Special sequences, including diffusion weighted imaging, perfusion MRI, and MR spectroscopy may be more helpful but are still not completely specific.
• When evaluating posttherapy gliomas, distinguishing recurrent/residual tumor cells from reactive astrocytes can be challenging and may be aided by using a panel of ancillary stains/tests.
• When evaluating lower-grade gliomas status post-chemoradiation, strict histologic criteria for spontaneous tumor necrosis and vascular endothelial proliferation should be used so as to not over-grade them.
• Many recurrence specimens will show a combination of both recurrent/residual tumor and radiation necrosis, and there is some evidence that the proportion of each may be prognostically significant and should therefore be reported.

1. Brat DJ, Ryken TC, Kalkanis SN, Olson JJ. The role of neuropathology in the management of progressive glioblastoma: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2014;118(3):461-78.
2. Delgado-Lopez PD, Rinones-Mena E, Corrales-Garcia EM. Treatment-related changes in glioblastoma: a review on the controversies in response assessment criteria and the concepts of true progression, pseudoprogression, pseudoresponse and radionecrosis. Clin Transl Oncol. 2018;20(8):939-53.
3. Forsyth PA, Kelly PJ, Cascino TL, et al. Radiation necrosis or glioma recurrence: is computer-assisted stereotactic biopsy useful? J Neurosurg. 1995;82(3):436-44.
4. Perry A, Schmidt RE. Cancer therapy-associated CNS neuropathology: an update and review of the literature. Acta Neuropathol. 2006(3);111:197-212.