A left basal ganglia lesion in a 57‐year‐old patient

Access at https://isn-slidearchive.org/?col=ISN&fol=Archive&file=6664274.svs. A 57-year-old woman with systemic hypertension and a remote history of melanoma of the left upper extremity presented with recent-onset headache, dizziness, and aphasia, and was found to have a left striatocapsular lesion. Due to persistent aphasia and development of right-sided weakness, a follow-up magnetic resonance imaging (MRI) was performed 2 months later, which showed significant interval enlargement in the area of restricted diffusion and increased signal intensity in the left basal ganglia with peripheral enhancement (Figure 1), for which a biopsy was performed. The biopsy revealed a highly cellular infiltrating neoplasm composed predominantly of cells with abundant, variably granular eosinophilic cytoplasm and round-to-ovoid, hyperchromatic nuclei. Up to 4 mitoses in a single high-power field and microvascular proliferation were identified, while no necrosis was seen (Figure 2A). By immunohistochemistry (IHC), the tumor was diffusely positive for GFAP (Figure 2B) and positive for OLIG2 in a subset (Figure 2B, inset), supporting a glial origin. The glioma cells also showed CD68 staining (Figure 2C), in keeping with the presence of abundant lysosomes and correlating with their granular appearance. CD163 highlighted microglia/macrophages in the background and was negative in tumor cells (Box 1, Figure 2D). IDH-R132H was negative and ATRX and H3K27me3 expression were retained (not shown). There was moderate p53 overexpression in a subset of cells (Figure 2E) and the Ki-67 proliferation index was ~5%–8% (Figure 2F). Targeted next-generation sequencing revealed a TERT promoter mutation (c.-124C>T, also known as C228T), while no IDH1 or IDH2 mutations were detected (not shown). Chromosomal microarray demonstrated combined gain of chromosome 7 and loss of chromosome 10 and CDKN2A/B homozygous deletion (Figure 2G). The tumor resulted positive for MGMT promoter methylation. Granular cell glioblastoma, IDH-wildtype, CNS WHO grade 4. Granular cell astrocytoma/glioblastoma (GCA/GBM) is a rare variant of diffuse glioma that typically presents in middle-aged and older adults and almost invariably arises in the supratentorial compartment, predominantly involving frontal and temporal lobes. On imaging, GCA/GBM can be unifocal or multifocal and shows variable features, ranging from well-defined, homogeneously enhancing lesions to peripherally enhancing, centrally necrotic masses [1]. Microscopically, GCA/GBM shows variable amounts of large cells with abundant granular, eosinophilic cytoplasm admixed with areas displaying conventional astrocytoma features. In some instances, granular cell features may comprise the majority or the entirety of the tumor resembling granular cell tumors arising from other sites or foamy histiocytes, which may be misinterpreted as macrophage-rich lesions such as demyelinating disorders, ischemic/hemorrhagic injury, infection, or histiocytic neoplasms. Since their granular appearance is due to their abundant lysosomal content, tumor cells are often immunoreactive for CD68. However, they also show variable expression of glial markers such as GFAP, OLIG2, and S100, and are negative for more specific histiocytic markers such as CD163 and IBA1, supporting a glial origin [2]. Previous studies have shown that, irrespective of the CNS WHO grade or the presence of partial vs. diffuse granular cell morphology, GCA/GBM usually shows aggressive clinical behavior. Therefore, the lack of significant nuclear atypia, mitotic activity, microvascular proliferation, and/or necrosis does not portend a more indolent course. Moreover, even when histologically low-grade, astrocytomas with granular cell morphology typically harbor one or more molecular features of glioblastoma, including TERT promoter mutation, gain of chromosome 7, EGFR gains, and loss of chromosome 10, in the absence of IDH1, IDH2, or ATRX mutations or whole-arm 1p/19q-codeletion [3]. This case illustrates the challenge of recognizing this unusual tumor. Considering the patient's history of systemic hypertension, her clinical presentation, and a lesion located in deep gray matter structures, the imaging features were initially most suggestive of a subacute infarct. However, her clinical and radiological progression prompted a biopsy to exclude other processes, including a glial neoplasm. Histologically, the presence of numerous cells with abundant granular cytoplasm, especially in this clinico-radiologic setting, may raise consideration for inflammatory/reactive processes such as infarct or abscess, which is additionally complicated by the presence of CD68 staining in the tumor. Nonetheless, the expression of glial markers, as seen in this patient, supported glial differentiation. In addition, next-generation sequencing and chromosomal microarray revealed a molecular profile supportive of glioblastoma, IDH-wildtype. In conclusion, it is important to consider GCA/GBM as part of the differential diagnosis of intra-axial lesions showing focal or prominent granular cell changes and to perform a thorough morphologic and immunohistochemical workup, since its bland appearance may be mistaken for low-grade neoplasms or a spectrum of macrophage-rich lesions. Distinguishing this unusual variant of infiltrating astrocytoma prevents misdiagnosis with other entities and is relevant due to its prognostic and therapeutic implications. Samir Atiya, Ashish Sonani, and M. Adelita Vizcaino contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Samir Atiya, Ashish Sonani, Jill Eickhoff, and M. Adelita Vizcaino. The first draft of the manuscript was written by Samir Atiya and Ashish Sonani, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. We thank the Mayo Clinic Cytogenetics Core and Director Patricia T. Greipp, D.O., and technologist Troy J. Gliem for their contributions to the analysis of the chromosomal microarray data. Some of the data in this publication were produced in the Mayo Clinic Cytogenetics Core Laboratory, which is supported, in part, by the Mayo Clinic Comprehensive Cancer Center Grant, funded by the National Cancer Institute (P30CA15083). The authors declare no conflicts of interest. This research was exempt from Institutional Research Board approval. The data that support the findings of this study are available from the corresponding author upon reasonable request.