Atypical Central Neurocytoma With Sarcomatous Differentiation.
REPORT OF A CASE
A 44-year-old white woman presented to the emergency department with a 1-day history of severe headache and intractable vomiting. Medical history included chronic Lyme disease, asthma, and polycystic ovarian disease. She reported no smoking or alcohol use. The family history was negative for malignancies. She had no occupational toxic exposures. A head computed tomography scan demonstrated a 3.4X3.0X3.1-cm mass with hemorrhage in the occipital horn of the left lateral ventricle and atrium. Magnetic resonance imaging (MRI) further revealed hydrocephalus and showed the aforementioned mass to be lobulated and heterogeneously enhancing (Figure 1). T2 and fluid-attenuated inversion recovery (FLAIR) images also showed mild periventricular and scattered deep white matter punctate areas of signal abnormality.
The patient underwent a left occipital-parietal craniotomy with an incomplete resection. Postoperative MRI demonstrated 2 small nodules of enhancement in the atrium of the left lateral ventricle, 0.5 and 1.0 cm in largest dimension, respectively. The patient received proton radiation therapy with concurrent temozolomide and bevacizumab after surgery. After a 4-week interval, MRI of the brain showed no evidence of residual mass. The patient's treatment proceeded with 6 cycles of adriamycin and ifosfamide (with mesna) and metronomic temozolomide. Two months following chemotherapy, MRI of the brain showed an area of enhancement in the left lateral ventricle, which was presumed to represent radiation changes. This was further supported by magnetic resonance spectroscopy. The patient was treated with 2 cycles (28 days each) of temozolomide. Subsequent MRI of the of the brain with spectroscopy, 14 months following the initial diagnosis, showed multiple enhancing lesions with the largest lesion (2.5 cm) situated within the left occipital lobe with subependymal/ependymal involvement. Additional lesions were located in the left posterior periventricular white matter, cerebellar vermis, and left cerebellar hemisphere. These findings were suggestive of recurrent/progressive disease with radiation changes. Magnetic resonance imaging of the entire spine and cerebrospinal fluid examination did not reveal evidence of leptomeningeal disease. The patient underwent stereotactic biopsy of the left periventricular white matter, which confirmed recurrent disease, and she subsequently underwent stereotactic radiosurgery of the left mediolateral parietal-occipital lobe and the left medial cerebellar region. Currently, she is being treated with carboplatin, irinotecan, and bevacizumab.
MATERIALS AND METHODS
A portion of the original resection specimen was received for frozen section diagnosis. The remaining unfrozen specimen from the irst resection was entirely submitted and routinely processed after formalin fixation. Histologic sections were prepared from 10% formalin-fixed, paraffin-embedded tissue and stained with hematoxylin-eosin for light microscopic examination.
Immunohistochemistry was performed on formalin-fixed, paraffin-embedded tissue sections by using an avidin-biotin-horseradish peroxidase system. The following commercially available antibodies were used: polyclonal antibodies against S100 protein and monoclonal antibodies against epithelial membrane antigen, vimentin, synaptophysin, glial fibrillary acidic protein (GFAP), CD34, p53, CD3, CD20, AE1/AE3, MIB-1 (all Ventana Medical Systems, Tucson, Arizona), CAM 5.2 (BD Biosciences, San Jose, California), and Neu-N (EMD Millipore Headquarters, Billerica, Massachusetts). In every case, formalin-fixed tissue was subjected to online platform retrieval using CC1 (Tris/borate/EDTA buffer) and pronase (protease-1) for vimentin and CAM 5.2. Neuron-associated HuC/HuD and smooth muscle actin tests were performed at an outside laboratory.
Fluorescence In Situ Hybridization Analysis
Deparaffinized tissue sections, after digestion and pretreatment along with appropriate controls, were incubated with a mixture of analyte-specific reagents containing 2 pairs of probes performed on 2 separate tissue sections. The irst probe set included a probe to the 1p36 region (orange) and 1p25 region (green) as a reference probe and the second probe set involved a probe to the 19q13 region (orange) and 19p13 region (green) as a reference probe. Quantitative analysis of both probe pair sets was performed manually by a pathologist on preselected areas of the paraffin tissue sections (Vitro Molecular Laboratories LLC, Miami, Florida).
O6-Methylguanine Methyltransferase Promoter Methylation Analysis
Genomic DNA was isolated from paraffin-embedded tissue sections. DNA methylation status of CpG islands at the O6methylguanine methyltransferase (MGMT) promoter was determined by methylation-specific polymerase chain reaction (PCR). DNA was subjected to PCR amplification with specific primers to amplify both methylated and unmethylated PCR products. Control methylated and unmethylated DNA samples as well as no template controls were used as positive and negative controls (Vitro Molecular Laboratories).
The specimen received for pathologic examination consisted of multiple fragments of soft tan-red tissue, 4.2 X 4.0 X 1.2 cm in aggregate. Histologically, the neoplasm contained 2 distinct morphologic components. One population of cells had round nuclei, some of which were haloed (Figure 2), along with interspersed multinucleated and binucleated cells. Scattered neurocytic rosettes were present. Mitotic figures were focally prominent (8 mitoses per 10 high-power fields). A second morphologically distinct population was composed of spindle cells arranged in a fascicular pattern with prominent mitotic figures (Figure 3, A and B). The 2 cell populations were distinct and largely well delineated throughout the tumor, but the specimen was fragmented. Focal necrosis was identified.
A synaptophysin immunostain was immunopositive, most notably in the areas with round, haloed nuclei (Figure 4). The spindle cells did not show immunonreactivity for synaptophysin. The round, haloed nuclei were also partially reactive for the neuron-associated HuC/HuD antigens. Rare mature neurons at the periphery of the tumor were immunoreactive for Neu-N and the adjacent tumor cells were negative. Vimentin and reticulin strongly labeled many spindle cell areas (Figure 5, A and B). The p53 antibody revealed scattered positive nuclei. GFAP had no significant immunoreactivity in either cell population, but it highlighted a few scattered glial cells. The S100 antibody showed patchy immunopositivity, but there were broad zones with no immunoreactivity in the spindle cell areas. Ki-67 nuclear labeling was focally 15.0% of the neoplastic nuclei (Figure 5, C). Tumor cells were negative for CD34, pankeratin, CAM 5.2, and epithelial membrane antigen. CD3 and CD20 antibodies revealed focally prominent clusters of T and B lymphocytes, respectively, within the tumor.
Owing to the presence of neoplastic cells with round, haloed nuclei, a block was submitted for 1p/19q fluorescence in situ hybridization testing. The tumor exhibited unusual genetic abnormalities including trisomy of chromosome 19 with superimposed deletion of 19q13 (2 copies instead of 3). There was no deletion of 1p36. Furthermore, the tumor was positive for methylation of MGMT by PCR.
A second brain biopsy of the left periventricular white matter was performed 14 months later owing to the evolving radiologic findings. The histomorphologic features were consistent with limited recurrent/residual neoplasm with scattered atypical cells, radiation necrosis, and reactive changes. The GFAP preparation appeared similar to the initial tumor resection, highlighting a few scattered glial cells. A few atypical nuclei labeled for p53. The Ki-67 preparation demonstrated focal 5.0% labeling of atypical pleomorphic nuclei. The spindle cell component was not represented in the second biopsy specimen.
Central neurocytoma is a primary WHO grade II neoplasm of the central nervous system (CNS) with an incidence of 0.1% to 0.5% of all primary brain neoplasms. (3,4)
Central neurocytomas have been reported as presenting as early as at 8 days of age, up to 67 years of age (mean age, 29 years). (2,3) Most cases are diagnosed in patients between 20 and 40 years of age, and males and females are equally affected. (2,3)
Central neurocytomas are typically midline supratentorial tumors located in the lateral or third ventricles in the region of the foramen of Monro, arising from the septum pellucidum, fornix, or walls of the lateral ventricles. (3) The most common sites are the frontal horn and corpus of the lateral ventricle, with the left lateral ventricle being more frequently involved. (2,9) The term extraventricular neurocytoma (EVN) is used when a histologically similar tumor arises outside the ventricles. Extraventricular neurocytoma has been reported to involve the cerebral hemispheres, thalamus, cerebellum, pons, amygdala, pineal gland, retina, and spinal cord. (3,4) The cerebral hemispheres are the most common location for EVNs, (predominantly the frontal lobe) followed by the spinal cord. (10) The mean age for EVN presentation is higher (39 years versus 29 years) than for CN and shows a slight female predominance. (10) Outside of the CNS, neurocytomas have been described within an ovarian teratoma as well as in the pelvic cavity. (3) Central neurocytoma is not commonly associated with intraventricular hemorrhage or acute clinical deterioration, yet multiple cases of CN with intratumoral hemorrhage have been reported. (11)
Neuroimaging of CN typically demonstrates a large, discrete, solitary lobulated mass near the foramen of Monro, which is often attached to the septum pellucidum, expanding into 1 or both lateral ventricles or the third ventricle. (3,4) Repeated imaging typically shows slow growth. (3) Calcifications are seen in up to 50.0% of CN and EVN cases, and microcysts and macrocysts may give the tumor a heterogeneous neuroimaging appearance. (4,10,12) In computed tomography scans, CN is isodense or mildly hyperdense with strong, uniform, contrast enhancement. (3,4,12) By MRI, CN is isointense or slightly hypointense in T1-weighted images and hyperintense in T2-weighted images and FLAIR with a well-defined margin. (4,12) Enhancement after gadolinium varies from mild to strong. (2,4)
Grossly, CNs are usually gray and soft, resembling gray matter. Areas of hemorrhage may be identified, and focal calcifications can impart grittiness during dissection. (2,3) Histologically, CN features uniform round to oval cells with scant cytoplasm and perinuclear halos, which imparts a "fried egg" appearance. Nuclei may appear finely speckled with evenly granular nuclear chromatin, and an occasional visible nucleolus can be seen. (3,4,12) Various architectural patterns have been described. Most commonly, CN takes on a honeycomb arrangement similar to oligodendroglioma or displays large fibrillary areas akin to the irregular rosettes of pineocytoma. (3,4) The tumor may also contain perivascular pseudorosettes such as those seen in ependymoma, and some tumor cells can be arranged in straight lines. (4) Capillaries are usually arranged in a linear, arborizing pattern, giving an organoid endocrine appearance. (3) Hyalinized vessels are a more frequent finding in EVN. (10) Rarer findings in CN include Homer-Wright rosettes, which are more frequently observed (up to 20%) in EVNs. (3,10)
Unusual differentiation has been recorded in a few central neurocytomas, including 7 examples with lipomatous areas, (12) melanocytic cells, (13) and rhabdomyomatous cells. (14) Ours is the first case of CN with a spindle cell sarcoma-like component. Several studies (2,3,12) noted that brisk mitotic activity, microvascular proliferation, and necrosis are seen in atypical CN. The proliferation rate for typical CN is usually less than 2.0% by a Ki-67 nuclear labeling. (2) The suggested cutoff criterion between conventional and atypical CN is a Ki-67 proliferation index exceeding 2.0% to 3.0%, with atypical CN increasing the risk for tumor recurrence. (3,4,12,15) Though there is a histomorphologic similarity between oligodendroglioma and CN, the characteristic 1p/19q codeletion seen in oligodendroglioma has not been shown to be associated with CN. However, the 1p/19q codeletion has been associated with EVN. (10)
Immunohistochemical evaluation of CN suggests that synaptophysin consistently stains the neuropil, especially in fibrillary zones and perivascular cell-free areas. (3,4) Neu-N and anti-Hu antibodies often have labeling to suggest neurocytic differentiation. (2,3) Chromogranin and neurofilament are absent except when ganglion cell-like morphology is present. (3,4) GFAP has been detected by some authors in CN, (2) which may indicate that the cell of origin may be a bipotential cell of germinal matrix origin. (3) GFAP immunopositivity is more frequently observed (30.8%) in EVNs. (10)
Most patients with CN have a favorable clinical outcome. Histologic atypia, cellular pleomorphism, vascular proliferation, and necrosis do not necessarily predict aggressive behavior or poor clinical outcome. (15) Involvement of the periventricular parenchyma has been associated with poorer outcome in some patients. (2) In some cases, the tumor recurs, (3,4,16) disseminates, (17) or causes death. (3,4,17) All factors that play a role in poor clinical results have not been fully elucidated, but subtotal resection and the proliferation rate seem to affect prognosis. (18,19)
Rades et al (7) reviewed the literature along with their own cases (n = 129) to determine if Ki-67 proliferation index had prognostic significance. Their findings showed significantly higher rates of recurrence (12.0% versus 48.0%) and tumor-related deaths (3.0% versus 25.0%) when the Ki-67 proliferation index was greater than 3.0%. Other studies (3,8,15) also showed poor outcome with Ki-67 nuclear labeling greater than 3.0%. The term atypical central neurocytoma may include other features such as vascular proliferation, increased mitotic activity, and necrosis. (3,4,20) In a series of 438 patients, (20) atypical CN had overall lower survival at 5 years (70.0% versus 97.0%) and worse local control at 5 years (46.0% versus 81.0%) than typical CN, respectively.
A retrospective review done by Kaur et al (18) noted that the extent of resection and Ki-67 proliferation index greater than 4% were predictive of recurrence. Their study showed that in patients with subtotal resection and Ki-67 nuclear labeling greater than 4%, tumor recurrence occurred at 23.5 months in 33% of patients, and at 4 years in 75% of patients. They did not observe recurrence of the tumor in patients with subtotal resection and Ki-67 nuclear labeling lower than 4%, suggesting that Ki-67 proliferation index is a useful prognostic tool. On the other hand, Vasiljevic et al (19) concluded that extent of tumor resection is the main prognostic indicator in patients with CN.
Our patient underwent radiation with concurrent chemotherapy since there was residual tumor along the left lateral ventricle. Magnetic resonance imaging after 4 months showed that the tumor had regressed; however, the tumor did recur 14 months following the initial diagnosis. This may indicate that a Ki-67 proliferation index of more than 3%, along with a sarcomatous component and incomplete resection, is correlated with a shorter disease-free interval. The second biopsy specimen from the left periventricular white matter showed radiation changes, scattered atypical neoplastic cells, and no evidence of the spindle cell component in the limited biopsy material submitted. The patient is currently receiving carboplatin, irinotecan, and bevacizumab. The most recent MRI of the brain showed stable disease.
In conclusion, this unusual case illustrates the diagnostic challenges of a previously undescribed entity. This tumor displays histologic features of atypical central neurocytoma admixed with malignant spindle cells. We do not believe this represents a collision tumor, but rather divergent differentiation of CN with sarcomatous differentiation. No previous cases are available to provide treatment options for atypical CN with sarcomatous features. Although this entity and its treatment have not previously been described, resection, radiation, and chemotherapy appear to have provided some benefit in this case. With an incomplete resection and high proliferative rate in this atypical CN with sarcomatous features, it is not surprising that the tumor exhibits more aggressive behavior and recurred 14 months after the original diagnosis.
We thank Marc Rosenblum, MD, for reviewing the histopathology of the case.
Please Note: Illustration(s) are not available due to copyright restrictions.
(1.) Hassoun J, Gambarelli D, Grisoli F, et al. Central neurocytoma: an electron-microscopic study of two cases. Acta Neuropathol. 1982; 56(2):151-156.
(2.) Figarella-Branger D, Soylemezoglu F, Burger PC. Central neurocytoma and extraventricular neurocytoma. In: Louis DN, Ohgaki H, Wiestler OD, et al, eds. World Health Organization Classification of Tumours of the Central Nervous System. Lyon, France: IARC Press; 2007:106-109. World Health Organization Classification of Tumours; vol 1.
(3.) Sharma MC, Deb P, Sharma S, Sarkar C. Neurocytoma: a comprehensive review. Neurosurg Rev. 2006; 29(4):270-285.
(4.) Choudhari KA, Kaliaperumal C, Jain A, et al. Central neurocytoma: a multidisciplinary review. Br J Neurosurg. 2009; 23(6):585-595.
(5.) Horoupian DS, Shuster DL, Kaarsoo-Herrick M, Shuer LM. Central neurocytoma: one associated with a fourth ventricular PNET/medulloblastoma and the second mixed with adipose tissue. Hum Pathol. 1997; 28(9):1111-1114.
(6.) Reifenberger G, Blumcke I, Pietsch T, Paulus W. Pathology and classification of tumors of the central nervous system. In: Tonn JC, Westphal M, Rutka JT, eds. Oncology of CNS Tumors. New York, NY: Springer; 2010:3-68.
(7.) Rades D, Schild SE, Fehlauer F. Prognostic value of the MIB-1 labeling index for central neurocytomas. Neurology. 2004; 62(6):987-989.
(8.) Soylemezoglu F, Scheithauer BW, Esteve J, Kleihues P. Atypical central neurocytoma. J Neuropathol Exp Neurol. 1997; 56(5):551-556.
(9.) Peltier J, Baroncini M, Le Gars D, Lejeune JP. Central neurocytomas of the lateral ventricle: a series of 35 cases with review of the literature [in French]. Neurochirurgie. 2011; 57(4-6):215-219.
(10.) Furtado A, Arantes M, Silva R, Romao H, Resende M, Honavar M. Comprehensive review of extraventricular neurocytoma with report of two cases, and comparison with central neurocytoma. Clin Neuropathol. 2010; 29(3):134-140.
(11.) Terakawa Y, Tsuruno T, Ishibashi K, Okada Y, Shimotake K, Murata T. Central neurocytoma presenting with massive hemorrhage leading to coma--case report. Neurol Med Chir (Tokyo). 2010; 50(2):139-143.
(12.) Jaiswal S, Vij M, Rajput D, et al. A clinicopathological, immunohistochemical and neuroradiological study of eight patients with central neurocytoma. J Clin Neurosci. 2011; 18(3):334-339.
(13.) Ng TH, Wong AY, Boadle R, Compton JS. Pigmented central neurocytoma: case report and literature review. Am J Surg Pathol. 1999; 23(9):1136-1140.
(14.) Pal L, Santosh V, Gayathri N, et al. Neurocytoma/rhabdomyoma (myoneurocytoma) of the cerebellum. Acta Neuropathol. 1998; 95(3):318-323.
(15.) Mackenzie IR. Central neurocytoma: histologic atypia, proliferation potential, and clinical outcome. Cancer. 1999; 85(7):1606-1610.
(16.) Bertalanffy A, Roessler K, Koperek O, Gelpi E, Prayer D, Knosp E. Recurrent central neurocytomas. Cancer. 2005; 104(1):135-142.
(17.) Ogawa Y, Sugawara T, Seki H, Sakuma T. Central neurocytomas with MIB-1 labeling index over 10% showing rapid tumor growth and dissemination. J Neurooncol. 2006; 79(2):211-216.
(18.) Kaur G, Kane AJ, Sughrue ME, et al. MIB-1 labeling index predicts recurrence in intraventricular central neurocytomas. J Clin Neurosci. 2013; 20(1): 89-93.
(19.) Vasiljevic A, Francois P, Loundou A, et al. Prognostic factors in central neurocytomas: a multicenter study of 71 cases. Am J Surg Pathol. 2012; 36(2): 220-227.
(20.) Rades D, Schild SE. Treatment recommendations for the various subgroups of neurocytomas. J Neurooncol. 2006; 77(3):305-309.
Lakshmi Vemavarapu, MD; Irene Czyszczon, DO; Joseph C. Parker Jr, MD; Stephanie Wagner, MD; Todd Vitaz, MD; John R. Parker, MD
Accepted for publication October 17, 2013.
From the Department of Pathology, University of Louisville, Louisville, Kentucky (Drs Vemavarapu, Czyszczon, J. C. Parker, and J. R. Parker); the Department of Hematology and Oncology, Indiana University Simon Cancer Center, Indianapolis (Dr Wagner); and the Department of Neurosurgery, Norton Neuroscience Institute, Louisville, Kentucky (Dr Vitaz).
The authors have no relevant financial interest in the products or companies described in this article.
Corresponding author: Lakshmi Vemavarapu, MD, Department of Pathology and Laboratory Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216 (e-mail: email@example.com).
Caption: Figure 1. T1-weighted magnetic resonance imaging with contrast showing a lobulated heterogeneously enhancing mass in occipital horn of the left lateral ventricle.
Caption: Figure 2. Tumor displaying round nuclei with perinuclear halos and scattered mitotic figures (arrows) (hematoxylin-eosin, original magnification X400).
Caption: Figure 3. A, Sarcomatous component with fascicles of spindle cells and nuclear pleomorphism. B, Mitotic figures (arrows) in sarcomatous component (hematoxylin-eosin, original magnifications X200 [A] and X 400 [B]).
Caption: Figure 4. Diffuse and strong synaptophysin immunopositivity in round, haloed tumor cell population (original magnification, X400).
Caption: Figure 5. A, Sarcomatous component with spindle cells diffusely positive for vimentin. B, Spindle cells with intercellular reticulin staining. C, Ki-67 nuclear positivity (original magnifications X200 [A through C]).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Case Reports|
|Author:||Vemavarapu, Lakshmi; Czyszczon, Irene; Parker, Joseph C., Jr.; Wagner, Stephanie; Vitaz, Todd; Parke|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Sep 1, 2014|
|Previous Article:||The Tahoe Study: Bias in the Interpretation of Papanicolaou Test Results When Human Papillomavirus Status Is Known.|
|Next Article:||Megakaryocytic Morphology and Clinical Parameters in Essential Thrombocythemia, Polycythemia Vera, and Primary Myelofibrosis With and Without JAK2...|