894

Degeneration and repair

Malignant transformation of DNETs: a case report and literature review Alan A. Moazzam, Naveed Wagle and Mark S. Shiroishi Dysembryoplastic neuroepithelial tumors (DNETs) have traditionally been thought of as lesions with a benign clinical course that generally do not undergo malignant transformation. However, cases are emerging of DNETs that progress to more malignant forms. We present a case of malignant transformation, and we compile and review all previously published cases to identify common characteristics that may confer a higher risk for malignant transformation. A PubMed search was performed of all English-language case reports of DNET transformations to malignant cancers. The demographic, clinical, and histologic features of these patients are compiled and analyzed. A separate case report with histologic descriptions is also presented. A total of 10 case reports of DNET malignant transformation were found. The majority of cases involved complex-type DNETs. A higher proportion of extratemporal lesions were observed. Almost all cases involved subtotal resection. Risk factors for malignant progression of DNET lesions include complex-type histologic features,

Introduction Dysembryoplastic neuroepithelial tumors (DNETs) are benign, slow-growing tumors thought to arise from cortical grey matter. They are currently given the classification by the WHO of grade I tumors of neuronal or mixed neuronal–glial tumor type. These lesions have classically been reported to cause intractable seizures in children, and can be treated definitively with surgical resection. DNETs were first described by Daumas-Duport et al. in 1988 [1]. They followed 39 patients with surgically curable partial seizures caused by cortically based tumors. Histologically, these lesions resembled gliomas but were benign and behaved in a stable manner. DNETs were later accepted as a distinct pathologic entity in the WHO classification system as grade I lesions in 1993 [2]. They have since been viewed as slow-growing lesions that typically do not undergo malignant transformation to higher-grade forms. In fact, previous studies have shown that, even in the presence of atypical histological features such as microvascular changes, nuclear atypia, necrosis, or mitoses, DNETs still behave in a benign manner [3]. However, since their relatively recent definition, several case reports have emerged describing DNETs undergoing malignant transformations. We present an additional case of a biopsy-proven DNET that progressed to a more malignant form. We also compiled and reviewed all previously described cases with the aim of finding 0959-4965 © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins

extratemporal location, and subtotal resection. Malignant dedifferentiation of astrocytic or oligodendrocytic cells within the glial nodule of complex DNETs may be the source of these transformations. There are no radiographic features that differentiate DNETs that are at higher risk for transformation. NeuroReport 25:894–899 © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. NeuroReport 2014, 25:894–899 Keywords: dysembryoplastic neuroepithelial tumors, malignant transformation, brain tumors USC Brain Tumor Center, USC Norris Comprehensive Cancer Center, Los Angeles, California, USA Correspondence to Alan A. Moazzam, MD, Department of General Internal Medicine, Keck Hospital of University of Southern California, 1500 San Pablo St Los Angeles, CA 90033, USA Tel: + 1 858 342 0874; fax: + 1 858 486 1099; e-mail: [email protected] Received 22 February 2014 accepted 8 April 2014

common characteristics that may portend a higher risk of undergoing malignant transformations.

Methods The PubMed electronic database was searched for all English-language case reports of DNET described to have undergone recurrence into a more malignant tumor. A broad search strategy was used, using numerous search keywords including but not limited to ‘DNET’, ‘malignant transformation’, and ‘high-grade glioma’. There were no constraints made on the date of publication. Search results were screened by title and abstract to select for case reports of DNETs with evidence of malignant transformation. Additional papers were drawn from references made within these case reports. Publications that met the following criteria were included: (a) patients with an initial pathology consistent with a DNET and a subsequent surgical pathology resembling a more malignant tumor form and (b) a well-described clinical course, including duration of symptoms before presentation, the extent of surgical resection, and the time until recurrence of symptomatic or radiographic disease. These case reports are summarized in Table 1.

Case report A 22-year old woman was referred to neurosurgery after work-up for her recurrent seizures since 19 years old revealed an intracranial mass. The seizures were DOI: 10.1097/WNR.0000000000000184

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Recent

5 years Not reported

47/F

29/M

12/F 7/M

14/M

56/M

7/F

7/F

Gonzales et al. [5]

Hammond et al. [6] Ray et al. [7] Ray et al. [7]

Rushing et al. [8]

Thom et al. [9]

Schittenhelm et al. [10]

Zakrzewski et al. [11]

R temporal

Temporal (side not reported) R frontal

R temporoparietal

R frontal L parieto-occipital

L frontal

L frontal

L frontal

R temporal

Location of lesion

Initial pathology

Complex DNET

Complex DNET

Complex DNET

Complex DNET

Simple DNET Complex DNET

a

DNET nonspecific type Complex DNET Ki-67: 0.3% Not reported Ki-67: 10% Complex DNET

+ Synaptophysin

Mild pleomorphism and rare mitoses Not applicable Scattered microcalcifications seen and no significant vascular changes Astrocytic and oligodendroglial. Mitotic activity, vascular changes, or necrosis were not seen No mitotic figures. No 1p/19q deletion No mitosis or necrosis. Strong + GFAP (astrocytes)

Astrocytic. Mild pleomorphism and rare mitoses 1p-19q deletion

Not applicable

Histopathologic description of cells found within glial nodule

STR

STR

STR

STR + radiation + lomustine

STR + radiation STR

STR

STR

STR

STR

Extent of resection

4

7

2

3

2 8

11

3

11

6

Time until recurrence (years)

Follow-up pathology

Increased cellularity and pleomorphism, microvascular proliferations, an elevated proliferative activity (MIB1-index focally up to 10%) and cellular atypia not typical for WHO grade I DNET Pilocystic astrocytoma. In DNET component: hypercellularity and slight anisonucleosis of oligodendroglioma-like cells were striking (Fig. 6). No mitotic activity was found

Anaplastic mixed glioneuronal tumor

Anaplastic astrocytoma

Grade III anaplastic astrocytoma Complex DNET with a component of pilocystic astrocytoma

Grade IV astrocytoma

Grade II oligoastrocytoma

Grade IV astrocytoma

Grade II oligoastrocytoma

a

DNETs, dysembryoplastic neuroepithelial tumors; F, female; GFAP, glial fibrillary acidic protein; L, left; M, male; R, right; STR, subtotal resection. The lack of a multinodular pattern suggested the simple form. However, this was not entirely certain, as the submitted surgical specimen was fragmented, potentially disrupting a more complex underlying architecture.

6 months

7 years

44 years

3 months

Not reported

7 years

29/M

Duggal et al. [4]

3 years

22/F

Age/sex

Current case report (2014)

Duration of symptoms before initial diagnosis

All previously described cases of DNETs with malignant transformation

References

Table 1

Malignant transformation of DNET tumors Moazzam et al. 895

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

896

NeuroReport 2014, Vol 25 No 12

Fig. 1

(a)

(b)

20 μm

20 μm

(c)

(d)

20 μm

20 μm

(e)

50 μm

Tumor histopathologic images. (a) High-power slide from pathologic specimen from 2001 resection with H&E stain: a moderately cellular tumor composed of mostly bare-appearing nuclei in a myxoid stroma. Most of the nuclei are round and of varying sizes. The larger nuclei frequently have a vesicular chromatin pattern and small distinct nucleoli (solid arrow), resembling neurons. The smaller nuclei tend to be darker with a dense, evenly dispersed chromatin pattern, resembling oligodendroglia. Occasionally, a larger nucleus has distinct amphophilic cytoplasm associated with it. Rare binucleated forms are seen (hollow arrow). There is no necrosis. (b, c) High-power slide from pathologic specimen from 2004 resection with (b) H&E stain and (c) synaptophysin: this specimen revealed a marked increase in cellularity with the majority of the cell nuclei appearing round (hollow arrow) and oligodendroglial, although there is increased fibrillarity in some areas of the stroma (solid arrow). Atypical, neuronal appearing cells are still identified. Synaptophysin staining revealing marked positivity. (d) High-power section of specimen from 2007 resection, (e) low-power view of same tumor: a highly cellular histologically heterogenous tumor. The background stroma ranges from eosinophilic and fibrillar to myxoid. In the myxoid areas, many of the cells are round to oval with eccentric nuclei in a small amount of distinct eosinophilic cytoplasm (mini gemistiocytes). Other cells have similarly round nuclei but with no apparent nucleoli that appear neuronal.

described as partial with no secondary generalization. An MRI brain was obtained; this revealed a 2.4 × 5.0 × 3.0 cm right temporal and inferior frontal mass, with low signal in T1 images and high signal in T2 images. There was no significant contrast enhancement and minimal mass effect. There were no other abnormalities within the brain parenchyma.

The patient underwent surgery, and a subtotal resection was achieved; only 75% of the tumor could be excised, as a portion of the lesion was abutting the Sylvian fissue. Binucelated neuronal forms and low-grade astrocytic proliferation on a myxoid background were identified in the tumor histology (Fig. 1a); these findings in the setting of a cortically confined temporal lobe lesion of a young

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Malignant transformation of DNET tumors Moazzam et al. 897

patient with long-standing epilepsy were interpreted to be consistent with a nonspecific type DNET. Cytogenetics were performed, and there were no 1p-19q chromosomal deletions. The patient subsequently had a recurrence of tumor in 2004 on surveillance imaging. This time there was some mass effect and midline shift seen on the MRI brain. Reresection was performed, and the specimen was again interpreted as a DNET (Fig. 1b and c). Finally, the patient had a third recurrence in 2007, as demonstrated by the MRI in Fig. 2. Pathology revealed a grade II anaplastic mixed-type oligoastrocytoma (Fig. 1d and e). The patient underwent her third resection successfully and remained seizure-free until late 2013 when she again demonstrated progression of the disease. She occasionally experiences partial seizures of her left upper extremity, despite receiving chemoradiation and taking three antiepileptic medications. She describes the frequency to be once every 2–3 days, as opposed to daily seizures when the mass was first discovered. She continues to be followed in our clinic.

Fig. 2

MRI scan of recurrent lesion from 2007: axial contrast-enhanced T1-weighted image with fat saturation shows a large infiltrative mass involving the right frontal lobe, basal ganglia, and right temporal lobe with some regions of contrast enhancement medially. There is marked mass effect on the right lateral ventricle and significant right-to-left midline shift.

Discussion Malignant transformations of DNET lesions are a welldescribed phenomenon; however, it is not entirely clear which features of these lesions induce a higher risk for subsequent malignant transformation. By compiling all previously described cases, we aimed to identify common characteristics that may confer such a higher risk. Since the original report in 1988, three subclassifications of DNETs have been described: simple, complex, and nonspecific [3]. The hallmark of simple DNETs is the presence of the ‘specific glioneuronal element’ (SGNE), a histopathological finding consisting of small, round oligodendroglial-like cells arranged in columns around a central core made of axon bundles running perpendicular to the cortical surface. Scattered between these columns are ‘floating neurons’ that sit in pools of mucopolysaccharide-rich, basophilic mucinous matrix. The larger floating neurons and the smaller oligodendroglial-like cells express neuronal markers such as S-100. They do not show evidence of glial differentiation. In addition to the SGNE, complex DNETs possess glial nodules and foci of adjacent cortical dysplasia. The cells within these glial nodules can be either of astrocytic or oligodendrocytic morphology and can resemble low-grade gliomas. Cytologic atypia may also be present. The ‘nonspecific’ subtype of DNETs remains somewhat controversial and not universally accepted. This form of DNET neither possesses the SGNE nor the glial nodules, but instead resembles a diffuse glioma confined to the cortex. Daumas-Duport et al. [3] have reported that the mitotic activity and high proliferative index in these variants may also be similar to high-grade gliomas. They therefore concluded that diagnosis of this subtype should only be made in the setting of supportive clinical history and radiographic features. Specifically, they list the following criteria for diagnosis: (a) onset of partial seizures occurring before 20 years of age; (b) no neurological deficit or a stable congenital neurological deficit; (c) lesions appear cortically based on MRI; and (d) no mass effect evident on neuroimaging, in particular, when there is no surrounding tumoral edema [3]. Almost all reported cases of malignant transformation in our review were reported as complex DNETs, and all underwent subtotal resection, as demonstrated in Table 1. It is possible that this observation may be the result of a causal association. As described above, the glial nodules of complex DNETs possess astrocyte or oligodendrocyte cells. It is possible that dedifferentiation of these particular cell populations is the source for malignant transformations. In a small case series, Duggal et al. [4] observed that many of their described patients had some sort of histologic or genetic aberrancy within the glial nodules, not typical of benign DNETs. They suggested that these features may be more influential in the long-term behavior of DNETs. This theory may also explain several observed cases where development of

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

898

NeuroReport 2014, Vol 25 No 12

astrocytomas or oligodendrogliomas occurred in or near otherwise benign DNET lesions. For instance, Josan et al. [12] described a case where an epileptogenic lesion was excised and pathology revealed pilocytic astrocytoma growing within a complex-type DNET. Duggal et al. [4] set out to correlate the aggressiveness of 14 DNET lesions to the relative balance between proliferative activity (as measured by Ki-67) and the rate of apoptosis (as measured by the TUNEL technique). Their study did not show any correlation between these indices and the presence of atypical histologic features. However, they did report a case that progressed to a more malignant form, and nuclear pleomorphism and mitotic figures were found in the astrocytic component of the glial nodule of this case. In reviewing the 13 additional cases, only one other case (patient 2) demonstrated cytological atypia, vascular proliferation, and mitoses, again confined to the glial nodule. Hammond et al. [6] reported a case of malignant recurrence where the initial lesion had glial nodules significant for moderate pleomorphism and rare mitotic figures. These observations seem to suggest that these atypical histologic findings within the glial nodule increase the risk for malignant transformation. Some authors have looked into the significance of cytogenetic abnormalities with respect to this phenomenon. There are no genetic abnormalities that are specific to the delineation of DNETs or their subtypes. However, genetic abnormalities can help distinguish these tumors from more malignant glial neoplasms. In particular, 1p19q chromosome deletions seen in oligodendrogliomas are not present in DNET. In addition, low-grade gliomas typically have an IDH1 mutation that is again atypical in DNETs. Gonzales et al. [5] followed 14 patients with recurrent lesions. Only one patient was found to have a subsequent WHO grade II oligoastrocytoma. This case was also the only lesion with a 1p-19q chromosomal deletion, leading them to conclude that this deletion may be useful in identifying DNETs with malignant potential. Ray and colleagues reported a patient with progression to a WHO grade III anaplastic astrocytoma. Although genetic testing was not performed on the original lesion, the remaining DNET components on the subsequent resection demonstrated 1p-19q deletions [7]. Such associations have not been demonstrated consistently throughout the literature. Extensive pathological study by Thom et al. [9] on 101 DNETs did not find any correlation between 1p-19q deletions and progression of malignancy. Our case also did not possess chromosomal deletions. Interestingly, of the 10 previously reported cases presented in Table 1, seven DNETs were of extratemporal location. This is in contrast with the typical site of DNET formation, which is the temporal lobe [9]. Although our compilation is subject to publication bias, it seems

plausible to suggest that extratemporal locations of DNETs may serve as an additional risk factor for progression to malignancy. From an imaging perspective, the classic MRI appearance of a DNET is a well-defined, cortically based mass, most often in the temporal lobe with minimal or no mass edema or mass effect [13–15]. They appear T1 hypointense and T2 hyperintense and often possess a hyperintense T2 FLAIR rim [16]. Bony remodeling of the inner table of the calvarium can be seen; however, this is dependent on the location of the tumor. Contrast enhancement is usually not seen, although in some series, contrast enhancement can be seen in 18–33% of patients [14]. Calcification and hemorrhage can also be seen in some patients. Our case demonstrated classical DNET MRI features of a well-circumscribed, cortically based, nonenhancing T1 hypointense mass. Mano et al. [14] recently reported two cases of DNETs, which developed dynamic changes over time. The first case involved a young child with a right parietal mass that demonstrated enlargement over time as well as development of a focus of contrast enhancement. Following subtotal resection, histopathological examination demonstrated the nonenhancing portion of the mass to be consistent with a WHO grade I DNET, whereas the enhancing portions were found to be anaplastic oligodendroglioma, WHO grade III. The second case involved a patient who underwent gross total resection of a DNET at an age of 3 years and then developed a recurrent enhancing mass, also diagnosed as a DNET, 11 years later. Gross total resection was again performed and the nonenhancing and enhancing components of the recurrent tumor were diagnosed as simple and complex forms of DNET, respectively. It does appear that DNETs may develop dynamic changes in appearance over time and that the development of enhancement may or may not indicate malignant transformation [17]. Daghistani et al. [13] recently reported the first systematic analysis of the prevalence of unusual behaviors of DNETs. They performed a retrospective review of 51 patients with pathologically proven DNET at two pediatric hospitals. They found that six of 29 patients (20.7%) who had follow-up imaging demonstrated enlargement preoperatively, six of 18 patients (33.3%) with subtotal resection showed enlargement of residual tumor, three of 30 (10%) patients with gross total resection developed recurrent tumor, and two of 51 (3.9%) patients developed lesions distant to the primary neoplasm. These results indicate that long-term follow-up is important for DNETs, even in patients with gross total resection.

Conclusion On the basis of our literature review of previously published cases, DNETs that progress to more malignant

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Malignant transformation of DNET tumors Moazzam et al. 899

forms appear to have a preponderance of extratemporal locations. They tend to occur exclusively in the setting of complex-type DNETs. It is possible that the degree of histopathologic atypia present within the glial component may affect the likelihood of malignant transformation. There are no radiographic features, such as areas of contrast enhancement, that help differentiate DNETs with a higher predilection for malignant transformation. In the case of nonspecific DNET subtypes, great care must be taken in making such a diagnosis, and it should only be made in the setting of corroborating clinical and radiographic data. This report emphasizes the potential of this tumor to transform into more malignant cancers, and therefore illustrates the need for continued follow-up [8,10,11].

6

7

8

9

10

Acknowledgements The authors thank Dr John Gruen for participating in this patient’s care.

11

12

Conflicts of interest

There are no conflicts of interest.

13

References 1 Daumas-Duport C, Scheithauer BW, Chodkiewicz JP, Laws ER, Vedrenne C. Dysembryoplastic neuroepithelial tumor: a surgically curable tumor of young patients with intractable partial seizures. Report of thirtynine cases. Neurosurgery 1988; 23:545–556. 2 Kleihues P, Burger PC, Scheithauer BW. The new WHO classification of brain tumours. Brain Pathol 1993; 3:255–268. 3 Daumas-Duport C, Varlet P, Bacha S, Beuvon F, Cervera-Pierot P, Chodkiewicz JP. Dysembryoplastic neuroepithelial tumors: nonspecific histological forms – a study of 40 cases. J Neurooncol 1999; 41:267–280. 4 Duggal N, Taylor R, Zou GY, Hammond RR. Dysembryoplastic neuroepithelial tumours: clinical, proliferative and apoptotic features. J Clin Pathol 2008; 61:127–131. 5 Gonzales M, Dale S, Susman M, Nolan P, Ng WH, Maixner W, Laidlaw J. Dysembryoplastic neuroepithelial tumor (DNT)-like oligodendrogliomas or

14

15

16

17

Dnts evolving into oligodendrogliomas: two illustrative cases. Neuropathology 2007; 27:324–330. Hammond RR, Duggal N, Woulfe JM, Girvin JP. Malignant transformation of a dysembryoplastic neuroepithelial tumor. Case report. J Neurosurg 2000; 92:722–725. Ray WZ, Blackburn SL, Casavilca-Zambrano S, Barrionuevo C, Orrego JE, Heinicke H, et al. Clinicopathologic features of recurrent dysembryoplastic neuroepithelial tumor and rare malignant transformation: a report of 5 cases and review of the literature. J Neurooncol 2009; 94:283–292. Rushing EJ, Thompson LD, Mena H. Malignant transformation of a dysembryoplastic neuroepithelial tumor after radiation and chemotherapy. Ann Diagn Pathol 2003; 7:240–244. Thom M, Toma A, An S, Martinian L, Hadjivassiliou G, Ratilal B, et al. One hundred and one dysembryoplastic neuroepithelial tumors: an adult epilepsy series with immunohistochemical, molecular genetic, and clinical correlations and a review of the literature. J Neuropathol Exp Neurol 2011; 70:859–878. Schittenhelm J, Mittelbronn M, Wolff M, Truebenbach J, Will BE, Meyermann R, Beschorner R. Multifocal dysembryoplastic neuroepithelial tumor with signs of atypia after regrowth. Neuropathology 2007; 27:383–389. Zakrzewski K, Biernat W, Liberski PP, Polis L, Nowoslawska E. Pilocytic astrocytoma as a predominant component of a recurrent complex type DNT. Folia Neuropathol 2009; 47:284–288. Josan V, Smith P, Kornberg A, Rickert C, Maixner W. Development of a pilocytic astrocytoma in a dysembryoplastic neuroepithelial tumor. Case report. J Neurosurg 2007; 106:509–512. Daghistani R, Miller E, Kulkarni AV, Widjaja E. Atypical characteristics and behavior of dysembryoplastic neuroepithelial tumors. Neuroradiology 2013; 55:217–224. Mano Y, Kumabe T, Shibahara I, Saito R, Sonoda Y, Watanabe M, Tominaga T. Dynamic changes in magnetic resonance imaging appearance of dysembryoplastic neuroepithelial tumor with or without malignant transformation. J Neurosurg Pediatr 2013; 11:518–525. Fernandez C, Girard N, Paz Paredes A, Bouvier-Labit C, Lena G, Figarella-Branger D. The usefulness of MR imaging in the diagnosis of dysembryoplastic neuroepithelial tumor in children: a study of 14 cases. Am J Neuroradiol 2003; 24:829–834. Parmar HA, Hawkins C, Ozelame R, Chuang S, Rutka J, Blaser S. Fluidattenuated inversion recovery ring sign as a marker of dysembryoplastic neuroepithelial tumors. J Comput Assist Tomogr 2007; 31:348–353. Jensen RL, Caamano E, Jensen EM, Couldwell WT. Development of contrast enhancement after long-term observation of a dysembryoplastic neuroepithelial tumor. J Neurooncol 2006; 78:59–62.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Malignant transformation of DNETs: a case report and literature review.

Dysembryoplastic neuroepithelial tumors (DNETs) have traditionally been thought of as lesions with a benign clinical course that generally do not unde...
493KB Sizes 0 Downloads 5 Views