Pilocytic astrocytoma as a predominant component of a recurrent complex type DNT

Introduction
Dysembryoplastic neuroepithelial tumour (DNT) is a benign glial-neuronal neoplasm of the cerebral hemispheres occurring in children and young adults. Patients with DNT usually present with seizures, which in some cases result in a long-standing and intractable epilepsy [1,2,5]. Although DNT used to be referred to as a non-neoplastic hamartomatous lesion by some pathologists, reports on its recurrence support an alternative view indicating its truly neoplastic nature [9,10,12]. Malignant transformation of DNT has also been reported [4,13,14]. We present a unique case of a child with subtotal resection of DNT that recurred four years later with predominant pilocytic astrocytoma and only small remnants of otherwise typical DNT.
Case presentation
In July 1999, a 7-year-old girl with a 6-month history of partial simple seizures (clonic convulsions of the left facial muscles) was admitted to the hospital. CT scans revealed a hypodense lesion within the right temporal lobe. Neither other complaints nor abnormalities on the neurological examination were disclosed at that time. On magnetic resonance imaging (MRI) scans, the lesion was 5.5 × 5 × 5 cm in diameter and appeared hypointense on T1-weighted images. Post-contrast scans revealed only small focal enhancement (8 mm in diameter) in the postero-lateral part of the lesion. On T2-weighted images the lesion was hyperintense with preserved gyrus-like configuration. Neither mass effect nor brain shift was noted, although the temporal bone covering the lesion was deformed, thinned and bulging (Fig. 1).
The lesion was removed by a standard right temporal craniotomy. The tumour was poorly vascularized. As the border between the lesion and the normal brain was not clearly demarcated, subtotal resection was achieved. The postoperative course was uneventful.
The microscopic evaluation showed typical complex form of DNT. It was composed of a specific glioneuronal element with floating neurons within the small mucoid lakes (Fig. 2). Oligodendrocyte-like cells (OLCs) showed expression of S100 protein and OLIG2, whereas the neurons expressed synaptophysin. The glial component surrounded the specific glioneuronal component in a diffuse manner. It was composed of OLCs, but in contrast to typical oligodendroglioma, no branching network of capillaries was present. No piloid cells were identified. The clinical result of the surgery was satisfactory. Anti-epileptic drugs were gradually withdrawn in the following 3 months. No seizures have been observed during the 4-year follow-up. Taking into account the minimal biological activity of DNT second-look surgery was not performed, and the residual tumour has been controlled by periodic MRI scans (Fig. 3).
Four years later the patient started to complain of periodic headache, vomiting and numbness of left limbs. The neurological examination did not reveal any abnormalities. On MRI, a new lesion in the area of previous surgery was identified. The tumour was cystic with a small peripheral irregular solid part. It was 7 × 6 × 5 cm in diameter, and appeared slightly hypointense on T1-weighted images and hyperintense on T2-weighted images. Post-contrast scans revealed intense enhancement of the cyst wall and the solid part of the tumour. Moderate mass effect and brain shift were also noted (Fig. 4).
The patient underwent surgery via the same route as for the previous operation. This time, the tumour was well demarcated and highly vascularized. Gross total resection was achieved and the postoperative course was uneventful.
The microscopic evaluation of the recurrent lesion disclosed two distinct morphological patterns. The prevailing one was typical for pilocytic astrocytoma (PA) and showed elongated, “piloid” cells dispersed within the mucoid stroma (Fig. 5A). Some of those cells had a hyperchromatic, enlarged nucleus, but mitotic activity was basically absent. Scattered eosinophilic granular bodies were present (Fig. 5A). Some areas were composed of oligodendroglioma-like cells (Fig. 5B). Piloid cells had cytoplasmic expression of GFAP, whereas oligodendroglioma-like cells were GFAP-negative. Synaptophysin-positive cells were not identified in this portion of the tumour. The second component was identified only in a small portion of the tumour (approximately 10% of the total tumour bulk). It shared the same histological features with the primary lesion (glioneuronal element of DNT). However, hypercellularity and slight anisonucleosis of oligodendroglioma-like cells were striking (Fig. 6). No mitotic activity was found in that component. Immunohistochemical analysis of that part was not performed due to the lack of this component in the subsequent slide sections.
Five years after the second operation the patient was in a good condition, with no complaints or neurological disorders. No recurrence or residual tumour was visible on the control MRI scans.
Discussion
Our case is the first showing development of PA as a main component within a regrowth of DNT. Both tumours had typical clinical and radiological features for the respective nosological entities. The radiological appearance of the primary tumour was characteristic of DNT [6,7,11], whereas that of the second tumour was pathognomonic for PA, the most common neuro-oncological entity in the paediatric population [3,8]. Clinical signs and symptoms of the two lesions also differed. The former, like all DNTs, caused epilepsy, while the latter was revealed due to the symptoms of intracranial hypertension. At the time of the original diagnosis of DNT, there were no radiological signs suggesting unusual clinical course of the disease.
DNT has areas composed of astrocytic, oligodendroglial and neuronal components. Theoretically, overgrowth of any of them may result in an independent tumour, as was disclosed in our case. This hypothetical sequence of events and the histological appearance also support neoplastic or preneoplastic nature of DNT. This was also confirmed by identification of DNT remnants identified within the recurrent lesion (Fig. 6). That component showed increased cellularity in comparison with the primary DNT, and this phenomenon was accompanied by slight cytological atypia of the oligodendroglioma-like cells. “Degenerative” atypia is a well defined phenomenon in “ancient” lesions, e.g., schwannoma or pilocytic astrocytoma. Nuclear atypia was also identified in the PA that recurred at the same location. All those findings may suggest that recurrent tumour developed from the remnants of a similar lesion that was not resected completely at the first surgery. This also suggests that resection of DNT may create a favourable local environment for astrocytic overgrowth, the DNT basically being effaced by the seemingly more vividly growing PA.
Slow radiological progression of untreated DNT as well as tumour recurrence after surgical excision, although rare, are well described [9,10,11]. However, “replacement” of DNT by another type of tumour is extremely rare. Only three cases of such a process have been described, all of them resulting in a malignant tumour. In the first case, the anaplastic astrocytoma in a 14-year-old boy arose at the site of a DNT three years after total resection followed by combined radiation and chemotherapy. Both treatment modalities seem to be main factors in the process of induction of the malignancy [13].
In the second patient, the 14-year-old girl, a recurrent tumour developed 7 years after subtotal resection of DNT, showing signs of atypia: increased cellularity and pleomorphism, microvascular proliferations, elevated proliferative activity and cellular atypia not typical for WHO DNT [14]. Another case was a 29-year-old man, who developed grade IV glioblastoma 11 years after subtotal resection of DNT. In that case neither radiotherapy nor chemotherapy was applied, and a pure coincidence of two biologically different neoplasms (collision tumours) could explain this phenomenon [4]. Our patient, according to standards, was not treated oncologically after the first subtotal tumour resection. Thus, we think that either collision tumour theory may elucidate the pathogenesis of the recurrent lesion or, hypothetically, we assume that incomplete DNT resection made suitable environmental conditions for overgrowth of the astrocytic component of DNT that resulted in full-blown PA. Another possibility is that the residual astrocytic component had more proliferative potential than the true DNT part of the neoplasm in our case.
Conclusion
Patients with incompletely removed dysembryoplastic neuroepithelial tumour may suffer local recurrence of that tumour. In rare cases development of
a neoplasm with a changed morphology may occur.
References
1. Daumas-Duport C. Dysembryoplastic neuroepithelial tumours. Brain Pathol 1993; 3: 283-295.
2. Daumas-Duport C, Scheithauer BW, Chodkiewicz JP, Laws ER Jr, Vedrenne C. Dysembryoplastic neuroepithelial tumor. A surgically curable tumor of young patients with intractable partial seizures. Report of thirty-nine cases. Neurosurgery 1988; 23: 545-556.
3. Finizio FS. CT and MRI aspects of supratentorial hemispheric tumors of childhood and adolescence. Child`s Nerv Syst 1995; 11: 559-567.
4. Hammond RR, Duggal N, Woulfe JM, Girvin JP. Malignant transformation of a dysembryoplastic neuroepithelial tumor. Case report. J Neurosurg 2002; 92: 722-725.
5. Kordek R, Waschnitz J, Biernat W, Saringer W, Czech T, Zakrzewski K, Polis L, Alwasiak J, Liberski PP, Budka H. Clinical, radiological and histological presentation of dysembryoplastic neuroepithelial tumors (DNT). Report of two cases. Folia Neuropathol 1996; 34: 199-205.
6. Kuroiwa T, Kishikawa T, Kato A, Ueno M, Kudo S, Tabuchi K. Dysembryoplastic neuroepithelial tumor. MR findings. J Comput Assist Tomogr 1994; 18: 352-356.
7. Kuroiwa T, Bergey GK, Rothman MI, Zoarski GH, Wolf A, Zagardo MT, Kristt DA, Hudson LP, Krumholz A, Barry E, Numaguchi Y. Radiologic appearance of the dysembryoplastic neuroepithelial tumor. Radiology 1995; 197: 233-238.
8. Lee YY, Van Tassel P, Bruner JM, Moser RP, Share JC. Juvenile pilocytic astrocytomas. CT and MR characteristics. Am J Neuroradiol 1989; 10: 363-370.
9. Maher CO, White JB, Scheithauer BW, Raffel C. Recurrence of dysembryoplastic neuroepithelial tumor following resection. Pediatr Neurosurg 2008; 4: 333-336.
10. Nolan MA, Sakuta R, Chuang N, Otsubo H, Rutka JT, Snead OC 3rd, Hawkins CE, Weiss SK. Dysembryoplastic neuroepithelial tumors in childhood. Long-term outcome and prognostic features. Neurology 2004; 62: 2270-2276.
11. Ostertun B, Wolf HK, Campos MG, Matus C, Solymosi L, Elger CE, Schramm J, Schild HH. Dysembryoplastic neuroepithelial tumors. MR and CT evaluation. Am J Neuroradiol 1996; 17: 419-430.
12. Prayson RA, Morris H, Estes ML, Comair YG. Dysembryoplastic neuroepithelial tumor. A clinicopathologic and immunohistochemical study of 11 tumours including MIB1 immunoreactivity. Clin Neuropathol 1996; 15: 47-53.
13. Rushing EJ, Thompson LD, Mena H. Malignant transformation of a dysembryoplastic neuroepithelial tumour after radiation and chemotherapy. Ann Diagn Pathol 2003; 7: 240-244.
14. 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.