Abstract

Background: Rosette-forming glioneuronal tumor (RGNT) is a rare tumor that arises primarily in the posterior fossa, with molecular features of FGFR1 mutation. A previous study reported that brainstem RGNT accounts for only 2.7% cases; therefore, midbrain RGNT is infrequent.

Case Description: The authors encountered two cases of RGNT located in the midbrain tegmentum (Case 1: 23-year-old woman and Case 2: 18-year-old boy), both exhibiting similar cystic components with gadolinium-enhanced cyst walls on preoperative magnetic resonance imaging, surgically resected through the occipital transtentorial approach. Histological findings in both cases comprised two characteristic architectures of neurocytic and glial components, typical of RGNT. Molecular assessment revealed no FGFR1 mutation in the initial specimen, but revealed FGFR1 K656E mutation in the recurrent specimen in Case 1 and showed no FGFR1 mutation but showed TERT C228T mutation in Case 2. Neither case revealed IDH1/2, BRAF, H3F3A K27, H3F3A G34, or HIST1H3B K27 mutations. DNA methylation-based classification (molecularneuropathology.org) categorized both cases as RGNT, whose calibrated scores were 0.99 and 0.47 in Cases 1 and 2, respectively.

Conclusion: Midbrain tegmentum RGNTs exhibited typical histological features but varied FGFR1 statuses with TERT mutation. RGNT in rare locations may carry different molecular alterations than those in other common locations, such as the posterior fossa.

Keywords: FGFR1, Midbrain, Rosette-forming glioneuronal tumor, Tegmentum, TERT

INTRODUCTION

Rosette-forming glioneuronal tumor (RGNT) is a rare tumor that arises primarily in the posterior fossa[ 21 ] with a hallmark of FGFR1 mutation.[ 4 , 12 , 13 , 25 , 29 ] This tumor entity was first described in 2002,[ 21 ] and classified as the World Health Organization Grade I in 2007. In addition to the posterior fossa, RGNT arises throughout the central nervous system, such as in the supratentorial region, pineal region, basal ganglia, and spinal cord;[ 2 , 7 , 9 , 14 , 20 , 22 , 30 , 31 ] its incidence in the brain stem is 2.7%.[ 31 ]

We encountered two cases of RGNT located in the midbrain tegmentum, both of which demonstrated a gadolinium-enhanced cyst wall and were radically resected using the occipital transtentorial approach. Molecular analyses of the rare midbrain tegmentum RGNT included FGFR1, TERT, IDH1, IDH2, H3F3A, HIST1H3B, BRAF, KIAA1549-BRAF fusion, and DNA methylation-based classifiers provided by MolecularNeuropathology.org.[ 8 ] Moreover, in a literature review, we illustrate the FGFR1 status in unusual tumor locations of RGNT.

CASE DESCRIPTION

Clinical summary

Case 1

A 23-year-old woman presented with sensory disturbances in the left face and hand. Gadolinium-enhanced T1-weighted (GdT1WI) magnetic resonance imaging (MRI) revealed enhancement suggestive of a thin cystic wall without nodule formation on the right side of the midbrain tegmentum [ Figure 1a ]. The patient underwent partial tumor resection through the occipital transtentorial approach without developing postoperative neurological deficits [ Figure 1b ]. GdT1WI MRI obtained 15-month postsurgery revealed spontaneous regression of the enhanced cystic tumor [ Figure 1c ]; however, GdT1WI MRI obtained 30-month postsurgery revealed regrowth of the enhanced tumor [ Figure 1d ]. She underwent a second surgery through the occipital transtentorial approach, subsequently achieving gross total tumor resection [ Figure 1e ]. No additional neurological deficits were noted. A central pathological review diagnosed both the initial and recurrent tumors as RGNT. Follow-up GdT1WI MRI 3 years after the second surgery revealed good local tumor control [ Figure 1f ] without neurological deficits.

Figure 1:

Gadolinium-enhanced T1-weighted magnetic resonance imaging (GdT1WI) of Cases 1 and 2. Case 1: Preoperative GdT1WI reveals an enhanced cystic tumor on the right side of the midbrain tegmentum (a). Postoperative magnetic resonance imaging indicates partial resection of the tumor (b). GdT1WI obtained 15 months after the initial surgery reveals tumor regression (c). After 30 months, follow-up GdT1WI reveals regrowth of the enhanced tumor (d). GdT1WI obtained immediately after the second surgery indicates gross total tumor resection (e). GdT1WI obtained 3 years after the second surgery reveals no recurrence (f). Case 2: Preoperative GdT1WI reveals an enhanced cystic tumor on the left side of the midbrain tegmentum (g). Postoperative GdT1WI indicates subtotal resection of the tumor (h). GdT1WI obtained 2 years after the second surgery reveals no recurrence (i).

Case 2

An 18-year-old boy presented with a tremor in the right hand. GdT1WI MRI revealed a cystic tumor with thin-wall enhancement on the left side of the midbrain tegmentum [ Figure 1g ]. The patient underwent subtotal tumor resection through an occipital transtentorial approach [ Figure 1h ]. Transient right oculomotor nerve palsy was observed postoperatively, with complete recovery. A central pathological review diagnosed the patient with RGNT. Follow-up MRI 3 years postsurgery revealed no tumor recurrence [ Figure 1i ].

Tumor specimens, DNA extraction, and pyrosequencing

Surgical tumor specimens of the initial and recurrent tissue samples of Case 1 and the initial tissue sample of Case 2 were subjected to the central pathological review by the Japan Children’s Cancer Group. Due to the tumor location in the midbrain, the tumor specimens were small. We were only able to obtain fresh frozen tissue from the recurrent tumor in Case 1, while paraffin-embedded specimens were available for the others. DNA from the frozen tumor tissue was extracted using a DNeasy Blood and Tissue Kit (Qiagen, Tokyo, Japan), and DNA from paraffin-embedded tumor tissue was extracted using a QIAamp DNA FFPE Tissue Kit (Qiagen, Tokyo, Japan). Hotspot mutations, including IDH1 R132, IDH2 R172, BRAF T599, BRAF V600, H3F3A K27, H3F3A G34, HIST1H3B K27, FGFR1 N546, FGFR1 K656, and C228T or C250T mutations in the promoter region of TERT (TERTp), were assessed through pyrosequencing[ 5 ] using the AQ assay of PyroMark Q96 (version 2.5.7) on a PyroMark ID pyrosequencer (Qiagen, Tokyo, Japan), according to the manufacturer’s instructions. Real-time polymerase chain reaction was performed to identify KIAA1549-BRAF fusion in Case 1. Sanger sequencing was conducted for FGFR1. The central nervous system tumor classification based on methylation profiling was performed using the methylation profiling classifier developed by the German Cancer Research Center (DKFZ)/University Hospital Heidelberg/German Consortium for Translational Cancer Research (DKTK) (the DKFZ classifier, molecularneuropathology.org).[ 8 ] We used an Infinium MethylationEPIC BeadChip array (Illumina, San Diego, CA, USA) and uploaded the IDAT files of the samples to their website to obtain the classification and copy-number profiles.

Histological findings

Cases 1 [ Figure 2a - f ] and 2 [ Figure 3a - e ] revealed similar histological findings. Hematoxylin and Eosin (HE) staining demonstrated a biphasic pattern of glial and neurocytic architecture [ Figures 2a - c and 3a-c ]. The former component, comprising an astrocyte-like structure, was positive for GFAP [ Figure 2d and 3d ], and the latter component, comprising neurocytic rosettes and perivascular pseudorosettes, was positive for synaptophysin [ Figure 2e and 3e ]. [ Figure 2f ] shows the HE staining of the recurrent tissue in Case 1 without malignant changes. Other immunohistochemical staining methods demonstrated IDH1 R132H-negativity, O6-methylguanine DNA methyltransferase-negativity, TP53 sparse positivity, and 2–3% on the MIB-1 labeling index; thus, RGNT was diagnosed.

Figure 2:

Histological findings of Case 1. Hematoxylin and Eosin (HE) staining demonstrate coexistence of glial and neurocytic components (a). Higher magnification reveals the glial component (b) and neurocytic component (c). The glial component is GFAP positive (d) and the neurocytic component is synaptophysin positive (e). HE staining of the recurrent tissue does not reveal malignant changes (f). The black bar indicates 100 um.

Figure 3:

Histological findings of Case 2. Hematoxylin and Eosin staining (a-c) demonstrates glial (b) and neurocytic (c) components, confirmed by GFAP (d) and synaptophysin (e) staining, respectively. The black bar indicates 100 um.

Molecular analysis

The results of molecular analyses are summarized in [ Table 1 ].

Table 1:

Summary of molecular analyses of two cases.

Case 1

FGFR1 mutation was assessed using the initial FFPE specimen, but neither N546 nor K656 mutations were detected [ Table 1 ]. Molecular analysis of fresh frozen tissue from the recurrent tumor specimen revealed FGFR1 K656E mutation [ Figure 4a ], but no other mutations in IDH1, IDH2, BRAF, H3F3A, H3F3A, HIST1H3B, TERTp, or KIAA1549-BRAF fusion [ Table 1 ]. DNA methylation-based classifier using the recurrent tissue specimen indicated that the case matched RGNT with a calibrated score of 0.99. The analysis of copy-number variations calculated from the DNA methylation array showed a balanced profile [ Figure 4b ].

Figure 4:

Molecular analysis of Case 1 (a and b) and Case 2 (c and d). Sanger sequencing reveals FGFR1 K656E mutation (c.1966A>G, arrow) in the recurrent specimen (a), and the DNA methylation-based classification reveals flattened copy-number alterations corresponding to those of RGNT (b). Pyrosequencing reveals a mutation at the C228T of TERT promoter (c, red arrow) and the DNA methylation-based classification reveals flattened copy-number alterations corresponding to those of RGNT (d).

Case 2

Molecular analyses using the initial FFPE specimen revealed no FGFR1 mutations but revealed TERTp C228T mutation [ Table 1 and Figure 4c ]. DNA methylation-based classifier classified the case as RGNT, but the calibration score was 0.47 (<0.5). The analysis of copy-number variations calculated from the DNA methylation array revealed a balanced profile [ Figure 4d ].

DISCUSSION

We have reported two cases of histologically confirmed RGNT located in the midbrain tegmentum using molecular analyses. Brainstem RGNT has a prevalence of 2.7%,[ 31 ] and RGNT extending to the tectum has a prevalence of 4.1%.[ 17 ] Only one RGNT case located in the midbrain tegmentum has been reported in the literature;[ 20 ] thus, our cases are second and third. These two cases demonstrated radiologically similar appearances with gadolinium-enhanced cyst walls and were radically resected using the occipital transtentorial approach. In general, midbrain tumors are challenging for neurosurgeons to operate on,[ 10 ] and a cystic tumor is not an optimal target for stereotactic biopsy. Therefore, sufficient tumor tissue cannot be obtained.

Diffuse midline gliomas or tectal gliomas can be differential diagnoses due to the midline, brainstem, or midbrain locations. A report of adult brainstem gliomas included nine lesions located in the midbrain tectum and two in the midbrain tegmentum. One case each exhibited H3K27M positivity on immunohistochemistry,[ 10 ] indicating that diffuse midline glioma can be observed in the midbrain. However, the present RGNT cases did not exhibit diffuse infiltrating features or harbor H3K27M mutations, which are prerequisites for diagnosing diffuse midline gliomas.[ 24 ] Another report of 22 cases of tectal gliomas found three cases (14%) that presented with a cystic component, and eight (36%) that exhibited contrast enhancement.[ 23 ] Although tectal gliomas radiologically resemble midbrain tectal RGNT,[ 14 , 28 , 31 ] DNA methylation profiling can differentiate tectal gliomas from RGNT.[ 23 ]

The initial tumors in Cases 1 and 2 did not harbor FGFR1 mutations, in contrast to the findings of Sievers et al., all of whose RGNT cases carried FGFR1 hotspot mutations.[ 29 ] The unique histological features of RGNT include the presence of neurocytic and glial components.[ 26 ] Kitamura et al. microdissected these two components separately and found two patients who harbored FGFR1 mutations in both components, whereas one harbored an FGFR1 mutation only in the glial component.[ 20 ] Of note, the case with discordance in FGFR1 mutation was located at the midbrain tegmentum. This case and our study indicate that the location in the midbrain tegmentum and the tumor component affected the FGFR1 status because a limited amount of tumor tissue might affect the proportion of the two components. Moreover, in 30 RGNT cases in Sievers’s study, 14 cases were located in the fourth ventricle, ten in the cerebellum, five in the diencephalon, and one in the mesencephalon (midbrain tectum),[ 29 ] but none were located in the midbrain tegmentum. Kitamura’s study included cases of RGNT without FGFR1 mutations located in the lateral ventricle, midbrain tectum, and frontal lobe. Furthermore, Bidinotto et al.,[ 7 ] Hamauchi et al.,[ 15 ] and Shibayama et al.[ 27 ] reported cases of spinal RGNT without FGFR1 mutations. Therefore, RGNT in rare tumor locations may present with various FGFR1 mutation statuses [ Table 2 ]. In terms of copy-number variations calculated from the DNA methylation array, both cases presented balanced profiles corresponding to Sievers’ series.

Table 2:

Summary of RGNT cases without the FGFR1 mutation in unusual tumor locations.

We found a TERTp mutation in the RGNT cases. Duan et al. analyzed TERTp mutations in spinal RGNTs found no mutations.[ 11 ] Our second case with TERTp mutation was classified as RGNT through DNA methylation profiling, but the calibration score was only 0.47. One interpretation is that the small tumor specimen resulted in a disproportion amount of glial and neurocytic components and might have affected the score. An alternative interpretation is that an atypical subset of RGNT with TERTp mutation was differentiated from a typical RGNT. In lower grade gliomas, TERTp mutation is a poor prognostic factor in IDH-wildtype gliomas, but not in IDH-mutated gliomas.[ 3 ] Moreover, TERTp mutation is a poor prognostic factor for glioblastoma.[ 6 , 19 ] At present, the clinical significance of TERTp mutation is unknown in RGNT. RGNT occasionally presents with spontaneous shrinkage,[ 16 ] as observed in Case 1 of the present study. However, RGNT also presents a malignant clinical course with dissemination[ 1 ] and malignant transformation to glioblastoma[ 18 ] which may be affected by TERTp mutation. Further molecular investigations and clinical follow-ups are required.

CONCLUSION

We have reported two cases of cystic tumor in the midbrain tegmentum, histologically diagnosed as RGNTs. Molecular analyses showed that both cases were negative for FGFR1 mutation in the initial tumor specimens, and one case presented with TERTp mutation. The present cases indicate that RGNT in uncommon tumor locations, including the midbrain tegmentum, may molecularly differ from RGNTs in other common tumor locations, and demonstrate the importance of molecular analyses for understanding the pathophysiology of RGNTs.

Declaration of patient consent

Institutional Review Board (IRB) permission obtained for the study.

Financial support and sponsorship

This work was in part supported by JSPS KAKENHI Grant Number 20K16873 (YN).

Conflicts of interest

There are no conflicts of interest.

Acknowledgements

The authors would like to thank central pathological and molecular review by Japan Children’s Cancer Group, and Enago (“ http://www.enago.jp ” www.enago.jp) for the English language review. We thank M. Kitahara and Y. Hibiya for their technical support.

References

1. Allinson KS, O’Donovan DG, Jena R, Cross JJ, Santarius TS. Rosette-forming glioneuronal tumor with dissemination throughout the ventricular system: A case report. Clin Neuropathol. 2015. 34: 64-9

2. Anan M, Inoue R, Ishii K, Abe T, Fujiki M, Kobayashi H. A rosette-forming glioneuronal tumor of the spinal cord: The first case of a rosette-forming glioneuronal tumor originating from the spinal cord. Hum Pathol. 2009. 40: 898-901

3. Aoki K, Nakamura H, Suzuki H, Matsuo K, Kataoka K, Shimamura T. Prognostic relevance of genetic alterations in diffuse lower-grade gliomas. Neuro Oncol. 2018. 20: 66-77

4. Appay R, Bielle F, Sievers P, Barets D, Fina F, Boutonnat J.editors. Rosette-forming glioneuronal tumours are midline, FGFR1-mutated tumours. Neuropathol Appl Neurobiol. 2022. p. e12813

5. Arita H, Narita Y, Fukushima S, Tateishi K, Matsushita Y, Yoshida A. Upregulating mutations in the TERT promoter commonly occur in adult malignant gliomas and are strongly associated with total 1p19q loss. Acta Neuropathol. 2013. 126: 267-76

6. Arita H, Yamasaki K, Matsushita Y, Nakamura T, Shimokawa A, Takami H. A combination of TERT promoter mutation and MGMT methylation status predicts clinically relevant subgroups of newly diagnosed glioblastomas. Acta Neuropathol Commun. 2016. 4: 79

7. Bidinotto LT, Scapulatempo-Neto C, Mackay A, de Almeida GC, Scheithauer BW, Berardinelli GN. Molecular profiling of a rare rosette-forming glioneuronal tumor arising in the spinal cord. PLoS One. 2015. 10: e0137690

8. Capper D, Jones DTW, Sill M, Hovestadt V, Schrimpf D, Sturm D. DNA methylation-based classification of central nervous system tumours. Nature. 2018. 555: 469-74

9. Chakraborti S, Mahadevan A, Govindan A, Bhateja A, Dwarakanath S, Aravinda HR. Rosette-forming glioneuronal tumor--evidence of stem cell origin with biphenotypic differentiation. Virchows Arch. 2012. 461: 581-8

10. Daoud EV, Rajaram V, Cai C, Oberle RJ, Martin GR, Raisanen JM. Adult brainstem gliomas with H3K27M mutation: Radiology, pathology, and prognosis. J Neuropathol Exp Neurol. 2018. 77: 302-11

11. Duan L, Zhang Y, Fu W, Geng S. Rosette-forming glioneuronal tumor originating from the spinal cord: Report of 2 cases and literature review. World Neurosurg. 2017. 98: 875.e1-7

12. Engelhardt S, Behling F, Beschorner R, Eckert F, Kohlhof P, Tatagiba M. Frequent FGFR1 hotspot alterations in driver-unknown low-grade glioma and mixed neuronal-glial tumors. J Cancer Res Clin Oncol. 2022. 148: 857-66

13. Gessi M, Moneim YA, Hammes J, Goschzik T, Scholz M, Denkhaus D. FGFR1 mutations in Rosette-forming glioneuronal tumors of the fourth ventricle. J Neuropathol Exp Neurol. 2014. 73: 580-4

14. Ghosal N, Furtado SV, Hegde AS. Rosette forming glioneuronal tumor pineal gland and tectum: An intraoperative diagnosis on smear preparation. Diagn Cytopathol. 2010. 38: 590-3

15. Hamauchi S, Tanino M, Hida K, Sasamori T, Yano S, Tanaka S. Spinal rosette-forming glioneuronal tumor: A case report. Medicine (Baltimore). 2019. 98: e18271

16. Haryu S, Saito R, Kanamori M, Sonoda Y, Kumabe T, Watanabe M. Rosette-forming glioneuronal tumor: Rare case presented with spontaneous disappearance of contrast enhancement. NMC Case Rep J. 2015. 2: 65-7

17. Hsu C, Kwan G, Lau Q, Bhuta S. Rosette-forming glioneuronal tumour: Imaging features, histopathological correlation and a comprehensive review of literature. Br J Neurosurg. 2012. 26: 668-73

18. Jayapalan RR, Mun KS, Wong KT, Sia SF. Malignant transformation of a rosette-forming glioneuronal tumor with IDH1 mutation: A case report and literature review. World Neurosurg X. 2019. 2: 100006

19. Kikuchi Z, Shibahara I, Yamaki T, Yoshioka E, Shofuda T, Ohe R. TERT promoter mutation associated with multifocal phenotype and poor prognosis in patients with IDH wild-type glioblastoma. Neurooncol Adv. 2020. 2: vdaa114

20. Kitamura Y, Komori T, Shibuya M, Ohara K, Saito Y, Hayashi S. Comprehensive genetic characterization of rosette-forming glioneuronal tumors: Independent component analysis by tissue microdissection. Brain Pathol. 2018. 28: 87-93

21. Komori T, Scheithauer BW, Hirose T. A rosette-forming glioneuronal tumor of the fourth ventricle: Infratentorial form of dysembryoplastic neuroepithelial tumor?. Am J Surg Pathol. 2002. 26: 582-91

22. Lin CC, Mansukhani MM, Bruce JN, Canoll P, Zanazzi G. Rosette-forming glioneuronal tumor in the pineal region: A series of 6 cases and literature review. J Neuropathol Exp Neurol. 2021. 80: 933-43

23. Liu AP, Harreld JH, Jacola LM, Gero M, Acharya S, Ghazwani Y. Tectal glioma as a distinct diagnostic entity: A comprehensive clinical, imaging, histologic and molecular analysis. Acta Neuropathol Commun. 2018. 6: 101

24. Louis DN, Giannini C, Capper D, Paulus W, Figarella-Branger D, Lopes MB. cIMPACT-NOW update 2: Diagnostic clarifications for diffuse midline glioma, H3 K27M-mutant and diffuse astrocytoma/anaplastic astrocytoma, IDH-mutant. Acta Neuropathol. 2018. 135: 639-42

25. Lucas CG, Gupta R, Doo P, Lee JC, Cadwell CR, Ramani B. Comprehensive analysis of diverse low-grade neuroepithelial tumors with FGFR1 alterations reveals a distinct molecular signature of rosette-forming glioneuronal tumor. Acta Neuropathol Commun. 2020. 8: 151

26. Matsumura N, Wang Y, Nakazato Y. Coexpression of glial and neuronal markers in the neurocytic rosettes of rosette-forming glioneuronal tumors. Brain Tumor Pathol. 2014. 31: 17-22

27. Shibayama C, Doai M, Matoba M, Morikawa M, Sato H, Okada N. Spinal rosette-forming glioneuronal tumor: First case in a young child. Radiol Case Rep. 2021. 16: 3982-6

28. Sieg EP, Payne R, Langan S, Specht CS. Case report: A rosette-forming glioneuronal tumor in the tectal plate in a patient with neurofibromatosis Type I. Cureus. 2016. 8: e857

29. Sievers P, Appay R, Schrimpf D, Stichel D, Reuss DE, Wefers AK. Rosette-forming glioneuronal tumors share a distinct DNA methylation profile and mutations in FGFR1, with recurrent co-mutation of PIK3CA and NF1. Acta Neuropathol. 2019. 138: 497-504

30. Solis OE, Mehta RI, Lai A, Mehta RI, Farchoukh LO, Green RM. Rosette-forming glioneuronal tumor: a pineal region case with IDH1 and IDH2 mutation analyses and literature review of 43 cases. J Neurooncol. 2011. 102: 477-84

31. Yang C, Fang J, Li G, Li S, Ha T, Wang J. Histopathological, molecular, clinical and radiological characterization of rosette-forming glioneuronal tumor in the central nervous system. Oncotarget. 2017. 8: 109175-90