Abstract
Background/Aim: Immature teratomas (IMT) are malignant germ cell tumours composed of immature embryonal tissue, mostly neuroectodermal tubules and rosettes. Meanwhile, embryonal tumours with multilayered rosettes (ETMR) are aggressive central nervous system tumours composed of neurocyte proliferation with rosette formation. The histopathological appearance of rosette formation in ETMR is the same as that in IMT. Recently, 19q13.42 amplification was reported as a specific genetic marker of ETMR. The aim of this study was to compare ETMR with IMT from histological, immunohistochemical and genetic perspectives. Materials and Methods: We retrospectively analysed tumour samples from 48 patients with IMT and 1 patient with ETMR. We performed fluorescence in situ hybridization (FISH) analysis, which revealed amplification of the 19q13.42 locus in the ETMR case. In addition, immunohistochemical analyses of LIN28A, β-catenin and p53 were performed. Results: In FISH analysis all 48 cases of IMT showed diploidy. By immunohistochemical analysis, LIN28A expression was observed in 54% of IMT cases (25/48 cases) and in the ETMR case. Nuclear staining of β-catenin was observed in 33% of IMT cases (16/48 cases). Meanwhile, aberrant expression of p53 was not identified in IMT nor ETMR cases. Conclusion: Genetic changes associated with IMT differ from those in ETMR, but LIN28A protein immunohistochemical expression, which is specific for ETMR, can be a biomarker for the immature neuroepithelial component in IMT.
Immature teratoma (IMT) is a malignant germ cell tumour known as a rare and rapidly progressing neoplasm. It is composed of mature and immature embryonal tissues from the tridermic component, mostly characterized by neuroectodermal tubules and rosettes. These rosettes are composed of mitotically active hyperchromatic cells with neuroepithelial differentiation. Notably, ovarian IMT frequently develops in younger women and its grading is based on the amount of immature neuroepithelium (1).
ETMR is an extremely rare and highly aggressive tumour of the central nervous system, which almost only affects the paediatric population and has a fatal outcome (2). It is composed of rosettes with neurocytes in a neuropil-like background. These rosettes contain mitotically active cells. The histopathological appearance of rosette formation in ETMR is the same as that in IMT.
Recently, genetic findings have begun to be applied to diagnose ETMR. 19q13.42 amplification was reported as a specific genetic marker of ETMR associated with tumorigenesis. Therefore, ETMR is regarded as a specific tumour subcategory of embryonal brain tumour in the current WHO classification (3). In addition, LIN28A, a known testicular germ cell tumour marker, was reported to be a highly specific diagnostic immunohistochemical biomarker for ETMR (4).
Despite the above findings, 19q13.42 amplification has never been explored in IMT through a large systematic investigation. Few genetic findings for IMT have been reported, although carcinogenic mutation of the TP53 gene was identified in IMT in small studies (5, 6). It was also reported that p53 immunostaining could be used as a prognostic indicator in CNS PNET (7). LIN28A, which regulates germ cell development, has been shown by immunohistochemical analysis to be expressed in ovarian germ cell tumours (8), although its specificity in IMT is unknown.
In this study, we reviewed cases of ETMR and ovarian IMT from histological, immunohistochemical and genetic perspectives in order to reveal the similarity of the neuroectodermal rosettes in these tumours.
Materials and Methods
This study was conducted in accordance with the principles embodied in the Declaration of Helsinki and was also approved by the Ethics Committee of the Kyushu University (No. 2020-530). We reviewed a series of 49 patients with histologically confirmed IMT (48 cases) and ETMR (1 case) at the Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan, from 1976 to 2018. All tumour samples were formalin-fixed and paraffin-embedded (FFPE). The histological grade of IMT was evaluated according to the grading system of the WHO Classification of Female Genital Tumours (1).
Histopathological analysis. Histopathologically, the following features of IMT and ETMR were evaluated: immature neuroectodermal components, other immature components, mature components in teratoma, rosette formation, cellular shape, nuclear atypia, and mitotic figures.
Immunohistochemistry. Sections were subjected to immunohistochemical staining of LIN28A, β-catenin, the critical downstream effector of the Wnt/β-catenin pathway, and p53 protein using the following primary antibodies: anti-LIN28A (rabbit polyclonal, clone A177, dilution 1:800; Cell Signaling Technology, Danvers, MA, USA), anti-β-catenin (mouse monoclonal, clone 14/β-catenin, dilution 1:200; BD Transduction Laboratories, Lexington, KY, USA) and anti-p53 (mouse monoclonal, clone DO-7, dilution 1:100; Novocastra, Newcastle, UK). IHC was performed using a DAKO EnVision Detection System. After immunohistochemical staining, two pathologists (NM, YY) who were blinded to the patient details evaluated the stained sections. Immunohistochemical evaluation was performed in the immature neuroectodermal area containing developing rosettes. LIN28A was evaluated using a categorical scoring system as follows (the percentage of tumour cells stained): score 0 (no staining), score 1 (<31%), score 2 (31%-60%), score 3 (61%-90%) and score 4 (>90%) (8). Cells with strong cytoplasmic immunostaining were considered to be positively stained. For the evaluation of β-catenin immunoreactivity, nuclear staining was scored as positive or negative and cytoplasmic staining was scored as present or absent (9-11). The expression of p53 was defined as aberrant, suggestive of TP53 mutation, when >80% of cells exhibited strong nuclear p53 staining or there was a complete absence of such nuclear staining (5, 12, 13).
Fluorescence in situ hybridization. To assay the amplification of genes at the 19q13.42 locus, dual-colour FISH was performed using commercially available FISH probes, in accordance with the manufacturer’s protocol (GSP Laboratories, Kawasaki, Japan). Gene amplification in the immature neuroectodermal component of ETMR was defined in accordance with a previous report (14).
Results
Clinicopathological findings. The clinical features of IMT are summarized in Table I. The age of the patients with IMT ranged from 0 to 40 years old (mean 17.9, median 20), while females were predominantly affected (male:female=9:39). Tumours were located in the ovary in 34 cases (71%), testis in 2 cases (4%) and extragonadally in 12 cases (25%), including 2 cases in the sacrococcygeal region, 4 cases in the retroperitoneum, 4 cases in the brain (1 case in the suprasellar region, 3 cases in the pineal region), 1 case in the stomach and 1 case in the neck. Clinical follow-up data were available for 39 cases (81%). The mean duration of follow-up was 95.8 months (1-276 months). Overall, 4 cases (10%) had recurrence, 3 cases (8%) showed peritoneal dissemination and 4 cases (10%) died of disease. A total of 36 cases (75%) were treated with surgical resection (for 11 cases, samples were not available, while for 1 case an autopsy sample was used). Eleven cases (23%) received adjuvant chemotherapy, which comprised bleomycin, etoposide, and cisplatin (BEP) or vincristine, actinomycin D and cyclophosphamide (VAC), and one case received adjuvant chemoradiotherapy involving etoposide and carboplatin.
Clinical and histological data from 48 patients with immature teratoma.
Histopathological findings. Representative histological figures are shown in Figure 1. ETMR was composed of highly proliferating undifferentiated neurocytes accompanied by true rosettes and the formation of perivascular pseudorosettes arranged in a neuropil-like background (Figure 1a). Mitotic figures and apoptosis were frequently seen.
Microscopic appearance of the embryonal tumour with multilayered rosettes and immature teratoma. (a) Embryonal tumour with multilayered rosettes contains neurocytes with numerous rosettes embedded within abundant neuropils (HE; bar: 50 μm). (b) Mixed germ cell tumour with yolk sac tumour (HE; bar: 100 μm). (c) Mixed germ cell tumour with embryonal carcinoma (HE; bar: 100 μm). (d) Mixed germ cell tumour with seminoma (HE; bar: 500 μm); inset component of seminoma (bar: 50 μm). (e) Mixed germ cell tumour with germinoma (HE; bar: 500 μm); inset component of germinoma (bar: 50 μm). (f) Immature teratoma contains immature neuroectodermal tissue with rosette formation (HE; bar: 100 μm).
IMT was found either in pure form or as a component of a mixed germ cell tumour. Overall, 40 cases involved pure IMT and 8 involved mixed germ cell tumours, including one or more of the following: yolk sac tumours in 4 cases (Figure 1b), embryonal carcinomas in 3 cases (Figure 1c), seminoma in 1 case (Figure 1d) and germinoma in 1 case (Figure 1e). IMT consisted of tissue derived from the three germ layers and contained varying amounts of immature tissue, most frequently with neural differentiation. Mature elements were also present in all cases. Immature neuroectodermal tissue was characterized by proliferation of neuroblastic cells with mitotic figures accompanied by frequent neuroectodermal rosettes, neural tubes, and neurofibrillary matrix (Figure 1f).
Immunohistochemistry. The detailed results of immunohistochemical staining are shown in Table II, while representative images of immunohistochemical staining are presented in Figure 2. LIN28A expression was observed in the immature neuroectodermal component in 54% of IMT cases (25/48 cases) (Figure 2a) and in the ETMR case (Figure 2b), while LIN28A expression was absent in the mature component. LIN28A was partially positive at various rates in ETMR and IMT. β-catenin was expressed predominantly in the cytoplasm of immature neuroectodermal tumour cells in 67% of IMT cases (32/48 cases) (Figure 2c) and in the ETMR case (Figure 2d). Nuclear staining was observed in 33% of IMT cases (16/48 cases) (Figure 2e). p53 immunostaining was very focally or weakly positive in immature neuroectodermal tumour cells in all 48 cases of IMT (Figure 2f) and in the ETMR case (Figure 2g).
Results of immunohistochemical staining of the 48 patients with immature teratoma.
Immunohistochemical patterns of LIN28A, β-catenin and p53. (a) Strong and partial expression of LIN28A in rosette-forming cells in IMT (bar: 100 μm). (b) Strong and partial expression of LIN28A in rosette-forming cells in ETMR (bar: 100 μm). (c) Cytoplasmic staining of rosette-forming cells for β-catenin in IMT (bar: 50 μm). (d) Cytoplasmic staining of rosette-forming cells for β-catenin in ETMR (bar: 50 μm). (e) Nuclear staining of rosette-forming cells for β-catenin in IMT (bar: 50 μm). (f) Wild-type p53 immunostaining in IMT (bar: 50 μm). (g) Wild-type p53 immunostaining in ETMR (bar: 50 μm).
Fluorescence in situ hybridization (FISH). Representative figures of 19q13.42 FISH are shown in Figure 3. Amplification of the 19q13.42 locus was observed in the immature neuroectodermal component of ETMR (Figure 3a); however, in all 48 cases, the immature neuroectodermal component of IMT showed diploidy (Figure 3b).
Fluorescence in situ hybridization (FISH) of chromosome 19q13.42 amplification. Dual-colour FISH was used for 19q13.42 probes labelled in red. The reference probe corresponding to the 19p13.11 locus is labelled in green. (a) Fluorescence in situ hybridization confirms high chromosome 19q13.42 amplification in embryonal tumour with multilayered rosettes. (b) Neuroectodermal component of immature teratoma did not show this amplification pattern.
Discussion
ETMR is an aggressive central nervous system primitive neuroectodermal tumour variant. It occurs primarily in infants and young children with typical survival of <1 year after diagnosis, despite intensive therapy (15, 16). Previous studies showed amplification of the miRNA cluster C19MC on chromosome 19q13.42 at a high frequency in ETMR using FISH (14, 17). C19MC exhibits restricted expression in undifferentiated/germinal tissues such as placenta and testis in non-neoplastic tissue (18) and hESCs (19, 20), indicating its function in cellular differentiation (21). Fusion of TTYH1 with C19MC leads to extreme overexpression of the miRNA cluster, which suggests that C19MC drives oncogenesis in part by facilitating the maintenance and transformation of a very early neural component (22). It was reported that overexpression of the 19q13.41 miRNA cluster modulated cell survival and enhanced the growth of untransformed human neural stem cells (hNSCs) in part by up-regulating WNT pathway signalling and restricting hNSC differentiation (21). A previous study using β-catenin IHC showed that the WNT/β-catenin pathway plays an important role in the pathogenesis of CNS PNETs (10). However, the relationship between 19q13.42 amplification and IMT has not been identified. In this study, the authors demonstrated that the neuroepithelium of rosettes in IMT harboured no definite 19q13.42 amplification in terms of copy number. To the best of our knowledge, the current investigation is the first to confirm the negativity of 19q13.42 amplification in a large series of IMT cases, although this was shown in a small number of cases in a previous study (17). It was also suggested that the genetic backgrounds of IMT and ETMR may differ despite their similarity in terms of histological features and that the 19q13.42 FISH method may provide critical information to distinguish ETMR from other cerebral tumours with neuroepithelial rosettes.
Ovarian IMT is the second most common malignant ovarian germ cell tumour (23). It has a favourable prognosis because treatment with surgery followed by systemic chemotherapy can achieve remission and/or complete cure in over 90% of cases (24). Survival rates are excellent across all stages, but older age at diagnosis, advanced stage and high-grade histology are associated with increased mortality (23). IMT can also arise in the central nervous system, with a higher frequency in children (25). In addition, although rare, brain metastasis of ovarian IMT was previously reported (26). IMT is generally treated surgically, while adjuvant chemotherapy is added to cases with a higher grade and stage (27). The current investigation also presented favourable survival rate and response to chemotherapy. However, the optimal treatment approach for ETMR has not been defined, due to the rarity of these tumours. Based on the treatment of other embryonal tumours, maximal resection, and age- and risk-adapted radiotherapy and chemotherapy are generally used (28). The clinical behaviour and treatment of ETMR are completely different from those of IMT.
In the current study, it was demonstrated that LIN28A was expressed in the immature neuroectodermal component of both ETMR and IMT. Moreover, the mature component of IMT presented no immunohistochemical positivity for LIN28A. LIN28A is an RNA-binding protein that regulates germ cell development, skeletal myogenesis and neurogenesis (29). LIN28A is widely expressed in embryonic stem cells and in early embryogenesis, but its expression is usually down-regulated in differentiated tissue. Considering the above, it is suggested that LIN28A may be a useful marker of the immature neuroepithelial component. When it is difficult to identify the immature neuroepithelial component of IMT, LIN28A may work as a biomarker for it.
In this study, LIN28A expression was observed in 54% of IMT cases. LIN28A functions as a negative regulator of the let-7 family of microRNAs, which may act as tumour suppressors. LIN28A has been implicated in stem cell pluripotency and metabolism (4). ETMR was also found to have high LIN28A expression and low let-7 miRNA expression and showed mTOR pathway activation. It is suggested that LIN28A may be not only a diagnostic marker but also a regulator of genes involved in growth and metabolism (30). Previous studies suggested that LIN28A is a potential therapeutic target for ETMR (31), ovarian cancer (32) and diabetes (33). Based on the present study, the authors considered that LIN28A may also have therapeutic potential for IMT resistant to the conventional therapeutic options.
β-catenin nuclear staining was observed in 33% of IMT cases but was negative in the ETMR case. Aberrant p53 immunostaining, suggestive of TP53 mutation, was not detected in either IMT or ETMR. It is suggested that immunostaining of β-catenin and p53 is not useful for detecting the immature neuroepithelial component of IMT and ETMR.
In conclusion, the genetic changes associated with IMT differ from those of ETMR, but LIN28A protein immunohistochemical expression, which is specific for ETMR, may be a biomarker for the immature neuroepithelial component in IMT.
Acknowledgements
We appreciate the technical assistance of staff at The Research Support Center, Kyushu University Graduate School of Medical Sciences. This study was supported by a JSPS KAKEN Grant (No. 18K06990). The Authors thank the National Hospital Organization Kyushu Cancer Center (Fukuoka, Japan) that kindly submitted the cases and provided clinical follow-up information when available.
Footnotes
Authors’ Contributions
Naomi Magarifuchi and Yuichi Yamada performed the research and wrote the article. Yoshihiro Oishi contributed to the research design and slide review. Kiyoko Kato and Kenichi Taguchi contributed to the sample collection and research design. Yoshinao Oda designed the research and gave final approval of the article. All Authors critically reviewed and approved the article.
Conflicts of Interest
The Authors declare that there are no potential conflicts of interest.
- Received June 1, 2022.
- Revision received June 29, 2022.
- Accepted July 8, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.