Germline loss-of-function variant in the E3 ubiquitin ligase TRAF2 in a young adult patient with medulloblastoma: a case report.

Introduction

Medulloblastoma (MB) is an embryonal neoplasm of the cerebellum that primarily affects children, adolescents, and young adults, with an incidence rate of 5 cases per 1 million individuals [ 1 ]. Although the overall survival rate in many patients with localized disease and/or favorable molecular subgroups is high, the quality of life for patients with MB, in general, remains suboptimal [ 1 , 2 ]. The long-term toxicity burden from current therapies underlines the urgent need for a deeper understanding of the biology of this malignancy to enable the development of more effective, less toxic therapies. Recent advancements in DNA methylation-based tumor profiling have unveiled the remarkable molecular heterogeneity of MB, identifying four distinct molecular subgroups: WNT (MB WNT ), Sonic Hedgehog (MB SHH ), Group 3, and Group 4 [ 1 , 3 ]. These subgroups are characterized by unique genomic alterations and clinicopathological features, underscoring the importance of subgroup-specific therapeutic approaches [ 1 , 3 ]. Elucidating the intricate molecular landscapes of each MB subgroup is imperative to identify novel therapeutic targets and optimize risk-stratified therapeutic strategies.

While MB is extensively documented within the pediatric population, there remains a significant gap in knowledge regarding the condition in adolescents and young adults (AYA) (ages 15–39) and adults (ages 40 and above). AYA/adults account for 10–30% of MB and most patients are diagnosed with MB SHH , MB WNT, or Group 4 [ 1 , 4 , 5 ]. Treatment approaches for AYA/adult patients with MB are largely extrapolated from pediatric protocols, which may not sufficiently address the unique characteristics of the disease in older patients [ 6 ]. Consequently, elucidating the distinct features of MB in AYA and adults is crucial for developing tailored treatment strategies and improving clinical outcomes.

Genetic tumor syndromes such as Gorlin syndrome ( SUFU , PTCH1 ), Li-Fraumeni syndrome ( TP53 ), Familial Adenomatous Polyposis ( APC ), and Fanconi anemia ( BRCA2, PALB2 ) have been identified as genetic risk factors for MB in children, adolescents, and young adults [ 7 ]. However, most genetic studies in MB have primarily focused on pediatric patients, leaving the role of genetic tumor syndromes in AYA and adult patients with MB largely unexplored. The discovery of novel germline variants associated with AYA/adult MB would have significant implications for genetic counseling and testing, potentially further improving clinical management for patients with MB.

In this study, we present a case of a young adult patient with MB associated with a germline loss-of-function (LOF) variant in the TRAF2 gene. Furthermore, we conducted a molecular comparison of this patient with our in-house cohort of 19 MB cases, highlighting the unique molecular characteristics of the TRAF2 -associated MB. To our knowledge, this is the first documented instance of a young adult patient with medulloblastoma harboring a germline TRAF2 variant. This case underscores the heterogeneity of the disease and the necessity for a deeper exploration of genetic tumor syndromes in the AYA and adult MB population.

Case presentation

A 21-year-old woman with no significant past medical history presented to her primary care provider with a 3-week history of worsening headaches accompanied by nausea. She also noted difficulty with walking and felt unsteady. She was referred to a local emergency department, where a computed tomography (CT) scan, and brain magnetic resonance imaging (MRI) revealed a 4.5 cm contrast-enhancing mass in the posterior fossa, which caused obstructive hydrocephalus with brainstem compression, tonsillar herniation, and moderate dilation of the third and lateral ventricles (Fig. 1 A). An MRI of the spine and cerebrospinal fluid analysis showed no evidence of metastatic disease. Her family history was not significant for malignancy. She underwent a gross total resection (GTR) of the mass via occipital craniectomy complicated by severe posterior fossa syndrome, including mutism, emotional lability, and ataxia. Histological examination of the tumor showed a small round blue cell tumor with brisk mitotic activity. The tumor was positive for synaptophysin, exhibited membranous but not nuclear immunoreactivity for β-catenin, displayed combined nuclear and cytoplasmic positivity for YAP1, was patchy positive for p75-NGFR, and negative for GAB1. The Ki-67 proliferation index was elevated (Fig. 1 B ) . These findings supported the diagnosis of medulloblastoma, classic histology, WHO grade 4. The overall risk classification was deemed "average risk" [ 8 ].

Next-generation sequencing, including mRNA sequencing and DNA sequencing, was conducted on a formalin-fixed, paraffin-embedded (FFPE) tissue block and matched peripheral blood (Materials and methods). In silico estimation revealed the tumor content to be very high (97%). Somatic mutations are presented in Table 1 . We further conducted subsequent analyses of the patient’s MB genome in conjunction with genomes from 19 other MB patients at our institution. At first glance, there were some genomic features suggestive of a Sonic hedgehog (SHH)-activated MB (Fig. 2 A). The tumor harbored a copy-neutral loss-of-heterozygosity (LOH) of 9q (Fig. 2 B), typically seen in the MB SHH subgroup and most frequent in the MB SHH-α and MB SHH-δ subtypes [ 9 ]. In addition, the presence of a somatic MAX hotspot mutation (R60Q) (Table 1 ) was consistent with MB SHH , as MAX hotspot mutations have previously been observed exclusively in the MB SHH subgroup (in two adults and one child) [ 10 ].

DNA methylation-based tumor classification of the index tumor was performed using two brain tumor classifiers (Materials and methods). The Heidelberg-DKFZ brain tumor classifier revealed a calibrated score of 0.837 that the tumor belongs MB SHH class, falling short of the 0.90 score required for definitive MB subgroup classification. The NCI (Bethesda) brain tumor classifier identified the tumor as a neuroblastic embryonal tumor with a high score of 0.992. However, it was unable to definitively determine the molecular subtype. The classifier's best match was MB SHH-δ , though this was associated with a low score of 0.253. Moreover, the tumor showed no significant mRNA expression of the curated markers associated with the four MB subgroups (Materials and methods, Fig. 2 A). Principal component analysis of the tumor's global gene expression profile was inconclusive based on the first two principal components (Fig. 2 C). Together, these findings suggested a medulloblastoma with atypical molecular features.

Other notable genomic findings included a heterozygous somatic MYC hotspot mutation (P74L) located within the MYC Box 1 phospho-degron, which enhances the stability of MYC by affecting protein degradation [ 11 ]. The somatic MAX hotspot mutation (R60Q) was homozygous via copy-neutral LOH of chromosome 14 (Table 1 , Fig. 2 B). Given that MYC and MAX form a heterodimer [ 12 ], double hits in MYC and MAX likely contributed to dysregulation of MYC signaling. In our patient, MYC expression was significantly elevated, ranking second among 20 samples in the MB cohort with exceeding levels relative to MB Group 3, where MYC overexpression is a known driver [ 9 , 10 ]. Although no focal MYC amplification was observed, this finding could imply a possible structural rearrangement that affected MYC enhancer that may not have been detected by targeted panel sequencing. Gene set enrichment analysis further demonstrated that several upregulated genes in this tumor, relative to the rest of the MB cohort, were enriched in pathways related to RNA processing and translation (Fig. 2 D). This pattern aligns with the known role of MYC in enhancing RNA processing and translation [ 13 ], suggesting that MYC signaling contributed to MB development in our patient.

Partial copy-neutral LOH of 9q is a less frequent cytogenetic event in MB SHH (Fig. 2 B). The LOH region did not encompass PTCH1 and ELP1 , two driver genes commonly affected by 9q LOH events in MB SHH [ 10 , 14 ]. A detailed analysis of all protein-coding genes within the 9q34.3 LOH region revealed a rare germline loss-of-function (LOF) variant in TRAF2 (p.Arg403Ter; c.1207C > T). No other rare germline LOF variants were identified in the 9q LOH region, apart from a rare, clinically benign germline missense variant in TSC1 (p.S334L; ClinVar VCV000041686). We further confirmed that the TRAF2 LOF variant is heterozygous in the patient’s germline genome (VAF 45%) and homozygous in the MB genome (VAF 99%) (Table 1 ). Further familial genetic testing was not possible; hence, we could not ascertain parental inheritance. We were, however, able to determine that TRAF2 p.Arg403Ter is exceedingly rare in humans with a germline carrier frequency of 1 in 806,922 (gnomAD database v4.1.0).

Tumor necrosis factor (TNF) receptor-associated factor 2 (TRAF2) is a ubiquitin ligase integral to various signaling pathways in innate immunity, inflammation, or apoptosis (Fig. 3 A) [ 15 , 16 ]. In pro-survival pathways, TRAF2 functions as a positive regulator of TNF-mediated signaling. Upon TNF binding to TNF receptor 1 (TNFR1), TRAF2 is recruited to TNFR1 by interacting with the adaptor protein TRADD and Receptor Interacting Protein 1 (RIPK1). Subsequently, TRAF2 recruits Cellular Inhibitor of Apoptosis 1 and 2 proteins (cIAP1/2, also known as BIRC2/3), which facilitates the polyubiquitination of RIPK1 and other components of the signaling complex. Ubiquitinated RIPK1 interacts with TGFβ-activated kinase 1 (TAK1) and other proteins to activate the canonical nuclear factor-κB (NF-κB) signaling pathway (Fig. 3 A), resulting in the expression of various genes involved in proliferation and inflammation [ 15 , 16 ]. Additionally, TRAF2 participates in TNF-mediated activation of the c-Jun N-terminal kinase (JNK) signaling pathway through RIPK1 and TAK1 (Fig. 3 A) [ 16 – 18 ]. The JNK signaling pathway can induce Activator Protein 1 (AP-1) dependent gene expression. This signaling cascade can also lead to apoptosis, although the precise mechanisms remain unknown [ 17 ].

Like other members of the TRAF family, TRAF2 possesses an N-terminal domain characterized by a RING structure and four zinc fingers [ 19 ]. The C-terminal TRAF domain is subdivided into the TRAF-N and TRAF-C subdomains. The TRAF-N subdomain features a coiled-coil structure that facilitates both homo- and hetero-oligomerization. The TRAF-C subdomain consists of antiparallel β-strands and interacts with receptor and adaptor proteins, including BIRC2 [ 20 ]. The stopgain variant identified in this patient significantly truncates the TRAF-C subdomain, thus potentially affecting its capacity to bind with other crucial signaling proteins (Fig. 3 B).

However, it has been observed that the TRAF2 protein with a partially truncated C-terminal domain may still retain some residual binding activity [ 21 ]. In a functional study, various mutants with progressively truncated TRAF-C domains were engineered and overexpressed in HEK293 cells [ 21 ]. These synthetic deletions demonstrated the ability to activate the downstream signaling pathway to varying degrees (Fig. 3 B). One of their constructs truncated TRAF2 at codon 400, closely resembling the truncation at codon 403 due to the germline LOF variant described in our index patient. Truncation of TRAF2 at codon 400 exhibited a milder activation of downstream signaling compared to the construct that completely eliminated the TRAF-C domain (Fig. 3 B). Strikingly, our patient further harbors a heterozygous somatic LOF mutation in the E3 ligase BIRC2 (Table 1 ). In this context, the somatic LOF BIRC2 mutation converges on the same pathway, further diminishing the residual binding efficacy of TRAF2, thereby potentially leading to additional inhibition of apoptosis (Fig. 3 A).

Transcriptome provided deeper insights into the oncogenic context of the signaling pathway involving TRAF2 and BIRC2 . Our patient exhibited significantly higher expression levels of pro-apoptotic genes, such as Fas receptor ( FAS ) and harakiri ( HRK ) [ 22 , 23 ], compared to other tumors in our MB cohort (Fig. 3 C). The tumor also demonstrated elevated expression level of chemokine (C-X-C motif) ligand 1 ( CXCL1 ), which is linked to inflammation [ 24 ]. These findings lead us to hypothesize that an inflammation-like phenotype could have contributed to tumorigenesis. Subsequently, biallelic loss of TRAF2 and monoallelic loss of BIRC2 could have collaborated in evading apoptosis and thus triggering MB development.

Discussion and conclusion

Despite the advancements facilitated by next-generation sequencing in elucidating the genomic diversity of medulloblastoma, our understanding of this complex disease remains incomplete. MB in AYA and adult patients introduces additional complexity due to its unique genomic and epigenomic features, which differ from pediatric cases. Group 3 medulloblastoma is rarely observed in adults. Approximately 60% of adult medulloblastoma cases are SHH-activated, 25% belong to Group 4, and 15% are WNT-activated [ 4 , 5 ]. MB SHH is further divided into four subtypes with distinct molecular profiles and clinical outcomes. Among them, MB SHH-δ is the most common in young adults and adults, with a median age of onset of 26 years, and have better prognosis [ 4 , 9 ].

Our 21-year-old patient with MB presents a challenge to the current central nervous system (CNS) tumor classification system. Immunohistochemistry revealed that the tumor lacked nuclear β-catenin, but showed both nuclear and cytoplasmic YAP1 positivity, as well as N75-NGFR, supporting its classification as MB SHH [ 25 ]. However, DNA methylation classification and gene expression clustering results remained inconclusive. Additionally, the tumor lacked somatic mutations in PTCH1 , SUFU , SMO , and the TERT promoter, which are typically seen in MB SHH [ 1 , 9 , 10 ]. Also, somatic and germline TP53 mutations are generally found in MB SHH patients aged 5–18 years [ 26 ]. Furthermore, MYC overexpression is more commonly associated with MB Group 3 rather than MB SHH [ 1 , 9 , 10 ].

MB has been linked to various genetic tumor syndromes. Gorlin syndrome, which arises from germline variants in PTCH1 and SUFU , is associated with an increased risk of the desmoplastic/nodular subtype of medulloblastoma. Li-Fraumeni syndrome, characterized by germline variants in TP53 , increases susceptibility to medulloblastoma and a range of other malignancies. APC -associated polyposis increases the risk of medulloblastoma, particularly the WNT subgroup [ 7 , 27 ]. Recently, pathogenic germline variants in BRCA2 , PALB2 , GPR161 , and ELP1 have also been implicated in medulloblastoma malignancy [ 7 , 14 , 28 ]. It has been estimated that 40% of pediatric patients with MB SHH harbor a genetic tumor syndrome [ 14 ]. The case presented in our study suggests that TRAF2 -associated medulloblastoma is a novel genetic tumor syndrome.

Somatic mutations in TRAF2 are observed in less than 6% of human cancers [ 29 ]. Somatic loss-of-function mutations and deletions in TRAF2 have been identified as drivers in hematological malignancies, including mantle cell lymphoma, diffuse large B-cell lymphoma, and multiple myeloma [ 30 , 31 ]. This highlights the critical role of TRAF2 as a negative regulator of the non-canonical NF-κB pathway in lymphoid cells [ 30 , 31 ]. However, the role of TRAF2 in solid tumors remains largely unexplored, further complicated by its dual function as both an anti-apoptotic and pro-apoptotic factor within a complex cellular signaling network [ 15 – 18 ]. In the context of CNS tumors, another member of the TRAF family and E3 ubiquitin ligase, TRAF7 , has been identified as a somatic driver in meningioma [ 32 ].

In addition to the germline TRAF2 loss-of-function variant and partial copy-neutral LOH of 9q, this patient’s tumor also acquired a somatic LOF mutation in BIRC2 , another E3 ubiquitin ligase that interacts with TRAF2. Similar to TRAF2, BIRC2 has been identified as a tumor suppressor in hematological malignancies [ 31 ], while it may function as an oncogene in other tumors through gene amplification [ 33 ]. BIRC2 and TRAF2 potentially act as tumor suppressors within a signaling axis that typically induces cell death, such as through JNK-induced apoptosis or TNF-stimulated necrosis [ 17 , 18 ]. Notably, somatic hotspot mutations in another E3 ligase, KBTBD4 , have been characterized in MB [ 34 ]. Our findings could have significant implications for understanding the broader role of dysregulated E3 ubiquitin ligases in medulloblastoma.

In conclusion, our case report suggests that germline loss-of-function TRAF2 alterations may represent a novel risk factor for medulloblastoma. Further studies in adolescent, young adult, and adult MB populations are warranted to determine their prevalence, full spectrum of co-occurring somatic driver alterations, and therapeutic and prognostic implications.

Treatment outcomes

The patient received craniospinal irradiation (CSI) treatment in accordance with standard-of-care protocols. She was administered a dose of 23.4 Gy to the entire neuraxis and an additional 30.6 Gy boost to the tumor bed in 30 fractions, accompanied by weekly intravenous vincristine. Furthermore, the patient underwent chemotherapy as per the Children's Oncology Group protocol ACNS0331. This regimen includes cisplatin, CCNU, vincristine, and cyclophosphamide, administered as maintenance cycles over a total of seven cycles. Maintenance chemotherapy was discontinued prematurely for this patient due to profound and prolonged cytopenias. During the course of therapy, the patient developed bilateral high-frequency hearing loss at 8000 and 12,000 Hz, necessitating a reduction in the cisplatin dose. The patient has been disease-free for 5 years and 4 months following surgery. Annual MRI brain and spine scans will continue for surveillance purposes. The patient and her family have been informed about the rarity of the germline variant and its potential association with hematological malignancies.

Materials and methods

Clinical sequencing was conducted following standard protocols in our Clinical Laboratory Improvement Amendments (CLIA)-certified sequencing laboratory [ 35 ]. For each patient, we performed targeted sequencing on DNA from the tumor and matched normal peripheral blood, using a panel covering 1,714 genes, including probes for the TERT promoter region. Tumor RNA sequencing was performed with an exome-capture transcriptome protocol developed in our lab, covering all coding genes. Bioinformatic analyses—including mutation calling, copy number analyses, and gene expression profiling—were carried out as previously described [ 35 ].

Gene expression signatures were derived from a published single-cell RNA sequencing (scRNA-seq) study, which performed UMAP clustering on 39,946 single cells from 28 MB patients [ 36 ]. For the WNT, SHH, and Group 3 clusters, the top 10 genes with the highest log fold change from differential expression analyses (as described in [ 36 ]) were selected. For Group 4, the top 20 genes were chosen due to the high heterogeneity within this group. Lowly expressed genes were removed to ensure reliable detection in bulk sequencing data, resulting in a final set of 36 expression markers (Fig. 2 A). Hierarchical clustering, based on Euclidean distance and the single-linkage method, was applied to the log2-transformed Fragments Per Kilobase of transcript per Million mapped reads (FPKM) values. This clustering approach identified tentative WNT, SHH, Group 3, and Group 4 subgroups, with the index patient identified as an outlier (Fig. 2 A heatmap). Genomic features, including mutations and copy number alterations, were curated from previous genomic studies and reviews [ 1 , 9 , 10 ], and aligned with RNA-seq-derived subgroup classifications (Fig. 2 A Oncoprint).

Additionally, transcriptome-wide principal component analysis (PCA) was performed on the log-transformed FPKM values. Gene variance was calculated across samples for each gene’s log-transformed expression data. Low-variance genes were filtered by removing the bottom 10% with the lowest variance, based on variance quantiles (Fig. 2 C). Gene set enrichment analysis of the index case, using a pre-ranked list of z-scores calculated among 20 patients, was performed on the Reactome Pathways Database with WebGestalt (Fig. 2 D).

DNA methylation analysis of the index case was performed using the Infinium MethylationEPIC v2.0 Kit. DNA methylation-based tumor classification of the index case was conducted with the Heidelberg-DKFZ brain tumor classifier v12.5 ( https://www.molecularneuropathology.org/mnp ) and the NCI (Bethesda) classifier v2.

Acknowledgements

Not applicable.

Author contributions

JNV and DRR conceptualized and outlined the report. JNV, AF, and SMW prepared the manuscript, with JNV conducting the bioinformatic analyses. NB performed histological staining and contributed to the methylation analysis. JNV, AF and NB made the figures. JNV, SMW, Y-MW, NB, and AMC engaged in data discussion and interpretation. All authors reviewed and approved the final manuscript.

Funding

JNV and AMC is supported by Early Detection Research Network (U2CCA271854-03, NIH). DRR is supported by NCI Outstanding Investigator Award (R35CA231996). Y-MW is supported by American Cancer Society CSDG-21-177-01.

Availability of data and materials

All sequencing data (FASTQ and BAM files) and methylation array data are available upon reasonable request.

Declarations

Ethics approval and consent to participate

All samples were acquired after patients provided written informed consent in compliance with the University of Michigan Institutional Review Board.

Patient consent

The investigators have obtained consent to publish the data and images contained within the following case report.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Josh N. Vo, Andrea Franson and Sebastian M. Waszak have contributed equally to this work.

References

1. Northcott PA, Robinson GW, Kratz CP, Mabbott DJ, Pomeroy SL, Clifford SC, Rutkowski S, Ellison DW, Malkin D, Taylor MD et al (2019) Medulloblastoma. Nat Rev Dis Primers 5:11

2. Chevignard M, Câmara-Costa H, Doz F, Dellatolas G (2017) Core deficits and quality of survival after childhood medulloblastoma: a review. Neurooncol Pract 4:82–97

3. Schwalbe EC, Lindsey JC, Nakjang S, Crosier S, Smith AJ, Hicks D, Rafiee G, Hill RM, Iliasova A, Stone T et al (2017) Novel molecular subgroups for clinical classification and outcome prediction in childhood medulloblastoma: a cohort study. Lancet Oncol 18:958–971

4. Wooley JR, Penas-Prado M (2021) Pediatric versus adult medulloblastoma: towards a definition that goes beyond age. Cancers. 10.3390/cancers13246313

5. Franceschi E, Giannini C, Furtner J, Pajtler KW, Asioli S, Guzman R, Seidel C, Gatto L, Hau P (2022) Adult medulloblastoma: updates on current management and future perspectives. Cancers. 10.3390/cancers14153708

6. Wong GC-H, Li KK-W, Poon MF-M, Ng H-K (2020) Is adult medulloblastoma merely the counterpart of pediatric medulloblastoma? Glioma 3:90

7. Waszak SM, Northcott PA, Buchhalter I, Robinson GW, Sutter C, Groebner S, Grund KB, Brugières L, Jones DTW, Pajtler KW et al (2018) Spectrum and prevalence of genetic predisposition in medulloblastoma: a retrospective genetic study and prospective validation in a clinical trial cohort. Lancet Oncol 19:785–798

8. Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, Hawkins C, Ng HK, Pfister SM, Reifenberger G et al (2021) The 2021 WHO classification of tumors of the central nervous system: a summary. Neuro Oncol 23:1231–1251

9. Cavalli FMG, Remke M, Rampasek L, Peacock J, Shih DJH, Luu B, Garzia L, Torchia J, Nor C, Morrissy AS et al (2017) Intertumoral heterogeneity within medulloblastoma subgroups. Cancer Cell 31:737-754.e6

10. Northcott PA, Buchhalter I, Morrissy AS, Hovestadt V, Weischenfeldt J, Ehrenberger T, Gröbner S, Segura-Wang M, Zichner T, Rudneva VA et al (2017) The whole-genome landscape of medulloblastoma subtypes. Nature 547:311–317

11. Schaub FX, Dhankani V, Berger AC, Trivedi M, Richardson AB, Shaw R, Zhao W, Zhang X, Ventura A, Liu Y et al (2018) Pan-cancer alterations of the MYC oncogene and its proximal network across the cancer genome atlas. Cell Syst 6:282-300.e2

12. Cascón A, Robledo M (2012) MAX and MYC: a heritable breakup. Cancer Res 72:3119–3124

13. Cole MD, Cowling VH (2008) Transcription-independent functions of MYC: regulation of translation and DNA replication. Nat Rev Mol Cell Biol 9:810–815

14. Waszak SM, Robinson GW, Gudenas BL, Smith KS, Forget A, Kojic M, Garcia-Lopez J, Hadley J, Hamilton KV, Indersie E et al (2020) Germline elongator mutations in sonic hedgehog medulloblastoma. Nature 580:396–401

15. Borghi A, Verstrepen L, Beyaert R (2016) TRAF2 multitasking in TNF receptor-induced signaling to NF-κB, MAP kinases and cell death. Biochem Pharmacol 116:1–10

16. Brenner D, Blaser H, Mak TW (2015) Regulation of tumour necrosis factor signalling: live or let die. Nat Rev Immunol 15:362–374

17. Dhanasekaran DN, Reddy EP (2008) JNK signaling in apoptosis. Oncogene 27:6245–6251

18. Ventura J-J, Cogswell P, Flavell RA, Baldwin AS Jr, Davis RJ (2004) JNK potentiates TNF-stimulated necrosis by increasing the production of cytotoxic reactive oxygen species. Genes Dev 18:2905–2915

19. Yin Q, Lamothe B, Darnay BG, Wu H (2009) Structural basis for the lack of E2 interaction in the RING domain of TRAF2. Biochemistry 48:10558–10567

20. Park HH (2018) Structure of TRAF family: current understanding of receptor recognition. Front Immunol 9:1999

21. Takeuchi M, Rothe M, Goeddel DV (1996) Anatomy of TRAF2: distinct domains for nuclear factor-ΚB activation and association with tumor necrosis factor signaling proteins. J Biol Chem 271:19935–19942

22. Strasser A, Jost PJ, Nagata S (2009) The many roles of FAS receptor signaling in the immune system. Immunity 30:180–192

23. Coultas L, Terzano S, Thomas T, Voss A, Reid K, Stanley EG, Scott CL, Bouillet P, Bartlett P, Ham J et al (2007) Hrk/DP5 contributes to the apoptosis of select neuronal populations but is dispensable for haematopoietic cell apoptosis. J Cell Sci 120:2044–2052

24. Zhou C, Gao Y, Ding P, Wu T, Ji G (2023) The role of CXCL family members in different diseases. Cell Death Discov 9:212

25. Pietsch T, Haberler C (2016) Update on the integrated histopathological and genetic classification of medulloblastoma—a practical diagnostic guideline. Clin Neuropathol 35:344–352

26. Zhukova N, Ramaswamy V, Remke M, Pfaff E, Shih DJH, Martin DC, Castelo-Branco P, Baskin B, Ray PN, Bouffet E et al (2013) Subgroup-specific prognostic implications of TP53 mutation in medulloblastoma. J Clin Orthod 31:2927–2935

27. Carta R, Del Baldo G, Miele E, Po A, Besharat ZM, Nazio F, Colafati GS, Piccirilli E, Agolini E, Rinelli M et al (2020) Cancer predisposition syndromes and medulloblastoma in the molecular era. Front Oncol 10:566822

28. Begemann M, Waszak SM, Robinson GW, Jäger N, Sharma T, Knopp C, Kraft F, Moser O, Mynarek M, Guerrini-Rousseau L et al (2020) Germline GPR161 mutations predispose to pediatric medulloblastoma. J Clin Oncol 38:43–50

29. Zhu S, Jin J, Gokhale S, Lu AM, Shan H, Feng J, Xie P (2018) Genetic alterations of TRAF proteins in human cancers. Front Immunol 9:2111

30. Siegmund D, Wagner J, Wajant H (2022) TNF receptor associated factor 2 (TRAF2) signaling in cancer. Cancers. 10.3390/cancers14164055

31. Vo JN, Wu Y-M, Mishler J, Hall S, Mannan R, Wang L, Ning Y, Zhou J, Hopkins AC, Estill JC et al (2022) The genetic heterogeneity and drug resistance mechanisms of relapsed refractory multiple myeloma. Nat Commun 13:1–13

32. Clark VE, Erson-Omay EZ, Serin A, Yin J, Cotney J, Ozduman K, Avşar T, Li J, Murray PB, Henegariu O et al (2013) Genomic analysis of non-NF2 meningiomas reveals mutations in TRAF7, KLF4, AKT1, and SMO. Science 339:1077–1080

33. Cancer Genome Atlas Network (2015) Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature 517:576–582

34. Chen Z, Ioris RM, Richardson S, Van Ess AN, Vendrell I, Kessler BM, Buffa FM, Busino L, Clifford SC, Bullock AN et al (2022) Disease-associated KBTBD4 mutations in medulloblastoma elicit neomorphic ubiquitylation activity to promote CoREST degradation. Cell Death Differ 29:1955–1969

35. Robinson DR, Wu Y-M, Lonigro RJ, Vats P, Cobain E, Everett J, Cao X, Rabban E, Kumar-Sinha C, Raymond V et al (2017) Integrative clinical genomics of metastatic cancer. Nature 548:297–303

36. Riemondy KA, Venkataraman S, Willard N, Nellan A, Sanford B, Griesinger AM, Amani V, Mitra S, Hankinson TC, Handler MH et al (2022) Neoplastic and immune single-cell transcriptomics define subgroup-specific intra-tumoral heterogeneity of childhood medulloblastoma. Neuro Oncol 24:273–286

Associated Data

Data Availability Statement

All sequencing data (FASTQ and BAM files) and methylation array data are available upon reasonable request.