High-yield points
- WHO 2021: classify by site and molecular group:
- Supratentorial: ZFTA (formerly C11orf95/RELA) or YAP1.
- Posterior fossa: PFA or PFB.
- Spinal: including MYCN-amplified.
- Also: myxopapillary ependymoma (now grade 2) and subependymoma .
- PFA:
- Infants and young children, laterally located, CIMP-positive.
- Subtotal resection is dismal.
- 1q gain (EFS 47% vs. 83%) and 6q loss (PFS 7%) mark high risk .
- PFB:
- Older patients; 5-year PFS about 90%.
- Some may be cured with gross total resection alone .
- Supratentorial:
- YAP1 does better than ZFTA.
- CDKN2A deletion in ZFTA–RELA is high risk .
- Extent of resection is the dominant modifiable factor. Gross total resection cut the hazard of death about sixfold (HR 0.16) .
- Adjuvant therapy:
- Focal RT, 59.4 Gy, including under age 3 .
- Maintenance chemotherapy gave no benefit in a randomized trial .
- Metastasis at diagnosis:
- About 2%, with poor outcome (5-year PFS 13–29%) .
- Depends on molecular group: ZFTA–RELA did better .
Landmark papers
- St. Jude conformal RT (Merchant et al., Lancet Oncol 2009): 153 children treated with conformal RT after surgery. 7-year EFS was 69.1% and OS 81.0%, and gross total resection independently improved OS (HR 0.16) .
- ACNS0121 (Merchant et al., J Clin Oncol 2019): a COG trial. Infratentorial tumors with 1q gain had a 5-year EFS of 47.4% vs 82.8% without it .
- ACNS0831 (Smith et al., Neuro Oncol 2026): a randomized trial (325 children). Maintenance chemotherapy after RT did not help: 5-year EFS 69.2% vs 63.7%, not significant .
Ependymoma is now classified by location plus molecular group, not by histologic grade alone. DNA methylation profiling identified nine molecular groups, three in each compartment: supratentorial, posterior fossa and spine. This molecular classification outperformed histopathology for predicting outcome . In children, the mainstays of treatment remain maximal safe resection and focal conformal radiotherapy .
Classification (WHO 2021)
- Types by site :
- Supratentorial ependymoma: ZFTA fusion-positive and YAP1 fusion-positive.
- Posterior fossa ependymoma: group PFA and group PFB.
- Spinal ependymoma: including the new spinal ependymoma, MYCN-amplified.
- Myxopapillary ependymoma and subependymoma, which remain separate types.
- Location-only diagnoses (with an NEC or NOS suffix) are used when molecular testing finds another alteration, fails, or isn't available.
- Key markers :
- Supratentorial: ZFTA, RELA, YAP1, MAML2.
- Posterior fossa: loss of H3 K27me3, EZHIP, plus methylome.
- Spinal: NF2, MYCN.
- ZFTA is the new name for C11orf95. It was chosen over RELA because ZFTA can fuse with partners other than RELA .
- In the original description, more than two-thirds of supratentorial ependymomas carried C11orf95–RELA fusions, generated by chromothripsis of 11q13.1 .
- Grading:
- "Anaplastic ependymoma" is no longer a listed type. Pathologists may still assign CNS WHO grade 2 or 3 to histologically defined ependymomas, though the reproducibility of grading has long been debated.
- Myxopapillary ependymoma is now grade 2 (previously grade 1) .
Infratentorial (posterior fossa) ependymoma
- PFA vs. PFB :
- PFA:
- Younger patients, more often laterally located, with a balanced genome.
- Much more likely to recur, metastasize at recurrence and cause death .
- CIMP-positive: poor-prognosis hindbrain ependymomas are genetically bland (no recurrent mutations), but show a CpG island methylator phenotype that silences targets of the Polycomb repressive complex 2 (PRC2) .
- Molecular group was the strongest predictor of outcome in 820 tumors, independent of age and treatment :
- Incompletely resected PFA: dismal, with 5-year PFS of only 26.1–56.8% across cohorts.
- Completely resected PFA: clearly benefits from adjuvant radiation.
- PFB: a substantial proportion are cured by surgery alone, and relapses can often be salvaged with delayed radiation. Gross total resection of PFB was proposed for a trial of observation .
- Cytogenetic risk within PFA:
- Pooled analysis of 663 PFA tumors :
- 1q gain in 18.9%, and 6q loss in 8.6%.
- 5-year PFS by copy number:
- balanced genome 50%;
- 1q gain alone 32%;
- 6q loss alone 7.3%;
- both 0%.
- 6q loss defines an "ultra high-risk" PFA.
- ACNS0121 (COG trial): infratentorial tumors with 1q gain had a 5-year EFS of 47.4% vs. 82.8% without it .
- PFB heterogeneity:
- Five subtypes (PFB1–5).
- Late relapses are common, but deaths are rare.
- Unlike in PFA, 1q gain is not a robust marker.
- Extent of resection remains the strongest predictor .
Supratentorial ependymoma
- ZFTA fusion-positive:
- The majority of supratentorial tumors, in older children.
- Relatively worse prognosis than YAP1 .
- YAP1 fusion-positive:
- Rarer, and enriched in younger children.
- Substantially more favorable prognosis .
- Prospective cohort (E-HIT2000 and registries) :
- 5-year PFS/OS:
- ZFTA: 64%/86%.
- YAP1 (only 4 patients): 50%/100%.
- Worse within ZFTA:
- Alternative (non-RELA) ZFTA fusions: 5-year PFS 36%.
- CDKN2A homozygous deletion in ZFTA–RELA: 4-year PFS 19% vs. 79%.
- Observation after complete resection:
- Classic supratentorial tumors, observed after gross total resection (ACNS0121): 5-year EFS 61.4% .
- Grade 2 supratentorial tumors, observed after GTR or complete response (ACNS0831): 5-year EFS 66.9% and OS 100%, so salvage at relapse was effective .
Prognosis with and without metastasis
- Without metastasis (localized disease):
- St. Jude (153 children, conformal RT after surgery) :
- 7-year outcomes: local control 87.3%, EFS 69.1%, OS 81.0%.
- With immediate postoperative RT: EFS was 76.9%.
- Gross total resection independently improved OS (HR 0.16).
- Anaplastic grade predicted worse OS and was the only predictor of distant failure.
- E-HIT2000 (291 children) :
- Overall: 5-year PFS 62% and OS 81%.
- By group: PFA 45%/77%, PFB 90%/100%.
- PFA without molecular risk factors, completely resected and irradiated: 75%/92%.
- PFA with molecular risk factors: 33%/64%, regardless of residual tumor.
- With metastasis at diagnosis (rare) :
- How rare:
- Only 2.2% of 453 children in HIT-2000, staged with spinal MRI and CSF cytology.
- CSF-positive-only (M1) disease was 0.7% .
- Outcome is poor overall:
- HIT-2000 cohort: 5-year PFS 13%, OS 58% .
- Multi-institutional series (40 patients, 90% posterior fossa): 5-year EFS 29%, OS 43% .
- Molecular group matters:
- ZFTA–RELA: all 4 metastatic patients were alive in remission 7.5–12.3 years later.
- Macroscopic metastases from a posterior fossa or multifocal tumor: all 4 patients died .
- Gross total resection of the primary still improved OS (59% vs. 32%) .
Adjuvant therapy
- Focal conformal radiotherapy is the mainstay :
- Dose: 59.4 Gy, or 54 Gy after gross total resection in children under 18 months, with a 1.0-cm clinical target volume margin.
- Age under 3: immediate postoperative RT gave EFS similar to older children in ACNS0121 .
- Late effects at 7 years: brainstem necrosis 1.6% and second brain tumors 2.3% .
- Chemotherapy: no proven benefit after RT.
- ACNS0831 (randomized, 325 children): maintenance chemotherapy after RT did not improve outcome .
- 5-year EFS: 69.2% vs. 63.7% with RT alone, not significant.
- 5-year OS: 88.3% vs. 86.9%.
- Subtotal resection (given RT plus chemotherapy): 5-year EFS 33.6% .
- ACNS0121, subtotal resection (chemotherapy, second surgery, then RT): 5-year EFS 37.2% .
- Bottom line: incomplete resection is the problem. Aim for gross total resection, including a second-look operation .
- Observation is reasonable after gross total resection of supratentorial grade 2 tumors .
- Proposed, not yet established: observation after gross total resection of PFB .
- Metastatic disease: treatment is individualized (craniospinal RT, chemotherapy, focal RT) .
Related topics
- Spinal Ependymoma in Children: spinal types, including myxopapillary and MYCN-amplified.
- Medulloblastoma: the other common posterior fossa tumor, with molecular groups.
- ATRT: another infant posterior fossa tumor.
- Neurofibromatosis: NF2-related spinal ependymoma.
- Posterior Fossa Tumor Resection: step-by-step procedure.
References
- Pajtler KW, Witt H, Sill M, et al. Molecular classification of ependymal tumors across all CNS compartments, histopathological grades, and age groups. Cancer Cell 2015;27(5):728–743. doi:10.1016/j.ccell.2015.04.002. PMID 25965575. verified
- Louis DN, Perry A, Wesseling P, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro Oncol 2021;23(8):1231–1251. doi:10.1093/neuonc/noab106. PMID 34185076. verified
- Merchant TE, Li C, Xiong X, et al. Conformal radiotherapy after surgery for paediatric ependymoma: a prospective study. Lancet Oncol 2009;10(3):258–266. doi:10.1016/S1470-2045(08)70342-570342-5). PMID 19274783. verified
- Merchant TE, Bendel AE, Sabin ND, et al. Conformal radiation therapy for pediatric ependymoma, chemotherapy for incompletely resected ependymoma, and observation for completely resected, supratentorial ependymoma. J Clin Oncol 2019;37(12):974–983. doi:10.1200/JCO.18.01765. PMID 30811284. verified
- Smith AA, Onar-Thomas A, Lensing S, et al. Phase 3 randomized trial of postirradiation chemotherapy in patients with newly diagnosed ependymoma: a report from the Children's Oncology Group. Neuro Oncol 2026;28(5):1286–1297. doi:10.1093/neuonc/noaf285. PMID 41423745. verified
- Parker M, Mohankumar KM, Punchihewa C, et al. C11orf95-RELA fusions drive oncogenic NF-κB signalling in ependymoma. Nature 2014;506(7489):451–455. doi:10.1038/nature13109. PMID 24553141. verified
- Witt H, Mack SC, Ryzhova M, et al. Delineation of two clinically and molecularly distinct subgroups of posterior fossa ependymoma. Cancer Cell 2011;20(2):143–157. doi:10.1016/j.ccr.2011.07.007. PMID 21840481. verified
- Ramaswamy V, Hielscher T, Mack SC, et al. Therapeutic impact of cytoreductive surgery and irradiation of posterior fossa ependymoma in the molecular era: a retrospective multicohort analysis. J Clin Oncol 2016;34(21):2468–2477. doi:10.1200/JCO.2015.65.7825. PMID 27269943. verified
- Mack SC, Witt H, Piro RM, et al. Epigenomic alterations define lethal CIMP-positive ependymomas of infancy. Nature 2014;506(7489):445–450. doi:10.1038/nature13108. PMID 24553142. verified
- Baroni LV, Sundaresan L, Heled A, et al. Ultra high-risk PFA ependymoma is characterized by loss of chromosome 6q. Neuro Oncol 2021;23(8):1360–1370. doi:10.1093/neuonc/noab034. PMID 33580238. verified
- Cavalli FMG, Hübner JM, Sharma T, et al. Heterogeneity within the PF-EPN-B ependymoma subgroup. Acta Neuropathol 2018;136(2):227–237. doi:10.1007/s00401-018-1888-x. PMID 30019219. verified
- Vallejo FA, Monsour MA, Fredricks NS, et al. Supratentorial ependymoma: a comprehensive review of molecular classification, management strategies, and clinical outcomes (Part II of pediatric ependymomas across compartments). J Neurooncol 2026;180(1). doi:10.1007/s11060-026-05818-7. PMID 42848093. verified
- von Hoff K, Obrecht-Sturm D, Ghasemi DR, et al. Developing an advanced risk stratification model for pediatric intracranial ependymoma based on the prospective trial E-HIT2000 and subsequent registries. Neuro Oncol 2026;28(2):520–534. doi:10.1093/neuonc/noaf218. PMID 41026848. verified
- Benesch M, Mynarek M, Witt H, et al. Newly diagnosed metastatic intracranial ependymoma in children: frequency, molecular characteristics, treatment, and outcome in the prospective HIT series. Oncologist 2019;24(9):e921–e929. doi:10.1634/theoncologist.2018-0489. PMID 30850560. verified
- Zacharoulis S, Ji L, Pollack IF, et al. Metastatic ependymoma: a multi-institutional retrospective analysis of prognostic factors. Pediatr Blood Cancer 2008;50(2):231–235. doi:10.1002/pbc.21276. PMID 17610266. verified
Verification (2026-11-11): each reference's authors, title, journal, volume, pages, DOI and PMID were checked on PubMed. Numbers were checked against the PubMed abstracts. The WHO 2021 types, markers, the ZFTA renaming, grading and the methylation statements come from the full text of reference 2 (PMC8328013), read via a fetch agent. Reference 12 has no page range on PubMed. The full text of reference 1 could not be retrieved (blocked by a bot check), so its subgroup-level survival details are not quoted; supratentorial and posterior fossa outcomes come from references 7, 8, 12 and 13 instead.
Caveats:
- The YAP1 survival estimate rests on only 4 patients.
- The trial cohorts span different eras and treatments.
- The figure is an original schematic.