Supratentorial ependymoma, or ST-EP, makes up approximately 30 to 40 percent of all ependymoma cases and tends to affect slightly older children than its infratentorial counterpart. The 2021 World Health Organization classification fundamentally reorganized the field by defining three molecular entities: ZFTA-fusion tumors, YAP1-fusion tumors, and a residual category labeled not elsewhere classified, or NEC. This reorganization matters because tumors that look nearly identical under the microscope can behave in radically different ways depending on which gene fusion drives them. The old histologic grading system, which divided ependymomas into grades 1 through 3, suffered from high interobserver variance even among experienced neuropathologists, and prognosis did not always track with grade. Molecular testing, through DNA methylation profiling and RNA sequencing, has now become the decisive diagnostic tool.

The dominant player is the ZFTA-fusion subtype, formerly known as RELA-fused ependymoma, which accounts for roughly 70 percent of supratentorial cases. The ZFTA gene, residing on chromosome 11q13 and formerly designated C11orf95, fuses most commonly with RELA, producing a chimeric oncoprotein that shuttles constitutively into the nucleus and drives transcription through NF-κB signaling, a pathway normally involved in inflammation and proliferation. These tumors typically affect children around eight years of age on average and carry a relatively worse prognosis than their YAP1 counterparts. Intriguingly, recent work has shown that NF-κB upregulation alone does not explain the malignancy: CRISPR knockout and pharmacologic inhibition of NF-κB fail to halt tumor growth in experimental models. Instead, ZFTA fusion proteins appear to hijack transcription factor programs involving PLAGL1 and PLAGL2, factors implicated in normal neuronal development, pushing cells along MAPK, SHH, and SWI/SNF pathways toward a dedifferentiated, anaplastic state.

The biology of ZFTA tumors extends even deeper into cellular metabolism. A recent study in Nature found that ZFTA-RELA ependymomas sustain their oncogenic fusion expression through epigenetically modulated metabolic pathways involving itaconate, an immune-signaling metabolite, suggesting that the tumor’s fuel supply is intertwined with its genetic engine. High levels of HDAC4 and SIRT1 messenger RNA have been associated with significantly worse outcomes, and computational drug screens have flagged HDAC4 and ABCG2 inhibitors as candidate therapeutics. On the cell surface, the adhesion molecule L1CAM is highly expressed and has been used as a surrogate marker for ZFTA fusion status, though prospective validation remains incomplete. These converging lines of evidence are beginning to sketch the outline of targeted therapies for the subtype that needs them most.

At the opposite end of the prognostic spectrum sit YAP1-fused tumors, which comprise fewer than 10 percent of supratentorial ependymomas but carry a dramatically more favorable outlook. These tumors arise in children with a median age under two years, and the largest study to date, published in 2019, reported near-complete five-year progression-free survival, with some patients requiring no post-surgical treatment at all. The molecular driver is most commonly a YAP1-MAMLD1 fusion: YAP1 is a principal downstream effector of the Hippo tumor-suppressor pathway, and when fused with MAMLD1 the resulting protein becomes a potent oncogenic transcriptional regulator that depends on TEAD and nuclear factor I-mediated programs. Because these tumors are rare, the evidence base remains thin, and the authors of the review caution that the spectacular survival figures come from a single, relatively small series pending confirmation in larger cohorts.

Between these two defined entities lies the NEC category, representing roughly 20 percent of supratentorial cases. These tumors resemble classic ependymomas histologically but carry molecular profiles matching no established subtype, encompassing both lesions with no identifiable defining alteration and incompletely characterized emerging entities. Their heterogeneity poses a genuine clinical problem, since they may not respond to treatment regimens calibrated for ZFTA or YAP1 biology. Better molecular characterization of this group is one of the field’s most pressing unmet needs, and the review’s authors call for standardized genomic interrogation across institutions to dissect it.

Even the radiology suite can now offer early molecular hints. ZFTA-fused tumors preferentially arise in the cerebral cortex and parenchyma, most often in the frontal lobe, extending from the pial surface to the ventricular margin as predominantly cystic masses with eccentric cysts, calcifications, hemorrhagic foci, and marked perilesional edema, showing heterogeneous contrast enhancement and restricted diffusion on MRI. At recurrence they tend to re-emerge along the dural flap. YAP1 tumors, by contrast, are characteristically intra- or periventricular, with multinodular solid components and a distinctive garland-shaped pattern of peripheral enhancement, and less prominent edema. These signatures are not substitutes for molecular analysis, but they help multidisciplinary teams anticipate the expected clinical course before genetic results return.

Across every molecular subtype, one principle towers above all others: the extent of surgical resection is the most important modifiable prognostic factor, and gross total resection offers the best chance of cure. Surgeons now deploy neuronavigation, ultrasound, and intraoperative MRI in nearly every case, and second-look operations have proven their worth, with a pan-Italian study showing that children who underwent a second procedure before radiotherapy achieved overall survival comparable to those completely resected at first surgery, without significantly higher morbidity. In some cases, chemotherapy given after the first operation has reduced tumor vascularity and facilitated complete resection at second look. Once resection is complete, patients undergo staging with cerebrospinal fluid cytology and gadolinium-enhanced spine imaging at least 10 to 14 days postoperatively before adjuvant decisions are made.

Radiation therapy follows maximal resection, with focal treatment of the tumor bed at cumulative doses classically ranging from 54 to 59.4 Gy, and the clinical target volume margin has shrunk from 10 millimeters in older trials to a standard 5 millimeters based on the excellent outcomes of the ACNS0831 study. Proton beam therapy is recommended where available because radiation dose to uninvolved brain structures is linked to long-term cognitive decline, and the conformal nature of protons can deliver full prescription dose to the tumor bed while leaving the rest of the brain at near-zero exposure. Perhaps the most consequential finding concerns observation alone: in the ACNS0831 trial, 37 children with grossly totally resected grade 2 supratentorial tumors were monitored without immediate radiotherapy, achieving 66.9 percent five-year event-free survival with no deaths from recurrence, since nearly all relapses were salvageable with repeat surgery and radiation. Notably, no patient in that observation arm harbored a CDKN2A deletion, a homozygous loss that the German HIT cohort identified as an ominous marker, conferring five-year progression-free survival of just 19 percent versus 81 percent in wild-type tumors within the E-HIT-2000 analysis.

Chemotherapy, by contrast, has repeatedly failed to earn its place. The ACNS0831 randomized comparison of radiotherapy alone versus maintenance chemotherapy after radiotherapy showed no significant difference in five-year event-free or overall survival, and a multi-institutional analysis of 108 ZFTA-fused tumors likewise found no benefit. For infants too young for radiation, management remains heterogeneous, with many clinicians bridging with chemotherapy until an appropriate age despite sparse supporting data. Roughly half of children with intracranial ependymoma eventually relapse, most often locally at the original tumor bed, and salvage hinges on repeat gross total resection and re-irradiation, while salvage chemotherapy has been associated with worse outcomes. With the SIOP Ependymoma II trial reporting soon and the prospective SIOP Ependymoma study continuing to enroll, the field is converging on a future in which each child’s tumor is profiled molecularly at diagnosis and treatment is tailored accordingly, a vision that seemed remote when these tumors were classified by appearance alone.

Subject of Research: Molecular classification and management of pediatric supratentorial ependymoma

Article Title: Supratentorial ependymoma: a comprehensive review of molecular classification, management strategies, and clinical outcomes (Part II of pediatric ependymomas across compartments)

Article References: Vallejo, F. A., Monsour, M. A., Fredricks, N. S., Tsang, D. S., Koutsouras, G., Ritzmann, T. A., Aquilina, K., Ramaswamy, V., & Dewan, M. C. (2026). Supratentorial ependymoma: a comprehensive review of molecular classification, management strategies, and clinical outcomes (Part II of pediatric ependymomas across compartments). Journal of Neuro-Oncology, 180(1), Article 8. https://doi.org/10.1007/s11060-026-05818-7

Image Credits: AI Generated

DOI: 10.1007/s11060-026-05818-7

Keywords: supratentorial ependymoma, ZFTA fusion, YAP1 fusion, pediatric brain tumor, WHO 2021 classification, gross total resection, proton beam therapy, CDKN2A deletion, NF-kB signaling, radiotherapy, NEC ependymoma, SIOP Ependymoma II

The central result is striking in its simplicity. Among the 100 enrolled patients, only 8 percent developed grade 2 acute genitourinary toxicity—defined in this trial as urinary frequency severe enough to require two or more medications—with a 95 percent confidence interval of 2.6 to 13.4 percent. Not a single patient experienced grade 3 or higher acute genitourinary toxicity, and no grade 2 or worse urinary retention, urinary pain, urinary urgency, or hematuria was recorded. Grade 1 urinary frequency, the mildest form, occurred in 46 percent of patients, but even this figure reflects a symptom burden that many patients with enlarged prostates already live with before treatment begins. No acute gastrointestinal adverse events were observed at all, a notable outcome given that rectal toxicity is often the limiting factor in prostate radiotherapy.

The radiobiological logic behind the regimen is what makes the result scientifically interesting. Prostate cancer cells carry an unusually low alpha/beta ratio—estimated at approximately 1.4 Gy—which describes how sensitive a tumor is to the size of each radiation fraction. A low alpha/beta value means the cancer is more vulnerable to large individual doses than the surrounding healthy organs are. Hypofractionation exploits this vulnerability: by delivering bigger doses per session, clinicians can inflict disproportionate damage on tumor cells while the nearby bladder and rectum, which have higher alpha/beta values, theoretically tolerate the schedule better. Conventional fractionated radiotherapy delivers 2.0 Gy per fraction across 38 to 40 sessions over nearly two months; the Japanese trial delivered 51.6 Gy in relative biological effectiveness terms across just 12 sessions in a median overall treatment time of 20 days, with no interruptions or delays in any patient.

Protons add a second layer of physical advantage on top of this radiobiological one. Unlike photon beams, which deposit radiation along their entire path through the body and exit on the far side, protons stop at a depth determined by their energy—the Bragg peak—releasing their maximum dose precisely at the tumor and falling off rapidly beyond it. This means minimal energy is deposited in normal tissue in front of the target and essentially none behind it. The trial used the IBA Proteus ONE system, a compact gantry with spot-scanning beam delivery, and planned treatments in RayStation version 10A using a single-field uniform dose algorithm applied to a beam-specific planning target volume, an approach shown to improve conformity and organ sparing compared with uniform-dose planning target volumes.

The technical preparation was equally meticulous. Before treatment, urologists implanted transperineal ultrasound-guided gold fiducial markers into each patient’s prostate and injected SpaceOAR, a hydrogel spacer that physically separates the prostate from the rectum, reducing the dose the rectal wall receives. Daily treatments began with image-guided bony alignment followed by fiducial-based registration, maintaining positional accuracy within a 1-millimeter margin. Patients were instructed to arrive with a comfortably full bladder and an empty rectum, immobilized supine on a vacuum pad. Simulation used 1-millimeter-thick CT slices fused with 3.0 Tesla MRI. The clinical target volume encompassed the prostate and, depending on risk group, part of the seminal vesicles—none for low-risk patients, the proximal third for intermediate-risk, and the proximal half for high-risk—expanded by 5 to 6 millimeters to form the planning target volume. Elective pelvic lymph node irradiation was deliberately omitted.

The trial’s design reflects a careful, staged approach to safety. Built on Simon’s minimax two-stage design with a significance level of 0.05 and 90 percent power, the protocol set acceptable and unacceptable toxicity thresholds at below 5 percent and above 15 percent respectively. In the first stage, 30 patients were enrolled; only two developed grade 2 genitourinary toxicity, satisfying the continuation criterion. In the second stage, the cohort grew to 83 patients with seven grade 2 events, again passing the checkpoint. A protocol-specified expansion then brought the total to 100 patients. Any grade 4 or 5 normal tissue toxicity would have mandated immediate suspension of the study, an event that never occurred. The researchers also adopted a modified grading system from earlier work by Nakajima and colleagues, because standard CTCAE version 5.0 criteria struggle to distinguish radiation-induced urinary symptoms from the baseline symptoms of benign prostatic hyperplasia, which many older prostate cancer patients already have. Under the modified criteria, one dysuria medication counted as grade 1 and two or more as grade 2.

Perhaps the most puzzling finding is a counterintuitive pattern in the subgroup analysis. Multivariate logistic regression revealed that high-risk patients—those with higher PSA levels, more advanced T stage, or Gleason scores of 8 or above—had a significantly lower likelihood of developing grade 2 genitourinary toxicity than lower-risk patients, with an odds ratio of 0.023 and a P value of 0.042. High-risk patients received both neoadjuvant and adjuvant androgen deprivation therapy, while intermediate-risk patients received neoadjuvant hormone therapy alone. One earlier hypothesis held that androgen deprivation shrinks the prostate and thereby reduces toxicity, but the intermediate-risk group, which also received hormone therapy, did not show the same protective pattern. The authors are candid that the mechanism behind this trend toward low toxicity in the high-risk group is unknown and warrants further investigation.

The analysis also confirmed benign prostatic hyperplasia as an independent risk factor for grade 2 genitourinary toxicity, with an odds ratio of 0.040 and a P value of 0.018. This aligns with clinical intuition: a gland already compromised by enlargement has less functional reserve when radiation inflames the urinary tract. The finding underscores why the modified grading criteria matter, since disentangling treatment effects from pre-existing conditions is essential for honest toxicity reporting. It also hints at a practical implication for future trials—baseline urinary function may need to be a stratification factor when comparing fractionation schedules across institutions.

How does this compare with photon-based alternatives? The evidence base for shorter prostate radiotherapy is already strong: randomized trials of hypofractionated and ultra-hypofractionated photon therapy, including the HYPO-RT-PC trial of ultra-hypofractionated treatment, have confirmed efficacy comparable to conventional schedules. But acute toxicity data have been less favorable. A pilot study of ultra-hypofractionated intensity-modulated radiation therapy delivering 54 Gy in 15 fractions reported acute grade 2 genitourinary toxicity in 21 percent of patients—more than double the 8 percent seen in the proton trial. Earlier proton work by Nakajima’s group using 60 to 63 Gy in 20 to 21 fractions reported a 5.9 percent rate, and carbon-ion radiotherapy studies using the same 12-fraction, 51.6 Gy regimen achieved rates of 3 to 4 percent. Because carbon ions and protons differ fundamentally in physical characteristics and biological modeling, the question of whether the carbon-ion schedule could be safely transplanted to proton therapy had lacked prospective evidence—until now. A related phase II effort by Iwata and colleagues reported a 6.3 percent incidence using similar modified criteria in a 300-patient proton cohort, closely mirroring the present result.

The study has acknowledged limitations. It was conducted at a single institution, and patients with severe baseline urinary symptoms were excluded, which ensures a clean read on radiation-induced toxicity but limits generalizability to men with poor baseline urinary function. The current report covers only acute toxicity within three months; late genitourinary and gastrointestinal effects, local control rates, and biochemical recurrence-free survival await the ongoing expansion cohort, which targets 300 evaluable patients. Still, the early signal is compelling: a three-week proton regimen that matches the safety profile of far longer schedules, spares the bowel entirely in the acute setting, and returns patients to their lives weeks earlier. If the long-term data hold, the combination of the Bragg peak’s physics, hypofractionation’s radiobiology, and hydrogel spacer technology may redefine what men can expect from curative prostate radiotherapy—fewer visits, fewer side effects, and no compromise on the medicine itself.

Subject of Research: Acute genitourinary toxicity of hypofractionated proton beam therapy for localized prostate cancer

Article Title: Analysis of acute genitourinary toxicity in hypofractionated proton beam therapy for localized prostate cancer

Article References: Yang, W., Mizoe, J., Osaka, Y., Miyabe, Y., Seki, H., Nakamura, H., Sugawara, H., Miyajima, N., Iwata, H., Hashimoto, T., & Komaki, R. (2026). Analysis of acute genitourinary toxicity in hypofractionated proton beam therapy for localized prostate cancer. Holistic Integrative Oncology, 5(1), Article 35. https://doi.org/10.1007/s44178-026-00246-z

Image Credits: AI Generated

DOI: 10.1007/s44178-026-00246-z

Keywords: prostate cancer, proton beam therapy, hypofractionation, genitourinary toxicity, radiation oncology, Bragg peak, CTCAE, androgen deprivation therapy, benign prostatic hyperplasia, SpaceOAR, clinical trial, urinary frequency