← Back

Ependymoma ACNS0121

Year of Publication: 2019

Journal: Journal of Clinical Oncology

Link: https://doi.org/10.1200/JCO.18.01765

PDF: https://ascopubs.org/doi/pdf/10.1200/JCO.18.01765

Clinical Question

Does stratum-based management (immediate postoperative CRT, chemotherapy for STR, observation for GTR supratentorial) optimize outcomes in pediatric intracranial ependymoma, and how do histologic grade, 1q gain, and methylation subgroups affect prognosis?

Bottom Line

In 356 children (ages 1-21) with newly diagnosed intracranial ependymoma treated according to tumor location, extent of resection, and histology (ACNS0121): 5-year EFS was 61.4% (observation after GTR of supratentorial classic), 37.2% (STR chemotherapy ± second surgery + CRT), and 68.5% (immediate postoperative CRT after NTR/GTR). Infratentorial 1q gain reduced EFS dramatically (47.4% vs 82.8%; p=0.0013). Established immediate CRT safety in children <3 years and validated 1q gain as a critical adverse biomarker.

        Major Points
        Children's Oncology Group phase 2 prospective multicenter trial at 115 institutions, 2003-2007 (ACNS0121, Merchant JCO 2019)
N=356 eligible children ages 1-21 with newly diagnosed nonmetastatic intracranial ependymoma
Non-randomized stratum-based treatment: observation (GTR supratentorial classic), chemo+second surgery+CRT (STR), or immediate CRT
Conformal radiation therapy 59.4 Gy (54 Gy if <18 months with GTR); 1-cm CTV margin
Overall 5-year EFS 62.7% (95% CI 57.4-70.0%); OS 83.8% (95% CI 79.7-87.9%)
Immediate postoperative CRT strata 3+4: 5-y EFS 68.5%; OS 86.2%
STR stratum poorest: 5-y EFS 37.2%, OS 70.2% despite chemotherapy + second surgery + CRT
Histology: classic 5-y EFS 74.6% vs anaplastic 60.7%; p=0.0044
Infratentorial 1q gain reduced 5-y EFS from 82.8% to 47.4% (p=0.0013); OS from 91.3% to 68.4% (p=0.0028)
PFA with 1q gain particularly poor: 5-y EFS 35.7% vs PFA without 1q gain 81.5%; p=0.0005
RELA-fusion supratentorial tumors did not confer poor prognosis when treated with immediate CRT
Validated immediate postoperative CRT as safe in children <3 years — landmark practice-changing finding

      

Design

Study Type: Children's Oncology Group phase 2 prospective multicenter non-randomized stratum-based trial

Randomization:

Blinding: Open-label

Follow-up Duration: 5 years (primary); follow-up continues

Sample Size: 378

Analyzed: 356

Analysis: Product-limit (Kaplan-Meier) survival estimation; log-rank comparisons; molecular correlates by FISH and DNA methylation

Baseline Characteristics

Characteristic	Control	Active
N	0	356
Notes	No pooled control; strata treated per protocol
Age median		5.6 years (range 1.0-21.0)
Female		42%
Classic ependymoma		60.4%
Anaplastic		39.6%
Infratentorial		258 (72%)
Supratentorial		96 (27%)

Arms

Field	Control	Stratum 2: STR chemotherapy + second surgery + CRT	Stratum 3: Supratentorial NTR/GTR immediate CRT	Stratum 4: Infratentorial NTR/GTR immediate CRT
N	11	64	118	163
Intervention	No adjuvant therapy; observation only	2 chemo cycles (VCR+carbo+cyclophosphamide, then VCR+carbo+oral etoposide) followed by possible second surgery then 59.4 Gy CRT	Immediate postoperative CRT 59.4 Gy (54 Gy if <18 mo GTR), 1-cm CTV margin	Immediate postoperative CRT 59.4 Gy (54 Gy if <18 mo GTR), 1-cm CTV margin
Duration	5+ years follow-up	~9-10 months	~6 weeks RT + follow-up	~6 weeks RT + follow-up

Outcomes

Outcome	Type	Control	Intervention	P-value
5-year event-free survival overall and by stratum (EFS = time to progression, recurrence, second neoplasm, or death)	Primary	Stratum 1 (observation, n=11): 61.4%	Stratum 2 (STR): 37.2%; Stratum 3 (supra CRT): 67.3%; Stratum 4 (infra CRT): 69.5%	Stratum comparisons descriptive; histologic grade p=0.0044
5-year OS overall	Secondary	Reference cohort	83.8%	Strata 3+4: 86.2%
EFS by histology (classic vs anaplastic, immediate CRT arms)	Secondary	Anaplastic 60.7%	Classic 74.6%	p=0.0044
Age <3 vs ≥3 years (immediate CRT)	Secondary	≥3 y EFS 70.5%	<3 y EFS 62.9%	p=0.23 (NS)
Infratentorial 1q gain vs no gain (EFS)	Secondary	No 1q gain 82.8%	1q gain 47.4%	p=0.0013
Infratentorial 1q gain vs no gain (OS)	Secondary	No 1q gain 91.3%	1q gain 68.4%	p=0.0028
Supratentorial RELA fusion impact	Secondary	Non-RELA	RELA-fusion positive	No significant difference with immediate CRT
PFA vs PFB methylation subgroup	Secondary	PFB	PFA	No overall difference; PFA+1q gain particularly poor (5-y EFS 35.7%)
Second surgery for STR (stratum 2)	Secondary	No second surgery 28.5%	Second surgery 50.5%	p=0.1163 EFS
Grade 3-4 chemo AE (Stratum 2)	Adverse	N/A	Common (myelosuppression)	Chemo-related
Ototoxicity (carboplatin)	Adverse	N/A	Reported	Platinum-related
Growth/endocrine deficits (CRT)	Adverse	Observation stratum low	Common in CRT-treated; dose-related	Expected for CRT
Cognitive/neurocognitive decline	Adverse	Minimal observation stratum	Less than conventional RT historically	CRT vs historical RT favorable
Hearing loss	Adverse	Low	Dose-dependent with posterior fossa RT	Predictable by dosimetry
Second malignancy (cumulative at 10 y)	Adverse	N/A	3.43% (95% CI 0.4-6.5%)	Rare but documented
RT necrosis	Adverse	N/A	Uncommon	CRT technique favorable
Posterior fossa syndrome (acute post-op)	Adverse	Pre-treatment event	Pre-treatment event	Surgical

Subgroup Analysis

Most significant findings: (1) 1q gain in infratentorial tumors strongly predicted worse EFS and OS (47.4% vs 82.8% EFS; 68.4% vs 91.3% OS). (2) PFA subgroup with 1q gain had especially poor 5-y EFS 35.7% vs PFA without 1q gain 81.5% (p=0.0005). (3) RELA fusion in supratentorial tumors (77% of supratentorial) did not confer poor prognosis when treated with immediate CRT — overturning prior retrospective reports. (4) Children <3 years achieved similar EFS to older children when treated with immediate CRT, a landmark finding that shifted pediatric ependymoma practice away from delaying RT.

Criticisms

Non-randomized stratum-based design — cross-stratum comparisons confounded by indication
Stratum 1 observation cohort very small (N=11) — 63% OS at 5 years with observation is encouraging but underpowered
Stratum 2 second surgery rate only 39% despite design intent — surgical feasibility limited planned intervention
Molecular analyses (RELA fusion, 1q, methylation) were retrospective — not prospectively stratified
Only 20 patients received proton therapy — limited conclusions about proton vs photon CRT
Long-term (10+ year) neurocognitive and endocrine outcomes require ongoing follow-up — ACNS0121 follow-up cohort continues

Funding

NCI Children's Oncology Group

Based on: Ependymoma ACNS0121 (Journal of Clinical Oncology, 2019)

Content summarized and formatted by NeuroTrials.ai.

Ependymoma ACNS0121

(2019)

Objective

Conformal radiation therapy (CRT) for pediatric ependymoma, chemotherapy for incompletely resected ependymoma, and observation for completely resected supratentorial classic ependymoma — stratum-based treatment to validate immediate postoperative CRT and molecular prognostic markers.

Study Summary

• 5-year EFS and OS for 356 patients: 62.7% and 83.8% overall; immediate postoperative CRT arm 68.5% EFS / 86.2% OS.
• Observation stratum (classic supratentorial GTR, n=11): 5-y EFS 61.4%, OS 100% — surgery alone viable for this select group.
• STR stratum (chemo ± second surgery + CRT): 5-y EFS only 37.2%, OS 70.2% — poorer prognosis despite intensification.
• Histology: classic ependymoma EFS 74.6% vs anaplastic 60.7% (p=0.0044).
• Infratentorial 1q gain reduced 5-year EFS from 82.8% to 47.4% (p=0.0013); OS 91.3% to 68.4% (p=0.0028).
• First prospective trial to use immediate postoperative CRT safely in children <3 years — validated landmark approach.

Intervention

Stratum 1: Observation after GTR of classic supratentorial. Stratum 2: STR — 2 cycles chemotherapy (VCR+carbo+cyclophosphamide, then VCR+carbo+oral etoposide) then second surgery then CRT. Strata 3-4: Immediate postoperative CRT (59.4 Gy, 54 Gy if <18 mo with GTR). 1-cm CTV margin, conformal/IMRT/proton.

Inclusion Criteria

Children 1-21 years with histologically confirmed nonmetastatic intracranial ependymoma, within 56 days of surgical resection, no prior chemotherapy or RT, MRI brain and spine.

Study Design

Arms: Stratum 1 observation; Stratum 2 chemotherapy + second surgery + CRT; Stratum 3 CRT after NTR supratentorial; Stratum 4 CRT after GTR/NTR infratentorial

Patients per Arm: Stratum 1: 11; Stratum 2: 64; Stratum 3: 118; Stratum 4: 163 (total eligible N=356; enrolled 378 at 115 institutions)

Outcome

• Overall 5-year EFS 62.7% (95% CI 57.4-70.0%); OS 83.8% (95% CI 79.7-87.9%)
• Stratum 3+4 combined immediate CRT: 5-y EFS 68.5% (95% CI 62.8-74.2%); OS 86.2% (95% CI 81.9-90.6%)
• Histology: Classic 5-y EFS 74.6% vs anaplastic 60.7% (p=0.0044)
• Age: <3 y 62.9% vs ≥3 y 70.5% EFS (p=0.23); OS similar 87.4% vs 85.8%
• Infratentorial 1q gain: 5-y EFS 47.4% vs 82.8% (p=0.0013); 5-y OS 68.4% vs 91.3% (p=0.0028)

Bottom Line

Major Points

Children's Oncology Group phase 2 prospective multicenter trial at 115 institutions, 2003-2007 (ACNS0121, Merchant JCO 2019)
N=356 eligible children ages 1-21 with newly diagnosed nonmetastatic intracranial ependymoma
Non-randomized stratum-based treatment: observation (GTR supratentorial classic), chemo+second surgery+CRT (STR), or immediate CRT
Conformal radiation therapy 59.4 Gy (54 Gy if <18 months with GTR); 1-cm CTV margin
Overall 5-year EFS 62.7% (95% CI 57.4-70.0%); OS 83.8% (95% CI 79.7-87.9%)
Immediate postoperative CRT strata 3+4: 5-y EFS 68.5%; OS 86.2%
STR stratum poorest: 5-y EFS 37.2%, OS 70.2% despite chemotherapy + second surgery + CRT
Histology: classic 5-y EFS 74.6% vs anaplastic 60.7%; p=0.0044
Infratentorial 1q gain reduced 5-y EFS from 82.8% to 47.4% (p=0.0013); OS from 91.3% to 68.4% (p=0.0028)
PFA with 1q gain particularly poor: 5-y EFS 35.7% vs PFA without 1q gain 81.5%; p=0.0005
RELA-fusion supratentorial tumors did not confer poor prognosis when treated with immediate CRT
Validated immediate postoperative CRT as safe in children <3 years — landmark practice-changing finding

Study Design

Study Type: Children's Oncology Group phase 2 prospective multicenter non-randomized stratum-based trial
Randomization: No
Blinding: Open-label
Sample Size: 378
Follow-up: 5 years (primary); follow-up continues

Primary Outcome

Definition: 5-year event-free survival overall and by stratum (EFS = time to progression, recurrence, second neoplasm, or death)

Control	Intervention	HR/OR	P-value
Stratum 1 (observation, n=11): 61.4%	Stratum 2 (STR): 37.2%; Stratum 3 (supra CRT): 67.3%; Stratum 4 (infra CRT): 69.5%	- (Stratum 1: 34.5-89.6%; Stratum 2: 24.8-49.6%; Strata 3+4: 62.8-74.2% EFS)	Stratum comparisons descriptive; histologic grade p=0.0044

Limitations & Criticisms

Non-randomized stratum-based design — cross-stratum comparisons confounded by indication
Stratum 1 observation cohort very small (N=11) — 63% OS at 5 years with observation is encouraging but underpowered
Stratum 2 second surgery rate only 39% despite design intent — surgical feasibility limited planned intervention
Molecular analyses (RELA fusion, 1q, methylation) were retrospective — not prospectively stratified
Only 20 patients received proton therapy — limited conclusions about proton vs photon CRT
Long-term (10+ year) neurocognitive and endocrine outcomes require ongoing follow-up — ACNS0121 follow-up cohort continues

Citation

View Original Publication →

Ependymoma ACNS0121

Clinical Question

Bottom Line

Major Points

Design

Baseline Characteristics

Arms

Outcomes

Subgroup Analysis

Criticisms

Funding

Ask about Ependymoma ACNS0121