Background
Updated long-term analysis of the NOA-04 phase III trial (318 patients, anaplastic gliomas). This 2016 Neuro-Oncology paper extends the original 2009 JCO report with longer follow-up and expanded molecular analysis including IDH1/2 and MGMT methylation, providing the most mature survival data from NOA-04 and clarifying the prognostic impact of molecular markers in sequential RT and chemotherapy strategies.
Interventions and follow up
Arm A: Radiotherapy first → PCV or TMZ at progressio
Primary endpoint: Time to treatment failure (updated long-term analysis)
mFollow up: Extended ≥10 years for many patient
Primary endpoint: Time to treatment failure (updated long-term analysis)
mFollow up: Extended ≥10 years for many patient
Results
Arm B1/B2: PCV or TMZ first → RT at progression
Updated OS: Still no significant difference between RT-first and chemotherapy-first sequences
IDH1 mutant vs WT (OS): Dramatic difference — IDH mutant median OS ~10+ years; IDH wild-type median OS ~1.5–2 years regardless of treatment sequence
MGMT methylation: Strong predictor of benefit in chemotherapy-treated patients
1p/19q codeletion: Best OS in IDH-mutant codeleted patients regardless of treatment arm
Updated OS: Still no significant difference between RT-first and chemotherapy-first sequences
IDH1 mutant vs WT (OS): Dramatic difference — IDH mutant median OS ~10+ years; IDH wild-type median OS ~1.5–2 years regardless of treatment sequence
MGMT methylation: Strong predictor of benefit in chemotherapy-treated patients
1p/19q codeletion: Best OS in IDH-mutant codeleted patients regardless of treatment arm
Adverse events
Main adverse events: No new safety signals reported in this long-term analysis; consistent with 2009 JCO primary report.
Conclusions
Long-term NOA-04 follow-up confirms that treatment sequence (RT vs chemotherapy first) does not affect final outcomes in anaplastic gliomas with combined sequential therapy. IDH wild-type anaplastic gliomas behave essentially like GBM with very poor prognosis; IDH-mutant anaplastic gliomas are a distinct, chemosensitive group with prolonged survival.
Key Limitations
Key Limitations: Retrospective molecular analysis on archival tissue — not all patients had tissue available. NOA-04 predates modern WHO 2021 classification; "anaplastic astrocytoma" in this cohort includes a mix of IDH-mutant grade 3 astrocytoma and IDH wild-type glioma. No concurrent chemoRT arm — the Stupp-backbone approach for IDH-mutant anaplastic glioma was not tested. Limited conclusions on optimal chemotherapy backbone (PCV vs TMZ) due to small arms.
Clinical Context
NOA-04 and its long-term update reinforce that molecular testing (IDH, 1p/19q, MGMT) is essential for treatment planning in anaplastic gliomas. IDH wild-type grade 3 gliomas should be treated aggressively as GBM equivalents. For IDH-mutant grade 3 astrocytoma, CATNON data support RT + adjuvant TMZ as the current standard. NOA-04 remains the basis for the "RT can be deferred in favor of initial chemotherapy" approach in selected IDH-mutant anaplastic gliomas.
References