Background
Phase III RCT (CCG A9961). 421 children (3–21 years) with standard-risk medulloblastoma. All received CSI 23.4 Gy + posterior fossa boost 55.8 Gy, then randomized to one of two maintenance chemotherapy regimens: CCNU-based vs carboplatin-based. Goal: determine if carboplatin substitution for CCNU improves outcomes or reduces toxicity.
Interventions and follow up
Arm A: CSI 23.4 Gy + PF boost 55.8 Gy → Maintenance: CCNU + vincristine + cisplatin (standard arm, 8 cycles)
Arm B: CSI 23.4 Gy + PF boost 55.8 Gy → Maintenance: Carboplatin + vincristine + cisplatin (experimental arm, 8 cycles)
Primary endpoint: EFS
mFollow up: 5.0 year
Arm B: CSI 23.4 Gy + PF boost 55.8 Gy → Maintenance: Carboplatin + vincristine + cisplatin (experimental arm, 8 cycles)
Primary endpoint: EFS
mFollow up: 5.0 year
Results
5-yr EFS (CCNU arm): 81% vs carboplatin arm 79%, HR 1.07, P=.68 — no difference
5-yr OS: 86% (CCNU) vs 85% (carboplatin), P=.83
Both arms exceeded historical outcomes: confirming superiority over RT alone
5-yr OS: 86% (CCNU) vs 85% (carboplatin), P=.83
Both arms exceeded historical outcomes: confirming superiority over RT alone
Adverse events
Main adverse events: CCNU arm: grade ≥2 ototoxicity 25%, myelosuppression. Carboplatin arm: similar ototoxicity from cisplatin, comparable myelosuppression. No significant difference in toxicity between arms. Neurocognitive decline on serial IQ testing in both arms.
Conclusions
Carboplatin substitution for CCNU provided equivalent EFS and OS in standard-risk medulloblastoma. Both regimens produced excellent outcomes (~80% 5-yr EFS) confirming CSI 23.4 Gy + boost + adjuvant chemotherapy as effective standard of care.
Key Limitations
Key Limitations: No molecular subgrouping. All patients received identical RT — no de-escalation arm. Long-term ototoxicity, neurocognition, and secondary malignancy risk were not fully evaluated. Cisplatin was present in both arms, making ototoxicity comparison between CCNU and carboplatin somewhat confounded.
Clinical Context
CCG A9961 validated the standard regimen (CSI 23.4 Gy + boost + vincristine/CCNU/cisplatin) and confirmed similar outcomes with carboplatin. The CCNU-based regimen remains widely used. Current investigations focus on molecular subgroup-directed therapy: WNT-activated tumors may tolerate reduced CSI; Group 3 (MYC-amplified) needs intensification.
References
References: Packer RJ et al, JCO 2006 (CCG A9961)