Background
Prospective pooled analysis of individual patient data from 6 randomized trials (TOSCA, SCOT, IDEA France, ACHIEVE, HORG, Alliance/SWOG 80702). 12,834 patients with stage III colon cancer treated with 3 vs 6 months of adjuvant oxaliplatin-based chemotherapy (FOLFOX or CAPOX). Pre-specified non-inferiority margin: HR ≤1.12 for 3-year DFS.
Interventions and follow up
Arm A: 3 months of FOLFOX (oxaliplatin 85mg/m² + 5-FU/LV q2wk × 6 cycles) or CAPOX (oxaliplatin 130mg/m² + capecitabine 1000mg/m² BID D1–14 q3wk × 4 cycles)
Arm B: 6 months of FOLFOX (12 cycles) or CAPOX (8 cycles)
Primary endpoint: 3-year DFS (non-inferiority, HR ≤1.12)
mFollow up: 72 mo
Arm B: 6 months of FOLFOX (12 cycles) or CAPOX (8 cycles)
Primary endpoint: 3-year DFS (non-inferiority, HR ≤1.12)
mFollow up: 72 mo
Results
DFS (overall, 3 vs 6 mo): HR 1.07, 95% CI 1.00–1.15 — overall non-inferiority NOT proven
DFS (CAPOX, T1–3/N1 low-risk): HR 0.85 — non-inferior
DFS (CAPOX, T4 or N2 high-risk): HR 1.02 — non-inferior
DFS (FOLFOX, low-risk): HR 1.10 — non-inferior borderline
DFS (FOLFOX, high-risk): HR 1.20 — NOT non-inferior
Neuropathy grade ≥2 at end of treatment: 16% vs 48% (3 vs 6 mo)
DFS (CAPOX, T1–3/N1 low-risk): HR 0.85 — non-inferior
DFS (CAPOX, T4 or N2 high-risk): HR 1.02 — non-inferior
DFS (FOLFOX, low-risk): HR 1.10 — non-inferior borderline
DFS (FOLFOX, high-risk): HR 1.20 — NOT non-inferior
Neuropathy grade ≥2 at end of treatment: 16% vs 48% (3 vs 6 mo)
Adverse events
Neurologic: Grade ≥3 neuropathy at end of treatment 3% vs 16% (3 vs 6 mo) — near-halving of clinically significant peripheral neuropathy with shorter duration
GI/hematologic: Nausea/vomiting grade ≥3 and myelosuppression both lower with 3 months
GI/hematologic: Nausea/vomiting grade ≥3 and myelosuppression both lower with 3 months
Conclusions
Three months of CAPOX is non-inferior to 6 months in stage III colon cancer across risk strata, while 3 months of FOLFOX is not clearly non-inferior in high-risk disease (T4 or N2). These findings established a treatment duration standard: 3-month CAPOX for all stage III, 6-month FOLFOX for high-risk (T4/N2) disease, substantially reducing long-term neuropathy without sacrificing efficacy in most patients.
Key Limitations
Key Limitations: Pooled analysis using pre-specified methodology rather than a single RCT introduces heterogeneity across trial designs, entry criteria, and FOLFOX/CAPOX allocation (allocation was not randomized — choice was per trial or physician). The overall non-inferiority boundary was not met for the entire cohort; conclusions are regimen- and risk-subgroup-specific. The interaction between regimen choice (CAPOX vs FOLFOX) and duration effect was not pre-specified in all contributing trials. Results in stage II high-risk disease are extrapolated from stage III data.
Clinical Context
IDEA established the current practice standard of 3 months of CAPOX (or cautiously 3 months of FOLFOX in low-risk T1-3/N1) for stage III colon cancer. ESMO guidelines recommend 3-month CAPOX for low-risk stage III (T1-3/N1) and 6-month FOLFOX or 3- or 6-month CAPOX for high-risk (T4/N2). The reduction in peripheral neuropathy with shorter treatment is a major patient benefit, particularly given that oxaliplatin neuropathy can persist for years post-treatment.
References