Background
Phase III randomized multicenter trial. 364 patients with resectable colorectal liver metastases (≤4 metastases per liver, no extrahepatic disease on CT, eligible for curative-intent resection). Conducted by EORTC at 78 centers in 11 European countries. Compared perioperative chemotherapy with FOLFOX4 (6 cycles before + 6 cycles after liver resection) to surgery alone, based on the hypothesis that perioperative chemotherapy could treat micrometastatic disease and downsize borderline-resectable lesions.
Interventions and follow up
Arm A: Perioperative FOLFOX4: 6 cycles (3 months) → liver resection → 6 cycles (3 months) adjuvant FOLFOX4 (oxaliplatin 85 mg/m² day 1 + leucovorin 200 mg/m² + 5-FU 400 mg/m² bolus + 600 mg/m² 22-h CI days 1–2, q2wk)
Arm B: Surgery alone (upfront liver resection without perioperative chemotherapy)
Primary endpoint: Progression-free survival (PFS) — in all eligible patients who underwent randomizatio
mFollow up: 3.9 year
Arm B: Surgery alone (upfront liver resection without perioperative chemotherapy)
Primary endpoint: Progression-free survival (PFS) — in all eligible patients who underwent randomizatio
mFollow up: 3.9 year
Results
3-yr PFS (all randomized): 35.4% vs 28.1%, HR 0.79 (95% CI 0.62–1.02), P=.058
3-yr PFS (eligible resected patients): 36.2% vs 28.1%, HR 0.73 (95% CI 0.55–0.97), P=.025
mPFS (eligible patients): 18.7 vs 11.7 months
OS (5-yr, updated): 51% vs 48%, HR 0.88 (95% CI 0.68–1.14), P=.34
R0 resection rate: 83% (perioperative) vs 84% (surgery alone)
Surgical complications: 25% vs 16% (more with perioperative FOLFOX)
3-yr PFS (eligible resected patients): 36.2% vs 28.1%, HR 0.73 (95% CI 0.55–0.97), P=.025
mPFS (eligible patients): 18.7 vs 11.7 months
OS (5-yr, updated): 51% vs 48%, HR 0.88 (95% CI 0.68–1.14), P=.34
R0 resection rate: 83% (perioperative) vs 84% (surgery alone)
Surgical complications: 25% vs 16% (more with perioperative FOLFOX)
Adverse events
Chemotherapy-related (grade ≥3): 11% pre-op, 23% post-op; neutropenia 7%, fatigue 7%, sensory neuropathy 3%
Surgical: Higher postoperative complication rate with perioperative FOLFOX (25% vs 16%, P=.04), primarily due to reversible hepatic sinusoidal obstruction syndrome; no increase in 60-day mortality
Surgical: Higher postoperative complication rate with perioperative FOLFOX (25% vs 16%, P=.04), primarily due to reversible hepatic sinusoidal obstruction syndrome; no increase in 60-day mortality
Conclusions
Perioperative FOLFOX4 produced a significant PFS benefit in eligible resected patients (HR 0.73, P=.025) and a borderline-significant benefit in the intent-to-treat population (HR 0.79, P=.058). No significant OS benefit was demonstrated at 5 years. Higher surgical complication rates with perioperative chemotherapy are an important consideration.
Key Limitations
Key Limitations: Primary endpoint (ITT PFS) did not meet statistical significance (P=.058) — the trial was arguably underpowered. Long-term OS benefit was not demonstrated (HR 0.88, P=.34). Higher surgical complication rate (25% vs 16%) with perioperative FOLFOX reflects hepatic sinusoidal injury from oxaliplatin. Modern surgical techniques and selective use of perioperative chemotherapy may attenuate this risk. No KRAS/RAS testing — the trial predates molecular-guided selection. Post-operative completion of chemotherapy was suboptimal (only 40% completed all 6 post-op cycles), reducing the absolute benefit of the strategy.
Clinical Context
EORTC 40983 established perioperative FOLFOX (pre- + post-operative) as a standard option for resectable CRC liver metastases, informing ESMO guidelines. However, subsequent evidence — including meta-analyses and the CHARISMA trial — has raised questions about whether perioperative chemotherapy vs surgery-first followed by adjuvant chemotherapy confers superior outcomes. Current practice varies by institutional preference; ESMO supports either perioperative or adjuvant FOLFOX for resectable CRLM, with a preference for systemic therapy in synchronous or high-risk disease.