Background
Phase III RCT (EORTC 22921). 1,011 patients with T3-T4 resectable rectal cancer randomized in a 2×2 factorial design to: preoperative RT (45 Gy) alone vs preoperative CRT, and no adjuvant chemotherapy vs adjuvant chemotherapy (four 5-FU/LV cycles after surgery). TME not mandated.
Interventions and follow up
Arm A: Preoperative RT 45 Gy in 25 fractions (no chemotherapy)
Arm B: Preoperative CRT: 45 Gy + 5-FU 350 mg/m² + LV 20 mg/m² days 1–5 and 29–33
Primary endpoint: Overall survival
mFollow up: 5.4 years (initial), 10.4 years (updated)
Arm B: Preoperative CRT: 45 Gy + 5-FU 350 mg/m² + LV 20 mg/m² days 1–5 and 29–33
Primary endpoint: Overall survival
mFollow up: 5.4 years (initial), 10.4 years (updated)
Results
±: Adjuvant chemotherapy: 5-FU 350 mg/m² + LV 20 mg/m² × 4 cycles (all 4 arms)
pCR (CRT arms): 13.7% vs 5.3% (RT alone), P<.001
5-yr local recurrence (CRT): 8.7% vs 17.1% (RT alone), P<.001
5-yr OS: ~65% all arms — no significant OS difference between any arms
Adjuvant chemo: No significant OS, DFS, or local control benefit
pCR (CRT arms): 13.7% vs 5.3% (RT alone), P<.001
5-yr local recurrence (CRT): 8.7% vs 17.1% (RT alone), P<.001
5-yr OS: ~65% all arms — no significant OS difference between any arms
Adjuvant chemo: No significant OS, DFS, or local control benefit
Adverse events
Main adverse events: CRT grade ≥3 acute toxicity: 13.9% vs 2.5% (RT alone). Grade ≥3 postoperative complications higher with preoperative CRT. No treatment-related deaths reported.
Conclusions
Preoperative CRT significantly improved pathologic complete response rates and local control compared to RT alone, but did not improve OS. Adjuvant chemotherapy after surgery provided no benefit in any arm. Established that the benefit of concurrent chemotherapy is limited to local control, not survival.
Key Limitations
Key Limitations: TME was not universally mandated — compromising surgical quality control. 2×2 factorial design with no single arm as current standard of care comparator. Adjuvant chemotherapy (5-FU/LV alone) was not adequate — modern oxaliplatin-based adjuvant therapy was not tested. No biologic agents tested. Median OS not significantly different across 4 arms — may be underpowered or follow-up too short to detect survival differences.
Clinical Context
EORTC 22921 established preoperative CRT over RT alone for local control. However, standard of care has evolved — TME is now mandatory, and modern regimens include oxaliplatin (ACCORD 12, PETACC-6) or total neoadjuvant therapy (TNT). The lack of adjuvant chemo benefit prompted debate resolved by newer studies (ADORE, capox adjuvant data).
References
References: Bosset JF et al, N Engl J Med 2006 (EORTC 22921)