Background
Phase III RCT in 1,608 patients with stage II (T3-4N0) or III (T1-4N1-2) rectal cancer, distal border <12 cm from the anal verge, undergoing neoadjuvant chemoradiation (45 Gy in 25 fractions plus a boost) with TME, using a 2x2 factorial design to test the fluoropyrimidine backbone and the addition of oxaliplatin.
Interventions and follow up
Arm A: 5-FU 225 mg/m²/d continuous infusion 7d/wk during RT
Arm B: 5-FU + oxaliplatin 50 mg/m² IV weekly x5 during RT
Arm C: Capecitabine 825 mg/m² BID 7d/wk during RT
Arm D: Capecitabine + oxaliplatin 50 mg/m² IV weekly x5 during RT
Primary endpoint: Local-regional control (also pCR, sphincter-sparing surgery, surgical downstaging)
Median follow up: NR
Arm B: 5-FU + oxaliplatin 50 mg/m² IV weekly x5 during RT
Arm C: Capecitabine 825 mg/m² BID 7d/wk during RT
Arm D: Capecitabine + oxaliplatin 50 mg/m² IV weekly x5 during RT
Primary endpoint: Local-regional control (also pCR, sphincter-sparing surgery, surgical downstaging)
Median follow up: NR
Results
pCR (oxaliplatin vs no oxaliplatin): 19.5% vs 17.8%, P=.42
Sphincter-sparing surgery: 57.8% vs 61.0%, P=.24
Surgical downstaging: 17.9% vs 23.5%, P=.20
5-FU vs capecitabine backbone: No significant difference in pCR or downstaging
3-yr local-regional recurrence (R0 resection): 2.9-4.6% across arms, no significant difference
Sphincter-sparing surgery: 57.8% vs 61.0%, P=.24
Surgical downstaging: 17.9% vs 23.5%, P=.20
5-FU vs capecitabine backbone: No significant difference in pCR or downstaging
3-yr local-regional recurrence (R0 resection): 2.9-4.6% across arms, no significant difference
Adverse events
GI (grade 3-5 diarrhea): 16.5% with oxaliplatin vs 6.9% without
Neurologic: Neuropathy 5.6% vs 0.6% (5-FU vs 5-FU+oxaliplatin); 2.2% vs 6.5% (capecitabine vs capecitabine+oxaliplatin)
Neurologic: Neuropathy 5.6% vs 0.6% (5-FU vs 5-FU+oxaliplatin); 2.2% vs 6.5% (capecitabine vs capecitabine+oxaliplatin)
Conclusions
Adding oxaliplatin to neoadjuvant fluoropyrimidine chemoradiation did not improve pCR, sphincter preservation, or downstaging but significantly increased toxicity. Capecitabine and continuous-infusion 5-FU were equivalent backbones.
Key Limitations
Surrogate surgical endpoints rather than survival as primary. Oxaliplatin added by protocol amendment, creating staggered enrollment. Concordant negative oxaliplatin results across STAR-01 and ACCORD 12 confirm the finding.
Clinical Context
R-04 established that oxaliplatin should not be added to neoadjuvant chemoradiation for rectal cancer as a radiosensitizer, with capecitabine an acceptable substitute for infusional 5-FU. Consistent with ESMO recommendations. Earlier work (O’Connell, NEJM 1994) showed infusional 5-FU superior to bolus during adjuvant RT.
References
O’Connell MJ et al, JCO, 2014, PMID: 24799484
STAR-01: Aschele C et al, JCO, 2011, PMID: 21606427
ACCORD 12/0405: Gérard JP et al, JCO, 2010, PMID: 20194850
Infusional vs bolus 5-FU: O’Connell MJ et al, NEJM, 1994, PMID: 8041415
STAR-01: Aschele C et al, JCO, 2011, PMID: 21606427
ACCORD 12/0405: Gérard JP et al, JCO, 2010, PMID: 20194850
Infusional vs bolus 5-FU: O’Connell MJ et al, NEJM, 1994, PMID: 8041415