CHICAGO – An effort to bolster standard preoperative chemoradiotherapy with capecitabine and oxaliplatin failed to improve surgical outcomes in a four-armed, phase III clinical study of 1,608 patients with stage II and III rectal adenocarcinoma.
Results were similar whether the preoperative regimen contained capecitabine (Xeloda) or continuous venous infusion (CVI) 5-fluorouracil (5-FU); the addition of oxaliplatin (Eloxatin) increased toxicity without changing outcomes, regardless of which regimen was used.
"For the surgical outcome analysis, administration of capecitabine with preoperative radiotherapy achieved rates similar to CVI 5-FU for surgical down-staging, sphincter-saving surgery, and [pathological] complete response," said Dr. Mark S. Roh, who presenting the results of the NSABP (National Surgical Adjuvant Breast and Bowel Project) R-04 trial at the annual meeting of the American Society of Clinical Oncology.
"Addition of oxaliplatin did not improve outcomes and added significant toxicity," added Dr. Roh, chairman of surgery at M.D. Anderson Cancer Center Orlando.
The current standard of care for T3-4 rectal cancer is preoperative chemoradiotherapy with CVI 5-FU. It has a distant relapse rate of 24%-30%, and compliance is suboptimal, he said. The trial was initiated, therefore, in the hope that more effective systemic chemotherapy would improve compliance, decrease the locoregional tumor relapse rate, and increase survival.
"The thought was that giving more effective systemic therapy up front perhaps would decrease the distant relapse rate," said Dr. Roh.
Evaluation of these end points is yet to come, with the final assessments of locoregional tumor relapse and disease-free and overall survival rates to be presented in 2013.
NSABP R-04 was activated in 2004 as a two-arm noninferiority study comparing CVI 5-FU with oral capecitabine, in both cases delivered with preoperative radiotherapy. With the Food and Drug Administration’s approval of oxaliplatin, the trial was amended in 2005 to include it, and the design was changed to a 2 x 2 factorial with the oxaliplatin comparison looking for superiority. During radiation therapy, 5-FU and capecitabine were reduced from 7 to 5 days each week. The trial was closed in August 2010.
All four groups received 4,600 cGy plus 540-1,080 cGy radiation. Chemotherapy dosages were 5-FU (CVI 225 mg/m2 on 5 days per week for 5-6 weeks) in group 1; the same dosage of 5-FU plus oxaliplatin (50 mg/m2 once weekly for 5 weeks) in group 2; oral capecitabine (825 mg/m2 twice daily on 5 days per week for 5-6 weeks) in group 3; and same dose of capecitabine plus oxaliplatin (50 mg/m2 once weekly for 5 weeks) in group 4.
More than half of the patients were stage II. About two-thirds of the population were men. At randomization, sphincter-sparing surgery was the intent in about three-quarters of the cases.
The only significant difference between arms of the study was that oxaliplatin increased grade 3/4 diarrhea (15.4% vs. 6.6% without oxaliplatin; P =.0001).
Surgical down-staging rates were similar in comparisons of 5-FU vs. capecitabine (20.7% and 23%, respectively) and oxaliplatin vs. no oxaliplatin (19.2% and 23%, respectively).
Nor was there any difference in sphincter-saving surgery rates by treatment (61.7% with 5-FU and 62.5% with capecitabine; 60.4% with oxaliplatin and 63.6% without oxaliplatin).
And there was no statistically significant difference in R-04 node positivity among the various treatment groups. The rates were about 70% with no positive nodes and about 90% with 0-3 positive nodes regardless of treatment?
Surgical complication rates also did not differ significantly among the groups; the rate of any complication was 34.92% with 5-FU, 37% with 5-FU plus oxaliplatin, 36.89% with capecitabine, and 40.2% with capecitabine plus oxaliplatin.
For pathological complete response, outcomes also were similar at 22.2% with capecitabine and 18.8% with 5-FU; they were 20.9% with oxaliplatin, and 19.1% without.
Discussant Dr. Robert Glynne-Jones of the Mount Vernon Centre for Cancer Treatment in Northwood, England, praised the study and its 2 x 2 factorial design as "a very nice study, very large numbers." He said the design allowed investigators to look at individual components and "see which is doing the work." He also said that this study would not change his clinical practice.
Dr. Roh stated that he had no relevant relationships to disclose. Dr. Glynne-Jones disclosed research funding from Roche.