Randomized multicenter phase II trial of oxaliplatin plus irinotecan versus raltitrexed as first-line treatment in advanced colorectal cancer.

Purpose: Irinotecan and oxaliplatin are two new agents with promising activity in advanced colorectal cancer. Based on preclinical and clinical evidence that both drugs act synergistically, a randomized phase II study was initiated to investigate the therapeutic potential and tolerance of this combination in the front-line setting.

Patients And Methods: Ninety-two patients with previously untreated, measurable disease were randomized to receive biweekly oxaliplatin 85 mg/m(2) plus irinotecan 175 mg/m(2) or raltitrexed 3 mg/m(2) given on day 1 every 3 weeks. Upon development of progressive disease, second-line treatment with the opposite arm was effected.

Results: Patients allocated to oxaliplatin/irinotecan had a significantly better radiologically confirmed response rate (43.5% v 19.6%; P =.0025) and longer progression-free survival (median, 7.1 v 5.0 months; P =.0033). Improvement in overall survival, however, did not reach the level of significance (median, 16.0 v 16.5 months; P =.3943). The response rate after cross-over was 33.3% (eight of 24) for assessable patients treated with oxaliplatin/irinotecan compared with 14.2% (three of 21) for those treated with second-line raltitrexed. Oxaliplatin/irinotecan caused more hematologic and gastrointestinal toxicities, necessitating dose reductions in 10 of the first 20 patients. After adjustment of the irinotecan starting dose from 175 to 150 mg/m(2), tolerance of treatment was acceptable; the most commonly encountered events (all grades) were neutropenia (81%), alopecia (65%), nausea/emesis (62%), peripheral sensory neuropathy (62%), and diarrhea (46%).

Conclusion: Oxaliplatin/irinotecan seems beneficial as first-line therapy in advanced colorectal cancer, with an acceptable toxicity profile at the reduced irinotecan dose level. Its promising therapeutic potential is supported by the high response activity noted in the raltitrexed control arm after cross-over, which may also explain the lack of a difference in overall survival.