Purpose: To review and assign attribution for the causes of early deaths on two National Cancer Institute-sponsored cooperative group studies involving irinotecan and bolus fluorouracil (5-FU) and leucovorin (IFL).
Patients and methods: The inpatient, outpatient, and research records of patients treated on Cancer and Leukemia Group B protocol C89803 and on North Center Cancer Treatment Group protocol N9741 were reviewed by a panel of five medical oncologists not directly involved with either study. Each death was categorized as treatment-induced, treatment-exacerbated, or treatment-unrelated.
Results: The records of 44 patients who experienced early deaths on C89803 (21 patients) or N9741 (23 patients) were reviewed. Patients treated with irinotecan plus bolus 5-FU/leucovorin had a three-fold higher rate of treatment-induced or treatment-exacerbated death than patients treated on the other arm(s) of the respective studies. For C89803, these rates were 2.5% (16 of 635) for IFL versus 0.8% (five of 628) for bolus weekly 5-FU and leucovorin. For N9741, these rates were 3.5% (10 of 289) for IFL, 1.1% (three of 277) for oxaliplatin plus bolus and infusional 5-FU and leucovorin, and 1.1% (three of 275) for oxaliplatin plus irinotecan. Multiple gastrointestinal toxicities that often occurred together were characterized into a gastrointestinal syndrome. Sudden, unexpected thromboembolic events were characterized as a vascular syndrome. The majority of deaths in both studies were attributable to one or both of these syndromes.
Conclusion: Close clinical monitoring, early recognition of toxicities and toxicity syndromes, aggressive therapeutic intervention, and withholding therapy in the presence of unresolved drug-related toxicities is recommended for patients receiving IFL or other intensive chemotherapy regimens.