Purpose: To examine the predictive value of baseline serum bilirubin measurement for chemotherapy-related toxicity or efficacy among patients receiving irinotecan for metastatic colorectal cancer.
Methods: We performed a secondary analysis of a cohort of 287 patients treated in a multicenter, phase III study with single-agent irinotecan administered either weekly or once every 3 weeks. Patients were grouped into three categories of baseline bilirubin measurements (0 to 0.4, 0.5 to 0.9, and 1.0 to 1.5 mg/dL). We performed analyses of overall survival, time to progression, and treatment-related toxicity based on bilirubin category, as well as using bilirubin as a continuous variable.
Results: With a median follow-up of 15.8 months, baseline serum bilirubin was not predictive of 1-year survival (42.4%, bilirubin 0 to 0.4; 42.3%, bilirubin 0.5 to 0.9; 48.1%, bilirubin 1.0 to 1.5 mg/dL), median overall survival (10.1, 9.7, and 15.6 months, respectively; P =.5), or median time to progression (2.8, 3.0, and 4.1 months, respectively; P =.5). Patients with elevated bilirubin had a significantly greater risk grade 3 to 4 neutropenia; however, this was limited to patients treated on a weekly schedule (P trend =.03) and not once every 3 weeks (P trend =.8). Other toxicities were not significantly different by initial bilirubin measurement.
Conclusion: Although modest elevations of bilirubin (1.0 to 1.5 mg/dL) are associated with increased grade 3 to 4 neutropenia in patients treated with weekly irinotecan, baseline serum bilirubin does not reliably predict overall irinotecan-related toxicity or efficacy. Additional methods, including potential application of pharmacogenetic information, are needed to optimize irinotecan dosing and tailor therapy to individual patients.