Analysis for prognostic factors of 60-day mortality: Evaluation of an irinotecan-based phase III trial performed in the first-line treatment of metastatic colorectal cancer

Background: Mortality rates in published irinotecan-based trials range between 1.7% and 5.0%. This analysis aimed to evaluate clinical and histopathologic factors associated with 60-day mortality in first-line therapy for metastatic colorectal cancer (mCRC). Patients and Methods: Sixty-day all-cause and disease-specific mortality rates from 479 patients who participated in a randomized phase III study comparing FUFIRI (5-fluorouracil [5-FU], leucovorin, irinotecan) (n = 238) vs. mIROX (modified irinotecan plus oxaliplatin (n = 241) were evaluated for association with prognostic factors such as platelet counts, alkaline phosphatase (AP) levels, white blood cell (WBC) counts, hemoglobin values, lactate dehydrogenase (LDH) levels, carcinoembryonic antigen (CEA) levels, and several other baseline parameters using univariate and multivariate logistic regression analyses applied to patients combined from both treatment groups. Results: The all-cause 60-day mortality rate was 5.0% (24/479). Thirteen patients (5.5%) in the FUFIRI arm died within the first 60 days of treatment compared with 11 (4.6%) patients in the mIROX arm (P =.68). Among the 24 patients in both treatment arms, mortality was qualified as disease related in 15 (63%) patients and treatment related in 7 (29%) patients (P =.695). In multivariate analyses, high LDH levels (P =.010) and an elevated WBC count (P =.006) remained as significant independent prognostic factors. Low Karnofsky performance status (KPS) showed a strong trend but failed to reach statistical significance (P =.057) as did AP levels and the number of metastatic sites. Conclusion: In this study 63% of the early deaths were disease related, whereas only 29% were possibly related to study medication. Independent prognostic factors for early mortality were LDH levels and WBC counts. KPS showed a strong trend in the multivariate analysis. Future investigation may consider LDH levels and WBC counts for exclusion criteria.