The value of patient-reported symptoms as prognostic factors for survival in advanced colorectal cancer

Abstract

3585 Background: The study aims to evaluate if patient’s baseline self-reported symptoms independently predict survival in advanced colorectal cancer patients. This study uses data from a recently completed international randomized controlled trial (RCT) conducted by the European Organisation for Research and Treatment of Cancer (EORTC) Gastrointestinal Tract Cancer Group. Methods: The overall sample consisted of 497 patients with previously untreated metastatic colorectal cancer who were randomly assigned to receive high-dose fluorouracil given as a weekly 24-hour infusion with or without leucovorin versus bolus fluorouracil plus leucovorin. Patients’ self reported symptoms at baseline were assessed using the EORTC QLQ-C30, and analyzed as continuous factors on seven pre-selected symptom related scales, scores ranging from 0 to 100. The Cox proportional hazards regression model was used for both univariate and multivariate analyses of survival. A stepwise (back and forward) variable selection procedure was used to identify independent factors prognostic for survival. Key biomedical variables were also included in the analysis. Results: The original RCT failed to show any differences between treatment arms in terms of overall survival (OS) and quality of life outcomes. The prognostic factor analysis was conducted on all available patients with baseline questionnaires (n=299). The final multivariate model identified the following clinical variables: white blood cell count (HR= 1.76; 95% CI=1.293 - 2.403; P<.001), alkaline phosphatase (HR=1.46; 95% CI=1.085–1.96; P=.01), performance status (HR=1.28; 95% CI=1.024–1.61; P=0.03) and number of sites involved (HR=1.09; 95% CI=1.009–1.181; P=.02). In addition to these clinical variables, fatigue also independently contributed to the prognostic information (HR=1.006; 95% CI=1.001–1.011; P=.01), resulting in a hazard increase of 6% for every 10-point shift on the fatigue symptom scale. Conclusion: Patients’ self reported fatigue levels appear to provide additional prognostic information beyond traditional medical data. Such information could help clinical management of individual cancer patients in routine clinical practice. No significant financial relationships to disclose.