The relationship of presenting symptom burden and survival in patients with head and neck cancer (HNC).

Abstract

5600 Background: Previous studies have shown that patient-reported outcomes with QOL instruments are predictive of survival. We investigate the relationship between HNC patient-reported symptoms and survival in addition to clinician ratings of ECOG performance status (PS). Methods: 129 patients with HNC participated in the study. Patients completed the M. D. Anderson Symptom Inventory - Head and Neck module (MDASI-HN) prior to radiotherapy or chemoradiotherapy. Cluster analysis was used to identify those who were symptomatic due to their disease prior to treatment. Cox regression was used to examine differences in survival between symptom burden groups with PS as covariate. Logistic regression was used to determine the relationship of overall survival status with symptom burden and performance status. Results: 46% (60/129) of HNC patients were identified to have symptoms prior to therapy. The top symptoms at presentation were fatigue, emotional distress, pain, disturbed sleep, drowsiness and sadness. Under Cox regression, symptom burden (OR=2.3, 95% CI: 1.1, 4.9, p<0.032) and performance status (OR=4.4, 95% CI: 2.1, 9.0, p<0.001) are significant predictors of survival. Under logistic regression, symptom burden (OR=3.2, 95% CI: 1.3, 7.9, p<0.011) and performance status (OR=7.0, 95% CI: 2.4, 20.4, p<0.001) are significant predictors of survival status. The logistic model with both performance status and symptom burden as predictors of survival showed better model fit than either predictor alone. Conclusions: Half of HNC patients are symptomatic prior to radiation therapy, but most have mild symptom severity. However, patients with high pretreatment symptom burden have poorer survival outcomes, and the combination of PS and symptom burden is a better predictor than either alone. No significant financial relationships to disclose.