Predictors of severe acute and late toxicities in patients with localized head-and-neck cancer treated with radiation therapy

Abstract

Purpose: Radiation therapy (RT) causes acute and late toxicities that affect various organs and functions. In a large cohort of patients treated with RT for localized head and neck cancer (HNC), we prospectively assessed the occurrence of RT-induced acute and late toxicities and identified characteristics that predicted these toxicities.

Methods and materials: We conducted a randomized trial among 540 patients treated with RT for localized HNC to assess whether vitamin E supplementation could improve disease outcomes. Adverse effects of RT were assessed using the Radiation Therapy Oncology Group Acute Radiation Morbidity Criteria during RT and one month after RT, and the Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer Late Radiation Morbidity Scoring Scheme at six and 12 months after RT. The most severe adverse effect among the organs/tissues was selected as an overall measure of either acute or late toxicity. Grade 3 and 4 toxicities were considered as severe. Stepwise multivariate logistic regression models were used to identify all independent predictors (p < 0.05) of acute or late toxicity and to estimate odds ratios (OR) for severe toxicity with their 95% confidence intervals (CI).

Results: Grade 3 or 4 toxicity was observed in 23% and 4% of patients, respectively, for acute and late toxicity. Four independent predictors of severe acute toxicity were identified: sex (female vs. male: OR = 1.72, 95% confidence interval [CI]: 1.06-2.80), Karnofsky Performance Status (OR = 0.67 for a 10-point increment, 95% CI: 0.52-0.88), body mass index (above 25 vs. below: OR = 1.88, 95% CI: 1.22-2.90), TNM stage (Stage II vs. I: OR = 1.91, 95% CI: 1.25-2.92). Two independent predictors were found for severe late toxicity: female sex (OR = 3.96, 95% CI: 1.41-11.08) and weight loss during RT (OR = 1.26 for a 1 kg increment, 95% CI: 1.12-1.41).

Conclusions: Knowledge of these predictors easily collected in a clinical setting could help tailoring therapies to reduce toxicities among patients treated with RT for HNC.