Laryngeal long-term morbidity after supraglottic laryngectomy and postoperative radiation therapy

Abstract

This study was performed to investigate factors associated with laryngeal morbidity when postoperative radiation therapy (RT) is added to supraglottic laryngectomy. From 1980 to 1994, 56 patients affected with T1 to 4 N0 to 2c supraglottic squamous cell carcinoma selected for standard (59%) or extended (41%) supraglottic laryngectomy at 2 different institutions were retrospectively analyzed. Most of the patients (91%) also underwent neck dissection. Approximately 80% of the patients had stage T4 primary lesions or N2 neck disease. Postoperative RT was added for presumed microscopic disease at the primary site (13 patients), regional nodes (23 patients), or both (20 patients). Median delivered doses to the larynx and to the neck were 50 Gy (range, 40 to 64 Gy) and 46 Gy (range, 40 to 64 Gy), respectively. Median follow-up for living patients is 11 years (range, 2.8 to 16.9 years). Laryngeal complication was defined as the appearance of grade 2 or higher toxicity according to the European Organization for Research and Treatment of Cancer (EORTC) and the Radiation Therapy Oncology Group (RTOG) scoring systems. Two- and 5-year actuarial locoregional control rates were 85+/-5% and 83+/-5%, respectively. Thirty patients (54%) developed laryngeal complications. However, just one patient experienced grade 4 laryngeal oedema requiring permanent tracheostomy. Estimated actuarial survival without laryngeal complications were 50+/-7%, 43+/-7%, and 39+/-7% at 2, 5, and 10 years, respectively. At univariate analysis, treated volumes (P = .03) and total dose to the larynx (P = .03) were significantly associated with local toxicity. A trend was observed also for the maximum dose to the neck (P = .06) and dose per fraction (P = .09). A multivariate Cox proportional hazards model showed total dose to the larynx to be the only independent predictor of toxicity (P = .03). The hazard ratio of laryngeal toxicity was 2.2 (95% confidence interval: 1.1/4.6), for a total dose to the larynx greater than 50 Gy. After supraglottic laryngectomy, postoperative RT to the neck does not affect local morbidity, but careful RT treatment planning is necessary to avoid delivering a total dose to the larynx greater than 50 Gy.