Why to start the concomitant boost in accelerated radiotherapy for advanced laryngeal cancer in week 3

Abstract

We analyzed toxicity and the local control rates for advanced laryngeal cancer, treated with two accelerated fractionation schedules. The main difference between the schedules was the onset of the concomitant boost, in Week 3 or Week 4. Overall treatment time and total dose were equivalent. In a prospective, nonrandomized study of T3, T4, and advanced T2 laryngeal cancer, concomitant boost schedules were used in 100 patients. Thirty patients received a schedule of twice daily 1.2 Gy in Weeks 1-3, followed by twice daily 1.7 Gy in Weeks 4 and 5; total dose was 70 Gy (the hyperfractionated accelerated schedule [HAS] regimen). Seventy patients were treated with 5 times 2 Gy in Weeks 1 and 2, followed by daily 1.8 Gy and 1.5 Gy (boost) in Weeks 3-5; total dose 69.5 Gy (the accelerated schedule only [ASO] regimen). Distribution of T stage was 47%, 40%, and 12% for T2, T3, and T4, respectively. In 24% of the patients, lymph nodes were positive. Pretreatment tracheotomy or stridor or both occurred in 8 patients. The distribution of prognostic factors was not significantly different between the two fractionation schedules. Acute and late toxicity was assessed. Results were estimated by the use of actuarial methods. For late toxicity and local control univariate and multivariate analyses were performed. Tumor control probability analysis was used to model cure rate differences. Overall acute mucositis score was equal for both schedules. Acute mucositis started and decreased significantly earlier in the HAS regimen. In all patients acute mucositis healed completely. The treatment was completed within 38 days in all patients. The regional control rate was 100% for clinical N0, and 75% for the clinical N+ patients. The 3-year local control rate was 59% and 78% for the HAS and ASO regimens, respectively (p = 0.05); the ultimate local control was 80% and 94%, respectively. In multivariate analysis, besides the fractionation schedule (relative risk [RR], 2.6 for HAS vs. ASO), pretreatment tracheotomy/stridor (RR 4.3, yes vs. no), and local tumor response 3-6 weeks after radiotherapy (RR 5.1, no vs. yes) were independent factors for local control. Tumor control probability analysis indicated that the onset of repopulation may be about 4-6 days earlier for the HAS regimen. The onset of repopulation in the HAS regimen is probably at the end of the second week or at the beginning of the third week. Severe late toxicity was observed in the HAS group and ASO group in, respectively, 11% and 16%. In multivariate analysis this toxicity related significantly to the field size and pretreatment tracheotomy/stridor. In our study the timing of the boost in accelerated radiotherapy for advanced laryngeal cancer was an independent factor for local control, favoring the use of a concomitant boost in Week 3. This finding may indicate that accelerated repopulation of tumor cells starts early in the treatment phase.