A prospective randomized trial of hyperfractionated versus conventional once daily radiation for advanced squamous cell carcinomas of the larynx and pharynx

Purpose: To examine the effects of an increased dose of radiation therapy (RT) delivered by a hyperfractionated schedule compared to conventional once daily RT on toxicity, locoregional control and survival in the treatment of squamous cell carcinomas (SCC) of the larynx and pharynx. Materials/Methods: Between 1988 and 1995 336 patients were randomized to receive RT with curative intent. Eligible patients had biopsy proven SCC of the larynx or pharynx, with TN stages (AJC-UICC 1987) T3 or T4 N0, or any T with any N+. All patients were M0. Patients were stratified by site (larynx/oropharynx/hypopharynx), node status (clinically positive/negative), and performance status. Patients were treated with either 51 Gy TAD/20 fractions/4 wk (2.55 Gy 1x/d, conventional RT=CRT) or 58 Gy TAD/40 fractions/4 wk (1.45 Gy 2x/d, hyperfractionated RT=HFRT). Patients underwent EUA and selective biopsies 10 wk after RT; surgical salvage was performed for residual or recurrent cancer whenever possible. Results: The primary cancer arose in the oropharynx (138), larynx (133) or hypopharynx (65). T stages were distributed T1 22, T2 72, T3 133 and T4 109. N stage distribution was N0 127, N1 74, N2 117 and N3 18. The proportion of patients with acute mucosal toxicity (RTOG Grade 3 or 4) was increased by HFRT (60% versus 40%), but other acute and late toxicity was not significantly different. There was no difference in the incidence of morbidity in the two treatment groups in those who underwent surgery following RT. The locoregional control rates at 3 yrs for all cases were 45% (HFRT) vs 40% (CRT) log rank p=0.16; for primary tumors <4 cm 54% (HFRT) vs 42% (CRT) p=0.04; for primary tumors ≥ 4 cm 38% (HFRT) vs 41% (CRT) p=0.73. Local control was improved to some extent in SCC which arose in all sites, but most noticeably in hypopharyngeal cancers. The disease free survival rates at 3 yr for all cases were 37% (HFRT) vs 30% (CRT) p=0.15; for primary tumors < 4 cm 47% (HFRT) vs 34% (CRT) p=0.10; for primary tumours ≥ 4 cm 31% (HFRT) vs 29% (CRT) p=0.88. Conclusion: The HFRT schedule used was tolerable and resulted in significantly improved locoregional control at 3 yr in patients with primary SCC less than 4 cm in size. Followup continues to determine whether this advantage is maintained and to monitor late toxicity