Once-a-Week Versus Once-Every-3-Weeks Cisplatin Chemoradiation for Locally Advanced Head and Neck Cancer: A Phase III Randomized Noninferiority Trial
Mené sur 279 patients atteints d'un carcinome épidermoïde de la tête et du cou de stade localement avancé (durée médiane de suivi : 22 mois), cet essai de non infériorité de phase III compare l'efficacité, du point de vue du contrôle locorégional de la maladie, et la toxicité de deux protocoles de chimioradiothérapie se différenciant par la fréquence d'administration du cisplatine (une fois par semaine ou une fois toutes les trois semaines)
Purpose : Chemoradiation with cisplatin 100 mg/m2 given once every 3 weeks is the standard of care in locally advanced head and neck squamous cell cancer (LAHNSCC). Increasingly, low-dose once-a-week cisplatin is substituted because of perceived lower toxicity and convenience. However, there is no level 1 evidence of comparable efficacy to cisplatin once every 3 weeks. Patients and Methods : In this phase III randomized trial, we assessed the noninferiority of cisplatin 30 mg/m2 given once a week compared with cisplatin 100 mg/m2 given once every 3 weeks, both administered concurrently with curative intent radiotherapy in patients with LAHNSCC. The primary end point was locoregional control (LRC); secondary end points included toxicity, compliance, response, progression-free survival, and overall survival. Results : Between 2013 and 2017, we randomly assigned 300 patients, 150 to each arm. Two hundred seventy-nine patients (93%) received chemoradiotherapy in the adjuvant setting. At a median follow-up of 22 months, the estimated cumulative 2-year LRC rate was 58.5% in the once-a-week arm and 73.1% in the once-every-3-weeks arm, leading to an absolute difference of 14.6% (95% CI, 5.7% to 23.5%); P = .014; hazard ratio (HR), 1.76 (95% CI, 1.11 to 2.79). Acute toxicities of grade 3 or higher occurred in 71.6% of patients in the once-a-week arm and in 84.6% of patients in the once-every-3-weeks arm (P = .006). Estimated median progression-free survival in the once-a-week arm was 17.7 months (95% CI, 0.42 to 35.05 months) and in the once-every-3-weeks arm, 28.6 months (95% CI, 15.90 to 41.30 months); HR, 1.24 (95% CI, 0.89 to 1.73); P = .21. Estimated median overall survival in the once-a-week arm was 39.5 months and was not reached in the once-every-3-weeks arm (HR, 1.14 [95% CI, 0.79 to 1.65]; P = .48). Conclusion : Once-every-3-weeks cisplatin at 100 mg/m2 resulted in superior LRC, albeit with more toxicity, than did once-a-week cisplatin at 30 mg/m2, and should remain the preferred chemoradiotherapy regimen for LAHNSCC in the adjuvant setting.