Sparing Irradiation vs. Conventional Irradiation to the Medial Retropharyngeal Space in Patients with Nasopharyngeal Carcinoma: An Open-Label, Non-Inferiority, Multicenter, Randomized Phase III Trial

Abstract

<h3>Purpose/Objective(s)</h3> Routinely complete coverage of the retropharyngeal space from skull base to cranial edge of the hyoid is the standard principle of radiotherapy in patients with nasopharyngeal carcinoma (NPC). However, involvement of the retropharyngeal lymph nodes (RLN) is mainly located at the lateral group, the medial RLN is rarely seen. We aimed to address whether medial retropharyngeal space (MRS) sparing irradiation provides uncompromising local relapse-free survival (LRFS) versus conventional irradiation in patients with NPC. <h3>Materials/Methods</h3> In this open-label, non-inferiority, randomized controlled, phase III trial, patients with newly diagnosed, non-keratinizing, non-distant metastatic NPC were recruited from three Chinese medical centers. Key inclusion criteria were age between 18 and 65 years, Karnofsky performance-status score of at least 70, and without medial RLN involved. Eligible patients were randomly assigned (1:1; block size of four) to receive sparing irradiation or conventional irradiation to MRS. Randomization was carried out centrally by the Clinical Trials Centre of one center using a computer-generated random number code, stratified by treatment center and treatment modality as follows: (1) radiotherapy (RT) alone; (2) concurrent chemoradiotherapy (CCRT); (3) induction chemotherapy plus RT/CCRT. The primary endpoint was LRFS; non-inferiority was shown if the upper limit of the 95% confidence interval (CI) of the difference in 3-year LRFS between sparing irradiation group and conventional irradiation group did not exceed 8%. This study is registered with ClinicalTrials.gov, number NCT03346109, and is currently in follow-up. <h3>Results</h3> A total of 568 patients were recruited in this trial (285 patients in the sparing irradiation group and 283 patients in the conventional irradiation group). At a median follow-up of 39 months [IQR] 36–42), intention-to-treat analysis showed that sparing irradiation group and conventional irradiation group had similar 3-year LRFS (95.8% [95% CI 93.3–98.3] vs. 96.4% [94.2–98.6], difference -0.6% [upper limit of the one-sided 95% CI 2.6], p_{non-inferiority} < 0.0001), and 3-year OS (95.5% [93.0–98.0] vs. 95.9% [93.5–98.3], difference -0.4% [upper limit of the one-sided 95% CI 2.9], p_{non-inferiority} < 0.0001). In the per-protocol population (n=564), sparing irradiation group had a lower incidence of acute radiation-related toxicities including grade≥1 mucositis (67.7% vs. 79.8%, p = 0.001), dysphagia (25.5% vs. 35.1%, p = 0.013), weight loss (46.8% vs. 57.8%, p = 0.009), and grade≥3 mucositis (10.6% vs. 16.7%, p = 0.037); and late toxicity including grade≥1 dysphagia (24.0% vs. 34.3%, p = 0.008). <h3>Conclusion</h3> MRS sparing irradiation achieves comparable local control and less radiation-related toxicity compared to conventional irradiation in patients with non-metastatic NPC.