A Randomized Trial of Radiotherapy vs. Trans-Oral Surgery for Treatment De-Escalation in HPV-Associated Oropharyngeal Squamous Cell Carcinoma (ORATOR2)

Purpose/Objective(s) Widespread oral human papillomavirus (HPV) infections have led to a rapid increase in the incidence of oropharyngeal squamous cell carcinoma (OPSCC). HPV-related OPSCCs have a better prognosis than conventional alcohol- and smoking-related OPSCCs, suggesting a role for treatment de-escalation. The goal of this phase II randomized trial was to assess survival, oncologic, and toxicity outcomes with two de-escalation approaches: primary reduced-dose radiotherapy (RT) vs. primary transoral surgery plus neck dissection (TOS + ND) with reduced-dose adjuvant therapy. Materials/Methods Weenrolled patients with T1-T2N0-2 (AJCC 8th edition) p16-positive OPSCC. After stratifying by smoking status, we randomized patients (1:1) to either the primary RT arm, which consisted of 60 Gy of RT and concurrent weekly cisplatin chemotherapy in node-positive patients vs. the TOS + ND arm, consisting of surgery and neck dissection, with adjuvant reduced-dose RT depending on pathologic findings. The primary endpoint was overall survival (OS), and secondary endpoints included progression-free survival (PFS), quality of life (QOL, using the MDADI and other metrics), and toxicity. The trial was closed to accrual in November 2020 due to excessive toxicity in the TOS + ND arm, consisting of two treatment-related deaths from known complications of TOS (one bleed and one cervical osteomyelitis following post-operative RT). After closure to accrual, all previously enrolled patients remained on follow-up. All analyses were pre-specified and intention-to-treat, unless otherwise specified. Due to these unexpected toxicity findings, the trial is being reported while survival outcomes remain immature, and therefore p-values are not reported for OS and PFS comparisons. Results BetweenFebruary 2018 and November 2020, 61 patients were randomized (n=30 in the RT arm and n=31 in the TOS + ND arm). Median age was 61.9 years, most patients (51%) were never-smokers, and the large majority of patients (n=51; 86%) were male. The arms were well-balanced. Median follow-up was 17 months (IQR: 15-20 months). Two-year estimates of OS were 100% in the RT arm (95% confidence interval [CI]: 100%-100%) and 89.1% (95% CI: 69.6%-96.4%) in the TOS + ND arm. Two-year PFS estimates were 100% in the RT arm (95% CI: 100%-100%) and 83.5% (95% CI: 60.8%-93.7%) in the TOS + ND arm. Grade 2-5 toxicities occurred in 67% of patients in the RT arm and 71% in the TOS + ND arm, with significantly more anorexia and dysgeusia in the RT arm. Mean (±SD) MDADI total scores at 1-year were similar between arms (85.7 ± 15.6 and 84.7 ± 14.5, respectively). One patient in each arm required a percutaneous feeding tube, and none required feeding tubes at 1-year. Conclusion The primary RT approach achieved excellent oncologic outcomes in treatment de-escalation, with a moderate toxicity profile, and should be tested in phase III trials. The primary TOS approach was associated with an upfront risk of treatment-related mortality and suboptimal PFS. (NCT03210103) Widespread oral human papillomavirus (HPV) infections have led to a rapid increase in the incidence of oropharyngeal squamous cell carcinoma (OPSCC). HPV-related OPSCCs have a better prognosis than conventional alcohol- and smoking-related OPSCCs, suggesting a role for treatment de-escalation. The goal of this phase II randomized trial was to assess survival, oncologic, and toxicity outcomes with two de-escalation approaches: primary reduced-dose radiotherapy (RT) vs. primary transoral surgery plus neck dissection (TOS + ND) with reduced-dose adjuvant therapy. Weenrolled patients with T1-T2N0-2 (AJCC 8th edition) p16-positive OPSCC. After stratifying by smoking status, we randomized patients (1:1) to either the primary RT arm, which consisted of 60 Gy of RT and concurrent weekly cisplatin chemotherapy in node-positive patients vs. the TOS + ND arm, consisting of surgery and neck dissection, with adjuvant reduced-dose RT depending on pathologic findings. The primary endpoint was overall survival (OS), and secondary endpoints included progression-free survival (PFS), quality of life (QOL, using the MDADI and other metrics), and toxicity. The trial was closed to accrual in November 2020 due to excessive toxicity in the TOS + ND arm, consisting of two treatment-related deaths from known complications of TOS (one bleed and one cervical osteomyelitis following post-operative RT). After closure to accrual, all previously enrolled patients remained on follow-up. All analyses were pre-specified and intention-to-treat, unless otherwise specified. Due to these unexpected toxicity findings, the trial is being reported while survival outcomes remain immature, and therefore p-values are not reported for OS and PFS comparisons. BetweenFebruary 2018 and November 2020, 61 patients were randomized (n=30 in the RT arm and n=31 in the TOS + ND arm). Median age was 61.9 years, most patients (51%) were never-smokers, and the large majority of patients (n=51; 86%) were male. The arms were well-balanced. Median follow-up was 17 months (IQR: 15-20 months). Two-year estimates of OS were 100% in the RT arm (95% confidence interval [CI]: 100%-100%) and 89.1% (95% CI: 69.6%-96.4%) in the TOS + ND arm. Two-year PFS estimates were 100% in the RT arm (95% CI: 100%-100%) and 83.5% (95% CI: 60.8%-93.7%) in the TOS + ND arm. Grade 2-5 toxicities occurred in 67% of patients in the RT arm and 71% in the TOS + ND arm, with significantly more anorexia and dysgeusia in the RT arm. Mean (±SD) MDADI total scores at 1-year were similar between arms (85.7 ± 15.6 and 84.7 ± 14.5, respectively). One patient in each arm required a percutaneous feeding tube, and none required feeding tubes at 1-year. The primary RT approach achieved excellent oncologic outcomes in treatment de-escalation, with a moderate toxicity profile, and should be tested in phase III trials. The primary TOS approach was associated with an upfront risk of treatment-related mortality and suboptimal PFS. (NCT03210103)