De-escalation of Primary Target and Elective Neck Doses in HPV-Positive Oropharyngeal Cancers

Abstract

Human papilloma virus-related oropharyngeal squamous cell carcinoma (OPSCC) is a highly curable malignancy that is increasing in incidence. Primary radiation therapy (RT) with or without chemotherapy currently achieves excellent long-term outcomes; however, it may be associated with significant acute and long-term side effects. We hypothesized that reductions in dose to the primary target (<69.3 Gy) and elective neck (<50 Gy) would result in similar control and may reduce acute toxicity. After IRB approval was obtained, a database of HPV or p16 positive non-metastatic OPSCC patients treated with definitive radiation therapy with or without chemotherapy was queried. Relevant features of patients in high-dose groups and low-dose groups were compared with Fischer Exact test. Locoregional control (LRC), regional control (RC), and overall survival (OS) were calculated from the end of RT and estimated via Kaplan-Meier method and comparisons made via log-rank test. A total of 387 patients were available for analysis with a median follow-up was 33 months. Standard doses of ≥69.3 Gy (median 70, range 69.3-75.2) were used in 298 patients, and <69.3 Gy (median 66 Gy, range 58-68 Gy) in 89 patients. Standard elective neck doses of ≥50 Gy (median 56 Gy, range 50-56 Gy) were used in 311 patients and <50 Gy (median 46, range 40-49.6) in 71 patients. Patients in the high-dose range to the primary target or elective neck were more likely to be higher AJCC 8th edition T stage, N stage, and overall stage (P < .05 for all comparisons). There was no difference in the 3-year LRC comparing <69.3 Gy and ≥69.3 Gy (95.2% and 91.8% respectively, P = .67), no difference in the 3-year RC comparing the <50 Gy and ≥50 Gy arms (94% vs 90% respectively P = .41). There was no difference in 3-year OS for both <69.3 Gy (95.3% and 87.3%, respectively, P = .13) and <50 Gy (95.6% and 87.9%, respectively, P = .20). When stratifying by AJCC 8th edition T, N, overall stages, or concurrent chemotherapy, there was no difference in LRC, RC, or OS at the different dose levels (P > .22 for all comparisons). The need for reactive gastrostomy tube (PEG) placements was significantly lower in patients receiving lower doses, 4.4% and 19.5% (P < .01). There were 14 (4.9%) grade ≥3 late effects in the ≥ 69.3 Gy arm compared to 1.1% in the <69.3 Gy; however, this was not statistically significant (P = .32). Mild de-escalation of doses to the primary tumor and elective neck does not appear to adversely affect LRC, RC, or OS in patients with AJCC 8th edition I-III HPV-related OPSCC, which remained true after stratification. Patients treated with lower doses had significantly reduced PEG tube rates compared to those in the high-dose group.