Radiation Dose and Volume De-Escalation for High-Risk HPV Positive Oropharyngeal Cancer

Abstract

<h3>Purpose/Objective(s)</h3> Human papillomavirus (HPV) associated oropharyngeal cancers (HPV+OPC) have a favorable prognosis with standard of care treatment paradigms. Recently published and ongoing clinical trials are investigating therapy de-escalation for patients with HPV+OPC with low-risk features with promising results, but largely exclude high risk (HR) patients. This is a secondary pooled analysis reporting outcomes of patients with HR HPV+OPC treated on consecutive prospective response-adaptive de-escalation trials. <h3>Materials/Methods</h3> Patients meeting HR criteria treated on the prospective phase II trials (NCT02258659, NCT03107182) as well as on a prospective cohort registry were included for analysis. HR criteria were defined by the NRG-HN005 exclusion criteria (T4, N2c-N3 (AJCC 7th ed.), and/or > 10 pack year smoking history). Patients were treated with induction chemo- or chemoimmunotherapy followed by RECIST 1.1 response adapted volume and dose de-escalated definitive chemoradiation. Patients with ≥ 50% response to induction therapy were treated either to 45 Gy in 1.5 Gy BID fractions with 3 cycles of TFHX in a week on week off fashion or to 50 Gy in 2 Gy daily fractions delivered over 5 weeks with concurrent cisplatin. Those with < 50% response received standard dose chemoradiotherapy. Descriptive statistics were used to summarize patient characteristics. Survival was estimated using the Kaplan Meier method for progression free survival (PFS), locoregional PFS (LRPFS), and overall survival (OS). Log rank analysis was used to investigate association between the number of NRG risk factors (RF) (1 vs > 1 RF) and LRPFS. <h3>Results</h3> 80 patients met HR criteria and were evaluable (15 registry patients). Median follow up was 38.5 (IQR 27.25-61.25) months. 60% smoked > 10 pack years. 25% of patients had T4 and 45% had N2c-N3 disease. 72.5%, 25%, and 2.5% had 1, 2, and 3 NRG RF, respectively. 56 of 80 (70%) patients were de-escalated. 3-year PFS, OS, and LRPFS was 92.9%, 93.5%, and 94.7%, respectively. Locoregional failure (LRF) was 1.7% in patients with 1 NRG RF, and 13.6% LRF in patients with 2 or 3 NRG RF (p=0.0265). 3-year LRPFS was 100% for patients with N0-N2b disease and 88.6% for patients with N2c-N3 disease; all 4 locoregional failures were in-field and in patients with N2c-N3 disease who were de-escalated. Only 1 patient had an isolated distant failure. <h3>Conclusion</h3> This prospective experience demonstrates feasible dose and volume de-escalation of HPV+OPC patients with NRG HR factors. The results for the overall cohort compare favorably to prior published results. Importantly, the failure rate in de-escalated patients is in line with acceptability criteria on prior phase II de-escalation studies for low-risk patients. No patients experienced regional out-of-field failures and distant failure rates were low in a traditionally HR population. Our pooled results call for renewed cooperative group focus on trial design for radiation dose de-escalation for HPV+ patients with higher-risk disease features.