Diminishing Toxicity Profiles in Oropharynx Cancer. Can We Safely Spare Nodal Levels IB and V in the Majority of Patients?

Elective treatment of level IB and level V for oropharynx cancer remains controversial and institution dependent. Recent consensus guidelines recommend treating these regions for a subset of oropharyngeal cancer patients based on specific criteria. We examined the overall rate of involvement of these regions at diagnosis and at the time of treatment failure. We performed a retrospective review of patients with oropharyngeal cancer treated with IMRT at our institution between 2002 and 2019. PET/CT scans were reviewed to identify nodal involvement within level IB and level V. In patients with recurrent disease, the PET/CT or diagnostic CT at the time of recurrence was evaluated to determine whether failure occurred within level IB or level V. Kaplan Meier method was used to calculate the rate of failure within level IB and V at 5 years. Fisher exact test was used to compare rates of nodal disease between p16 positive and negative patients. Two-hundred and forty patients were identified with oropharyngeal cancer treated with IMRT at our institution. Of these, 190 had PET/CT at the time of diagnosis available and were included in this study. Median follow-up was 48 months. There were 109 patients with p16 positive disease and 81 with p16 negative disease. Rates of N2c/N3 disease were similar between p16 positive and negative patients at 35% and 32%, respectively (p = 0.75). Locoregional control was 85.0% (95% CI: 77.9%-92.1%) at 5 years in p16 positive patients and 84.0% (95% CI: 75.7%-92.3%) in p16 negative patients (p = 0.83, log rank). At diagnosis, the overall rate of level V involvement was 4.6% and 2.4% in p16 positive and p16 negative patients, respectively (p = 0.47). The rate of level IB involvement at diagnosis was 7.3% and 7.4% in p16 positive and p16 negative patients, respectively (p = 1.0). The rate of failure in level IB at 5 years was 0.6% (95% CI: 0.0%-1.75%) with a single failure in the p16 positive group. The rate of level V failure was 1.9% (95% CI: 0.1-4.1%) with one p16 positive recurrence and two p16 negative. None of the failures were isolated. The failures occurred in previously uninvolved level IB and level V. Despite not generally electively treating levels IB or V at our institution, low rates of failure in level IB (< 1%) and V (< 2%) were identified in our series with no isolated failures in these stations. This suggests that with high quality PET/CT imaging, occult metastatic spread to level IB and V are reliably detected. Because the rates of initial disease were 4-7%, high quality PET/CT is crucial if this volume is to be omitted. Omission of level IB and V treatment volumes may represent potential targets for safe volume de-escalation to help reduce normal tissue toxicity profiles.