Survival outcomes for head and neck cancer patients with N3 cervical nodal metastases.

Abstract

Management of metastatic N3 nodal disease from primary head and neck squamous cell carcinoma (HNSCC) is controversial. Recently, there has been a move to observation of the neck for those who achieve complete response (CR) after chemoradiotherapy (CRT). We sought to determine survival outcomes for N3 nodal disease, particularly for patients with human papilloma virus (HPV)-positive HNSCC. We carried out a systematic search of MEDLINE and Embase for articles between 01/2008 and 12/2017. Articles on N3 nodal disease in HNSCC patients treated with CRT or surgery+adjuvant RT/CRT were included if they reported on oropharyngeal or HPV+ subgroups. Local control (LC), nodal control (NC), distant metastasis-free survival (DMFS), disease-free survival (DFS) or overall survival (OS) was assessed. Nine studies met the inclusion criteria. Eight of these studies (N=5304) allowed further comparison: 4 were on CRT, 2 on surgery+RT/CRT and 2 on both. Four of these eight studies and the remaining included study reported on residual nodal disease on histology after neck dissection (ND) following CRT. Patients treated with CRT achieved LC rates of 77%-94% at 2-3years; those who had a CR had LC of >90%. Better NC was noted in patients who achieved a CR. Three-year OS was better for HPV+ HNSCC (range, 55.2%-81%). Patients with CR had better survival outcomes (DMFS 77% at 3years vs 69.8% for HPV+; OS 68.9% at 3years vs 55.2% for HPV+). Primary surgery demonstrated similar survival for HPV+ vs HPV- and better survival in oropharyngeal cancers. Five-year DFS rates varied from 30% to 87%, and OS from 26.6% to 84%. For patients with non-CR, positive histology rates varied from 27.3% to 100%, with average positive histology rates of 27.3% in HPV+ patients with non-CR vs 60% for HPV- patients with non-CR. The current literature does not support the de-escalation of treatment with HPV- N3 disease. Observation of HPV+ patients who achieve a CR post-CRT is reasonable but further prospective studies are required given the heterogeneity and risk of bias within these current studies. Planned ND should remain standard of care for non-CR cohort.