Utility of positron emission tomography-computed tomography in identification of residual nodal disease after chemoradiation for advanced head and neck cancer

OBJECTIVES: Planned neck dissection after chemoradiation (CR) is often advocated in patients with head and neck squamous cell cancer (HNSCC) with advanced nodal disease who demonstrate a clinical complete response to CR because identification of residual occult nodal disease is difficult. We sought to investigate the utility of positron emission tomography-computed tomography (PET-CT) in identifying patients with occult nodal disease after CR. STUDY DESIGN: Nonrandomized retrospective cohort analysis. MATERIALS AND METHODS: The medical records of all patients treated with primary CR for advanced HNSCC with N2 or N3 disease from December 2003 to June 2005 were reviewed. Patients with a clinical complete response were eligible for inclusion if PET-CT performed at 8 to 10 weeks after CR showed no evidence of distant disease and they were treated with a planned neck dissection. RESULTS: Seventeen patients met study criteria. PET-CT was positive for residual nodal disease in 11 (64.7%) patients, with a standardized uptake value (SUV) range of 1.7 to 3.8. Pathologic examination revealed residual viable carcinoma in five (29.4%) patients, with tumor size ranging from 2.0 to 9.5 mm. Carcinoma was present in 2 of 11 (18.2%) patients with positive PET-CT scans and 3 of 6 (50%) patients with negative PET-CT scans. The sensitivity and specificity of PET-CT in predicting occult nodal disease was 40% and 25%, respectively. There was no correlation between PET-CT findings and histologic findings (P = .26) or between SUV and size of viable tumor (P = .67). CONCLUSIONS: A significant proportion of HNSCC patients with advanced neck disease harbor residual occult metastases after CR. PET-CT is not sufficiently specific or sensitive to reliably predict the need for posttreatment neck dissection.