Selective Neck Dissection May Be Reasonable for Head/Neck Carcinoma

Background: The need for neck dissection after primary chemoradiation for patients with >N2 disease at presentation is a well-accepted concept. However, it is less known whether selective neck dissection (SND) is sufficient to control the neck in these patients. Objective: To evaluate one institution's experience with performing selective neck dissections after primary chemoradiation for patients with N2 or N3 disease. Design: Retrospective review. Participants: 58 patients were identified with advanced cervical nodal metastases (N2-N3) from squamous cell carcinoma of the head and neck. Patients were treated with primary radiation (XRT) or chemoradiation (CRT) followed by a staged SND. Methods: Patients were analyzed for efficacy of SND in terms of locoregional recurrence, survival, and pathological status of the lymph nodes. Results: 70 neck dissections were performed in 58 patients. Primary site was the tongue base in 15 (26%), tonsil in 16 (28%), hypopharynx in 12 (21%), larynx in 11(19%), and unknown in 4 (6%). Of patients, 31 (53%) were N2a, 10 (17%) N2b, 12 (22%) N2c, and 5 (8%) N3. Interestingly, 22 (38%) received XRT, while 36 (62%) received CRT. SNDs were performed on all patients 3 to 6 weeks after completing therapy. Disease specific survival at 34 months (median time of follow-up) was 86.7%. Of patients, 9 died due to distant metastases, 3 had regional recurrence, and 1 had local recurrence. When comparing the XRT to the CRT group, there was no difference between the local and regional control rates, and there was a benefit in the CRT group with regards to distant control rates (72% versus 92%). On pathological exam, 13 (22.4%) specimens had residual tumor: 5 were treated with XRT, and 8 were treated with CRT. There was no statistically significant difference in survival when accounting for treatment type and nodal stage. Conclusions: A staged SND is sufficient for patients receiving primary nonsurgical treatment for advanced head and neck squamous cell carcinoma. Reviewer's Comments: As we try to refine our treatment algorithms, there is an attempt to reduce potential morbidity. Radical neck dissections are now quite uncommon. As such, if a planned neck dissection is considered after XRT or CRT for >N2 disease, why not do a SND? This study doesn't really tell us if there was level V involvement prior to treatment; clearly, for most tumors, levels II-IV are the most likely to be involved, but not always. The authors did not focus on imaging modalities, as it was assumed that the neck dissections would be performed. Overall, I think it's a reasonable option to do a SND, but would obviously think twice if there were bulky level V disease on presentation.