Background: Treating the neck after organ-preservation treatment with radiotherapy or chemoradiotherapy can be problematic.
Methods: To develop management guidelines, we reviewed the results of a 100-patient phase-3 trial that had compared outcome after radiotherapy alone with outcome after chemoradiotherapy for head and neck cancer. Patients were randomly assigned to receive radiotherapy alone or concurrent chemoradiotherapy. After completing therapy, patients were reassessed, and surgery was recommended for persistent disease at the primary site or neck and for all patients with stage N2-3 neck nodes regardless of clinical response.
Results: Of the 47 patients with stage NO-1, 43 had a complete response (CR); of the 18 N1 patients, all but 4 had a CR. One of these 4, as well as 5 others among the NO-1 patients, underwent neck dissection (n = 6). No disease was found on pathologic examination, and no patient had neck recurrence. Of the remaining 41 N0-1 patients, 3 had disease progression and received no further therapy. Of the 38 others, 4 had neck recurrence, with 3 recurring at the primary site. Of the 53 with stage N2-3, 23 had less than a complete response (<CR), and 30 had a CR. In 35 N2-3 patients, neck dissection was performed as planned. Of these 35, 18 had a CR in the neck; 4 had positive nodes on pathologic examination. The other 17 had a <CR in the neck; 8 had positive nodes on pathologic examination. One patient in this group of 17 had regional recurrence after a pathologically negative neck dissection. Of the 18 N2-3 patients who did not undergo planned neck dissection, 6 had tumor progression and had no further therapy. The other 12, all with a CR in the neck, were followed, and 3 had neck recurrence; none successfully salvaged. Despite a CR in 30 N2-3 patients, 7 had persistent disease or eventual neck recurrence. Adding neck dissection minimized neck recurrence (p = .05). In N2-3 patients, disease-specific survival was significantly better in patients with a CR in the neck (p = .002). Disease-specific survival was not affected by neck dissection (p = .40) but was significantly affected by viable tumor in the specimen (p = .03).
Conclusion: Based on these results and the realization that it is difficult to follow patients for recurrent neck cancer, that salvage is often unsuccessful, and that patients dying from uncontrollable neck disease have an extremely poor quality of life, we recommend neck dissection for all N2-3 patients regardless of the neck response and for N1 patients without a CR.