Is a planned neck dissection necessary after chemoradiotherapy in head and neck cancer patients with advanced nodal stage

Abstract

The use of planned neck dissection (ND) in patients with advanced nodal disease in head and neck squamous cell carcinoma (HNSCC) after treatment with chemoradiotherapy is controversial. The purpose of this study is to clarify the role for neck dissection in patients treated with combined chemoradiotherapy (CRT) on 2 similar organ preservation protocols at our institution. The records of 90 patients with N2-N3 neck nodes who were treated between 1991 and 2001 at Stanford University on two organ preservation CRT protocols (OSP2 and OSP3) were reviewed. All patients received 2 cycles of cisplatin and 5-Fluorouracil (5-FU) induction chemotherapy, followed by concurrent CRT with similar chemotherapeutic agents. Patients on OSP3 were randomized to receive either CRT alone versus CRT and Tirapazamine for 8 doses. Radiotherapy was delivered at conventional fractionation at 2 Gy/fraction to a total dose of 66–70 Gy to the gross target volume. Patients with persistent neck nodes either clinically or radiographically at a planned evaluation at 50 Gy proceeded to a neck dissection following completion of CRT. Patients treated on the OSP3 protocol (n = 54) also received a single dose of 5 Gy delivered via 9–16 MeV electrons to the largest nodal mass prior to treatment for the comet study. The median follow up was 3.6 years. Overall, 63% (n = 57) of the patients attained a clinical complete response (cCR) in the neck; of these, 8 patients had a ND and all 8 had a pathologic complete response (pCR). Of the remaining 49 cCR patients whose necks were observed, 13 relapsed and 2 had a neck relapse without a recurrence at the primary site (1 with an isolated neck recurrence, 1 with a neck and distant failure). Of the 33 patients (37%) with < cCR in the neck, 2 had progressive disease and died. The remaining 31 patients had NDs with a pCR rate of 52% (n = 16). Outcomes of the 3 groups: (1) cCR, (2) < cCR/pCR and (3) < cCR/< pCR are summarized in table 1. The cCR rates were similar for the 2 OSP protocols (61% and 65%). There was a trend for higher pCR rate patients who underwent ND in OSP3 group (58% vs. 42%, p = 0.3, χ2 Test) Based on our experience, in patients with N2-N3 neck nodes who have achieved a clinical and radiographic cCR in the neck following CRT, planned ND benefited only 4% (2/49) and is therefore not routinely recommended. Patients with a <cCR should proceed to ND. Patients with pathologically persistent tumors in the neck on ND specimens have poor prognosis and will need more aggressive therapy.