CyberKnife Fractionated Stereotactic Radiosurgery for the Treatment of Primary and Recurrent Head and Neck Cancer

Abstract

Purpose/Objective: The treatment of head and neck cancer with radiation therapy is challenging given the proximity of target volumes to critical structures. The management of recurrent disease is a therapeutic dilemma since many patients have had prior treatment which often limits the feasibility of reoperation or conventional reirradiation without unacceptable morbidity. With the Cyberknife, it is possible to treat extracranial sites with stereotactic radiosurgery (SRS) without the need for external fixation and with the ability to easily fractionate treatment. The purpose of this study was to report our initial experience using the Cyberknife to treat primary and recurrent head and neck cancers and to evaluate its feasibility, toxicity and local control. Materials/Methods: The CyberKnife is a frameless image-guided SRS system which utilizes a 6-MV linear accelerator mounted on a robotic arm with 6 degrees of freedom. From December 2002 to March 2005, 37 sites were treated in 32 patients using the Cyberknife. Sixteen patients received concurrent chemotherapy. Ten patients were treated as a component of primary treatment. Six patients with nasopharyngeal carcinoma received a boost after conventional radiation. Three patients were treated post-operatively to the submandibular fossa, paranasal sinuses, and nose. One patient was treated pre-operatively for a nasal cavity tumor. Twenty patients were treated for locoregionally recurrent disease. The treated sites included the neck (8), paranasal sinuses (5), skin (5), parapharyngeal space (3), nasopharynx (2), and parotid fossa (1). All of these patients had previous external beam radiation to a median dose of 65 Gy (range 44–120). Seven underwent initial surgical resection (3 subtotally resected, 4 with positive margins), 10 were felt to be unresectable, and 3 had failed multiple prior surgeries. Two patients were treated for metastatic disease to the head and neck. The median follow up was 5 months (range 1-31). Acute and late toxicity were assessed using RTOG grading scales. Local control was assessed with follow-up imaging and clinical examination. Results: The median tumor dose was 25 Gy to the 75% isodose line (range 15–30 Gy) in a median of 5 fractions (range 2–10). The median tumor volume was 52.9 cc (range 4.1–723.2). The median cumulative radiation dose to the treated site was 84 Gy (range 28–158). At the time of analysis, 26 patients were alive, 2 are dead, and 4 have been lost to follow up. The local control rate was 81% overall, 100% in patients treated for primary disease, and 69% in those treated for recurrent disease. Of the 26 patients with measurable disease at the time of treatment, 7 had a complete response, 8 had a partial response, 3 had stable disease, 3 had recurrent/progressive disease, 1 died of complications from infection, and 4 were lost to follow up. Of the 6 patients with no measurable disease at the time of treatment, 4 are NED and 2 have recurred locally. Acute toxicities included 2 patients with Grade 3 GI toxicity, 1 with Grade 3 mucositis, and 3 with Grade 4 skin reactions over areas of tumor necrosis. There was no observed Grade 3 or higher late toxicity. Conclusions: Our early experience demonstrates the efficacy and safety of Cyberknife fractionated SRS for select patients with head and neck cancer. This technique offers an effective treatment option in patients with tumors that are unresectable or incompletely resected, previously irradiated, or located in proximity to radiosensitive critical structures. The local control rate is encouraging with an acceptable toxicity profile.