A preliminary analysis of patterns of failure in patients treated with intensity modulated radiotherapy (IMRT) for head and neck cancer: The University of Nebraska Medical Center experience

Abstract

To analyze the patterns of failure in patients with head-and-neck cancer treated with IMRT. Between May 2000 and December 2003, 188 patients received IMRT for head-and-neck cancer at the University Nebraska Medical Center. Most patients (97%) were treated with the Peacock system using a dynamic multivane intensity-modulating multileaf collimator (MIMiC), and the remainder patients were treated with MLC-based step-and-shoot IMRT. Thirty patients who either received IMRT for palliation (23 patients) or did not complete the planned treatments (7 patients) were excluded from analysis. One hundred and fifty-eight patients received IMRT with curative intent. Of the 158 patients, 46 were women and 112 were men (median age 60 years, range 11–87). One hundred and twenty-five patients (79%) had squamous cell carcinoma. Twenty-six patients (16%) were treated for recurrent disease. In patients who had primary disease, 112 (85%) were stage III or IV. Seventy-seven patients (49%) received definitive IMRT, and 81 (51%) received adjuvant IMRT. Eighty-eight patients (56%) also received chemotherapy; 56/77 (73%) of definitive cases and 32/81 (40%) of postoperative cases. Twenty-five (16%) patients had previously received conventional radiation therapy. The IMRT was used only in the upper neck for most patients (96%). A conventional AP low-neck field was added when comprehensive neck treatment was indicated. A two-step treatment planning approach was used; namely an initial comprehensive field for all targets followed by a boost field for high-risk areas to ensure adequate fraction size for all intended targets. The dose was prescribed to the isodose that provided adequate coverage for each defined target (GTV, CTV1 and CTV2) depending upon the probability of tumor burden. In definitive IMRT cases, the median dose to the GTV/CTV1 was 70Gy (range 50-74Gy), and 50Gy (range 45-54Gy) to CTV2, whereas in the postoperative cases the median dose was 64Gy (range 50.4–72.4Gy) for CTV1 and 50Gy (range 45Gy-54Gy) for CTV2. The patterns of failure were analyzed. The probability of loco-regional control, distant metastasis, disease-free survival, and overall survival were calculated using the Kaplan-Meier method. The median follow-up was 17 months. A total of 137 patients (87%) had minimum follow-up of 6 months. Ten patients (6%) developed local or regional recurrences. Three of the loco-regional failures had persistent disease. Eight (5%) had loco-regional and distant failures, and 19 (12%) patients developed distant metastases as the only site of failure. All but one loco-regional failure occurred within the GTV or CTV1. One recurred in CTV2 area in the contra-lateral neck. The 3-year actuarial loco-regional control rate was 88%, and distant metastasis-free rate was 82%. The 3-year overall survival and disease-free survival rates were 80% and 77% respectively. We report one of the largest single institutional IMRT experiences in patients with head-and-neck cancer. IMRT is highly effective and offers excellent loco-regional control. Distant failure remains the predominant site for treatment failure. More effective systemic treatment is critical to improve overall disease-free survival