Predictive Factors Related to Salvage External Beam Re-irradiation for RecurrentHead and Neck Squamous Cell Carcinoma after Primary Radical Therapy

Background: Recurrent orsecond-primary tumor inhead and neck cancer (re-HNC) is achallenge. Curative approaches include definitive new course of RT (re -RT) combined to surgery and/or chemotherapy. Methods: We evaluated the data from 36 patients presenting re HNC who were treated between 2007 and 2011. Results:Twenty two (61.1%) patients had surgery and re -RT using IMRT. The median first radiation dose and interval between re-RT and the initial RT course were 60.0 Gy and 28 months. The median follow-up was 24 months. The 2 -and 5year actuarial OS, PFS and LC rates were 58.6%, 83.8%, 75.0%, 24.4%, 25.9% and 13.5%, respectively. On univariate analysis disease free interval > 24 months and free surgical margins, p=0.005 and p=0.012, where related to LC. Free surgical margin and re-RT with concurrent CHT were related to PFS, p=0.029 and p=0.001, respectively. IMRT when compared to other techniques showed LC and PFS advantages, p= 0.047 and p=0.050, respectively.Multimodality treatment (p=0.027) and free surgical margin (p=0.016) were related to improved OS. Cox regression multivariate analysis confirmed that patients who underwent re-RT with techniques other than IMRT, HR=8.68 (p= 0.003, 95% CI: 0.029 �0.491) and recurrence free interval < 24 months, HR=6.71 (p= 0.010, 95% CI: 0.039 �0.637) had an inferior PFS. Gross tumor after or absence of surgery were related to worse LC rates, HR=4.18 (p= 0.041, 95% CI: 0.040 � 0.934).Severe late complicat ions (Grade"e 3) occurred in 14 (38.8%) patients. Conclusion:Re-RT should be offered for patients who are not suitable for surgery or for those with marginal resections, with a clear understanding that severe toxicity is associated and