(OA31) Patterns of Failure for Oral Cavity Squamous Cell Carcinoma Patients Receiving Post-Operative Radiotherapy +/- Chemotherapy

Abstract

To analyze patterns of failure and evaluate treatment outcomes for oral cavity cancer patients undergoing surgery and post-operative intensity-modulated radiotherapy (IMRT) with or without chemotherapy. We performed an analysis on patients with stage I-IVB oral cavity squamous cell carcinoma, treated with curative intent using primary surgery followed by radiotherapy +/- chemotherapy at our institution. When patients were treated with surgical flap reconstruction (pedicle or free flap), our institutional convention was to provide full coverage of the original tumor bed without requiring high dose coverage of the entire reconstruction graft. We analyzed patient survival as well as local and regional patterns of failure. This study was approved by our institutional review board. Survival data was available for 99 of 100 eligible patients, treated between 2000-2016, with median follow-up time of 39.4 months. The median survival for the entire cohort was 92.5 months (range 1.5-111.8). The 2-year overall survival, local control, and loco-regional control was 70%, 85% and 79%, respectively. The primary sites represented in this study were: oral tongue (43%), floor of mouth (18%), alveolus and hard palate (16%), retromolar trigone (12%) and lip (2%). Oral tongue (44.3%) and floor of mouth (18.6%) accounted for the majority of recurrences. Recurrence within the primary tumor bed alone was the most common pattern of failure (10 out of 26, 38.5%), while primary and ipsilateral neck recurrence occurred in 5 patients (19.2%) and 4 occurred in the contralateral neck alone (15.4%). On logistic regression, tumor grade, degree of alcohol use, floor of mouth primary, N2b nodal stage, and external beam dose were associated with recurrence (either local, regional or both) (p <0.05). Smoking pack-years approached significance (p = 0.06). Over half of the cohort (62 patients) had pedicle or free flaps to reconstruct the primary tumor bed, with 14 local failures. All failures occurred at the interface between the tumor bed and grafted tissue, with no failures beyond this interface. The loco-regional control rate following post-operative IMRT +/- chemotherapy in patients with oral cavity cancer in this series approaches 80%. The majority of patients who fail have a component of recurrence within the primary tumor bed. Routine treatment of the entire flap reconstruction may require large volume irradiation of normal tissues within the oral cavity with resultant toxicity. With a predominant failure pattern for oral cavity tumors within or directly adjacent to the original tumor bed, and rare recurrence within the flap reconstruction donor tissue itself, there is rationale to limit routine coverage of the entire flap particularly for aspects distant from the primary tumor. This is the approach we have used at our institution for 27 years with very low rates of tumor recurrence within flap reconstruction tissues.