Abstract 3677: Prediction of oral cancer recurrence at post-surgery follow-up using fluorescence visualization

Abstract

Abstract Background: Recurrence is observed in up to 30% of surgically treated high-risk oral lesions (HRLs, severe dysplasia, carcinoma in situ and cancer) and is associated with poor prognosis. Reactive tissue change observed post-treatment often masks recurrence and makes early detection difficult. New advances using fluorescence visualization (FV) represent a promising approach to this problem that may facilitate early detection of disease recurrence. Objectives: 1) To identify FV alterations post-treatment of HRLs and 2) To determine whether a relationship exists between FV alterations and local recurrence. Methods: In the BC Oral Cancer Longitudinal Study, we have recruited ∼400 HRLs with surgical treatment as the primary modality. Patients eligible for this analysis included those that had an initial follow-up appointment within 6 months of surgery with at least 2 follow-up appointments within the first year of treatment, with each visit involving FV examination of the treatment site. Recurrence was defined as the presence of biopsy-proven HRLs. The ‘plateau’ of the FV during the follow-ups is defined as the change of FV measurement in width within ± 1 mm (superior-inferiorly) in various time intervals of 3, 6, 9, or 12 months. Results: A total of 198 patients were identified of which 24 (12%) had lesion recurrence at the previously treated site. There was no difference in gender, age, ethnicity, smoking habit, anatomical site, primary diagnoses, and follow-up time between the recurrence and non-recurrence groups. The duration of plateau was longer in non-recurrence group compared to those in recurrence group (P = 0.03). When we examined the duration ‘plateau’ at various intervals of follow-ups, among 166 patients/lesions with at least 9 months follow-ups, we found out that the presence of plateau was more frequent in the non-recurrence group than those recurrent case (P = 0.03). Using linear mixed effects and logistic regression analyses, there was a significant difference of the individual slopes between recurrence and non-recurrence group (P = 0.001), adjusted for the individual intercepts (i.e., the original FV width). Conclusion: The stability, i.e., timing and duration of the plateau, of the FV alteration during post-surgical follow-ups can be potentially used to predict local recurrence of HRLs. (Supported by Supported by grant R01 DE17013 from the National Institute of Dental and Craniofacial Research and grant CCSRI-20336 from Canadian Cancer Society Research Institute; Canadian Institute for Health Research and Michael Smith Foundation from Health Research) Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3677. doi:10.1158/1538-7445.AM2011-3677