Objectives: In resected buccal cancer patients, an unexpected close surgical margin has been observed to correlate with poor clinical outcomes. However, close surgical margin alone does not independently guide post-operative therapies, revealing a clinical debate. Hence, the present study intended to explore epigenetic-based bio-predictors for further stratifying this debating patient population.
Materials and methods: Between 2000 and 2008, we retrospectively recruited 44 resected buccal cancer patients with a close surgical margin of ≤5 mm. All patients had post-operative radiotherapy. Genomic DNA was extracted from tumor-enrich areas that contained cancer cells of >70%. Methylation-specific PCR was performed to detect promoter methylation of four tumor suppressor genes, including RASSF1A, DAPK, IRF8, and SFRP1. Post-irradiation locoregional control was defined as the primary end point.
Results: There were 40 males and 4 females, with a median age of 53.5 years (range, 32-82 years). Multivariate analysis identified two independent predictors for locoregional recurrence: very close margin of ≤1 mm (HR: 4.96; 95% CI, 1.63-15.09; P=0.018) and promoter hypermethylation of DAPK (HR: 2.83; 95% CI, 1.05-7.63; P=0.042). The highest risk of locoregional recurrence was observed in patients with both of the two factors (HR, 8.05; 95% CI, 2.56-25.82; P=0.002) when compared with patients with none. Shorter disease-free survival, but not overall survival, was also observed.
Conclusion: More aggressive managements should be considered in resected buccal cancer patients with both very close margin and DAPK promoter hypermethylation rather than post-operative observation or radiotherapy alone.
Copyright © 2012 Elsevier Ltd. All rights reserved.