Background: Resistance to apoptosis is a hallmark of cancer and proteins regulating apoptosis have been proposed as prognostic markers in several malignancies. However, the prognostic impact of apoptotic markers has not been consistently demonstrated in oral squamous cell carcinoma (OSCC). This inconsistency in reported associations between apoptotic proteins and prognosis can be partly attributed to the intrinsic low resolution and misclassification associated with manual, semi-quantitative methods of biomarker expression measurement. The aim of this study was to examine the association between apoptosis-regulating proteins and clinical outcomes in oral squamous cell carcinoma (OSCC) using the quantitative fluorescence immunohistochemistry (IHC) based AQUAnalysis technique.
Methods: Sixty-nine OSCC patients diagnosed between 1998-2005 in Calgary, Alberta, Canada were included in the study. Clinical data were obtained from the Alberta Cancer Registry and chart review. Tissue microarrays (TMAs) were assembled from triplicate cores of formalin-fixed paraffin embedded pre-treatment tumour tissue. Bax, Bcl-2 and Bcl-XL protein expression was quantified using fluorescent IHC and AQUA technology in normal oral cavity squamous epithelium (OCSE) and OSCC tumour samples. Survival was analyzed using Kaplan-Meier plots and the Cox proportional hazard model.
Results: Bax expression was predominantly nuclear in OCSE and almost exclusively cytoplasmic in OSCC. No similar differences in localization were observed for Bcl-2 or Bcl-XL. Only Bax expression associated with disease-specific survival (DSS), with 5-year survival estimates of 85.7% for high Bax versus 50.3% for low Bax (p = 0.006), in univariate analysis. High Bax expression was also significantly associated with elevated Ki67 expression, indicating that increased proliferation might lead to an improved response to radiotherapy in patients with elevated Bax expression. In multivariate analyses, Bax protein expression remained an independent predictor of DSS in OSCC [HR 0.241 (0.078-0.745), p = 0.013].
Conclusions: The AQUA technique used in our study eliminates observer bias and provides reliable and reproducible estimates for biomarker expression. AQUA also provides essential measures of quality control that cannot be achieved with manual biomarker scoring techniques. Our results support the use of Bax protein expression as a prognostic marker in conjunction with other clinico-pathological variables when designing personalized treatment strategies for OSCC patients.