OP230: Multi-marker discrimination of dysplasia grades in potentially malignant oral disorders

Abstract

Purpose Early diagnosis and targeted therapy are crucial to mitigating the morbidity and mortality of oral squamous cell carcinoma (OSCC). Among the potentially malignant oral disorders (PMOD), epithelial dysplasia has known association with malignant transformation. Variance in clinical manifestation and rates of transformation of dysplasia are likely to be underpinned by discrete genetic, viral and environmental pathologies. Cross-disciplinary research is needed to link causal factors with PMOD aetiology and prognosis. The aim of this component of our ongoing research programme was to identify immuno-histochemical (IHC) markers that categorically discriminate diagnoses of mild and severe dysplasia. Materials and Methods Two discrete cohorts of archival formalin-fixed paraffin embedded biopsies of oral lesions (combined n = 378 cases) classified as normal, mild dysplasia, moderate/severe dysplasia (including carcinoma in situ) and OSCC were studied by IHC for 14 antigens loosely grouped as: putative stem cell markers (ALDH1, p75, CD44, LGR5, CD24), cancer associated antigens (CLSP, ELF3, IFI44, USP18, CXCL13) and DNA mismatch repair associated proteins (hMLH1, hMSH2, hMSH6 and hPMS2). IHC score data was compiled along with demographic data, and analysed by regression analysis. Results Overall, whilst staining for many of the antigens (ALDH1, LGR5, CD24, CLSP, ELF3, IFI44, hPMS2) could delineate disease severity from normal through dysplasia to OSCC, none could reliably discriminate mild from moderate/severe dysplasia. Conclusion Even with a very broad span of target antigens, IHC techniques alone have been unable to categorically discriminate mild from moderate/severe dysplasia lesion grades. More detailed insight into the underlying pathology behind the various clinical manifestations of dysplasia and the effect of these on PMOD aetiology and prognosis is required. To meet this challenge, genomic screening of selected lesions with known progression data is being pursed.