MicroRNAs (miRs) control cell growth, apoptosis and differentiation, and thus play a key role in carcinogenesis. Identification of a set of miRs that demonstrate differential expression in oral squamous cell carcinoma (OSCC) patients with poor prognosis has potential for utility as a prognostic marker. A retrospective study of miR expression was conducted in 20 tissue samples from early stage (Stages I & II) OSCC patients with known clinical outcome (10 from those who had 5-year disease free survival and 10 who died of disease within 5 years) using genome-wide deep sequencing analysis. The promising miR candidates were then validated in 80 tissue samples using quantitative real-time PCR (qRT-PCR). The deep sequencing and qRT-PCR analysis identified two promising miRs, miR-375 and miR-214-3p. Combining the two miRs as a panel with age and gender had a predictive value for the area under the curve (AUC) of 0.932, with a sensitivity of 87.5% and a specificity of 87.2% (p<0.0001) to identify patients with poor prognosis. A miR-based prognostic risk score model was constructed, which included the miR-214-3p, miR-375, age and gender, each weighed by relative contribution. The risk score model was able to identify high-risk individuals who had significantly shorter time to relapse (p<0.001) and time to death (p<0.001). The model consisting of a two-miR panel with age and gender may be useful in prognostication of early stage OSCC patients, which can aid in identifying patients with poor prognosis who will benefit from a subsequent aggressive treatment regimen.
Keywords: deep sequencing; microRNA; oral squamous cell carcinoma; prognosis; qRT-PCR; risk score.