Prognostic and predictive roles of circulating microRNAs as biomarkers in head and neck cancer patients in the Northern Territory, Australia undergoing chemoradiotherapy-a pilot prospective cohort study.

Background: Head and Neck cancers are the fourth most common cancers worldwide with approximately 1.8 million new cases each year. The 5-year relative survival rates have increased significantly for most cancers in Australia, however, some head and neck cancers including laryngeal and lip cancers have shown no significant change over time. Therefore, there is an urgent need to identify novel prognostic biomarkers to tailor the treatment. MicroRNAs (miRNAs) are endogenous, short, non-coding single-stranded RNAs that regulate gene expression at the post-transcriptional level, controlling a wide range of biological and pathological processes. Increasingly, associations between miRNA and cancer diagnosis, prognosis, and treatment response have been reported. In this study, we investigated serum miRNA expressions in primary invasive head and neck cancers to evaluate their effectiveness in predicting prognosis and therapeutic response. Methods: MiRNA expression in serum samples from 11 consecutive head and neck cancer patients undergoing chemoradiotherapy (CRT) was evaluated prospectively using miRNA microarray and quantitative reverse transcription polymerase chain reaction (qRT-PCR) at baseline, end of CRT, and 12 weeks after treatment. Participants were followed up with standard clinical and radiological follow-up procedures as per clinical guidelines. The relationship of miR-335-5p to clinicopathological parameters was examined by a Chi-squared test, and the Kaplan-Meier survival curve was generated following a log-rank test to determine overall survival (OS) differences between patients with upregulated and/or downregulated miRNA levels. Results: A customised miRNA microarray identified 15 global miRNAs that were highly altered in a total of 33 serum samples collected from 11 head and neck cancer patients undergoing curative intent CRT treatment. MiR-335-5p was seen to be downregulated after radiation treatment and positively correlated with p16 status (P=0.02) of head and neck squamous cell carcinoma (HNSCC) subtype. Acknowledging the relatively short clinical follow-up and small sample size, no significant association between miRNA expression and patient survival was observed. Univariate and multivariate analyses were not able to associate miR-335-5p expression as an independent prognostic factor, most likely due to the small sample size. Conclusions: A three-point sampling method may be useful to reduce intra-individual differences while evaluating the changes in the expression of miRNAs during the period of curative intent CRT for HNSCC patients. Exposure to CRT altered the expression of several serum miRNAs which may predict therapy response and prognosis. However, due to small sample size, no statistically significant results could be derived from this study. Further larger studies are required to validate and extend our findings.