BACKGROUND

Molecular biomarkers assisting risk-group assignment and subsequent treatment stratification are urgently needed for patients with squamous cell cancer of the head and neck region (HNSCC). Aberrant methylation is a frequent event in cancer and, therefore, a promising source for potential biomarkers. Here, the methylation status of the paired-like homeodomain transcription factor 3 () was evaluated in HNSCC.

METHODS

Using a quantitative real-time PCR, methylation was assessed in a cohort of 326 HNSCC patients treated for localized or locally advanced disease (training cohort). The results were validated with Infinium HumanMethylation450 BeadChip data from a 528 HNSCC patient cohort (validation cohort) generated by The Cancer Genome Atlas (TCGA) Research Network.

RESULTS

methylation was significantly higher methylated in tumor compared to normal adjacent tissue (NAT; training cohort: median methylation NAT 32.3%, tumor 71.8%,  < 0.001; validation cohort: median methylation NAT 16.9%, tumor 35.9%,  < 0.001). methylation was also significantly correlated with lymph node status both in the training ( = 0.006) and validation ( < 0.001) cohort. methylation was significantly higher in HPV-associated (p16-positive) tumors compared to p16-negative tumors (training cohort: 73.7 vs. 66.2%,  = 0.013; validation cohort: 40.0 vs. 33.1%,  = 0.015). Hypermethylation was significantly associated with the risk of death (training cohort: hazard ratio (HR) = 1.80, [95% confidence interval (CI) 1.20-2.69],  = 0.005; validation cohort: HR = 1.43, [95% CI 1.05-1.95],  = 0.022). In multivariate Cox analyses, added independent prognostic information. Messenger RNA (mRNA) expression analysis revealed an inverse correlation with methylation in the TCGA cohort.

CONCLUSIONS

DNA methylation is an independent prognostic biomarker for overall survival in patients with HNSCC and might aid in the process of risk stratification for individualized treatment.