Nanopore sequencing-derived methylation biomarker prediction for methylation-specific PCR in patients with head and neck squamous cell carcinoma.

Abstract

Background: DNA methylation of CpG islands is altered in cancer cells. Hypermethylation of single CpG islands in the promoter regions of tumor-suppressor genes occurs already in the early stages of cancer. These methylation changes are cancer-type specific and therefore can serve as early cancer biomarker. Identifying good and reliable biomarkers is crucial for the development of diagnostic tests and their application in clinical practice and remains the most significant challenge to date.

Results: Here, we present a generic workflow for the discovery and design of DNA methylation-specific PCR (MSP) biomarkers using nanopore sequencing. We show that nanopore sequencing of three control and three tumor tissue samples was sufficient to predict differentially methylated regions between head and neck squamous cell carcinoma (HNSCC) and healthy control tissue samples and to design functional MSP primers. When applied to a validation cohort of 48 HNSCC and 46 control samples, four out of six designed MSP singleplex assays achieved good sensitivity and specificity with an AUC above 0.8.

Conclusion: Our resulting DNA methylation-based workflow demonstrates how long-read methylation data enable the design of adaptable, clinically relevant epigenetic assays, even with low coverage and small initial sample numbers.