Background: Differentiated thyroid cancer (DTC) is usually associated with an excellent prognosis. The main cause of death is due to distant metastases, but distant metastatic DTC (dmDTC) presents an interpatient heterogeneity. Some patients with distant metastases live with stable disease for many years, while others die very early. Still no effective biomarkers are available to predict either which patients will eventually develop distant metastases or what will be the final outcome of patients with dmDTC. The key for the future development of predictors probably lies in a better understanding of the (epi)genetic wiring of dmDTC. DNA methylation, one of the most studied epigenetic mechanism, affects CpG sites and is associated with transcriptional repression. While major progress has been made in understanding DNA methylation in DTC, the epigenetic architecture of dmDTC is completely unknown.
Objective: The aim of this study was to characterize the DNA methylomes of dmDTC and to identify biomarkers that predict the development of distant metastases.
Methods: We profiled DNA methylation of a series of 98 formalin-fixed paraffin-embedded tissues including 30 low risk non-metastatic DTC, 35 dmDTC, 18 metastases and 15 adjacent normal tissues using the Illumina Infinium HumanMethylationEPIC platform. We selected candidate biomarkers using the Simple Logistics classifier implemented in RWeka, and validated them by bisulfite-PCR-sequencing, bisulfite pyrosequencing and MethylQuant.
Results: We identified a signature of 156 CpGs associated with dmDTC independently of histology and mutations in BRAF and RAS. DNA methylation differed between distant metastatic (synchronous and metachronous) and low-risk non-metastatic primary tumors, while the paired primary tumor and distant metastases were similar, suggesting that molecular alterations of the primary tumors may dictate the ability to metastasize. The signature was enriched in hypomethylations and over half were located in promoters and enhancers, pointing out their role in the regulation of gene expression. We selected 10 independent CpGs with high classification power, and quantified DNA methylation by different techniques to develop a simplified quantitative and cheap assay implementable in the routine clinical practice. Based on preliminary results MethylQuant analysis appeared as the best alternative for on its simplicity, performance and cost-effectiveness.
Conclusion: We identified a 10-epigenetic biomarker panel associated with distant metastases in thyroid cancer and established new quantitative DNA methylation assays.
19 May 2018 - 22 May 2018