Back

Discovery and Validation of RUNX1 DNA Methylation in Differentiating Papillary Thyroid Cancer from Benign Nodules

Li, J.; Yin, Y.; Huang, H.; Li, M.; Li, H.; Zhang, M.; Jiang, C.; Yang, R.

2023-04-11 genetics
10.1101/2023.04.10.536270 bioRxiv
Show abstract

Although most thyroid nodules can be diagnosed preoperatively by thyroid ultrasonography and fine-needle aspiration biopsy, it remains a challenge to accurately identify malignancy of thyroid nodules when the biopsy is indeterminate. This study aims to explore a novel biomarker to distinguish benign and malignant thyroid nodules. Tissue samples from patients with Stage I&II papillary thyroid carcinoma (PTC) and benign thyroid nodules (BTN) were collected for genome profiling by methylation EPIC 850K array and RNA-Sequencing. Genes with significantly differential DNA methylation and inverse mRNA expression were filtered out. The altered methylation of RUNX1 gene was validated in two independent case-control studies with a total of 699 formalin fixed paraffin-embedded (FFPE) samples using mass spectrometry and calculated by binary logistic regression analysis. Hypomethylation of RUNX1 gene in PTC patients compared to BTN subjects was verified in Validation (140 PTC vs. 189 BTN, ORs [≥] 1.50 per-10% methylation, P [≤] 4.40E-05, for all measurable CpG sites) and Validation (184 PTC vs. 186 BTN, ORs [≥] 1.72 per-10% methylation, P [≤] 2.38E-11, for all measurable CpG sites). Besides, RUNX1 methylation achieved good accuracy in differentiating early-stage PTC from BTN in Validation (AUC: 0.74) and Validation (AUC: 0.79). Gender- and age-stratified analysis revealed RUNX1 hypomethylation as an important risk factor for thyroid disease in younger women. We disclosed a significant association between RUNX1 hypomethylation and PTC, suggesting RUNX1 methylation based on FFPE tissue samples as a potential biomarker for predicting malignancy of thyroid nodules.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.