Genome-wide discovery of circulating cell-free DNA methylation signatures for the differential diagnosis of triple-negative breast cancer.

BACKGROUND

Preoperative identification of breast cancer (BC) subtypes is essential for optimizing treatment strategies and improving patient outcomes. This study aimed to identify circulating cell-free DNA (cfDNA) methylation signatures to differentiate triple-negative breast cancer (TNBC) from other BC subtypes (non-TNBC).

METHODS

We initially performed a genome-wide analysis to identify differentially methylated CpG sites (DMCs; |Δ| > 0.10 and < 0.05) between five TNBC and nine non-TNBC tissues using the Infinium HumanMethylationEPIC BeadChip. These DMCs were further validated using large-scale data from the Cancer Genome Atlas (TCGA, = 774; |Δ| > 0. 25 and < 0.05), and only CpG sites with average values > 0.90 or < 0.10 in white blood cells (GSE50132, = 233) were retained to minimize potential background methylation interference. Least absolute shrinkage and selection operator (LASSO) regression was applied to select optimal markers. Diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC), and prognostic value was evaluated using Cox regression analysis. A multiplex digital droplet PCR (mddPCR) assay was developed to simultaneously detect cg06268921 and cg23247845 in cfDNA from TNBC ( = 33) and non-TNBC ( = 80) patients.

RESULTS

We identified 113 DMCs, of which eight were selected as optimal markers. They effectively discriminated TNBC from non-TNBC tissues. Then an eight-marker diagnostic panel was developed with an AUC of 0.922 in TCGA and 0.875 in GSE69914. Among them, cg06268921 was significantly associated with overall survival (hazard ratio = 0.249, = 0.044) and disease-free survival (hazard ratio = 0.194, = 0.015) in the TCGA-TNBC cohort. In the cfDNA cohort, cg06268921 significantly differentiated TNBC from non-TNBC ( < 0.001), and the combination of both markers yielded an AUC of 0.728. The findings demonstrated the potential of methylation signatures as non-invasive diagnostic tools for TNBC. Future research with larger cohorts is warranted.