A1 Refereed original research article in a scientific journal
Blood DNA methylation and breast cancer risk: a prospective nested case–control study; 
Authors: Rodriguez, Juan; Grassmann, Felix; Kaukonen, Damien; Hägg, Sara; Humphreys, Keith; Eriksson, Mikael; Zhang, Yuqi; Morawitz, Katharina; Gabrielson, Marike; Hall, Per; Czene, Kamila
Publisher: Elsevier BV
Publication year: 2026
Journal: EBioMedicine
Article number: 106352
Volume: 129
eISSN: 2352-3964
DOI: https://doi.org/10.1016/j.ebiom.2026.106352
Publication's open availability at the time of reporting: Open Access
Publication channel's open availability : Open Access publication channel
Web address : https://doi.org/10.1016/j.ebiom.2026.106352
Self-archived copy’s web address: https://research.utu.fi/converis/portal/detail/Publication/526945294
Self-archived copy's licence: CC BY
Self-archived copy's version: Publisher`s PDF
Background
Blood tests that predict breast cancer (BC) risk before diagnosis could complement mammography screening. Whole-blood DNA methylation is stable and may capture early systemic changes related to tumour development.
Methods
We conducted a prospective matched case–control study nested within Sweden’s mammography screening programme, profiling whole-blood DNA methylation on the Illumina MethylationEPIC array in 377 female participants who later developed BC and 378 age-matched female controls. We identified methylation sites (CpG sites) in Stockholm and validated them in Skåne, after stringent quality control and adjustment for technical and epidemiological confounders. We then constructed a methylation risk score (MRS).
Findings
We identified 87 CpG sites in Stockholm (FDR < 0.05; |β| > 0.015), of which 22 validated in Skåne (P < 0.05; concordant direction) and formed the MRS. The MRS demonstrated significant associations with BC risk, was consistent across different BC subtypes, and independent of established risk factors. The association of MRS with BC remained relatively stable over a pre-diagnostic window of up to five years. MRS exhibited predictive performance comparable to a polygenic risk score, and it significantly improved the 3–5-year prediagnosis discrimination of established BC risk models (P < 0.05, DeLong’s test). Several CpG sites correlated with immune-related genes, suggesting mechanisms involving immune modulation and tumour progression.
Interpretation
Our methylation signature was significantly associated with BC risk and may complement existing risk models, particularly for estimating 3–5-year risk. These findings suggest that blood DNA methylation may support BC risk stratification, although further validation and assessment of clinical utility are needed.
Downloadable publication This is an electronic reprint of the original article. |
Funding information in the publication:
This work was supported by the Swedish Research Council (Grant 2025-02467), the Swedish Cancer Society (Grants 25 4712, 21 1571 & 24 3696), the Stockholm County Council (Grant FoUI-955110), Cancerföreningen i Stockholm (Grant 231003), Erling-Persson Foundation (Grant 2024 013), Sjöberg Foundation (Grant 2024-998), and EU funds through the Investitionsbank des Landes Brandenburg (ILB, project number: 86000694).