A new blood test based on DNA methylation can identify patients with type 2 diabetes at the highest risk of heart attack or stroke of years in advance, beating traditional risk scores and opening the door to more previous and more targeted prevention.
Study: epigenetic biomarkers predict macrovascular events in people with type 2. Image credit: Dragon Claws / Shutterstock
In a recent study published in the journal Cell reports medicineResearchers have identified epigenetic biomarkers predicting incident macrovascular events (IME) in people with type 2 diabetes (T2D).
People with T2D have a risk of cardiovascular disease (CVD) of two to four times higher than non -diabetic individuals. The identification of individuals at risk of macrovascular events (MES) is crucial for disease prevention. However, the prediction of me in people with T2D is sub-optimal. Although risk stratification scores for MCVs are available, they have moderate capacity to stratify T2D patients. As such, it is necessary to identify new biomarkers to improve the prediction of MCV and ME in T2D subjects.
Study and results
In this study, researchers have identified blood -based epigenetic biomarkers to predict the IME in newly diagnosed with T2D. They included 752 participants with newly diagnosed T2D data with available DNA methylation data and without my known diabetic cohorts in scania (Andis) and the county of Uppsala (Andiu) in the “prospective cohort for my in the T2D”. Among these, 102 individuals have developed MES on an average follow-up of about four years and a maximum of seven years, while 650 did not do so.
The researchers analyzed DNAM of more than 853,000 blood sites to identify epigenetic biomarkers associated with future SEM. They found that the DNAM of 461 sites was associated with the Imes using a cox regression model adjusted for sex, age, body mass index (BMI) and glycated hemoglobin (HBA1C). These sites, annotated at 422 genes, were distributed through the genome.
GARCı´a-Calzo´ n et al. Identify a blood -based epigenetic biomarker tool that helps predict future cardiovascular diseases, which people with type 2 diabetes are more likely to develop. Their results support the use of epigenetic biomarkers to improve risk stratification and guide more personalized prevention strategies in diabetes care.
In addition, the results remained consistent, with around 453 sites associated with the IME, after an additional adjustment for drugs, smoking, lipid profiles and cell composition. Then, the team examined if a risk of methylation risk score (MRS) would predict the future MES among people newly diagnosed with T2D. Methylation sites with absolute methylation differences ≥ 2% between witnesses and individuals with the IME were filtered for inclusion in the MRS.
The MRS included 87 methylation sites; Most sites (74%) have been hypomethylated among those with IME compared to witnesses. In addition, the MRS was significantly different between witnesses and the individuals with my. Then, the team carried out a cross validation five times using logistics models to assess whether the MRS could make the difference between witnesses and the individuals with the IME in the prospective cohort of my in T2D.
For comparison, a cross validation analysis was carried out with only clinical risk factors of MES. In addition, the combination of clinical factors and MRS has been examined. The receiver’s operating characteristic curves revealed that the MRS predicts the IMEs with an area under the curve (AUC) of 0.81, while the clinical risk factors predicted the Imes solely with an AUC of 0.69. The combination of the MRS and clinical risk factors gave an ASC of 0.84.
Statistical tests have shown that the MRS has been significantly better than clinical risk factors alone (p = 0.001), and the combined model has significantly surpassed clinical risk factors (p = 1.7 × 10⁻¹). Precision-recording analysis, appropriate for unbalanced data, has also shown improved sensitivity and precision for MRS-based models.
Then, MRS’s ability to predict EMS has been compared to established risk scores, a polygenic risk score (PRS) based on 204 monocleotide polymorphisms associated with coronary arteries (SNP) and epigenetic mortality and aging clocks. The study of prospective diabetes of the United Kingdom (UKPDS) and the risk scores of diabetes score2 which predict the MCV in diabetic individuals had AUC of 0.54 and 0.62, respectively, in the forecast of MES.
The PRS, epigenetic clocks and mortality and other risk scores (multi -thnical study of atherosclerosis, atherosclerotic MCVs and framingham risk scores) were significantly worse than the MRS or MRS plus clinical factors, with AUCs going between 0.61 and 0.68. The optimal cutting point for the combined biomarker tool, which showed the best prediction, was estimated at 0.023, with a sensitivity and a specificity of 0.804 and 0.728, respectively. During this threshold, the model obtained a high negative predictive value of 95.9% and a moderate positive predictive value of 31.8%, which means that it could reliably identify individuals that are unlikely to feel IMEs but were less precise to confirm those in danger. Analyzes for improving net reclassify showed a continuous improvement of 28.2% and 90.2% compared to clinical risk factors alone. The test was estimated at the cost of around $ 200 per sample, a factor suggesting that the authors could be possible for clinical screening if they were selectively used in high -risk populations.
In addition, researchers have evaluated whether the 64 genes annotated at 87 methylation sites in the MRS have a differential expression in carotid plates of symptomatic and asymptomatic patients. They found the differential expression of four genes in symptomatic patients compared to asymptomatic patients. They also pointed out that 72% of the MRS genes had prior links with MCV in the data in the literature or the Gwas, and that several methylation sites overlapped with those differentially methylated in the fabric of the aortic plate.
Then, the team carried out validation analyzes of the methylation sites associated with the IMES in optimized and epic-Potsdam cohorts. They validated 43 and 32 methylation sites in optimized and epic-updam cohorts, respectively, using models of regression of cox adjusted for BMI, age, sex and HBA1C. In addition, five sites out of 87 in the MRS were associated with the IME in the optimal cohort.
An MRS developed using these five sites (MRS5SITES) was significantly different between witnesses and individuals with the IME in the optimal cohort and the prospective cohort of my T2D. The combination of clinical risk factors and MRS5 had an ASC of 0.8 in the optimal cohort and 0.78 in the prospective cohort of MES in T2D. In particular, newly diagnosed T2D participants included, while Epic-Potsdam was a general population cohort, supporting the broader applicability of the results.
Conclusions
In short, the study has identified and validated an epigenetic biomarker based on blood predicting the risk of the first ME, in combination with and regardless of clinical risk factors, in newly diagnosed with T2D. T2D subjects presenting an MRS in combination with clinical risk factors of more than 0.023 were likely to develop MES on the maximum monitoring of seven years. Overall, the epigenetic biomarker better predicts established risk scores, supporting its future clinical use.
However, the authors note that an external validation in other ethnic groups is necessary and that the moderate PPV probably reflects the low prevalence of events in this newly diagnosed population. They also warn that environmental factors such as food, physical activity and drugs can influence DNA methylation and that they were not entirely captured in current cohorts.