Photo Credit: Freepick

The following is a summary of “Identification of hypertension subtypes using microRNA profiles and machine learning,” published in the March 2025 issue of European Journal of Endocrinology by Reel et al. 

Hypertension is a major cardiovascular risk factor, with endocrine hypertension (EHT) comprising 10% of cases, often misdiagnosed as primary hypertension (PHT), leading to delayed treatment and reduced QoL.  

Researchers conducted a retrospective study to investigate the potential of circulating microRNA as a biomarker for differentiating EHT from PHT.  

They systematically identified the most distinguishing circulating microRNA features that classified and differentiated EHT and its subtypes, including primary aldosteronism (PA), pheochromocytoma or paraganglioma (PPGL), and Cushing’s syndrome (CS), from PHT. They applied 8 different supervised machine learning (ML) methods for prediction, ensuring an objective approach to classification.  

The results showed that the trained models accurately classified PPGL, CS, and EHT from PHT with an AUC of 0.9. The PA was distinguished from PHT with an AUC of 0.8 using the test set. The most significant circulating microRNA features for identifying hypertension across different disease combinations were hsa-miR-15a-5p and hsa-miR-32-5p.  

Investigators concluded that circulating microRNAs demonstrated potential as diagnostic biomarkers for EHT, and ML effectively identified the most informative microRNA species. 

Source: academic.oup.com/ejendo/advance-article/doi/10.1093/ejendo/lvaf052/8086791