Identification of hypertension subtypes using microRNA profiles and machine learning.

in European journal of endocrinology by Smarti Reel, Parminder Singh Reel, Josie Van Kralingen, Casper K Larsen, Stacy Robertson, Scott M MacKenzie, Alexandra Riddell, John D McClure, Stelios Lamprou, John M C Connell, Laurence Amar, Alessio Pecori, Martina Tetti, Christina Pamporaki, Marek Kabat, Filippo Ceccato, Matthias Kroiss, Michael C Dennedy, Anthony Stell, Jaap Deinum, Paolo Mulatero, Martin Reincke, Anne-Paule Gimenez-Roqueplo, Guillaume Assié, Anne Blanchard, Felix Beuschlein, Gian-Paolo Rossi, Graeme Eisenhofer, Maria Christina Zennaro, Emily Jefferson, Eleanor Davies

TLDR

  • The study used machine learning algorithms to identify circulating microRNA biomarkers for endocrine hypertension and its subtypes, achieving high accuracy in classification and confirming the potential of microRNAs as diagnostic biomarkers.
  • The most prominent biomarkers identified were hsa-miR-15a-5p and hsa-miR-32-5p.

Abstract

Hypertension is a major cardiovascular risk factor affecting about 1 in 3 adults. Although the majority of hypertension cases (∼90%) are classified as 'primary hypertension' (PHT), endocrine hypertension (EHT) accounts for ∼10% of cases and is caused by underlying conditions such as primary aldosteronism (PA), Cushing's syndrome (CS), pheochromocytoma or paraganglioma (PPGL). EHT is often misdiagnosed as PHT leading to delays in treatment for the underlying condition, reduced quality of life and costly, often ineffective, antihypertensive treatment. MicroRNA circulating in the plasma is emerging as an attractive potential biomarker for various clinical conditions due to its ease of sampling, the accuracy of its measurement and the correlation of particular disease states with circulating levels of specific microRNAs. This study systematically presents the most discriminating circulating microRNA features responsible for classifying and distinguishing EHT and its subtypes (PA, PPGL, CS) from PHT using 8 different supervised machine learning (ML) methods for the prediction. The trained models successfully classified PPGL, CS and EHT from PHT with AUC 0.9 and PA from PHT with AUC 0.8 from the test set. The most prominent circulating microRNA features for hypertension identification of different disease combinations were hsa-miR-15a-5p and hsa-miR-32-5p. This study confirms the potential of circulating microRNAs to serve as diagnostic biomarkers for EHT and the viability of machine learning as a tool for identifying the most informative microRNA species.

Overview

  • The study aimed to identify and classify circulating microRNA biomarkers for endocrine hypertension (EHT) and its subtypes using machine learning algorithms.
  • The study utilized 8 different supervised machine learning methods to predict EHT and its subtypes from primary hypertension (PHT) and presented the most discriminating circulating microRNA features for classification.
  • The primary objective of the study was to develop a diagnostic biomarker for EHT using circulating microRNAs, with a focus on accuracy, ease of measurement, and correlation with disease states.

Comparative Analysis & Findings

  • The trained models successfully classified PPGL, CS, and EHT from PHT with an AUC of 0.9, and PA from PHT with an AUC of 0.8 from the test set.
  • The most prominent circulating microRNA features for hypertension identification of different disease combinations were hsa-miR-15a-5p and hsa-miR-32-5p.
  • The study confirmed the potential of circulating microRNAs as diagnostic biomarkers for EHT and the viability of machine learning as a tool for identifying informative microRNA species.

Implications and Future Directions

  • The study's findings have the potential to improve diagnostic accuracy and reduce delays in treatment for patients with EHT, leading to improved quality of life and reduced healthcare costs.
  • Future studies can focus on validating these findings in larger cohorts and exploring the use of circulating microRNAs as biomarkers for other clinical conditions.
  • The application of machine learning algorithms to analyze circulating microRNAs may lead to the development of personalized diagnostic and therapeutic strategies for EHT and other diseases.
Read Full Article