Searching for peripheral proteomic markers of primary aldosteronism

Primary aldosteronism (PA) is prevalent among hypertensive patients, and associated with worsened cardiovascular outcomes compared to essential hypertension (HT). Screening and diagnostics for PA are currently complicated and invasive, why new methods are needed. Unilateral PA (uPA) is best treated surgically, and bilateral PA (bPA) - medically. No validated proteomic diagnostic test has been found yet. Our aim was to explore proteomic markers in peripheral serum to discriminate between HT, PA, uPA and bPA.

Eighty-eight hypertensive individuals were evaluated for PA, and diagnosed with HT (n = 30); bPA (n = 29); and uPA (n = 29). Serum samples from these study groups were analyzed by Olink® Explore 384 Cardiometabolic Panel. A machine learning model based on ridge logistic regression with a stratified 5-fold cross-validation was used to identify HT, PA, bPA and uPA.

In the study groups, 56 circulating proteins were significantly different, and some of them specifically: 4 between PA vs. HT; 3 between bPA vs. uPA; 1 between bPA vs. HT; 9 between uPA vs. HT; 1 between HT vs. bPA vs. uPA. Three proteins with strongest differentiation (Coagulation factor IX for PA vs. HT; dipeptidyl peptidase 4 for uPA vs. HT and bPA; heat shock protein B1 for bPA vs. uPA) were validated by enzyme-linked immunosorbent assay. Our machine learning model could successfully identify 95% of HT, bPA, and uPA samples.

Serum protein biomarkers may serve as a tool for discriminating HT, PA, uPA and bPA. Further studies are needed to support our results.