Background: The ENSAT-HT project aims to discover biomarkers for forms of adrenal hypertension (AHT) based on a multi-omics approach, including untargeted plasma NMR metabobolomics. Our goal is to define the metabolic signature of AHT patients suffering from either Primary Aldosteronism (PA, n=106), Cushing’s Syndrome (CS, n=33) or Pheochromocytoma and Paraganglioma (PPGL, n=96), and compare it to primary hypertensives (PHT, n=40).
Methods: Plasma samples were collected and analyzed using proton Nuclear Magnetic Resonance spectroscopy (1H-NMR). The NMR spectra were converted to a readable table and the resulting peak data were subsequently analyzed by multivariate statistical analysis (MVA) methods; Principal Component Analysis (PCA) was applied to check for outliers and trends within the dataset and Partial Least Squares-Discriminant Analysis (PLS-DA) to build sample classification models for the following scenarios: AHT vs PHT, PPGL vs CS vs PA vs PHT, PPGL vs PHT, PA vs PHT and CS vs PHT. Finally, ASCA (ANOVA Simultaneous Component Analysis) was used for the investigation of unwanted sources of variation and their influence on sample classification.
Results: The metabolomics method was able to classify samples as either Primary Hypertensives or Adrenal Hypertensives with high accuracy. Namely, 75% of samples were classified correctly as either AHT or PHT, PA or PHT with 77%, PPGL or PHT with 75%, CS or PHT with 70%. In the AHT vs PHT model, a 77% sensitivity was achieved and a 75% specificity on average. The most important NMR signals for highlighting the overall AHT signature included those of lactate, creatine, pyruvate, ornithine, ketone bodies and acetate.
Conclusions: Based on the results presented here, all forms of adrenal hypertension have a significant impact on metabolism. By using plasma NMR metabolomics and a novel data processing pipeline, we were able to stratify patients belonging to each group with high accuracies. Future perspectives include validation of the metabolic signature and investigation of relative pathways.