Exploratory Study of Prognostic Plasma Biomarkers in Patients with Pulmonary Arterial Hypertension

Figure 1. Increased plasma levels of oncology- and neurology-related proteins in patients with PAH. A: Volcano plot representations of the plasma proteins (oncology panel) differentially expressed between patients with PAH and controls (CTRLs), and patients with PAH categorized as decompensated right ventricle [dRV; cardiac index ≤ 2.2 L/(min·m2)] and controls. Green and orange points represent up-regulated proteins that meet both criteria for significant change (absolute fold change ≥1.5 and adjusted P_value ≤ 0.05) between the indicated groups. B: Venn diagram showing the distribution and overlap of up-regulated plasma proteins belonging to the oncology panel between the identified groups. C: Volcano plot representations of the plasma proteins (neurology panel) differentially expressed between patients with PAH and controls, and patients with PAH categorized as dRV and controls. Blue and purple points represent up-regulated proteins that meet both criteria for significant change (absolute fold change ≥1.5 and adjusted P_value ≤ 0.05) between the indicated groups. D: Venn diagram showing the distribution and overlap of up-regulated plasma proteins belonging to the neurology panel between…

Figure 2. Expression levels [normalized protein expression (NPX)] of FGF23, ATOX1, EDA2R, WFDC2, ADM, KLK4, MDK, CALCA, TNFRSF6B, THBS2, SFRP1, GGT1, TNFRSF10B, and SPINK1 in control (CTRL) patients and patients with PAH. A: According to contemporary cardiac index (CI) measured by right heart catheterization at the time blood sample was drawn, patients with PAH were categorized as compensated right ventricle [cRV; CI > 2.2 L/(min·m2)] and decompensated right ventricle [dRV; CI ≤ 2.2 L/(min·m2)]. B: Expression levels (NPX) of indicated proteins in patients with PAH classified as idiopathic PAH (IPAH) or connective tissue disease associated with PAH (CTD-PAH). C: Correlation matrix showing how the identified protein biomarkers correlate with one another in patients with PAH and how they correlate with sex, age, body mass index (BMI), PAH type, mean pulmonary artery pressure (mPAP), cardiac output (CO), CI, pulmonary vascular resistance (PVR), estimated glomerular filtration rate (eGFR), and N-terminal pro B-type natriuretic peptide (NT-proBNP; measured at the hospital). The color reflects the magnitude and direction of the Spearman rank correlation coefficient, ranging from red (positive…

Figure 3. Receiver operating characteristic (ROC) curves assessing the prognostic accuracy of FGF23, EDA2R, KLK4, ATOX1, WFDC2, ADM, MDK, TNFRSF6B, THBS2, CALCA, GGT1, SFRP1, TNFRSF10B, and SPINK1; and associated Kaplan-Meier survival curves estimating transplant-free survival. ROC curves were analyzed for death or lung transplantation in patients with PAH. In Kaplan-Meier survival curves, patients with PAH were divided into two groups based on optimal cutoff values identified by the ROC curves. n = 60 patients with PAH. AUC, area under the curve.

Figure 4. Incremental predicted performance of identified biomarkers and expression in animal models. A: Performance of the logistic regression models with 2015 European Respiratory Society (ERS)/European Society of Cardiology (ESC) risk score, REVEAL 2.0 risk score, and the refined four-strata risk assessment method in patients with PAH for death and lung transplantation. Graph of C-statistics with 95% CI for each biomarker and their combinations. C-statistics were compared using the DeLong test with each reference. B: Heat map showing RNA-sequencing expression data of the indicated genes [in transcripts per million (TPM)] obtained from the Genotype-Tissue Expression (GTEx version 8) database (https://gtexportal.org/home, last accessed August 13, 2024) in lung, heart, artery, kidney, liver, and adipose tissue. C: Western blot analyses and corresponding quantifications of WFDC2 (HE4) in right ventricles and dissected pulmonary arteries (PAs) from monocrotaline (MCT)–injected and Sugen/hypoxia (Su/Hx)–exposed rats. D: Western blot analyses and corresponding quantifications of WFDC2 (HE4) in kidney from MCT-injected and Su/Hx-exposed rats. E:Wfdc2 mRNA expression in right ventricle (RV),…

Figure 5. Combination of the four 384-plex panels focused on diseases and biological processes and functional analysis of deregulated plasma proteins. A: Volcano plot representations of the plasma proteins (cardiometabolic and inflammatory panels) differentially expressed between patients with PAH and controls (CTRLs), and patients with PAH categorized as decompensated right ventricle (dRV) and controls. Pink and red points represent up-regulated proteins that meet both criteria for significant change (absolute fold change ≥1.5 and adjusted P_value ≤ 0.05) between the indicated groups. The vertical dashed lines represent the cutoff levels of fold change ≥1.5, and the horizontal dashed lines signal statistical significance threshold (adjusted P_value ≤ 0.05). B: Venn diagram showing the distribution and overlap of up-regulated plasma proteins belonging to these panels between the identified groups. C: Quality control for Olink plasma proteomics data. Correlation (Pearson) between normalized protein expression (NPX) values of the same protein targets [IL-6, tumor necrosis factor (TNF), and CXCL8] measured across all four protein panels. D: Kyoto Encyclopedia of Genes and Genomes (KEGG)…

Figure 6. Plasma proteins independently associated with death or lung transplantation in patients with PAH after adjusting for risk assessment tools. A: Multivariate Cox proportional hazard analyses for death or lung transplantation of patients with PAH. Each biomarker was adjusted for 2015 European Society of Cardiology (ESC)/European Respiratory Society (ERS) risk score, age, and sex (turquoise), REVEAL 2.0 risk score (purple), and refined four-strata risk assessment, age, and sex (red). Data are presented as hazard ratio and 95% CI. B: Heat map (top left) and correlation matrix (top right and bottom) for the 23 proteins up-regulated in the plasma of patients with PAH and showing independent prognostic value after adjusting for age, sex, and 2015 ESC/ERS risk score, the REVEAL 2.0 risk score, or age and sex; and the refined four-strata risk assessment. For each protein biomarker, the area under the curve (AUC) and P value of the prognostic receiver operating characteristic (ROC) curve analysis are shown. The Spearman correlation matrix shows how the protein biomarkers correlate with one another (top right) and how they correlate with sex, age, hemodynamic parameters, and hospital…

Supplemental Figure S1. Receiver operating characteristic (ROC) curves assessing the prognostic accuracy of N-terminal pro B-type natriuretic peptide (NT-proBNP) measured at the hospital (clinical) or by Olink proximity extension assay. ROC curves were analyzed for death or lung transplantation in patients with PAH. A: ROC curve using NT-proBNP was measured at the hospital using a flex reagent cartridge (clinical). B: ROC curve using NT-proBNP measured by Olink. n = 60 (A and B). AUC, area under the curve; NPX, normalized protein expression.

Supplemental Figure S2. Transcript and protein levels of EDA2R and TNFRSF10B in animal models of pulmonary hypertension. A: Western blot analyses and corresponding quantifications of EDA2R and TNFRSF10B in right ventricles from monocrotaline (MCT)–injected and Sugen/hypoxia (Su/Hx)–exposed rats. B: Representative Western blot analyses and corresponding quantifications of EDA2R and TNFRSF10B in dissected pulmonary arteries (PAs) from MCT-injected and Su/Hx-exposed rats. C: Western blot analyses and corresponding quantifications of EDA2R and TNFRSF10B in kidney from MCT-injected and Su/Hx-exposed rats. Protein expression was normalized to amido black (AB). D:Eda2r and Tnfrsf10b mRNA expression in right ventricle (RV), lung, and kidney of control (CTRL) and MCT-injected or Su/Hx-exposed rats. Scatter dot plots show individual values. Statistical analyses were performed using t-test. Data are given as means ± SEM (A–D). n = 6 to 16 (D). ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001.

Supplemental Figure S3. WFDC2 promotes proliferation of control pulmonary artery smooth muscle cells (PASMCs) and control cardiac fibroblasts (CFs). A: Representative fluorescent images of Ki-67–labeled control PASMCs and control CFs, exposed or not to recombinant human WFDC2 (rhWFDC2; 40 nmol/L) for 24 and 48 hours, respectively. The corresponding quantifications are shown. B: Gene expression of FOXM1, BIRC5, and IL1b in control PASMCs exposed or not to rhWFDC2 (40 nmol/L) for 24 hours, as determined by real-time quantitative PCR (qPCR). C: Gene expression of WFDC2 in control PASMCs exposed or not to tumor necrosis factor (TNF)-α (10 ng/mL) or platelet-derived growth factor (PDGF)-BB (30 ng/mL) for 24 hours, as determined by qPCR. D: mRNA levels of Wfdc2 in the lung of monocrotaline-injected rats treated or not with macitentan and tadalafil. Scatter dot plots show individual values and. Statistical analyses were performed using t-test. Data are given as means ± SEM (A–D). n = 3 control PASMCs (A); n = 4 control CFs (A); n = 5 control PASMCs (B); n = 4 control PASMCs (C); n = 5 to 11 (D). ∗P < 0.05. Scale bars = 100 μm (A). Veh, vehicle.

Supplemental Figure S4. A multiprotein prognostic model. A: Receiver operating characteristic (ROC) curves for death or lung transplantation in patients with PAH. The least absolute shrinkage and selection operator (LASSO) model identified selected four biomarkers (PTN, SPINK1, EDA2R, and GGT1). The area under the curves (AUC) of the combined model was significantly higher than that of N-terminal pro B-type natriuretic peptide (NT-proBNP; AUC, 0.856 versus 0.713). B: Patients with PAH were divided into groups based on the scores based on the optimal cutoff value of each ROC curve of the selected four biomarkers (PTN, SPINK1, EDA2R, and GGT1) for death or lung transplantation in patients with PAH. n = 60 (A). NPX, normalized protein expression.