+1 Recommend
0 collections
      • Record: found
      • Abstract: not found
      • Article: not found

      Discovery of Distinct Immune Phenotypes Using Machine Learning in Pulmonary Arterial Hypertension

      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.


          Accumulating evidence implicates inflammation in pulmonary arterial hypertension (PAH) and therapies targeting immunity are under investigation, though it remains unknown if distinct immune phenotypes exist. Identify PAH immune phenotypes based on unsupervised analysis of blood proteomic profiles. In a prospective observational study of Group 1 PAH patients evaluated at Stanford University (discovery cohort, n=281) and University of Sheffield (validation cohort, n=104) between 2008–2014, we measured a circulating proteomic panel of 48 cytokines, chemokines, and factors using multiplex immunoassay. Unsupervised machine learning (consensus clustering) was applied in both cohorts independently to classify patients into proteomic immune clusters, without guidance from clinical features. To identify central proteins in each cluster, we performed partial correlation network analysis. Clinical characteristics and outcomes were subsequently compared across clusters. Four PAH clusters with distinct proteomic immune profiles were identified in the discovery cohort. Cluster 2 (n=109) had low cytokine levels similar to controls. Other clusters had unique sets of upregulated proteins central to immune networks– cluster 1 (n=58)(TRAIL, CCL5, CCL7, CCL4, MIF), cluster 3 (n=77)(IL-12, IL-17, IL-10, IL-7, VEGF), and cluster 4 (n=37)(IL-8, IL-4, PDGF-β, IL-6, CCL11). Demographics, PAH etiologies, comorbidities, and medications were similar across clusters. Non-invasive and hemodynamic surrogates of clinical risk identified cluster 1 as high-risk and cluster 3 as low-risk groups. Five-year transplant-free survival rates were unfavorable for cluster 1 (47.6%, CI 35.4–64.1%) and favorable for cluster 3 (82.4%, CI 72.0–94.3%)(across-cluster p<0.001). Findings were replicated in the validation cohort, where machine learning classified four immune clusters with comparable proteomic, clinical, and prognostic features. Blood cytokine profiles distinguish PAH immune phenotypes with differing clinical risk that are independent of World Health Organization Group 1 subtypes. These phenotypes could inform mechanistic studies of disease pathobiology and provide a framework to examine patient responses to emerging therapies targeting immunity.

          Related collections

          Most cited references36

          • Record: found
          • Abstract: found
          • Article: not found

          Inflammation and immunity in the pathogenesis of pulmonary arterial hypertension.

          This review summarizes an expanding body of knowledge indicating that failure to resolve inflammation and altered immune processes underlie the development of pulmonary arterial hypertension. The chemokines and cytokines implicated in pulmonary arterial hypertension that could form a biomarker platform are discussed. Pre-clinical studies that provide the basis for dysregulated immunity in animal models of the disease are reviewed. In addition, we present therapies that target inflammatory/immune mechanisms that are currently enrolling patients, and discuss others in development. We show how genetic and metabolic abnormalities are inextricably linked to dysregulated immunity and adverse remodeling in the pulmonary arteries.
            • Record: found
            • Abstract: found
            • Article: not found

            Discovery of meaningful associations in genomic data using partial correlation coefficients.

            A major challenge of systems biology is to infer biochemical interactions from large-scale observations, such as transcriptomics, proteomics and metabolomics. We propose to use a partial correlation analysis to construct approximate Undirected Dependency Graphs from such large-scale biochemical data. This approach enables a distinction between direct and indirect interactions of biochemical compounds, thereby inferring the underlying network topology. The method is first thoroughly evaluated with a large set of simulated data. Results indicate that the approach has good statistical power and a low False Discovery Rate even in the presence of noise in the data. We then applied the method to an existing data set of yeast gene expression. Several small gene networks were inferred and found to contain genes known to be collectively involved in particular biochemical processes. In some of these networks there are also uncharacterized ORFs present, which lead to hypotheses about their functions. Programs running in MS-Windows and Linux for applying zeroth, first, second and third order partial correlation analysis can be downloaded at: http://mendes.vbi.vt.edu/tiki-index.php?page=Software. Supplementary information can be found at: URL to be decided.
              • Record: found
              • Abstract: found
              • Article: not found

              Modern age pathology of pulmonary arterial hypertension.

              The impact of modern treatments of pulmonary arterial hypertension (PAH) on pulmonary vascular pathology remains unknown. To assess the spectrum of pulmonary vascular remodeling in the modern era of PAH medication. Assessment of pulmonary vascular remodeling and inflammation in 62 PAH and 28 control explanted lungs systematically sampled. Intima and intima plus media fractional thicknesses of pulmonary arteries were increased in the PAH group versus the control lungs and correlated with pulmonary hemodynamic measurements. Despite a high variability of morphological measurements within a given PAH lung and among all PAH lungs, distinct pathological subphenotypes were detected in cohorts of PAH lungs. These included a subset of lungs lacking intima or, most prominently, media remodeling, which had similar numbers of profiles of plexiform lesions as those in lungs with more pronounced remodeling. Marked perivascular inflammation was present in a high number of PAH lungs and correlated with intima plus media remodeling. The number of profiles of plexiform lesions was significantly lower in lungs of male patients and those never treated with prostacyclin or its analogs. Our results indicate that multiple features of pulmonary vascular remodeling are present in patients treated with modern PAH therapies. Perivascular inflammation may have an important role in the processes of vascular remodeling, all of which may ultimately lead to increased pulmonary artery pressure. Moreover, our study provides a framework to interpret and design translational studies in PAH.

                Author and article information

                Circulation Research
                Circ Res
                Ovid Technologies (Wolters Kluwer Health)
                March 15 2019
                March 15 2019
                : 124
                : 6
                : 904-919
                [1 ]From the Division of Pulmonary and Critical Care Medicine (A.J.S., M.R.N., R.T.Z.), in the Department of Medicine, Stanford University, CA
                [2 ]Vera Moulton Wall Center for Pulmonary Vascular Disease, Stanford University, CA (A.J.S., A.H., M.R.N., M.R., R.T.Z.)
                [3 ]Quantitative Sciences Unit (H.K.H., V.B.), in the Department of Medicine, Stanford University, CA
                [4 ]Division of Immunology and Rheumatology (L.K.B., W.H.R.), in the Department of Medicine, Stanford University, CA
                [5 ]Division of Cardiovascular Medicine (F.H.), in the Department of Medicine, Stanford University, CA
                [6 ]Stanford Cardiovascular Institute (F.H.), in the Department of Medicine, Stanford University, CA
                [7 ]Department of Infection, Immunity, and Cardiovascular Disease, University of Sheffield, United Kingdom (P.M.H., A.L.)
                [8 ]Sheffield Pulmonary Vascular Disease Unit, Royal Hallamshire Hospital, United Kingdom (R.C.).
                [9 ]Institute for Immunity, Transplantation, and Infection (M.R.N., P.K.), in the Department of Medicine, Stanford University, CA
                [10 ]Department of Pediatric Cardiology, Stanford University, CA (M.R.)
                [11 ]Division of Biomedical Informatics Research (P.K.) in the Department of Medicine, Stanford University, CA
                © 2019


                Comment on this article