Bayesian Inference Associates Rare KDR Variants with Specific Phenotypes in Pulmonary Arterial Hypertension

Abstract

Background - Approximately 25% of patients with pulmonary arterial hypertension (PAH) have been found to harbor rare mutations in disease-causing genes. To identify missing heritability in PAH we integrated deep phenotyping with whole-genome sequencing data using Bayesian statistics. Methods - We analyzed 13,037 participants enrolled in the NIHR BioResource - Rare Diseases (NBR) study, of which 1,148 were recruited to the PAH domain. To test for genetic associations between genes and selected phenotypes of pulmonary hypertension (PH), we used the Bayesian rare-variant association method BeviMed. Results - Heterozygous, high impact, likely loss-of-function variants in the Kinase Insert Domain Receptor (KDR) gene were strongly associated with significantly reduced transfer coefficient for carbon monoxide (KCO, posterior probability (PP)=0.989) and older age at diagnosis (PP=0.912). We also provide evidence for familial segregation of a rare nonsense KDR variant with these phenotypes. On computed tomographic imaging of the lungs, a range of parenchymal abnormalities were observed in the five patients harboring these predicted deleterious variants in KDR. Four additional PAH cases with rare likely loss-of-function variants in KDR were independently identified in the US PAH Biobank cohort with similar phenotypic characteristics. Conclusions - The Bayesian inference approach allowed us to independently validate KDR, which encodes for the Vascular Endothelial Growth Factor Receptor 2 (VEGFR2), as a novel PAH candidate gene. Furthermore, this approach specifically associated high impact likely loss-of-function variants in the genetically constrained gene with distinct phenotypes. These findings provide evidence for KDR being a clinically actionable PAH gene and further support the central role of the vascular endothelium in the pathobiology of PAH.

Keywords: computed tomography.

Figure 1.

Design of the genetic association study. A, Overview of the analytical approach. Using deep phenotyping, data tags were assigned to patients who shared phenotypic features. Rare sequence variants, called from whole-genome sequencing data, were filtered, and explained cases were labeled. BeviMed was applied to a set of unrelated individuals to estimate the posterior probability of gene-tag associations. B, Consort diagram summarizing the size of the study cohort. C, Schematic representation of the definition of cases, exemplified by the transfer coefficient for carbon monoxide (KCO) lower tertile tag. Cases were defined as individuals carrying a particular tag, whereas patients with missing information or those without a tag were removed from the gene-tag association testing. Individuals from non-pulmonary arterial hypertension (PAH) domains served as controls. CADD indicates combined annotation dependent depletion; and MAF, minor allele frequency.

Figure 2.

Rare variant association study results revealing established and novel genotype-phenotype links. A, Figure showing phenotype tags on the x axis and corresponding posterior probability of genotype-phenotype association on the y axis, as calculated by BeviMed. The definitions of the tags are listed in Table 1. Shape and colour of points indicate the mode of inheritance and impact/consequence type of variants driving the association. Box-and-whisker plots showing the distribution of (B) the transfer coefficient for carbon monoxide (KCO) and (C) the age at diagnosis stratified by genotype across the pulmonary arterial hypertension (PAH) domain. The 2-tailed Wilcoxon signed-rank test was used to determine differences in the medians of the distributions, which are indicated by the bars at the top of the figures providing the respective P values. ACVRL1 indicates activin-like kinase 1; AQP1, aquaporin 1; bial, biallelic; BMPR2, bone morphogenetic protein receptor type 2; EIF2AK4, eukaryotic translation initiation factor 2 alpha kinase 4; FPAH, familial PAH; GDF2, growth differentiation factor 2; I/HPAH, idiopathic/hereditary pulmonary arterial hypertension; lof, loss-of-function; mis, missense; PAH, pulmonary arterial hypertension; PH, pulmonary hypertension; PVOD/PCH, pulmonary veno-occlusive disease/pulmonary capillary hemangiomatosis; and TBX4, T-box transcription factor 4.

Figure 3.

Summary of rare single nucleotide variants (SNVs) and small insertions and deletions (indels) identified in the novel pulmonary arterial hypertension (PAH) candidate gene KDR (kinase insert domain receptor). A, Only rare predicted deleterious variants in KDR are shown (minor allele frequency [MAF] <1/10 000 and combined annotation dependent depletion [CADD] ≥10). SNVs and indels are represented by colored lollipops on top of the protein sequence. The domain annotations were retrieved from Uniprot (accession number P35968). Lollipop colors indicate the consequence type, and sizes represent the variant frequency within a cohort. Missense variants that are predicted to be deleterious (Sorting Intolerant From Tolerant prediction score [SIFT]) and damaging (PolyPhen-2) are colored in red, otherwise in yellow (ie, SIFT and PolyPhen-2 disagree). High impact variants are labelled with the respective Human Genome Variation Society notation. The number of variants by predicted consequence type and cohort is provided in the table. B, Familial segregation of KDR nonsense variant c.183G>A (p.Trp61*) with PAH (ie, reduced KCO and late onset) from father (W000229) to daughter (W000229.d). Sanger sequencing results are shown in the chromatograms. NBR indicates NIHR BioResource—Rare Diseases; and USBB, US PAH Biobank.

Figure 4.

Chest computerized tomography (CT) scans of patients carrying high impact kinase insert domain receptor (KDR) mutations. A, Axial image of CT pulmonary angiogram at the level of the right ventricle (RV) moderator band, showing flattening of interventricular septum, leftwards bowing of the interatrial septum and the enlargement of the right atrium (RA) and RV, indicative of RV strain; bilateral pleural effusion, larger on the right side. B, Axial image of a pulmonary CT angiogram demonstrating enlarged pulmonary artery and mild central lung ground-glass opacity (GGO). C, Axial high-resolution CT slice of the chest in the lung window showing a trace of non-specific GGO with a central distribution. D, Coronal image showing the trace of central GGO and enlarged central pulmonary arteries. Axial high-resolution CT slice of the chest in the lung window showing apical subpleural fibrosis (E), and very minor subpleural fibrosis at the lung bases (F). Axial high-resolution CT slice of the chest in the lung window showing subpleural GGO at apical level (G), and mild GGO at mid-thoracic level (H). Patients: E001392 (A and B), E003448 (C and D), W000229 (E and F), W000274 (G and H).