Proprotein Convertase Subtilisin/Kexin Type 6: A Risk Factor for Survival in Pulmonary Fibrosis?

Genetic risk factors explain a large portion of attributable risk (1) in idiopathic pulmonary fibrosis (IPF) and represent important targets for early detection before the lung is irreversibly scarred, and therapeutic intervention for this disease with limited treatment options (2, 3). Common variants in 17 genetic loci have been associated with IPF through genome-wide association studies (GWASs) (4–9). This includes the most strongly associated and most replicated common variant in the promoter of the airway mucin MUC5B gene (10). Rare variants in telomere pathway–related genes, surfactant genes, and KIF15 (Kinesin Family Member 15) have also been associated with familial forms of pulmonary fibrosis (11–13) and with IPF more generally through recent whole-genome sequencing (WGS) studies (14, 15). Risk variants in telomerase-related genes (15) and shorter telomeres (16) have also been associated with reduced survival, whereas the MUC5B promoter variant has been shown to offer a survival advantage (17).

With the goal of identifying additional genetic risk factors for poor outcomes in IPF, in this issue of the Journal (pp. 1515–1524) Oldham and colleagues report the results of a two-stage, multicenter, international GWAS of transplant-free survival (18). Because of the inadequate sample size, the authors were unable to perform a true discovery and/or replication analysis, so they performed a meta-analysis of the two groups of samples, one with genotyping array data and the other with WGS data available. A stage I GWAS was performed on 1,481 IPF cases from three cohorts of European ancestry with previously published genotyping array data: 1) United States (118 cases from the University of Chicago and 200 cases from the University of Pittsburgh); 2) United Kingdom (119 from the University of Edinburgh, 210 from the Trent Lung Fibrosis Study, 175 from the PROFILE Study, and 61 from Hull and Papworth); and 3) a combined cohort from the United States, United Kingdom, and Spain (187 from the University of Chicago [independent of those from the U.S. cohort], 299 from the PROFILE study, 84 from the University of California, and 28 aggregated patients from centers in Spain). A stage II GWAS was performed on 397 cases with WGS data from Vanderbilt.

Using this staged analysis approach, the authors identified four genetic variants associated with transplant-free survival, with consistent effect direction in the two stages of genetic association analyses. Three variants remained significant in two models that adjusted for known confounders in IPF survival analysis: baseline gender-age-physiology (GAP) stage (model I); and sex, baseline age, FVC (% predicted), and Dl_CO (% predicted) (model II). Of these, rs35647788, an intronic variant in the PCSK6 (proprotein convertase subtilisin/kexin type 6) gene, showed the strongest association across stage I and stage II cohorts and crossed the genome-wide significance threshold in the meta-analysis (hazard ratio, 4.11; 95% confidence interval, 2.54–6.67; P = 9.45 × 10⁻⁹). To investigate the potential functional role of this novel genetic candidate in IPF, the authors showed that 1) the PCSK6 protein was highly expressed in IPF lung tissue—most highly in the airway epithelium, adventitial fibroblasts, and lymphatic endothelial cells—and was negatively correlated with survival; and 2) expression of the PCSK6 transcript in peripheral blood, as well as 3) plasma concentration of the PCSK6 protein were associated with reduced transplant-free survival.

Overall, this is an important study that identified a novel risk variant that appears to affect survival and a protein that may become another important therapeutic target in this devastating and largely untreatable disease. However, there are also a number of unanswered questions that will require further investigation. Among the most important next steps would be a true replication study, especially given the rare nature of this variant (<1% minor allele frequency in the study population) and inconsistencies in its presence across the cohorts included in the stage I analysis. Furthermore, given its rarity and intronic location, this variant is not likely to be the only genetic factor that is important in regulating the expression of the PCSK6 transcript and protein. It would be essential to perform a careful expression quantitative trait locus mapping and colocalization analysis of all genetic variants in the region to better understand the link between genetics and gene expression in the lung. Understanding cell specificity of this signal (airway epithelium, for example) would also be important. Finally, using cell and animal models to decipher the functional role of PCSK6 in the pathobiology of IPF will be critical. PCSK6, also referred to as PACE4, encodes a widely expressed calcium-dependent serine endoprotease with an unknown role in lung fibrosis. PCSK6 or PACE4 has been linked to patterning the early mouse embryo through regulation of TGFβ signaling (19); this link to TGFβ signaling in adult lungs should be explored. Tools such as PCSK6-deficient mice are available and have been used to demonstrate the key role of this protease in vascular remodeling, for example (20). The study by Oldham and colleagues opens up an entire line of investigation into a new target for this devastating and largely untreatable disease.