Bronchoalveolar lavage transcriptome characterise IPF endotypes with prognostic impact

Background: Given the plethora of pathophysiological mechanisms described in IPF, we hypothesize that the mechanisms driving fibrosis in IPF might be different from one case to the other. Research question: Do IPF endotypes exist and do they associate with outcome? Methods: Using a publicly available gene expression dataset retrieved from BAL samples of IPF patients and controls (GSE70867), we clustered IPF samples based on a dimension reduction algorithm specifically designed for -omics data, called DDR Tree. After clustering, gene set enrichment analysis was performed for functional annotation, associations with clinical variables and prognosis were investigated and differences in transcriptional regulation were determined using motif enrichment analysis. The findings were validated in 3 independent publicly available gene expression datasets retrieved from IPF blood samples. Results: 176 IPF samples from 3 centres were clustered in 6 IPF clusters, with distinct functional enrichment. Although clinical characteristics did not differ between the clusters, one cluster conferred worse GAP-score corrected survival, while another showed a numeric trend towards worse survival (p=0.08). The first was enriched for increased epithelial and innate and adaptive immunity signatures, while the other showed important telomere and mitochondrial dysfunction, loss of proteostasis and increased myofibroblast signatures. The existence of these 2 endotypes, including the impact on survival of the immune endotype, was validated in 3 independent validation cohorts. Finally, we identified transcription factors regulating the expression of endotype-specific survival-associated genes. Interpretation: Gene expression-based endotyping in IPF is feasible and can inform clinical evolution. As endotype-specific pathways and survival-associated transcription factors were identified, endotyping might open the horizon for endotype-tailored therapy.