C71-04 Bronchoalveolar Lavage Fluid Proteomics Identifies Distinct Inflammatory and Fibrotic Subgroups in Idiopathic Pulmonary Fibrosis and Scleroderma Associated Interstitial Lung Disease

Abstract Rationale Interstitial lung disease (ILD) encompasses a broad spectrum of conditions, including idiopathic pulmonary fibrosis (IPF) and autoimmune-associated ILDs such as systemic sclerosis (SSc)-related ILD. Both represent heterogeneous diseases characterized by variable fibrotic and inflammatory profiles. Prior studies have identified variable protein signatures in SSc-ILD, with antifibrotic profiles in fibrotic phenotypes and inflammatory profiles in scleroderma-associated pulmonary arterial hypertension2. SSc-ILD is typically managed with immunomodulatory therapy, though treatment responses are variable. Conversely, IPF lung tissue exhibits enrichment of immune and chemotactic proteins, yet immunosuppression has been shown to be unsafe when applied to IPF populations3. In the absence of validated diagnostic tools to guide management, pulmonologists face uncertainty in predicting therapeutic response. We aimed to utilize bronchoalveolar lavage fluid (BALF) proteomics to identify molecular endotypes across IPF and SSc-ILD that may guide individualized management. Methods We conducted a cross-sectional study of BALF from 42 participants (21 IPF, 21 SSc-ILD). Global proteomic profiling of cell-free BALF supernatant was performed using LC-MS/MS. Of 2697 proteins isolated, 1077 proteins (≥80% of samples, noncontaminants) were identified. Protein abundances were normalized (VSN), transformed and scaled prior to principal component analysis (PCA), clustering, differential expression analysis (|log2FC| ≥ 1, q ≤ 0.05), and pathway analysis. Results PCA revealed no distinct segregation by clinical diagnosis (IPF vs SSc-ILD) but visualized two mixed molecular clusters. K-means clustering (k = 2) confirmed these groupings (Image). Differential expression analysis of the two molecular clusters revealed 152 significant differentially expressed proteins—85 upregulated in Cluster 1 and 67 in Cluster 2. Pathway enrichment analysis identified Cluster 1 as a proinflammatory phenotype, overrepresented by proteins crucial for pro-inflammatory cytokine regulation, leukocyte migration, and immune activation networks. Cluster 2 represented a fibrosis phenotype, characterized by fibroblast activation, extracellular matrix remodeling, DNA repair, and YAP/TAZ mechanosensitive signaling pathways. Conclusion BALF proteomics identified two unique subtypes of protein expression profiles - inflammatory and fibrotic, spanning both IPF and SSc-ILD. These findings illustrate the molecular overlap within IPF and among autoimmune associated ILD subtypes. The presence of underlying inflammatory molecular endotypes across ILD subtypes could explain differential responsiveness to immunosuppressive versus antifibrotic strategies. Currently, BALF is not routinely assessed in IPF or autoimmune associated ILD outside of superimposed or concomitant infection4. This study supports BAL use for diagnosis and personalized therapy decision making based on individual patient’s protein profile. These results suggest that molecular phenotyping, as opposed to clinical phenotyping, may improve precision medicine across ILDs. This abstract is funded by: None