Background/Purpose: In systemic sclerosis (SSc), interstitial lung disease (ILD) remains the leading cause of mortality. A decline in forced vital capacity (FVC) is considered a key measure of profibrotic activity in SSc. We adopted the rate of FVC decline as an endpoint to analyse a retrospective longitudinal cohort with available serum samples to identify soluble markers associated with FVC decline over time.  

Methods: Serum samples with matched FVC results (+- 3 months) were analysed employing the Olink-1536 panel to identify prognostic biomarkers (pBMs). Longitudinal analysis of clinical outcome was implemented to identify patients with a monthly rate of FVC% predicted deterioration of over 0.5% (rapid progressors [rP]). Random Forest analysis (RFA) was employed to identify and rank pBMs that best separated the two groups (rP vs non-rP). Logistic regression and standard ROC analysis were built to determine the predictive value of the pBMs, with identification of the highly significant ones using backward elimination.

All predictors were assessed with univariate logistic regression models, and then a multivariate model was built using only significant predictors to assess for independence. Literature analysis of highest ranked pBMs was performed to prioritise proteins for functional studies. Expression levels of epithelial to mesenchymal transition (EMT) markers were measured in the A549 lung epithelial cell (LEC) line treated with human recombinant proteins and relevant drug inhibitors.   

Results: 100 patients were enrolled (33 VEDOSS, 21 LcSSc, 46 DcSSc) and 20 patients (20%) met the progression criteria (3 VEDOSS, 2 LcSSc and 15 DcSSc). Anti-centromere antibody, Scl70, ILD, immunosuppressants and baseline FVC% were significantly associated with lung progression at univariate logistic regression analysis. RFA of the proteomic dataset prioritised 18 pBMs as the best predictors of progression independent of disease subset or duration. Backward elimination identified a model using 4 pBMs (GDF15, SFTPA2, CR2, ACAA1) carrying a very strong statistical significance (AUC of 0.848). GDF15 is a stress-responsive cytokine with previous studies showing increased levels in SSc. GDF15 levels were indeed increased in rP vs non-rP (NPX mean 0.85 [-0.02, 1.73] vs -0.12 [-0.66, 0.41] p = < 0.001). GDF15 remained associated with progression in multivariate analysis independently of the clinical features found to be predictive in the univariate analysis (OR: 4.03, 95% CI:1.63-11.5, P = 0.004). In vitro experiments indicated that GDF15 was able to induce an EMT response in A549 cells at the RNA and protein levels in a dose dependent manner (3.7- and 3.1-fold increase at RNA level and 2.2- and 1.9-fold increase at protein level was seen with vimentin and N-cadherin). GDF15 EMT effects were not associated with SMAD3 phosphorylation, and they were independent of canonical TGF-b signalling.

Conclusion: GDF15 is a pBM of progressive fibrosis in SSc independently of clinical features. GDF15 can act directly on LEC’s contributing to EMT independently of TGF-b and as such may be directly implicated in the pathogenesis of progressive fibrotic ILD. Replication in an independent cohort will be crucial to verify the clinical value of our findings.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/growth-differentiation-factor-15-is-a-serum-biomarker-and-pathogenic-factor-of-progressive-fibrosis-in-systemic-sclerosis/