Background/Purpose: Sarcoidosis is a heterogeneous disease with diverse manifestations and varied clinical courses.  Currently there are no reliable non-invasive biomarkers of disease activity or severity. The aim of this study was to identify “pathway-specific” blood gene expression signatures in subjects with sarcoidosis and to ascertain whether these signatures identify distinct subsets of patients and correlate with measures of disease severity.

Methods: We analyzed whole blood RNA (PAXgene RNA tubes) from 38 patients with sarcoidosis and 20 healthy controls using Affymetrix U133 Plus 2.0 microarrays (Affymetrix Inc., Santa Clara, CA).   Correcting with the Benjamini-Hochberg method, we found differentially expressed (DE) genes that were upregulated in sarcoidosis compared to controls.  We performed Ingenuity Pathway Analysis (IPA) and then hierarchical clustering (MultiExperiment Viewer) using selected genes representative of canonical pathways highly up-regulated in sarcoid.  Standardized expression levels of DE genes were correlated with measures of pulmonary function testing (PFT).   Whole blood DE genes expression levels were compared to levels found in publically available sarcoidosis tissue gene expression datasets (Gene Expression Omnibus GSE32887, GSE19976).

Results: 193 DE genes associated with a B-statistic > 0 (adjusted p value <0.006) were upregulated in sarcoidosis.  These genes were associated with interferon (IFN), TNFα, and TGF-β1 signaling pathways.  Hierarchial clustering of subjects using genes representative of IFN and TNFα pathways revealed different patient subsets, including subjects with IFN-high/TNFα-low or IFN-low/TNFα-high signatures, and subjects with dual IFN/TNFα-high or dual IFN/TNFα low signatures.  Expression levels of TGF-β1-associated genes (SKIL, USP15, LPCAT2)  correlated with PFT measures of restrictive lung volumes and decreased oxygenation, including FVC% predicted (R² = 0.39, p<0.0001), TLC% predicted (R² = 0.26, p = 0.001), and DLCO% predicted (R² = 0.2, p = 0.004).  Expression levels of TNFα-inducible genes (TNFAIP2, PLAUR, ICAM1) correlated with degree of pulmonary obstruction as measured by FEV1/FVC (R²= 0.26, p = 0.001).  Expression of these pathway-specific genes was elevated in lesional sarcoid skin and progressive fibrotic sarcoid lung biopsy gene expression datasets.  

Conclusion: Based on these data we hypothesize that IFN-, TNFα-, and TGF-β1-specific gene expression biomarkers reflect distinct subgroups of patients with sarcoidosis and signaling pathways that may play a pathogenic role in sarcoidosis.  TGF-β1-associated DE gene expression levels may reflect severity of pulmonary fibrosis.  We plan to test this hypothesis in a larger longitudinal cohort to 1) validate our groupings, 2) determine whether these signatures are stable over time or vary with disease activity, 3) determine whether these groupings have prognostic value.

---

« Back to ACR/ARHP Annual Meeting 2013

ACR Meeting Abstracts - https://acrabstracts.org/abstract/transcriptomic-blood-markers-and-molecular-stratification-in-sarcoidosis/