Identifying Inflammatory Gene Expression Signatures for Skin and Soft Tissue Infections

Abstract

Background: Misdiagnosis of skin and soft tissue infections (SSTIs) due to clinical mimics can result in delay of care, unnecessary antibiotic exposure, and inappropriate hospitalization. Comprehensive screening of inflammatory genes in SSTIs could identify biomarkers to distinguish SSTIs from mimics. Methods: We performed a search of the MGH James Homer Wright Pathology Laboratories database from 2008-2018 for diagnoses of necrotizing fasciitis, cellulitis, and stasis dermatitis, yielding 103 cases. Diagnoses were verified by chart review and categorized by discharge diagnosis. Three samples from each category, along with three controls from location-matched skin were selected for further study. mRNA isolated from paraffin-embedded skin biopsies was analyzed by Nanostring, with 594 inflammatory genes profiled. Results: We identified differentially expressed genes between necrotizing fasciitis, cellulitis, and infectious cases (necrotizing fasciitis and cellulitis) compared to non-infectious stasis dermatitis. Differentially upregulated genes in SSTIs included those with known roles in inflammation (CXCR2, IL6, IFI16, TNFRSF1B) and transcriptional regulation (BCL3, MBP). We also identified differential upregulation of genes not previously associated with SSTIs including S100A8, S100A9, MCL1, CD14, and LTF. Conclusions: We characterized transcriptomic signatures of severe and moderate SSTIs compared to stasis dermatitis and normal skin from the lower extremities. Though limited by small sample size, these data support the utility of a prospective study analyzing outcomes of patients diagnosed with SSTIs based on gene expression signatures. Identifying SSTI-specific gene expression signatures could help differentiate true skin infections from non-infectious inflammatory skin conditions, facilitating more accurate diagnoses and improving patient care.