Altered Vulvar Microbiomes in Vulvar Lichen Sclerosus

BACKGROUND: Vulvar lichen sclerosus (VLS) is a chronic and progressive dermatosis that most commonly affects the anogenital region of postmenopausal women. VLS etiology is incompletely understood, though a role for immune system dysregulation has been suggested given associations with autoimmune disease. The skin microbiome plays an important role in cutaneous immune function and may contribute to VLS pathogenesis or expression. OBJECTIVE: In this prospective, longitudinal study, we investigated the association between bacterial and fungal microbiomes at the vulva and vagina in patients with VLS. METHODS: First, we recruited postmenopausal women with a clinical diagnosis of VLS and healthy postmenopausal controls from the same general population. Experimental samples were obtained from the clitoral hood, interlabial sulcus, labia minora, perineum, and vagina. Additionally, patients with VLS were re-sampled 10-14 weeks after once-daily treatment with ultrapotent topical corticosteroid. Negative control samples of the clinical sampling environment were obtained with each encounter. Microbial DNA was extracted from samples. Bacterial and fungal libraries were prepared with primer sets targeting the 16S V1-V3 and ITS1 regions respectively. These libraries were sequenced at the UT Southwestern Genomics Core. QIIME2 with a divisive amplicon denoising algorithm 2 (DADA2) pipeline was used to process and analyze sequences. Shannon index was used to quantify alpha diversity. Unweighted UniFrac (bacteria) and Bray-Curtis dissimilarity (fungi) were used to quantify beta diversity. Relative abundance testing was performed at sites with significant community structural alterations using Linear discriminant analysis Effect Size (LEFSE) based on disease status. Pairwise difference and distance comparisons were used to evaluate changes in bacterial alpha and beta diversity with treatment. A supervised learning classifier was used to determine whether disease status could be predicted as a function of baseline vulvar bacterial microbiome composition. RESULTS: A total of six subjects with VLS and twelve healthy postmenopausal controls were enrolled. Bacterial analysis at the community level revealed a loss of alpha diversity at each site analyzed; this loss was significant at the interlabial sulcus (p = 0.030, Kruskal-Wallis). Commensal bacteria including Staphylococcus epidermidis, Lactobacillus gasseri, and Lactobacillus jensenii were of significantly higher relative abundance in the control group (LEFSE) at this site. Longitudinal analysis of bacterial community diversity after treatment revealed the greatest recovery of alpha diversity at the interlabial sulcus. Supervised learning with a random forest classifier on bacterial vulvar data revealed an accuracy of 0.923 to predict disease status and redemonstrated several amplicon sequence variants including Staphylococcus epidermidis to be important in this prediction. CONCLUSIONS: Taken together, we identified significant alterations in the diversity and composition of skin bacterial communities in VLS that are dependent on anatomic location. We also demonstrated indicators of disease status using machine learning techniques. Altogether, findings support a relationship between the skin microbiome and VLS and underscore the potential for the skin microbiome to serve as a non-invasive biomarker of disease.

Author dc:creator

• Pyle, Hunter James

Contributors dc:contributor

• Harris-Tryon, Tamia A.
• Mauskar, Melissa M.
• Vandergriff, Travis W.

• Vulvar Lichen Sclerosus
• Microbiota

Language dc:language

en

Identifier

1522122311

OAI identifier oai:identifier

oai:tdl-ir.tdl.org:2152.5/10625