This study assessed whether extracellular microRNAs (miRNAs) isolated from serum and serum-derived extracellular vesicles (EVs) could serve as candidate biomarkers to predict risk of Chlamydia trachomatis reinfection. The principal aim was to compare miRNA detection at the baseline treatment visit among women who later were or were not found to be reinfected at a 3-month follow-up visit.
Participants were non-pregnant women presenting to the Jefferson County Department of Health Sexual Health Clinic after a positive C. trachomatis nucleic acid amplification test (NAAT). Enrollment spanned March 27, 2012 through October 11, 2018. At the baseline visit clinical information and specimens were collected, and all participants received directly observed azithromycin 1 g orally as the treatment administered in the parent study.
Participants were scheduled for a 3-month follow-up visit that included repeat clinical interview and repeat specimen collection with NAAT to determine reinfection status. The study protocol and participant consent procedures were approved by the University of Alabama at Birmingham Institutional Review Board and the Jefferson County Department of Health.
Samples were excluded from miRNA testing if participants had concurrent co-urogenital infections (gonorrhea, candidiasis, trichomoniasis, and/or bacterial vaginosis) or if serum samples showed red blood cell contamination, to avoid confounding of miRNA expression.
Samples that passed quality criteria were categorized a priori into four analysis groups:
For EV-derived miRNA testing, 53 samples from 42 women were included (24 baseline samples: 13 NRB and 11 RB; 29 follow-up samples: 19 NRF and 10 RF). For whole serum miRNA testing, 49 samples from 35 women were included (26 baseline samples: 14 NRB and 12 RB; 23 follow-up samples: 15 NRF and 8 RF). When limited sample volume allowed testing in only one fraction, EV testing was prioritized over whole serum testing.
Serum was processed for EV isolation and RNA extraction following the study’s laboratory protocol. EVs were isolated from serum and RNA was recovered for miRNA profiling. MiRNA expression was assayed using Bruker nCounter microarrays. MiRNAs were processed with positive ligation normalization and were categorized as detected or not detected for downstream categorical analysis.
The primary analysis compared baseline miRNA detection between NRB and RB groups to identify candidate biomarkers predictive of reinfection risk. Secondary analyses compared RB versus RF to evaluate changes from baseline to follow-up in reinfected participants, and RF versus NRF to compare follow-up miRNA profiles between participants with and without reinfection.
Nominal significance for categorical miRNA detection was evaluated using Fisher’s exact test. Predicted mRNA targets for candidate miRNAs were interrogated using Qiagen’s Ingenuity Pathway Analysis (IPA) to assess potential biological pathways and immune cell–related associations.
Distinct candidate miRNAs were identified in EVs versus whole serum that associated with reinfection status at the 3-month follow-up. Detection of EV-derived miR-888-5p at baseline was associated with reinfection at follow-up. In whole serum, detection at baseline of miR-1285-5p, miR-548aa + 548t-3p, and miR-575 was associated with absence of reinfection at follow-up.
These associations emerged from categorical detection analyses using the nCounter platform and Fisher’s exact test; the authors present these miRNAs as preliminary biomarker candidates for further validation.
Predicted mRNA target analysis using IPA suggested differential immune-related associations for the candidate miRNAs. miR-888-5p mapped with stronger alignment to mRNA targets associated with CD8+ T-cell functions. In contrast, miR-548aa aligned largely with mRNA targets related to CD4+ T-cell responses. The authors interpret these pathway connections as hypothesis-generating links between extracellular miRNA detection and immune processes that could influence reinfection risk.
The authors emphasize that these findings are preliminary and require validation in independent cohorts. Study limitations cited include limited sample availability, stringent exclusion criteria that reduced sample numbers, and the prioritization of EV testing when sample volume was limited, all of which influenced the number of samples available for EV versus whole serum analysis and baseline versus follow-up comparisons.
All miRNA data generated in this study are deposited in the NCBI Gene Expression Omnibus (GEO) under accession number GSE343618. The authors note that no commercially available miRNA biomarkers for C. trachomatis exist at present and that future work should assess whether these candidate miRNAs can robustly predict reinfection risk and inform targeted testing strategies.