Tissue and plasma expression patterns and bioinformatic analysis of miR-15a-5p, miR-145-5p, miR-221-3p, and miR-455-3p in primary pterygium

Abstract Pterygium is a fibrovascular ocular surface disorder driven by proliferative, angiogenic, and fibrotic processes, with high postoperative recurrence and no established molecular markers. While several microRNAs (miRNAs) have been implicated in pterygium tissue, whether these alterations extend to the circulation has not been established. This exploratory study evaluated the tissue and plasma expression of miR-15a-5p, miR-145-5p, miR-221-3p, and miR-455-3p in primary pterygium and explored their candidate molecular targets through integrated bioinformatic analysis. In this prospective case-control study, 25 patients with primary, nasal, grade 3 pterygium and 25 cataract-surgery controls were enrolled. Paired tissue and plasma samples were obtained from each participant, and miRNA expression was quantified by RT-qPCR using the 2 −ΔΔCt method; statistical comparisons were based on − ΔCt values. Discriminatory performance was assessed by receiver operating characteristic (ROC) curve analysis. Validated and predicted target genes were retrieved from miRTarBase and miRDB, followed by Kyoto Encyclopedia of Genes and Genomes pathway enrichment, Gene Ontology enrichment, and protein-protein interaction network analyses. All four miRNAs were significantly upregulated in pterygium tissue compared with control conjunctiva. miR-15a-5p showed the greatest tissue upregulation (approximately 11-fold) and the strongest discriminatory performance (area under the curve [AUC] = 0.919; sensitivity 0.72, and specificity 0.96). A concordant upregulation was detected in plasma, where miR-15a-5p and miR-455-3p retained significant, albeit moderate, discriminatory value (AUC = 0.723 and 0.722). The selected miRNAs showed different correlation patterns in the pterygium and control groups. Target genes were significantly enriched in cancer-associated, PI3K-AKT, cell cycle, apoptosis, and transforming growth factor-beta pathways, with MYC, PTEN, KRAS, BCL2, and VEGFA identified as candidate hub genes. The four miRNAs we examined were upregulated in pterygium tissue and a similar, though weaker, pattern was present in plasma. miR-15a-5p separated pterygium from control tissue well and in plasma both miR-15a-5p and miR-455-3p showed moderate discriminatory value. Their candidate targets cluster in pathways related to proliferation, apoptosis, angiogenesis, and TGF-β signalling. Given the cross-sectional design and the inclusion of grade 3 lesions only, these findings should be seen as hypothesis-generating. Larger studies covering different clinical grades, with longitudinal follow-up, are needed before any role in disease monitoring can be considered.