OBJECTIVE: Dopamine agonist (DA) treatment is effective for hyperprolactinemia and reduces tumor size in patients with prolactinoma; however, prolonged DA administration without prolactinoma causes fibrosis around tumor tissues. Therefore, we aimed to identify circulating microRNAs (miRNAs) as potential biomarkers to predict prolactinoma in patients with hyperprolactinemia and pituitary tumors. DESIGN: Plasma samples were collected from 3 comparison groups: (1) patients clinically diagnosed with prolactinoma vs nonfunctioning pituitary adenoma (NFPA) based on response to cabergoline treatment, (2) patients with surgically confirmed prolactinoma vs NFPA, and (3) patients before and after cabergoline treatment. Candidate miRNAs from the initial nCounter assay were validated using quantitative reverse transcription-polymerase chain reaction (qRT-PCR) in a larger cohort of 247 patients with hyperprolactinemia and 37 controls. METHODS: The nCounter assay was used for miRNA expression profiling, and the qRT-PCR validated the candidate miRNAs in the plasma and tumor tissue samples. Total RNA sequencing was conducted on pituitary tumor tissues to identify transcriptomic alterations. Furthermore, candidate miRNA target genes and their biological roles were analyzed using prolactinoma cell lines. RESULTS: Three miRNA candidates (miR-20a-5p, miR-424-5p, and miR-514a-5p) were selected by analyzing 3 sets of expression comparisons between the 2 groups. Furthermore, the relative miR-20a-5p expression significantly increased in prolactinoma compared with that in normal pituitary glands, NFPA, growth hormone-secreting pituitary adenoma, and adrenocorticotropic hormone-secreting pituitary adenoma. In MMQ and GH4 cells, miR-20a-5p inhibition decreased prolactinoma cell proliferation and prolactin secretion. CONCLUSIONS: Circulating miR-20a-5p is a potential biomarker for prolactinoma, which could be associated with responsiveness to DAs.