Prediction of lymphocyte-predominant breast cancer using microRNA expression in cytology specimens

Tumor-infiltrating lymphocytes are prognostic and predictive biomarkers of breast cancer; however, conventional assessment is hindered by invasiveness and subjective evaluation, limiting clinical reproducibility. This study investigated the utility of microRNA (miRNA) profiles in predicting lymphocyte-predominant breast cancer (LPBC).

This study included 56 patients with breast cancer who underwent fine-needle aspiration cytology at the time of diagnosis (28 LPBC and 28 non-LPBC cases). Total RNA was extracted from cytology smear specimens, and the expression levels of six candidate miRNAs (miR-30a-3p, miR-187, miR-196b, miR-1247, miR-4485, and miR-6510) were quantified using reverse-transcriptase polymerase chain reaction. The performance in identifying LPBC was evaluated using receiver operating characteristic curve analysis and logistic regression models.

All six miRNAs showed significantly lower expression levels in the LPBC group than in the non-LPBC group (all p < .05). Receiver operating characteristic curve analysis demonstrated moderate to high diagnostic accuracy, with an area under the curve of 0.708 for miR-30a-3p, 0.739 for miR-187, 0.783 for miR-196b, 0.777 for miR-1247, 0.790 for miR-4485, and 0.673 for miR-6510. Multivariate analysis further identified each miRNA as an independent diagnostic predictor for LPBC.

miRNA profiling of cytologic samples enabled the objective identification of LPBC. These findings suggest that specific miRNA expression patterns reflect the tumor immune microenvironment and can be used to evaluate the status of tumor-infiltrating lymphocytes in breast cancer.