BACKGROUND: MicroRNAs (miRNAs) are small non-coding RNAs that regulate a broad array of cellular and disease processes. Several miRNAs are differentially expressed in cancer and many are being considered as biomarkers for predicting clinical outcomes. Here we quantified the expression of three miRNAs, miR-21, miR-141, and miR-221, from prostate cancer surgical specimens and evaluated their association with disease recurrence after primary therapy. METHODSA pilot nested case-control study was designed from a large cohort of men who underwent radical prostatectomy between 1993 and 2001. Total RNA was extracted from malignant prostate tissue of 59 cases (recurrence) and 59 controls. Cases and controls were matched on age, race, pathologic stage, and grade. The relative expression of each miRNA was then determined for each sample by quantitative real-time RT-PCR. Conditional logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) of recurrence for tertiles of miRNA expression. We noted block storage time effects and thus, used separate tertile cutpoints based on the controls by calendar year of prostatectomy. RESULTSLower miR-221 expression was associated with a higher risk of recurrence; the ORs were 3.21 for the lowest tertile and 2.63 for the middle tertile compared with the highest tertile of expression (P-trend=0.02). This pattern was unchanged after multivariable adjustment (P-trend=0.05). No statistically significant trends were observed for miR-21 or miR-141 after multivariable adjustment. CONCLUSIONSBased on this small pilot study, men with localized prostate cancers with lower miR-221 expression may have a greater risk for recurrence after surgery