BACKGROUND AND AIMS: Brush cytology is a widely used diagnostic method in conjunction with intraductal biopsies during endoscopic retrograde cholangiopancreatography, but its diagnostic yield remains a limitation. This study evaluated the efficacy of biliary cytology using a newly developed brush device with rapid on-site cytological evaluation (ROSE) for detecting malignant biliary strictures (MBSs). METHODS: In total, 58 patients with suspected intrinsic MBS identified by intraductal ultrasound were enrolled. After achieving tissue sampling with ROSE through a maximum of two brushing passes, a transpapillary forceps biopsy (TPB) was performed. The primary outcome was diagnostic accuracy, and the secondary outcomes were technical success, sampling adequacy, and procedure-related adverse events. RESULTS: Biliary cytology with ROSE was technically successful in all patients (58/58), with a sampling adequacy of 96.6% (56/58). The technical success and sampling adequacy of TPB were 94.8% (55/58) and 91.4% (53/58), respectively. Brush cytology with ROSE and TPB yielded sensitivity rates of 91.8% and 85.7%, specificity rates of 88.9% for both, and accuracy rates of 88.9% for both. The receiver operating characteristic curve comparing the diagnostic accuracies of brush cytology with ROSE and TPB combined versus TPB alone showed a significantly higher value for the combined approach (0.93) than TPB alone (0.87) (p = 0.010). CONCLUSION: Biliary brush cytology using a novel brush device with ROSE is effective and can be used complementarily to TPB in patients with suspected MBS.