ĐÁNH GIÁ MỐI TƯƠNG QUAN GIỮA HỆ THỐNG PHÂN LOẠI PI-RADSv2.1 VÀ UNG THƯ TUYẾN TIỀN LIỆT CÓ Ý NGHĨA LÂM SÀNG

Abstract

Purpose: To evaluate the correlation between PI-RADSv2.1 classification and clinically significant prostate cancer. Materials and Methods: A study was conducted at Vinmec Times City International Hospital from July 2018 to August 2025. The study included 69 patients with 88 prostate lesions who underwent multiparametric MRI. Risk assessment was performed using the PI-RADSv2.1 scoring system, followed by transrectal ultrasound (TRUS)-guided biopsy. Histopathological grading was based on the ISUP- International Society of Urological Pathology- grade. Results: No significant prostate cancer was identified in lesions scored as PI-RADS 1–2. The PI-RADS 3 group predominantly corresponded to low-grade histology, with only one case of ISUP grade 3. In the PI-RADS 4 group, the incidence of clinically significant cancer increased markedly to 18.9%. Notably, the PI-RADS 5 group showed a high concentration of ISUP grades 3–5 (54.2%). Regarding the diagnostic performance of the PI-RADS score for csPCa, using a cut-off of PI-RADS ≥ 3 yielded a sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of 100%, 14.3%, 48.1%, 100%, and 52.3%, respectively. When the cut-off increased to PI-RADS 4, the values are 89.7%, 46.9%, 57.4%, 85.2%, and 65.9%. At a cut-off of PI-RADS 5, the values are 51.3%, 91.8%, 83.3%, 70.3%, and 73.0%, respectively. Conclusion: PI-RADSv2.1 classification demonstrates a strong correlation with clinically significant prostate cancer and serves as a highly valuable tool for both the diagnosis and screening of this disease