The aim of this study is to compare performance of two biopsy approaches in patients with at least one previous negative prostate biopsy (PB): classical transrectal biopsy (ClTB) versus cognitive registration biopsy (COG-TB). A retrospective study of 205 patients with at least one negative PB. 144 (70.2%) patients underwent a prior mpMRI and 61 (29.8%) patients did not. Nodule classification was carried out according PI-RADS version 2. Peripheral zone (PZ) grouped pZa, pZpl and pZpm areas, transition zone (TZ) Tza, Tzp and Cz areas, and anterior zone (AZ) AS areas. COG-TB was conducted in patients with previous mpMRI (144); while in the remaining 61 (29.8%) patients a ClTB of PZ and TZ was performed. Statistical analysis was performed using Chi square and T-student tests for qualitative and quantitative variables, respectively. Multivariate analysis was carried out in order to identify predictive variables of prostate cancer. Median patient age was 68 (IQR 62-72) years, median PSA was 8.3 (IQR 6.2-11.7) ng/ml and median previous biopsies was 1 (IQR 1-2). Digital rectal examinations (DRE) findings were normal in 169 (82.4%) patients and suspicious in 36 (17.6%) patients (cT2a-b in 34 patients and cT2c in 2). Median prostate volume was 48 (IQR 38-65) cc. Statistically significant differences in PSAD between both groups were found (P=.03). Transrectal ultrasound (TRUS) showed hypoechoic nodules in 8 (13.1%) ClTB patients and in 62 (43.1%) COG-TB patients (P=.0001). The median number of biopsy cylinders per set of prostate biopsies was 10 (IQR 10-10) in ClTB group and 11 (IQR 9-13) in COG-TB group (P=.75). Cancer was diagnosed in 74 (36.1%) patients: of them, 10 (16.4%) were ClTB patients and 64 (44.4%) COG-TB (P=.0001). Tumors classification was as follow: ISUP-1: 34 (45.9%), ISUP-2: 21 (28.4%), ISUP-3: 9 (12.2%), ISUP-4: 7 (9.5%) and ISUP-5: 3 (4.1%). No significant statistical differences were found (P=.89). The median number of biopsy cylinders impaired per set of prostate biopsies was 1 (IQR 1-5) in ClTB group and 2 (IQR 1-4) in COG-TB group (P=.93). Regarding independent predictive variables for prostate cancer the results were: age (OR=12.05; P=.049), suspicious DRE (OR=2.64; P=.04), hypoechoic nodule (OR=2.20; P=.03) and mpMRI +COG-TB sequence (OR=3.49; P=.003). In patients with at least one negative PB, mpMRI +COG-TB sequence improves 3.5 (OR=3.49) times the diagnosis prostate vs. ClTB. Copyright © 2018 AEU. Publicado por Elsevier España, S.L.U. All rights reserved.