The concordance rates of transperineal (TP) versus transrectal (TR) prostate biopsies with radical prostatectomy (RP) specimen have been assessed poorly in men diagnosed with magnetic resonance imaging (MRI)-targeted biopsy (TBx).To evaluate International Society of Urological Pathology (ISUP) concordance rates between the final pathology at RP and MRI-TBx or MRI-TBx + random biopsy (RB) according to the biopsy approach.A multi-institutional database included patients diagnosed with TP or TR treated with RP.TP-TBx or TR-TBx of the prostate.The ISUP grade at biopsy was compared with the final pathology. A multivariable logistic regression analysis (MVA) was performed to assess the association between the biopsy approach (TP-TBx vs TR-TBx) and ISUP upgrading, downgrading, concordance, and clinically relevant increase (CRI).Overall, 752 (59%) versus 530 (41%) patients underwent TR versus TP. At the MVA, TP-TBx was an independent predictor of upgrading (odds ratio [OR] 0.6, 95% confidence interval [CI] 0.4-0.9, p < 0.01) and improved concordance relative to the final pathology (OR 1.7, 95% CI 1.2-2.5, p < 0.01) after adjusting for age, cT stage, Prostate Imaging Reporting and Data System, number of targeted cores, prostate-specific antigen, and prostate volume. Moreover, TP-TBx was associated with a lower risk of CRI than TR-TBx (OR 0.7, p < 0.01). This held true when considering patients who underwent MRI-TBx + RB (OR 0.6, p < 0.01). The inclusion of men who had RP represents a potential selection bias.The adoption of TP-TBx compared with TR-TBx may reduce the risk of upgrading and improve the concordance of biopsy grade with the final pathology. The TP approach decreases the odds of CRI with improved patient selection for the correct active treatment.In this report, we evaluated whether transperineal (TP) targeted biopsy (TBx) may improve the concordance of clinically significant prostate cancer with the final pathology in comparison with transrectal (TR) TBx in a large worldwide population. We found that TP-TBx might increase concordance compared with TR-TBx. Adding random biopsies to target one increases accuracy; however, concordance with the final pathology is overall suboptimal even with the TP approach.