Abstract Purpose Multiparametric MRI (mpMRI) has an emerging role in prostate cancer (PC) diagnostics. In addition, clinical information are reliable predictors for significant PC (sPC). We analyzed if the negative predictive value (NPV) of mpMRI to rule out sPC could be improved by using clinical factors, especially prostate specific antigen (PSA)-density. Methods 1040 consecutive men with suspicion of PC underwent mpMRI first, followed by transperineal systematic and MRI/TRUS-fusion-guided biopsy. Logistic regression analyses were performed to test different clinical factors as predictors of sPC and to build nomograms. To simplify these for clinical use, patients were stratified to three PSA-density groups (1: <0.07, 2: 0.07-0.15; 3: >0.15). After stratification, NPVs for PI-RADS Likert score<3 were calculated. sPC was defined as Gleason score (GS)≥3+4. High-grade PC was defined as GS≥4+3. Results Overall, 451 men were diagnosed with sPC and 187 had a GS≥4+3. In ROC curve analyses the predictive power of the developed nomogram for sPC showed a higher AUC compared to PI-RADS alone (0.79 vs. 0.75, p<0.001). The NPV to harbor sPC increased for men with unsuspicious MRI from 79% up to 89% when these men had a PSA-density ≤0.15. In the repeat biopsy setting the NPV for sPC increased from 83% to 93%. The NPV to harbor high-grade PC increased from 92% up to 98% in the entire cohort. Conclusion Using PSA-density in combination with mpMRI improves the NPV of PI-RADS scoring. By increasing the probability of ruling-out sPC, approximately 20% of unnecessary biopsies could be avoided safely.