36 Background: Prostate-specific antigen (PSA) is used to screen for prostate cancer but suffers from a high false-positive rate that translates into unnecessary prostate biopsies and over-diagnosis of low-risk prostate cancers. We aimed to develop a model for prostate cancer screening with better test characteristics than PSA. Methods: STHLM3 is a prospective, population-based, paired screen-positive diagnostic study. It investigates whether the predefined STHLM3 model, a combination of 6 plasma protein biomarkers (PSA, free-PSA, intact-PSA, HK2, MIC-1, and MSMB), genetic polymorphisms (232 SNPs), and clinical variables (age, family history, previous biopsies, DRE, and prostate volume) can substantially reduce the proportion of men biopsied while maintaining the same sensitivity to diagnose Gleason score ≥ 7 prostate cancer as PSA ≥3 ng/ml. In addition, a health economic evaluation was performed comparing the STHLM3 model with current clinical practice in Stockholm. Results: 58,818 men without prostate cancer aged 50-69, participated in the STHLM3 study, with 6,700 undergoing subsequent prostate biopsy. The STHLM3 model performed significantly better (p<0.001) than PSA for detecting GS ≥7 cancers, increasing the Area Under the Curve from 0.56 to 0.74. All variables used in the STHLM3 Model were significantly associated with Gleason score ≥7 prostate cancers (p<0.05) in a multiple logistic regression model. Using the same sensitivity as PSA ≥3 ng/ml to diagnose Gleason score ≥7 prostate cancer, the STHLM3 model reduced the number of biopsies by 32% (95% CI 24%-39%) and avoided 44% (95% CI; 35%-54%) of the negative biopsies. The number of Gleason score 6 cancers was reduced by 17% (95% CI; 7%-26%). Positive ICER for the STHLM3 model was seen in all models of the health economic analysis. Conclusions: The STHLM3 model reduces unnecessary biopsies without compromising the ability to diagnose Gleason score ≥7 prostate cancer in a cost-efficient way. This is a significant step towards personalized risk-based prostate cancer diagnostic programs. Clinical trial information: ISRCTN84445406.