Enhanced Detection of Clinically Significant Prostate Cancer: Insights from the Stockholm3-MRI Screening Study Two-Year Follow-up

Highlight

1. The Stockholm3 blood-based multivariable test outperforms PSA alone for detecting clinically significant prostate cancer (csPC) in population screening.
2. Stockholm3 combined with MRI-guided biopsy reduces false-negative rates and unnecessary biopsies compared with PSA-based screening.
3. With 2-year follow-up, Stockholm3 achieves a sensitivity of 90% for detecting csPC versus 74% for PSA, without compromising specificity.
4. Implementation of Stockholm3 in screening protocols may optimize early prostate cancer detection and reduce overdiagnosis.

Study Background

Prostate cancer is among the most common cancers in men worldwide. Early detection of clinically significant prostate cancer (csPC)—cancer with histologic grade group 2 or higher—is critical for improving outcomes, while minimizing overtreatment of indolent tumors. Prostate-specific antigen (PSA) testing is a widespread screening tool but remains controversial due to suboptimal specificity and sensitivity, leading to unnecessary biopsies and missed significant cancers.

The Stockholm3 test is an advanced multivariable risk model combining PSA levels, additional plasma protein biomarkers, genetic risk scores, and clinical information to improve prostate cancer risk stratification. The recent STHLM3-MRI trial evaluated the clinical utility of Stockholm3 versus PSA testing in population-based prostate cancer screening, integrating magnetic resonance imaging (MRI) to guide biopsies.

Study Design

This is a secondary analysis of the baseline round of the prospective, randomized STHLM3-MRI screening trial conducted in Stockholm, Sweden, from 2018 to 2020. The study enrolled 12,670 men aged 50 to 74 years who underwent both PSA and Stockholm3 screening tests.

Abnormal screening was defined as PSA ≥3 ng/mL or Stockholm3 risk score ≥11. Men with abnormal results were randomized in a 2:3 ratio to either systematic biopsy or MRI-targeted and systematic biopsy when lesions had a Prostate Imaging Reporting and Data System (PI-RADS) score ≥3. Prostate cancer diagnoses within 2 years of baseline were identified through the Swedish National Cancer Register. Cancers detected after a negative baseline screening were classified as false negatives.

Key Findings

Among 12,670 men screened, 443 (3.5%) were diagnosed with clinically significant prostate cancer (grade group ≥2) within 2 years. Key diagnostic performance metrics include:

• False-negative rate: Stockholm3 (≥11) had a 10% rate versus 26% for PSA (≥3 ng/mL), showing fewer missed csPC cases with Stockholm3.
• False-positive rate: Comparable between Stockholm3 (11%) and PSA (10%), indicating similar specificity profiles.
• Sensitivity: Stockholm3 demonstrated substantially higher sensitivity at 90% (95% CI, 87% to 93%) compared to 74% (CI, 69% to 78%) for PSA.
• Specificity: Both tests performed similarly with Stockholm3 at 89% and PSA at 90%.

Decision curve analysis demonstrated a higher net clinical benefit for Stockholm3 screening, reflecting reduced unnecessary biopsies alongside improved detection of significant cancers.

The addition of MRI allowed targeted biopsies of suspicious lesions, enhancing diagnostic precision beyond systematic biopsies alone.

Expert Commentary

This study provides compelling evidence supporting the integration of multivariable risk prediction models like Stockholm3 with advanced imaging to optimize prostate cancer screening. The substantial reduction in false negatives is particularly important to reduce missed clinically actionable cancers. Although PSA remains widely used due to simplicity and cost, its limitations in sensitivity have long been recognized.

Combining protein biomarkers, polygenic risk, and clinical variables enhances risk stratification. The relatively low false-positive rate with Stockholm3 compared to PSA suggests no increase in unnecessary procedures, alleviating a major concern in prostate cancer screening.

Study limitations include only 25% participation among invited men, which might introduce selection bias, the short 2-year follow-up, and predominance of European descent in the cohort, potentially limiting applicability to more diverse populations. Furthermore, cost-effectiveness, accessibility, and implementation logistics of Stockholm3 and MRI-based screening warrant further investigation.

Conclusion

In this large population-based screening cohort with a two-year follow-up, Stockholm3 combined with MRI-based biopsy protocols outperformed PSA testing alone for detection of clinically significant prostate cancer. Enhanced sensitivity with comparable specificity supports Stockholm3 as a superior screening tool, reducing both missed significant cancers and unnecessary biopsies.

These findings represent an important advance in prostate cancer early detection strategies, with potential to improve clinical outcomes and reduce harms associated with conventional PSA screening. Further long-term studies and translational efforts are needed to confirm benefits in diverse populations and real-world settings.

Funding and Trial Registration

This study was funded by the Swedish Research Council, Swedish Prostate Cancer Society, Stockholm Region, and the Swedish Cancer Society. The trial is registered on ClinicalTrials.gov (NCT03377881).