314 Background: Diagnostic accuracy estimates for prostate cancer tests are often biased when biopsy verification depends on the index test. Using population registry follow-up, we compared two-year detection of clinically Grade Group 2 or higher (GG2) prostate cancer between PSA and a Stockholm3 pathway in the STHLM3-MRI trial. Methods: In the prospective STHLM3-MRI trial (NCT03377881), men aged 50-74 years provided blood for PSA and Stockholm3. Cancer diagnoses within 2 years of blood draw were ascertained by linkage to the Swedish National Cancer Registry; men alive and not emigrated at the end of the 2 years were considered for the analysis. The prespecified thresholds were PSA ≥ 3 ng/mL and Stockholm3 ≥ 11. Sensitivity, specificity, PPV, and NPV were estimated with 95% CIs; paired differences were tested with McNemar tests. Results: Among 12,670 men, 443 (3.5%) were diagnosed with GG2, including 40 interval cancers (men diagnosed after a negative test during follow-up). At Stockholm3 ≥ 11 the sensitivity was 90% (95% CI, 87–93), specificity 89% (95% CI, 89–90), PPV 24%, and NPV 99%. At PSA ≥ 3 ng/mL the sensitivity was 74% (95% CI, 69–78), specificity 90% (95% CI, 90–91), PPV 21%, and NPV 99%. Per 1000 men screened, Stockholm3 detected 31.6 GG2 prostate cancers vs. 25.8 with PSA (absolute difference 5.8/1000), with false positives 101.8 vs. 95 per 1000 and missed GG2 3.4 vs 9.2 per 1000. Differences in sensitivity were significant (p<0.001). Conclusions: In a population-based screening setting with registry follow-up that reduces verification bias, the Stockholm3 pathway detects more GG2 prostate cancers within 2 years than PSA while maintaining comparable specificity. Clinical trial information: NCT03377881 . Operating characteristics of Stockholm3 and PSA, reporting sensitivity, specificity, PPV, NPV and AUC for diagnosis of GG2 prostate cancer within the 2-year follow-up with 95% CI for Stockholm3 (≥11 and ≥15) and PSA ( ≥ 3 and ≥ 4 ng/mL). Metric Stockholm3 ≥ 11 Stockholm3 ≥ 15 PSA ≥ 3 PSA ≥ 4 Sensitivity 0.90 (0.87–0.93) 0.75 (0.71–0.79) 0.74 (0.69–0.78) 0.52 (0.47–0.57) Specificity 0.89 (0.89–0.90) 0.95 (0.94–0.95) 0.90 (0.90–0.91) 0.95 (0.94–0.95) PPV 0.24 (0.22–0.26) 0.33 (0.30–0.36) 0.21 (0.19–0.24) 0.26 (0.23–0.29) NPV 0.99 (0.99–0.99) 0.99 (0.99–0.99) 0.99 (0.99–0.99) 0.98 (0.98–0.98) AUC 0.96 (0.95–0.96) 0.96 (0.95–0.96) 0.93 (0.92–0.94) 0.93 (0.92–0.94)
Palsdottir et al. (Sun,) studied this question.