Implementation of Optimized urine smart test in the diagnostic pathway of prostate cancer: prospective observational study in biopsy-naive patients

Prostate cancer (PCa) stands as the second most prevalent malignancy in men globally. Despite declining mortality rates, the incidence of PCa in Europe is on the rise, posing a significant impact on the healthcare system. Two urine biomarkers, uPSA and uZinc, have individually shown promising results in identifying clinically significant PCa in the diagnostic path of prostate cancer. This study aims to determine whether the use of both these urine biomarkers, individually and in combination with the standard of care and mpMRI, enhances the diagnostic capabilities in diagnosing clinically significant PCa (ISUP ≥ 3) compared to the current standard of care (SOC). The patients selected (n=247) were to undergo their first prostate biopsy at the urology department of San Luigi Gonzaga hospital in Turin, Italy. Indications for biopsy followed the current indications recommended by the European Association of Urology guidelines based on DRE evaluation, PSA levels, and mpMRI. The urine samples were collected following a DRE and prostatic massage prior to undergoing biopsy, which were retrieved under either MRI-US fusion or standard approach. Biopsy samples were then analysed, and patients were classified into three groups: healthy patients, positive samples with GS ≤ 7 (3+4) – ISUP < 3 and GS ≥ 7 (4+3) – ISUP ≥ 3. Ultimately, 238 urine samples were analysed. Our results demonstrate that both urine PSA and Zinc (Urine model) exhibit decreased mean levels in PCa (p<0.0001 and p=0.01, respectively), with lower values correlating with increasing histological grades. ROC curves for the diagnostic accuracy of uPSA for GS ≤ 7 (3+4) and ≥ 7 (4+3) yielded AUC of 0.768 and 0.795, respectively, implying that this biomarker is a reliable predictor of PCa. Concerning uZinc the AUC were 0.52 and 0.634, respectively, yielding milder diagnostic accuracy. ROC curves were then employed to assess the different models diagnostic accuracy, revealing an AUC of 0.804 for the Urine model compared to 0.677 for the SOC model alone, confirming its diagnostic superiority. Combining the urine model with SOC and mpMRI yielded the greatest improvement in detecting PCa, with AUC values of 0.843 (SOC+Urine) and 0.868 (MRI+Urine). The highest enhancement in diagnostic accuracy was observed in the SOC+MRI+Urine model, yielding an AUC of 0.868. Additionally, simulation scenarios estimated a 42% reduction in unnecessary biopsies with the use of the SOC+Urine and a 57% decrease with the SOC+Urine+MRI model, with less than 3% PCa overlooked. Despite the study's limitations, these promising results suggest that the use of uPSA and uZinc as biomarkers improves the diagnostic accuracy of the currently employed diagnostic modalities. Further research is warranted to improve the robustness of these findings and assess the use of these biomarkers in the screening of patients for prostate cancer.