Development and Validation of the Utlife-Pc Algorithm for Noninvasive Detection of Prostate Cancer in Urine: A Prospective, Observational Study

Prostate biopsy is the gold standard for prostate cancer diagnostics and is an invasive procedure. Overbiopsy is a serious health issue caused by the low specificity of PSA. A urine tumor DNA multidimensional bioinformatic algorithm, utLIFE, was developed for urine signal detection. This study (ChiCTR2300071837) was designed to apply the utLIFE algorithm in prostate cancer early detection. The objective is to discriminate between any grade group of prostate cancer and non-cancer with high accuracy. A case-control cohort was used for model construction, and an independent prospective cohort was used to validate the model blindly. Of the 801 participants recruited in this study, 580 participants were selected for subsequent analysis. The median age was 67 (34-90) years. In the training cohort (n = 237), utLIFE-PC got an AUC of 0.967 (95% CI, 0.947-0.988) and a sensitivity of 85.59% (95% CI, 76.97%-91.88%) at 95% (95% CI, 89.97%-97.97%) specificity. In the validation cohort (n = 343), utLIFE-PC had an AUC of 0.929 (95% CI, 0.897-0.961), sensitivity of 84.24% (95% CI, 77.77%-89.44%), and specificity 93.26% (95% CI, 88.25%-96.47%). The model showed better performance than PSA (p < 0.001) or the single-dimensional biomarkers (methylation, p < 0.001; CNVs, p< 0.001; mutation, p < 0.001). Moreover, the model could avoid 93.26% (166/178) of unnecessary prostate biopsies compared with the current criteria. Multidimensional biomarkers may increase the performance of cancer detection beyond that of single-dimensional biomarkers. The utLIFE-PC model has the potential to optimize the PCa diagnostic process and avoid unnecessary biopsies noninvasively and conveniently.