Systemic inflammatory indices in biopsy-naïve men with PI-RADS 3 lesions: added value over PSA density and the free-to-total PSA ratio

INTRODUCTION AND OBJECTIVES: A PI-RADS 3 lesion on multiparametric magnetic resonance imaging (mpMRI) carries an intermediate risk of prostate cancer that existing parameters characterize poorly, and the biopsy decision remains unsettled. This study evaluated whether the pre-biopsy systemic immune-inflammation index (SII) and neutrophil- to-lymphocyte ratio (NLR) provide diagnostic information beyond two prostate-specific parameters, PSA density (PSAD) and the free-to-total PSA ratio (%fPSA), in biopsy-naïve men. METHODS: Eighty-two biopsy-naïve men with an index PI-RADS 3 lesion, evaluated between January 2022 and December 2024, were retrospectively reviewed. All patients underwent transrectal ultrasonography (TRUS)-guided 12-core systematic biopsy with additional cognitively targeted cores from the index lesion. SII, NLR and the platelet-to-lymphocyte ratio (PLR) were calculated from pre-biopsy complete blood counts. Cut-off values were determined by receiver operating characteristic (ROC) analysis at the maximum Youden index, except for PSAD, for which a predefined cut-off of 0.15 ng/mL/mL was used. Independent predictors were identified by multivariable logistic regression. RESULTS: Prostate cancer was detected in 20 men (24.4%) and clinically significant cancer (ISUP grade group ≥ 2) in 13 (15.9%). Age, total PSA and prostate volume did not differ between men with and without cancer. Men with cancer had a higher PSAD (0.16 ± 0.08 vs. 0.12 ± 0.06 ng/mL/mL, p = 0.024) and a lower %fPSA (14.2 ± 5.8% vs. 19.6 ± 7.4%, p = 0.004). SII and NLR were also significantly higher in the cancer group (p = 0.042 and p = 0.016), whereas PLR did not differ. The area under the ROC curve (AUC) was highest for %fPSA (0.743) and PSAD (0.712), and lower for NLR (0.651) and SII (0.634). In multivariable analysis, only PSAD ≥ 0.15 ng/mL/mL (OR 2.86, 95% CI 1.12-7.32, p = 0.028) and %fPSA (OR 0.94, 95% CI 0.89-0.98, p = 0.012) remained independent predictors; SII and NLR did not. CONCLUSIONS: In biopsy-naïve men with a PI-RADS 3 lesion, SII and NLR were higher in those with cancer on biopsy, but they were not independent predictors after adjustment for PSAD and %fPSA. The statistical power of the model to detect a modest effect was limited. In this setting, the biopsy decision should still be based on prostate-specific parameters. SII and NLR may complement these parameters, and their additional value should be tested within multivariable models rather than in isolation.