Does size matter? Tumor density as a predictor of clinically significant prostate cancer on MRI-targeted...

Prior studies of tumor burden as a predictor of clinically significant prostate cancer (csPCa) consist of per-patient analyses or are limited by imprecise size measurements. This study evaluates tumor density (TD) as a lesion-level predictor of csPCa on magnetic resonance imaging (MRI)-targeted biopsy.

We reviewed patients who consecutively underwent MRI-ultrasound fusion transperineal biopsy for suspicion of csPCa from July 2023 to August 2025 at a tertiary academic center. TD was calculated using exact lesion volume generated by fusion software divided by MRI prostate volume. The primary outcome was per-lesion detection of csPCa, defined as International Society of Urological Pathology Grade Group (GG) ≥2. We assessed the association between TD and csPCa using generalized estimating equation logistic regression models and provided estimates of discrimination and clinical benefit.

The cohort included 349 patients with 496 PI-RADS 3-5 lesions, of which 34.7% harbored csPCa. Lesions with csPCa had greater median TD than those without (0.013 vs. 0.006, p<0.001). On univariate analysis, greater TD was associated with higher odds of csPCa (OR 1.59, 95% CI 1.39-1.83, p<0.001). In multivariable analysis adjusting for age, race, family history, PSA, PI-RADS score, lesion location, and biopsy history, TD remained independently associated with csPCa (OR 1.44, 95% CI 1.20-1.71, p<0.001). Adding TD to the baseline multivariable model also led to improvement in AUC (ΔAUC 0.024, 95% CI 0.017-0.029; p=0.004).

TD was independently associated with csPCa on targeted biopsy and showed moderate discrimination. TD is a readily available metric which may improve lesion-level risk stratification beyond standard clinical and imaging information.