2. UroToday Home
  3. Recent Abstracts
  4. Urologic Oncology
  5. Prostate Cancer

Development of a Prognostic Prediction Model for Clinically Significant Prostate Cancer Based on Lesion Zone and Apparent Diffusion Coefficient Value Quantification - Beyond the Abstract

The Prostate Imaging and Reporting Data Scoring System (PI-RADS v2.1) has improved the localization of disease burden within the prostate and the likelihood of identifying, through targeted biopsy, clinically significant prostate cancer (csPCa).1

Despite its well-validated and standardized scoring, including considerations for lesion zone, limitations remain.2 Among others, unique transitional zone (TZ) features, and competing diseases (such as benign prostatic hyperplasia (BPH)) may mimic prostate cancer and thus pose additional challenges for reviewing radiologists. In this article, Development of a Prognostic Prediction Model for Clinically Significant Prostate Cancer Based on Lesion Zone and Apparent Diffusion Coefficient Value Quantification, published in Urology in February 2026, we explore the impact of lesion zone and quantified apparent diffusion coefficient (ADC) value within a multivariate model on the probability of targeted csPCa.

We sought to better equip the practicing clinician with meaningful data to guide biopsy decision-making, primarily focusing on readily available pre-biopsy findings. Radiographically, we focused on PI-RADS lesion zone (peripheral zone – PZ versus transition zone – TZ) and quantified the lesions' apparent diffusion coefficient (ADC) value. This was added to other predictors of csPCa, including prostate-specific antigen density (PSAD). We developed a pre-biopsy multivariate model to identify key predictors that impact the likelihood of detecting csPCa. In our model, we substratify PI-RADS scoring by these findings as well as standard patient clinical data. While prior risk stratification models exist and are supported in guidelines such as the ERSPC-RC3, this model is novel in including these specific parameters.

This study analyzes 546 PI-RADS 3-5 lesions in a cohort of 429 men in a retrospective fashion with a primary endpoint of detecting csPCa on biopsy, defined as grade group 2 or above. The lesion zone played a substantial role in predicting csPCa, with PZ lesions being 2.6x more likely to harbor csPCa as compared to TZ lesions. Interestingly, this outcome was identified despite TZ lesions harboring a higher proportion of PI-RADS 5 lesions and proportionally fewer PI-RADS 3 lesions as compared to PZ lesions. As such, the proportion of PI-RADS 5 lesions being csPCa in the TZ was lower than that of PZ lesions.

Quantification of ADC value has been used as a predictive feature for csPCa detection previously, but is challenging to broadly implement due to heterogeneity in MRI technology, and is not often lesion-specific. We sought to quantify ADC values of lesions and analyze differences stratified by zone. Therein, the denser TZ demonstrated less variability in ADC quantification between benign/indolent disease and csPCa, a feature that was not duplicated within the less dense PZ. Despite this encouraging identification, our model suggests that lesion zone and ADC quantification, in addition to other clinical features, could not reduce the probability of csPCa to less than 10%, which was our cutoff for biopsy validation. As such, while these individual factors do influence the probability of csPCa and warrant significant consideration with our patients, they are not to be used as sole predictors of whether to biopsy or not.

This finding should empower Urologists to engage in well-informed, data-compelled shared decision making that prostate biopsy, in most instances of positive imaging findings (PI-RADS 3-5), is necessary, but that TZ lesions are less likely to be associated with csPCa. This study should also encourage future studies amongst Radiologists to strongly consider the impact of zone and competing diseases that mimic csPCa when deciding on the ultimate PI-RADS score.

Written by: Rani Ashouri,1 Penny S. Reynolds,2 Shay Rajavel,3 Kunal Hanchate,3 William Snead,1 John M. DiBianco,1 Jason Joseph,1 Paul Crispen,1 Padraic O'Malley,1 Joseph R. Grajo,4 Sara M. Falzarano,5 Thomas F. Stringer,1 Li-Ming Su,1 Christopher J. Weight,6 and Tarik Benidir1

  1. University of Florida, UF Health, Department of Urology, Gainesville, FL.
  2. University of Florida, Department of Anesthesiology, Gainesville, FL.
  3. University of Florida, College of Medicine, Gainesville, FL.
  4. University of Florida, UF Health, Department of Urology, Gainesville, FL; University of Florida, UF Health, Department of Radiology, Gainesville FL.
  5. University of Florida, UF Health, Department of Urology, Gainesville, FL; University of Florida, UF Health, Department of Pathology, Immunology and Laboratory Medicine, Gainesville, FL.
  6. Glickman Department of Urology, Cleveland Clinic, Cleveland, OH.
References:

  1. Kasivisvanathan V, Rannikko AS, Borghi M, et al. MRI-Targeted or Standard Biopsy for Prostate-Cancer Diagnosis. N Engl J Med. 2018;378(19):1767-1777. doi:10.1056/NEJMOA1801993
  2. Park BK. Prostate Imaging and Reporting and Data System Version 2.1: Limitations for Clinical Use. Journal of Urologic Oncology. 2023;21(2):106-111. doi:10.22465/JUO.234600320016
  3. Lin DW, Carlsson S, Filson CP, et al. Updates to Early Detection of Prostate Cancer: AUA/SUO Guideline (2026). Journal of Urology. Published online February 25, 2026. doi:10.1097/JU.0000000000004995
Read the Abstract