Introduction:

Prostate cancer (PCa) is the most commonly diagnosed male malignancy with there being nearly 36,000 deaths projected in the US in 2024. The core initial risk stratification tool is magnetic resonance imaging (MRI) of the prostate and assessment with the Prostate Imaging Reporting and Data System (PIRADS). Prostate biopsy is then performed under MRI guidance for positive lesions (defined as PIRADS 3 or above). Prior studies have shown that although cancer detection rates remain high for PIRADs 5 lesions they are relatively modest for PIRADs 3 lesions, up to 90% and 30%, respectively. Given the significant heterogeneity in cancer detection rates with the PIRADs system alone, we aimed to identify additional clinical predictors to help identify patients most likely to be diagnosed with clinically significant prostate cancer (csPCA) with PIRADs 3 lesions.

Methods:

We conducted a retrospective chart review for patients who had a PIRADS 3 lesion on mpMRI from 2015-2023. All patients were biopsy naïve prior to MRI and had a prostate biopsy within 1.5 years of their MRI. Both transrectal and transperineal biopsies were included. Clinical-pathologic variables including age, race, PSA at diagnosis, PSA density, MRI prostate volume, location of MRI, lesion location, # of lesions, lesion volume, and Gleason pattern were captured. We calculated cancer detection rates to evaluate presence of csPCa, defined as Gleason Grade Group (GG) 2 or higher. Univariable logistic regression was utilized to determine predictors of csPCA for PIRADs 3 lesions.

Results:

A total of 285 patients with a mean age of 64.1 (SD 7.2) met the inclusion criteria to be included in the analysis. Overall, prostate cancer was detected in 155 patients (54.4%). The majority of cancer lesions biopsied were GG 1, representing 40% of all cancer lesions detected. A total of 93 patients were diagnosed with csPCA (32.6% of the total cohort). Interestingly, the majority of patients with csPCA had lesions discovered on a systematic biopsy (40.9%). Figure 1 shows detection of csPCA stratified by various predictors. Patients with PSA density of < 0.1 had modest cancer detection rates when compared to PSA density > 0.2 at 19.4% and 30.2%, respectively. On univariable logistic regression black race, PSA density > .20, MRI prostate volume < 30cc, and MRI performed at tertiary center were all predictive of diagnosis of csPCA.

Conclusion:

In order to provider further risk stratification in the heterogeneity of PIRADs 3 cancer lesions we identified strong  predictors that alter the risk of csPCA in these patients. Our results show that patients with PSAD < 0.1, MRI prostate volume >80cc, and non-black race have significantly lower rates of csPCa. This suggests that these patients may forgo an initial biopsy and continue to be monitored closely for changes in their PSA. In patients planned for biopsy, we recommend that they undergo both a systematic biopsy and a fusion biopsy given that the majority of the significant cancers in our cohort were discovered on the template cores.

Funding: N/A