# Diagnostic Accuracy of Combination of Multiparametric MRI PI-RADS Score v2.1 and Prostate-Specific Antigen Density for Prostate Cancer Detection

## TL;DR

This study shows that combining MRI scores and PSA density improves accuracy in detecting prostate cancer compared to using MRI scores alone.

## Contribution

The study identifies a new optimal PSA density cutoff (0.18) for prostate cancer detection when combined with MRI scores.

## Key findings

- Combining PI-RADS v2.1 score ≥ 3 and PSAD ≥ 0.18 achieved 96.97% sensitivity and 71.43% specificity for prostate cancer detection.
- PSAD with a cutoff of 0.18 showed good diagnostic performance with an AUC of 0.897.
- Lesion volume was not a significant predictor of prostate cancer in this study.

## Abstract

Introduction

Prostate cancer (PCa) is the second most commonly diagnosed cancer in men worldwide. The Prostate Imaging-Reporting and Data System version 2.1 (PI-RADS v2.1) scoring system using multiparametric magnetic resonance imaging (mp-MRI) increases the accuracy for the assessment of clinically significant PCa. This study evaluates the diagnostic accuracy of a combination of PI-RADS v2.1 scores with prostate-specific antigen density (PSAD) for the detection of PCa, using biopsy outcomes as the gold standard, as well as the diagnostic accuracy of the combination of PI-RADS 3 lesion volume and PSAD.

Methods

This is single-center cross-sectional retrospective study including 54 subjects with serum PSA values > 4 ng/mL, who were referred for prostate mp-MRI. All patients underwent subsequent transrectal ultrasound (TRUS)-guided biopsy. Data collected includes PSA value, mp-MRI characteristics of the lesion, and histopathological findings. PI-RADS v2.1 score and PSAD were used to evaluate the diagnostic accuracy of this combination.

Results

In our study, the optimal PSAD cutoff was >0.18 with an area under the curve (AUC) of 0.897, indicating good diagnostic performance. The combination of PI-RADS v2.1 score ≥ 3 and PSAD ≥0.18 increased diagnostic accuracy, with a sensitivity of 96.97% and specificity of 71.43%. However, lesion volume was not a significant predictor of PCa.

Conclusion

In summary, our study demonstrates that the combination of PI-RADS score and PSAD yields higher diagnostic accuracy for the detection of PCa (p < 0.001) than using the PI-RADS score alone. We found an optimal PSAD cutoff of 0.18, which differs from the international consensus of 0.15. However, it cannot be used as a substitute for definitive pathological diagnosis but can be used in combination for better risk stratification, counselling, and management of patients with elevated PSA levels. Combining PI-RADS 3 lesion volume and PSAD did not have statistically significant results in our study.