PI-RADSv2.1 combined with PSA density for optimizing prostate biopsy decisions: a retrospective analysis

Yuchun Li, Siran Wang, Jianqiu Wang, Xin Qi, Tianran Liu, Xiaodong He, Yu Zhang, Yi Zhu, Yunfu ZengFrontiers in Oncology 2025 · DOI

Abstract

The use of the PI-RADS v2.1 score combined with Prostate-Specific Antigen Density (PSAD) is recommended as a refined indication for prostate biopsy. Based on a retrospective analysis, specific thresholds were identified: (1) For patients with PI-RADS scores of 1-2, biopsy can be avoided if PSAD is < 0.30 ng/(mL·cm³). (2) For patients with a PI-RADS score of 3, biopsy can be avoided if PSAD is < 0.15 ng/(mL·cm³). (3) Patients with a PI-RADS score of 4 are recommended for biopsy in all cases. (4) For patients with a PI-RADS score of 5, biopsy is standard, though empirical treatment may be considered if PSAD ≥ 2.00 ng/(mL·cm³). Using these criteria, 40.3% (186/462) of patients in this study could potentially avoid unnecessary prostate biopsies.

Introduction

Prostate cancer (PCa) screening remains a clinical challenge, often resulting in a high volume of unnecessary, invasive biopsies due to the low specificity of traditional PSA blood tests. While multiparametric MRI (mpMRI) and the PI-RADS scoring system have significantly improved diagnostic precision, "gray area" cases—particularly those with PI-RADS 3 scores—continue to present diagnostic uncertainty. This study investigates whether combining PI-RADS v2.1 scores with PSAD can provide a more reliable clinical signal for detecting clinically significant prostate cancer (csPCa) while reducing the burden of over-diagnosis.

Methods and Datasets

The researchers conducted a retrospective analysis of 462 consecutive patients at the First People’s Hospital of Yibin, China, between January 2020 and October 2024. All included patients underwent both mpMRI and prostate biopsy. The study population was divided into a PCa group (n=233) and a Non-PCa group (n=229), with a further subgroup of 211 patients identified as having csPCa (Gleason score ≥ 3+4).

Flowchart of the patient enrollment process.

Diagnostic performance was evaluated using univariate and multivariate logistic regression to identify independent risk factors. The study specifically compared the PI-RADS v2.1 score, PSAD, total PSA (tPSA), and the free-to-total PSA ratio (f/tPSA) using Receiver Operating Characteristic (ROC) curve analysis.

Key Results and Insights

The study demonstrated that while PI-RADS and PSAD are useful individually, their combination offers superior diagnostic accuracy. The combined model achieved an Area Under the Curve (AUC) of 0.966, significantly outperforming PI-RADS alone (0.926) and PSAD alone (0.915).

ROC curve analysis for csPCa detection.

Tiered Biopsy Recommendations

By integrating PSAD thresholds with PI-RADS scores, the researchers established "safe zones" where the risk of csPCa was negligible:

  • PI-RADS 1-2: Biopsy can be safely avoided if PSAD is < 0.30 ng/(mL·cm³).
  • PI-RADS 3: Biopsy can be avoided if PSAD is < 0.15 ng/(mL·cm³).
  • PI-RADS 4: Biopsy remains recommended for all patients.
  • PI-RADS 5: Biopsy is standard, but for patients with PSAD ≥ 2.00, the cancer detection rate was 100%, suggesting potential for accelerated treatment pathways.

This synergy is particularly effective because PSAD corrects for the confounding effects of Benign Prostatic Hyperplasia (BPH), where larger prostates naturally produce higher PSA levels without indicating malignancy.

Conclusion and Limitations

The combined PI-RADS v2.1 and PSAD approach achieved a sensitivity of 92.4% and a specificity of 91.6%. Most importantly, applying these criteria would have allowed 40.3% of the study cohort to avoid a biopsy entirely.

However, the study is limited by its retrospective, single-center design. Because prostate characteristics and PSA levels can vary by ethnicity, these specific PSAD thresholds (0.15 and 0.30) require validation in broader, multi-ethnic global populations. Future research should focus on prospective validation and the potential integration of AI-driven image analysis to further refine these diagnostic models.

Summary updated

Abstract

The use of the PI-RADS v2.1 score combined with PSAD is recommended as an indication for prostate biopsy: (1) For patients with PI-RADS scores of 1-2, biopsy can be avoided if PSAD is < 0.30 ng/(mL·cm³), whereas biopsy is recommended if PSAD is ≥ 0.30 ng/(mL·cm³). (2) For patients with a PI-RADS score of 3, biopsy can be avoided if PSAD is < 0.15 ng/(mL·cm³), but it is recommended if PSAD is ≥ 0.15 ng/(mL·cm³). (3) Patients with a PI-RADS score of 4 are recommended for biopsy in all cases. (4) For patients with a PI-RADS score of 5, biopsy is recommended if PSAD is < 2.00 ng/(mL·cm³), but empirical initiation of treatment without biopsy may be considered if PSAD ≥ 2.00 ng/(mL·cm³), subject to ethics committee approval. Using these criteria, 40% (186/462) of patients in this study could potentially avoid prostate biopsy.
BibTeX
@inproceedings{li2025pi,
  author = {Yuchun Li and Siran Wang and Jianqiu Wang and Xin Qi and Tianran Liu and Xiaodong He and Yu Zhang and Yi Zhu and Yunfu Zeng},
  title = {{PI-RADSv2.1 combined with PSA density for optimizing prostate biopsy decisions: a retrospective analysis}},
  booktitle = {Frontiers in Oncology},
  year = {2025},
  doi = {10.3389/fonc.2025.1602412},
}
Figure