# Diagnostic value of prostate magnetic resonance imaging in men with prostate-specific antigen levels ≥ 15 ng/mL for biopsy decision-making

**Authors:** Samuel Trappe, Lars Schimmöller, Patrick Althoff, Karla Johanna Schero, Sebastian Berg, Jan Philipp Radtke, Irene Esposito, Florian Roghmann, Peter Albers, Gerald Antoch, Rouvier Al-Monajjed, Matthias Boschheidgen

PMC · DOI: 10.1186/s13244-026-02235-2 · Insights into Imaging · 2026-03-02

## TL;DR

This study shows that MRI can accurately detect significant prostate cancer in men with very high PSA levels, helping avoid unnecessary biopsies.

## Contribution

The study demonstrates that MRI can reliably exclude significant prostate cancer in men with PSA ≥ 15 ng/mL, reducing the need for biopsies.

## Key findings

- 280 patients underwent biopsy, and 42% showed no presence of prostate cancer despite elevated PSA.
- High PI-RADS scores (4 and 5) were strongly associated with significant prostate cancer detection.
- No significant prostate cancer was found in patients with PI-RADS 2 MRI findings.

## Abstract

To determine the value of MRI in men with highly elevated PSA values for the exclusion of clinically significant prostate cancer (csPC).

In this retrospective bicenter cohort study, consecutive men with PSA values ≥ 15 ng/mL and multiparametric (mp) MRI were included. We excluded patients with acute prostatitis and patients without histopathology or follow-up. Examinations were evaluated regarding MRI quality, PSAD, and PI-RADS classification. For all patients with subsequent biopsy, PC and csPC detection rates were determined. In a subgroup analysis, patients with and without the presence of csPC were compared regarding clinical and MRI parameters.

Finally, 376 patients (median PSA 20 ng/mL) were included. MRI quality was excellent (median PI-QUAL 3). 26% of the patients revealed an MRI with a PI-RADS category 2, 16% were classified as category 3, 12% PI-RADS 4, and 46% showed a PI-RADS 5. A total of 280 patients underwent systematic screening with or without targeted prostate biopsy. Among these, 42% with PSA values ranging from 15 to 116 ng/mL (median 19.5 ng/mL) showed no presence of PC. Overall, csPC detection rates were 94% for PI-RADS 5 and 51% for PI-RADS 4. No csPC were identified in PI-RADS 2, and 8% in PI-RADS 3. Comparative analysis between patients with and without csPC revealed significant differences in age, PSA, PSAD, and PI-RADS (p ≤ 0.05).

mpMRI demonstrated excellent performance in the detection of csPC in this high-risk cohort with PSA levels ≥ 15 ng/mL. High-quality MRI helps to exclude csPC in cases with significantly elevated PSA levels to avoid unnecessary prostate biopsies.

mpMRI demonstrated a high diagnostic accuracy for csPCs in men with PSA ≥ 15 ng/mL, and in cases of non-suspicious MRI findings, it can avoid unnecessary biopsies in these patients at risk.

MpMRI demonstrated high diagnostic accuracy in men with PSA values of ≥ 15 ng/mL.MpMRI enables the reliable exclusion of csPC in cases with non-suspicious MRI findings in these patients.In patients with significantly elevated PSA levels, mpMRI provides an effective risk stratification to avoid unnecessary biopsies.

MpMRI demonstrated high diagnostic accuracy in men with PSA values of ≥ 15 ng/mL.

MpMRI enables the reliable exclusion of csPC in cases with non-suspicious MRI findings in these patients.

In patients with significantly elevated PSA levels, mpMRI provides an effective risk stratification to avoid unnecessary biopsies.

## Linked entities

- **Diseases:** prostate cancer (MONDO:0005159)

## Full-text entities

- **Genes:** KLK3 (kallikrein related peptidase 3) [NCBI Gene 354] {aka APS, KLK2A1, PSA, hK3}
- **Diseases:** Cancer (MESH:D009369), PC (MESH:D011471), inflammation (MESH:D007249), PC (MESH:D015324), infection (MESH:D007239), benign prostate hyperplasia (MESH:D011470), AS (OMIM:612348), hyperplasia (MESH:D006965), ISUP 1 (MESH:C538557), PI-RADS 5 (MESH:D011472), significant (MESH:D065309)
- **Chemicals:** DCE (-)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12953813/full.md

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12953813/full.md

## References

5 references — full list in the complete paper: https://tomesphere.com/paper/PMC12953813/full.md

---
Source: https://tomesphere.com/paper/PMC12953813