Composite Imaging Score Improves Prostate Cancer Detection in PSA Gray Zone

A composite scoring system combining multiparametric magnetic resonance imaging and fluorine-18 prostate-specific membrane antigen positron emission tomography/computed tomography improved the detection of clinically significant prostate cancer in men within the prostate-specific antigen gray zone. In a two-center retrospective study published in BMC Medical Imaging, the composite P score yielded a significantly higher area under the receiver operating characteristic curve than magnetic resonance imaging alone for a junior reader, while offering comparable performance for a senior reader.

Biopsy triage presents persistent clinical challenges in men with prostate-specific antigen (PSA) levels between 4.0 and 10.0 ng/mL. Multiparametric magnetic resonance imaging (mpMRI) in this gray zone can frequently yield equivocal findings and remains vulnerable to reader-dependent interpretation. To supplement anatomical imaging, prostate-specific membrane antigen (PSMA) PET/CT has emerged as a modality capable of providing complementary molecular information to guide diagnostic decisions.

To evaluate this combined approach, researchers conducted a two-center retrospective analysis of 178 men with PSA levels between 4.0 and 10.0 ng/mL who underwent mpMRI, 18F-PSMA PET/CT, and prostate biopsy between January 2021 and December 2024. Clinically significant prostate cancer was defined pathologically as a Gleason score of 3 + 4 or greater on prostate biopsy. Among the 178 men evaluated in the study cohort, clinically significant prostate cancer was confirmed in 125 patients, representing an overall detection rate of 70.2 percent.

Diagnostic image interpretation was structured to compare reader experience levels and multimodal scoring systems. Junior and senior radiologists independently reviewed the mpMRI examinations and assigned Prostate Imaging Reporting and Data System (PI-RADS) scores. For 18F-PSMA PET/CT scans, two nuclear medicine specialists independently interpreted the images, adjudicating discrepancies to determine a single PRIMARY score. The composite P score integrated each radiologist’s individual PI-RADS assessment with the adjudicated PRIMARY score.

When evaluating overall diagnostic accuracy using receiver operating characteristic analysis, the composite P score demonstrated clear advantages for less experienced readers. For the junior radiologist, the composite P score achieved an area under the curve (AUC) of 0.769, significantly outperforming PI-RADS alone, which achieved an AUC of 0.661 (P = 0.014). For the senior radiologist, the composite P score attained an AUC of 0.805 compared with 0.744 for PI-RADS alone, a difference that was not statistically significant (P = 0.167).

At a diagnostic cutoff threshold of 3 or higher, the composite P score demonstrated perfect sensitivity, identifying all 125 clinically significant prostate cancer cases (125/125, 100 percent) for both junior and senior readers. Net reclassification improvement (NRI) was statistically significant for the junior reader at 0.175 (P = 0.010), demonstrating meaningful risk restratification. In contrast, net reclassification improvement did not reach statistical significance for the senior reader.

Clinical utility across decision thresholds was further evaluated through decision curve analysis (DCA). The composite P score yielded a higher estimated net benefit than PI-RADS alone across clinically relevant threshold probabilities for both the junior and senior readers. This analytical metric indicates that integrating molecular and anatomical imaging modalities offers added clinical value regardless of reader experience when determining biopsy indications.

The study authors concluded that the composite P score demonstrated favorable diagnostic performance for detecting clinically significant prostate cancer among men in the PSA gray zone. Combining 18F-PSMA PET/CT with mpMRI provides complementary diagnostic information that may refine biopsy triage strategies in selected patient populations where conventional magnetic resonance imaging results are equivocal or reader-dependent.

Subject of Research: Diagnostic performance of the composite P score in prostate cancer biopsy triage

Article Title: Integrating mpMRI and 18F-PSMA PET/CT via the composite P score to detect clinically significant prostate cancer in the PSA gray zone

Article References: Liu, B., Liu, Y., Chen, Y., Xu, H., Ding, X., Li, W., Wang, Z., Zhang, X., Bai, X., Zhao, J., Wen, C., Wang, R., & Wang, H. (2026). Integrating mpMRI and 18F-PSMA PET/CT via the composite P score to detect clinically significant prostate cancer in the PSA gray zone. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-03042-3

Image Credits: AI Generated

DOI: 10.1186/s12880-026-03042-3

Keywords: prostate cancer, clinically significant prostate cancer, multiparametric MRI, 18F-PSMA PET/CT, composite P score, PSA gray zone, prostate-specific antigen, PI-RADS, PRIMARY score, biopsy triage, Gleason score, area under the curve, net reclassification improvement, decision curve analysis, radiology, nuclear medicine, two-center retrospective study, diagnostic performance