Application Value of Prostate-Specific Magnetic Resonance Imaging Features Combined with Nonspecific Serum Markers in Risk Stratification of Clinically Significant Prostate Cancer.

OBJECTIVE

In this retrospective study, 150 cases of prostate cancer were evaluated to explore the ability of combined magnetic resonance imaging (MRI) features and nonspecific serum markers to distinguish clinically significant prostate cancer (csPCa) from nonclinically significant prostate cancer (ncsPCa) in patients with confirmed prostate cancer and to evaluate their application value in risk stratification.

METHODS

This retrospective study analysed 150 patients with prostate cancer treated at our institution between May 2022 and May 2025. The patients were divided into csPCa (Gleason score ≥7) and ncsPCa groups (Gleason score = 6) according to Gleason score of pathology. Baseline clinical data and routine haematological and coagulation markers, including neutrophil count (NEU), lymphocyte count (LYM), fibrinogen (FIB), D-dimer and prostate-specific antigen (PSA) were collected. All patients subsequently underwent prostate-specific MRI following enrolment.

RESULTS

Significant difference in Prostate Imaging Reporting and Data System (PI-RADS) V2.1 score distribution was observed between the two groups (p < 0.05). The csPCa group also had higher neutrophil-to-lymphocyte ratio (NLR), FIB, D-dimer and PSA levels than the ncsPCa group (p < 0.05). Multivariate analysis confirmed these indicators as independent predictors of csPCa (p < 0.05). Receiver operating characteristic curve analysis showed the following area under the curve (AUC) values in diagnosing csPCa: 0.677 (95% confidence interval (CI): 0.571-0.784) for PI-RADS V2.1 score, with an optimal cutoff of 3.00; 0.738 (95% CI: 0.638-0.838) for NLR, with an optimal cutoff of 3.67; 0.769 (95% CI: 0.680-0.858) for FIB, with an optimal cutoff of 4.01; And 0.745 (95% CI: 0.639-0.852) for D-dimer, with an optimal cutoff of 0.595. The combined diagnostic model yielded an AUC of 0.839 (95% CI: 0.757-0.920) for identifying csPCa.

CONCLUSIONS

The combined use of prostate-specific MRI features and nonspecific serum markers (NLR, FIB and D-dimer) can effectively improve the diagnostic accuracy of csPCa.