Assessing the Clinical Value of Positive Multiparametric Magnetic Resonance Imaging in Young Men with a Suspicion of Prostate Cancer

Abstract

Background: There is a lack of evidence on the ability of magnetic resonance imaging (MRI) of the prostate to detect clinically significant prostate cancer (csPCa) in young patients.

Objective: We hypothesised that the diagnostic performance of MRI for csPCa varies according to patient's age. To address this, we assessed the variation in the csPCa detection rate of MRI targeted biopsy (MRI-TBx) versus systematic random biopsy (SBx) across different patient ages.

Design, setting, and participants: We retrospectively identified 930 patients who underwent prostate MRI and subsequent biopsy at two referral centres between 2013 and 2018. The Prostate Imaging Reporting and Data System (PI-RADS) was used for MRI reporting.

Intervention: A lesion with a PI-RADS score of ≥3 detected at MRI received an MRI-TBx in addition to an SBx during the same session.

Outcome measurements and statistical analysis: The outcome of our study was the relationship between age and csPCa detection rate at MRI-TBx and SBx, respectively. Clinically significant prostate cancer (PCa) was defined as the presence of PCa with Gleason score ≥3+4. Multivariable logistic regression analyses (MVAs) predicting csPCa detection were assessed for both MRI-TBx and SBx. Covariates were age, prostate-specific antigen density, PI-RADS score, previous biopsy status, digital rectal examination, and the number of targeted and systematic cores. The hypothesis that MRI accuracy in detecting csPCa differed by age was finally tested with a nonparametric loess analysis.

Results and limitations: The overall rate of csPCa was 54% (n=506). Overall, 325 (35%) and 461 (50%) patients had csPCa at SBx and MRI-TBx, respectively. The median numbers of SBx and MRI-TBx cores were 12 (interquartile range [IQR]: 10-13) and 5 (IQR: 4-7), respectively. At MVA, age at biopsy was an independent predictor of csPCa at MRI-TBx only (odds ratio: 1.05), after accounting for confounders. In men aged less than roughly 50yr, SBx had a higher probability of detecting csPCa relative to MRI-TBx (25% vs 16% at 40yr). Conversely, in patients aged >50yr, the probability of csPCa was higher in MRI-TBx than in SBx, reaching the highest difference for very elderly patients (48% vs 68% at 80yr). The main limitations were the retrospective design and the small number of young patients.

Conclusions: In this study, we reported the performance of MRI and MRI-TBx in detecting csPCa changes according to patients' age.

Patient summary: In young patients, the performance of a systematic random biopsy in detecting clinically significant prostate cancer (csPCa) is higher relative to magnetic resonance imaging targeted biopsy (MRI-TBx), reflecting the lower accuracy of MRI in younger men. Conversely, in older patients, MRI-TBx showed a clinical benefit with a higher csPCa detection rate compared with SBx, suggesting an increase of MRI accuracy with the increase of age.

Keywords: Age; Biopsy; Diagnosis; Magnetic resonance imaging; Prostate cancer; Targeted biopsy.