Prostate cancer: mpMRI cuts overdiagnosis by 40% compared with TTPM

Takeaway

• Multiparametric magnetic resonance imaging (mpMRI)-targeted biopsy has a good detection rate of clinically significant prostate cancer (csPCa).
• 81% agreement between mpMRI and transperineal template mapping (TTPM) biopsies.
• Compared with TTPM, mpMRI reduced overdiagnosis of insignificant PCa by 40%.

Why this matters

• The Imaging for Significant Prostate Cancer Risk Evaluation (PICTURE) study was designed to overcome methodological issues in previous studies that used TRUS-biopsy or radical prostatectomy as the reference standard.

Key results

• 9% with csPCa on MRI-biopsy had been designated benign or clinically insignificant by TTPM-biopsy.
• 17% with csPCa on TTPM-biopsy were benign or clinically insignificant by MRI-biopsy.
• TTPM-biopsies detected 84% of csPCa and 20.5% of clinically insignificant cancers.
• MRI-biopsies detected 61% and 11%, respectively.
• 5.5% with Gleason >/=3+4=7 on MRI had benign or clinically insignificant cancers on TTPM-biopsies. 18.5% had Gleason >/=3+4=7 detected on TTPM-biopsy that was benign or clinically insignificant by MRI-biopsy.
• 4.5% with definition 2 csPCa on MRI-biopsy had been benign or clinically insignificant by TTPM-biopsy.
• 20% with definition 2 csPCa on TTPM-biopsy were benign or clinically insignificant with MRI.
• 1.5% with Gleason >/=4+4=8 on TTPM-biopsies were not identified by MRI-biopsy.
• 2% classified as insignificant or benign by TTPM were found to have Gleason >/=4+4=8 disease by MRI-biopsy.
• Exact agreement for the presence or absence of any cancer between the two modalities was 81% (95% CI 74%-87%).
• Agreement for presence or absence of definition 1 csPCa was 76% (95% CI 69%-82%).

Study design

• Prospective cohort study of 249 men who had undergone prior TRUS-biopsy.
• All underwent transperineal template prostate mapping (TTPM) biopsies.
• Funding: US National Institutes of Health, Advanced Medical Diagnostics SA.

Limitations

• Single-centre study.
• Only largest lesion on mpMRI was included.
• Limit on number of targeted samples per lesion.