Introduction

MRI-targeted prostate biopsy (MRI-TB) is a promising diagnostic test option for the detection of clinically significant prostate cancer. There is limited data on what the key radiological, surgical and patient factors are that influence detection of clinically significant cancer by MRI-TB. Knowledge of these factors will help optimise the conduct of this diagnostic strategy.

Methods

604 men with clinical suspicion of prostate cancer underwent multiparametric MRI (scored on a 1-5 Likert scale) followed by cognitively registered transperineal MRI-TB at a single centre in a 30-month period. Multi-parametric MRI included T2-weighted, diffusion-weighted and dynamic-contrast enhanced sequences reported by an expert uro-radiologist. Factors influencing detection of clinically significant cancer by MRI-TB were investigated with a multivariate logistic regression using STATA. Multiple imputation was carried out to account for missing data.

Results

In the 604 men, mean age was 65 and median was PSA 7.1. Significant cancer was detected in 390 men (65%). The multivariate analysis adjusting for key confounders showed that factors significantly associated with clinically significant cancer detection included MRI-Likert score (p<0.0001), MRI coil strength (p=0.0013), prior biopsy status (biopsy naive, previous negative biopsy, previous positive biopsy, p=0.0016) and PSA (p=0.0135). Factors not associated with detection of significant cancer included anaesthetic (general versus local,p=0.1274), abnormal digital rectal examination (p=0.0918), surgeon (p=0.1724) and location of tumour (anterior vs posterior, p=0.5825; basal vs apical, p=0.9204).

Conclusions

Key factors that influence the odds of detection of clinically significant cancer have been identified. This study validates the MRI-Likert score as one of the strongest predictors of clinically significant cancer detection. It highlights the importance of presenting PSA and MRI coil strength. Notably, this data also highlights that digital rectal exam finding is not very reliable, consistent with previous studies. Of particular importance, this data supports the feasibility of a local anaesthetic-only approach for transperineal targeted biopsy which has major healthcare delivery and health resource use benefits.

Funding

National Institute for Health and Research UK