Multiparametric magnetic resonance imaging for pre-treatment local staging of prostate cancer: A Cancer Care Ontario clinical practice guideline

INTRODUCTION: The utility of T2-weighted magnetic resonance imaging (MRI) in the local staging of prostate cancer is controversial. Due to the success of multiparametric MRI in cancer localization, there is renewed interested in MRI (± functional sequences) for local staging. Guidance on pre-treatment local staging of prostate cancer by MRI was developed using systematic review methodology and expert consultation. METHODS: MEDLINE, EMBASE, Cochrane Database of Systematic Reviews, and other databases were searched to identify studies comparing: (1) MRI staging vs. radical prostatectomy staging on diagnostic accuracy outcomes; and (2) MRI staging vs. routine clinical staging on clinical and patient outcomes. Studies meeting inclusion criteria were synthesized by outcome and sensitivity/specificity analysis by tumour location was performed. Evidence quality of included studies was assessed and considered in recommendation formulation. RESULTS: The literature search identified 2510 citations; 62 studies were included. Analysis of MRI ≥1.5 T plus endorectal coil (ER) (± functional sequences) in the detection of extraprostatic extension or seminal vesicle invasion showed modest sensitivities (≥50%) and excellent specificities (>85%) among patients scheduled for radical prostatectomy. MRI upstaging was shown in 20/21 studies, with large variation in correctness (11-85%). Scarcity of clinical and patient outcomes among studies limited synthesis and evaluation. Quality assessment found non-trivial biases. CONCLUSIONS: Modest imaging performance was shown for MRI (1.5 T + ER and 3 T ± ER) ± functional sequences in regards to sensitivity. Limitations in study design, reporting of clinical and patient outcomes, and the heterogeneous use of MRI tempered the strength of the recommendations.