PLoS ONE (Jan 2022)

A protocol for the VISION study: An indiVidual patient data meta-analysis of randomised trials comparing MRI-targeted biopsy to standard transrectal ultraSound guided bIopsy in the detection of prOstate cancer.

  • Veeru Kasivisvanathan,
  • Vinson Wai-Shun Chan,
  • Keiran D Clement,
  • Brooke Levis,
  • Masoom Haider,
  • Ridhi Agarwal,
  • Mark Emberton,
  • Gregory R Pond,
  • Yemisi Takwoingi,
  • Laurence Klotz,
  • Caroline M Moore,
  • VISION study collaborators

DOI
https://doi.org/10.1371/journal.pone.0263345
Journal volume & issue
Vol. 17, no. 2
p. e0263345

Abstract

Read online

BackgroundTransrectal ultrasound (TRUS) guided biopsy for prostate cancer is prone to random and systemic error and has been shown to have a negative predictive value of 70%. PRECISION and PRECISE are among the first randomised studies to evaluate the new MRI-targeted biopsy (MRI-TB) pathway with a non-paired design to detect clinically significant prostate cancer and avoid unnecessary treatment. The trials' results individually demonstrated non-inferiority of MRI-TB compared to TRUS biopsy. An individual patient data (IPD) meta-analysis was planned from the outset of the two trials in parallel and this IPD meta-analysis aims to further elucidate the utility of MRI-TB as the optimal diagnostic pathway for prostate cancer.Methods and materialsThis study is registered on PROSPERO (CRD42021249263). A search of Medline, Embase, Cochrane Central Register of Registered Trials (CENTRAL), Web of Science, and ClinicalTrials.gov was performed up until 4th February 2021. Only randomised controlled trials (PRECISE, PRECISION and other eligible trials) comparing the MRI-targeted biopsy pathway and traditional TRUS biopsy pathway will be included. The primary outcome of the review is the proportion of men diagnosed with clinically significant prostate cancer in each arm (Gleason ≥ 3+4 = 7). IPD and study-level data and characteristics will be sought from eligible studies. Analyses will be done primarily using an intention-to-treat approach, and a one-step IPD meta-analysis will be performed using generalised linear mixed models. A non-inferiority margin of 5 percentage points will be used. Heterogeneity will be quantified using the variance parameters from the mixed model. If there is sufficient data, we will investigate heterogeneity by exploring the effect of the different conducts of MRIs, learning curves of MRI reporting and MRI targeted biopsies.Trial registrationThis systematic review is registered on PROSPERO (CRD42021249263).