We summarised baseline characteristics (including presenting PSA, prostate volume, PSA density [PSAD], and MRI lesion counts) using simple statistics such as medians, interquartile ranges (IQRs), and proportions. PSAD was calculated by dividing PSA by the MRI-derived prostate volume (ellipsoid method). We hypothesised that MRI lesions in men with significant disease (definitions 1 and 2) differ from lesions seen in men without significant cancer (no cancer/insignificant cancer) in terms of their prevalence, count, location (peripheral zone [PZ], transition zone, or both), laterality (right, left, or bilateral), focality (focal or diffuse), overall Likert scores, per-sequence Likert scores (T2-weighted imaging [T2WI], diffusion-weighted imaging [DWI], and dynamic contrast-enhanced [DCE] sequences), volume, and apparent diffusion coefficients (ADCs; nonstandardised mean ADCs derived from axial images demonstrating the highest restriction within each lesion). Nonparametric tests (Wilcoxon rank sum and Kruskal-Wallis analysis of variance [ANOVA]) were used to test differences between groups. In men with Likert 3 index lesions, we hypothesised that PSAD and index lesion ADC are predictors of significant disease (definition 1 or 2) in a multivariable binary logistic regression model.

In order to visualise the false positive mpMRI phenotype and further understand its morphology, the prostate borders, transition zone outlines, and any lesions with overall Likert ≥3 in the TPM-negative group were manually contoured in all axial slices of positive mpMRI sequences (ie, individual sequence Likert score ≥3) using the Osirix platform (Pixmeo SARL, Geneva, Switzerland) and the PROMIS pictorial report as a reference. The surfaces of the manually segmented prostate capsule and the transition zones were aligned in a common space using a feature-based, group-wise registration algorithm that iteratively produced a “mean prostate shape” on which lesions can be superimposed, in line with previous work [7]. This algorithm iteratively updates a mean point cloud based on pair-wise alignment between each case and the mean shape until convergence (with apex and base landmarks guiding nonrigid registration). The R statistical software (R Foundation for Statistical Computing, Vienna, Austria; http://www.R-project.org/) was used for all exploratory and statistical analyses, whereas Matlab (MathWorks Inc, Natick, MA, USA) was used for producing lesion density maps. All p values were considered significant at the 0.05 level.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.