Digital histopathology and clinical outcomes in PROMIS

  • Research type

    Research Study

  • Full title

    Digital histopathology metrics and baseline clinical-MRI information as predictors of prostate cancer outcome in the PROMIS study.

  • IRAS ID

    329659

  • Contact name

    Pushpsen Joshi

  • Contact email

    uclh.randd@nhs.net

  • Sponsor organisation

    University College London

  • Clinicaltrials.gov Identifier

    11/LO/0185, REC reference for PROMIS trial; 21/NE/0139, REC reference for additional tissue work on PROMIS; Z6364106/2024/01/115, UCL Data Protection Registration

  • Duration of Study in the UK

    5 years, 0 months, 1 days

  • Research summary

    The overarching aim of the proposed work will be to obtain clinical outcomes for men who participated in the PROMIS study and digitise all PROMIS prostate biopsy H&E slides to examine whether digital pathology-based biomarkers predict outcome.

    PROMIS was conducted at UCL between 2012-2015. This unique, never-to-be-repeated study was the first to prove that MRI is a superior test for prostate cancer detection compared to standard transrectal biopsy.

    In total, 574 participants with raised PSA underwent MRI in multiple UK institutions, followed by combined transrectal-transperineal template mapping (TPM) biopsies at 5 mm intervals, resulting in the most extensively characterised prostate cancer diagnosis cohort in existence.

    All biopsies were reported by study pathologists, who looked at H&E slides and recorded standard information such as cancer grade (Gleason), maximum cancer core length and presence or absence of inflammation.

    Although these are essential for clinical decisions, modern image analysis offers unprecedented opportunities to use automated computational tools to examine prostate tissue morphology in previously unattainable detail. Furthermore, clinical outcomes for PROMIS patients are yet to be collected and linked to other information.

    This project will have two basic objectives:
    1. Create a PROMIS digital pathology repository by retrieving all standard H&E PROMIS biopsy slides and digitising them at the UCLH Biobank.
    2. Collect linked hospital data (overall survival, death from prostate cancer) to test whether existing baseline clinical-histological-MRI information from PROMIS and newly generated digital pathology-based features predict cancer progression or death.

    If successful, our work will lead to the creation of the most comprehensive repository for prostate cancer diagnosis in the world, incorporating MRI and pathology report data, clinical outcomes and digital H&E from a unique, never-to-be-repeated study. This will be an invaluable resource for prostate cancer researchers worldwide and could produce entirely new standards of prostate biopsy interpretation and cancer risk prediction.

  • REC name

    Yorkshire & The Humber - Sheffield Research Ethics Committee

  • REC reference

    24/YH/0106

  • Date of REC Opinion

    6 Jun 2024

  • REC opinion

    Favourable Opinion