CANDIDATE

  • Research type

    Research Study

  • Full title

    A study to develop an algorithm utilising artificial intelligence to predict the likelihood of patients completing the dose limiting toxicity assessment period of early phase cancer clinical trials. A study limited to working with retrospective data.

  • IRAS ID

    323831

  • Contact name

    Louise Carter

  • Contact email

    louise.carter@manchester.ac.uk

  • Sponsor organisation

    University of Manchester

  • Duration of Study in the UK

    1 years, 0 months, 26 days

  • Research summary

    Clinical trials of new cancer treatments have strict eligibility criteria that patients are required to meet to be able to participate in the trial. In addition, it important that patients remain on trial for at least 28 days after starting the drug to assess any side effects they experience so they do not need to be replaced by other patients in the trial as this causes delays in how quickly new drugs are developed. Currently, the clinical team rely on their expertise and experience to decide whether patients are likely to be suitable for a trial and remain on study for at least this minimum period. However, patients who consent for clinical trials are regularly found to be unsuitable for trials or discontinue early with associated distress and risk of side effects without chance of benefit for these patients. Here, we propose to use recent advances in computer science to develop an artificial intelligence tool that can assess clinical data to predict how likely patients are to be eligible for trial and continue beyond this minimum period.

    This is a retrospective data only study using data from patients with advanced cancer who have already attended a consultation regarding suitability for trials with early phase cancer clinical trials teams at either The Christie NHS Foundation Trust, Manchester, UK or The Centre Léon Bérard, Lyon, France. The study will use both English and French language patient data (letters and proformas) created by the clinicians as standard during these appointments to develop a predictive tool. There will be no direct benefit to the patients involved in this study and it will not impact their care. It is estimated it will take up to 1 year to complete the processing of this data and the production and optimisation of the artificial intelligence tool.

  • REC name

    East of England - Cambridge East Research Ethics Committee

  • REC reference

    25/EE/0014

  • Date of REC Opinion

    25 Feb 2025

  • REC opinion

    Further Information Favourable Opinion