Identifying Cancer Recurrence within Patient Care Pathways

  • Research type

    Research Study

  • Full title

    Identifying Cancer Recurrence within Patient Care Pathways across Linked National Clinical Datasets

  • IRAS ID

    310352

  • Contact name

    Kate Walker

  • Contact email

    kate.walker@lshtm.ac.uk

  • Sponsor organisation

    London School of Hygiene and Tropical Medicine

  • Duration of Study in the UK

    2 years, 11 months, 31 days

  • Research summary

    For patients having curative for their cancer, the aim of treatment is to get rid of the cancer without it coming back years later. For these patients, cancer recurrence is the most important outcome of care. Cancer recurrence is a key outcome in research of primary cancer treatments. Treatment options for cancer are increasingly complex and there are growing gaps in evidence on the best treatment combinations. Data on cancer
    recurrence is needed to address these gaps and to ensure all patients receive the best care.

    Routinely collected digital healthcare data can be used to identify cancer patients but cannot currently
    be used to detect cancer recurrence. Studies using these types of data are therefore unable to use cancer
    recurrence as an outcome.
    National clinical data on cancer recurrence is needed to be able to compare treatment options in all
    patients, not just the selected groups who are included in randomised clinical trials. It would also allow
    comparisons of treatments that are unlikely to be assessed in randomised clinical trials.

    With more information collected in national clinical datasets, we can build a detailed picture of patient
    care. Patients who are treated with an aim to cure them of their cancer tend to have predictable patterns of care. If their cancer returns, this will appear in the data as a change in frequency and type of
    hospital visits, tests and treatments. A group of clinical experts will use evidence from research, clinical
    guidelines and clinical audit to help to develop clinical rules which will identify cancer recurrence in the
    data. Sophisticated computer science methods, known as ‘machine learning’, offer alternative ways to
    detect patterns of care that tell us it is very likely the cancer has recurred.

    The research will develop methods to identify cancer recurrence for bowel cancer and find out how well the methods work for identifying breast and prostate cancer
    recurrence.

  • REC name

    London - Queen Square Research Ethics Committee

  • REC reference

    21/HRA/5559

  • Date of REC Opinion

    20 Jan 2022

  • REC opinion

    Favourable Opinion