Virtual Clinical Outcome Ascertainment in a Prostate Cancer Treatment Trial

ABSTRACT

Introduction/Objective: The COVID-19 pandemic stopped site visits for clinical outcome data collection in March 2020. We utilized several remote methods to collect data and assessed their relative effectiveness. Background(s) The ProtecT randomized trial (Prostate cancer testing and treatment trial) aimed to determine the effectiveness of active monitoring (surveillance), radiation and surgery for localized disease. The primary outcome is prostate cancer mortality with clinical secondary outcomes of disease progression and metastasis. There was no difference in the primary outcome at 10 years (published in 2016) between groups (differences in metastasis and functional problems) so followup was extended to 15 years (November 2020). Method(s) The 10-year analysis used annual paper case report forms (CRFs) completed by research nurses based at UK hospitals. In extended follow-up, it was intended that National Health Service routine data would identify participants with potential disease progression. Prior to the pandemic the research nurse reviewed electronic health records at eight English hospitals and completed an eCRF in REDCap software. It became unlikely that site visits were going to be possible in 2020. A shortened eCRF was created focussed on essential outcome data and site staff agreed to help collect clinical data in July 2020. Result(s) Ethical approval for extending the study end date and the sponsor updating GDPR terms of site agreements were delayed by COVID-19 research taking priority. This also prevented the research nurse updating their NHS Research Passport for Honorary Contracts to access sites. Approvals were gained in December 2020. At four sites, local staff completed REDCap eCRFs with support from the data manager and research nurse by email and virtual calls. The research nurse gained remote access to hospital electronic health records at three sites by April 2021, which required extensive research governance approvals, training on hospital IT systems and their software on multiple laptops. At one site, from December 2020, 2-h virtual calls were held with local staff who interrogated electronic medical records as trial staff completed REDCap eCRFs. On average, 15 note reviews were conducted each call which were planned around clinical commitments. Secondary clinical outcomes were collected remotely for 94% of participants in follow-up (1395/1474). There was no difference in the three methods: remote hospital record access 594/601 (98.8%);local staff completion 575/600 (95.8%);and online calls 215/221 (97.3%) although less data cleaning was required as data queries were resolved during calls. Benefits also included savings on time traveling to sites and accommodation and local clinical staff could access a wider range of health records and information outside their hospital. However, enabling remote data capture delayed data analysis by 6 months. Conclusion(s) In a prostate cancer treatment trial remote data capture of clinical outcomes was successful as site visits became impossible due to the pandemic. Online methods were tailored to sites requirements but required substantial preparation and governance approvals.