Advancing Survivorship after Cancer: Outcomes Trial (ODR1819_039)

Data for this study has previously been shared when the data was controlled and managed by Public Health England (PHE). PHE facilitated data release via its Office of Data Release service (ODR). ODR was responsible for providing a common governance framework for responding to requests to access PHE data for secondary purposes, including service improvement, surveillance and ethically approved research. All requests to access data were reviewed by the ODR and were subject to strict confidentiality provisions. The responsibility for the management of the National Disease Registration Service of which the National Cancer Registration and Analysis Service is a part, transferred from PHE to NHS England on 1st October 2021.

Smoking, poor diet, low levels of physical activity and obesity are established risk factors for the development of cancer, but there is now a growing body of evidence linking them with survival after diagnosis.

The weight of scientific evidence has led to consensus statements regarding lifestyle change for cancer survivors. The most comprehensive of these recommendations are by the World Cancer Research Fund (WCRF) who recommend that cancer survivors be physically active for 30 minutes every day, limit consumption of? energy dense foods, limit consumption of red meat and avoid processed meat, limit consumption of salty foods, eat more plant foods, be as lean as possible without being underweight, limit alcoholic drinks to one or two per day for women and men respectively, and to avoid tobacco. These recommendations are for cancer survivors defined as ‘all people who are living with a diagnosis of cancer, and those who have recovered from the disease’, which is the definition University College London (UCL) will adopt for this study.

While these recommendations are an important development for the field, they now need to be translated into effective behaviour change programmes that are appropriate for use with cancer survivors and can be disseminated cost-effectively through existing care pathways.

A number of studies carried out in the US, Canada and Australia have demonstrated that lifestyle interventions for prostate, breast and colorectal cancer are not only acceptable to patients but can successfully improve more than one health behaviour. However, although these results are promising, there is a lack of behaviour change interventions in a UK healthcare setting. Existing lifestyle interventions are fairly labour intensive, and in order to benefit enough cancer survivors to have a significant public health impact, an intervention must be cost-effective and feasible to implement in the current clinical pathway.

A potential strategy is to use standard print materials containing advice on guidelines and simple behavioural skills as the ‘first-line’ intervention. This maximises coverage and takes advantage of patient preferences for print materials and for simple advice. A systematic review of the literature by the study group identified that interventions in breast, prostate and colorectal cancer survivors can be successful in changing multiple health behaviours.

Advancing Survivorship after Cancer: Outcomes Trial (ASCOT) is funded by Cancer Research UK and sponsored by UCL. Recruitment to the trial began in 2015. Cancer patients were identified by hospital sites and were sent an ASCOT survey. If patients chose to complete the survey, they then had the option to leave their details to then hear more about the ASCOT trial. Surveys were returned by patients to UCL where the study team then checked patient eligibility and if eligible sent them the information sheet for the trial. All trial participants had completed their 2-year follow-up by the end of November 2021. Participants consented to the trial team keeping their personal information up to ten years after they consented in case access to data up until that time was required for follow up research purposes.

UCL is responsible for determining the scope of the study specification. These specifications set out the purpose of the study, the patient groups and survey topics to evaluate. The sponsor, Cancer Research UK (CRUK) will not process the data nor direct UCL in the processing of these data. CRUK is not a controller in this instance and do not set out the purpose of this study and will have no ability to suppress or otherwise limit the publication of findings.

The aims of the Advancing Survivorship after Cancer: Outcomes Trial (ASCOT) are:

1) To carry out a survey of UK breast, prostate and colorectal cancer survivors to assess current health behaviours (e.g. diet and physical activity) and desire and preference for lifestyle advice (ASCOT patient survey). A secondary aim is to assess current levels of fatigue, sleep, happiness, social isolation and quality of life.

2) To carry out a two-year, two-arm individually randomised controlled trial to compare the effectiveness of a low-intensity behaviour change intervention on improving health behaviours among breast, colorectal or prostate cancer survivors, compared with usual care (ASCOT pilot trial).

Consenting participants complete their baseline assessments (the survey) then eligible participants are individually-randomised to receive either the ASCOT intervention (a self-guided printed booklet designed to help cancer survivors make healthy lifestyle behaviours habitual) or to receive ‘usual care’. The current understanding of usual care is that patients do not receive behavioural/ lifestyle health advice unless they specifically ask for it. Follow-up assessments (the pedometer, dietary recalls by telephone, and a shortened version of the ASCOT patient survey were repeated at 3, 6 months and 2 years. The results will help determine the proportion of those who are meeting WCRF recommendations for physical activity, diet, smoking, alcohol and whether the ASCOT intervention assists patients diagnosed with breast, prostate and colorectal cancer to make healthier habitual life choices.

Participants were chosen where records show there was a diagnosis of breast, prostate or colorectal cancer. UCL invited around 2500 people to take part in this trial of which 1348 enrolled with 1152 still active at the time the study team approached participants about accessing their NHS England records. Of the 1152 approached about NHS England data access 987 (86%) of these consented. Of the 987 who consented to NHS England record access: 3 of these withdrew from the study before the last NHS England data dissemination. 2 more participants have subsequently withdrew from the study. Participation is on a voluntary basis with freedom to withdraw at any time and without giving a reason.

Participants often do not know or are unable to remember, details of their cancer diagnosis. Therefore, data requested from the National Cancer Registration and Analysis Service will be combined with data collected within the trial to enable the UCL to characterise the sample based on their cancer type and stage at diagnosis. This Agreement does not permit the Controllers to further disseminate NHS England data and any outputs will only contain aggregated data with small numbers supressed.

Specifically, UCL would like to access National Cancer Registration and Analysis Service data on date of diagnosis, site of cancer, and cancer stage, in order to characterise the sample and compare participants outcomes across different groups of cancer patients. Information is requested for any cancers that a participant has had since 2005 as this will include the cancer (Prostate, Breast or Colorectal) that is the reason for their recruitment to the trial and their most recent cancer before they started the trial. This amendment requests to adjust the data specification to include the addition of mortality data and to extend the dates for which UCL previously received disseminated data on any cancers diagnosed (01 January 2005- 31 December 2017) to the 31 December 2020. The data requested under this agreement is to be used to evaluate and compare the experimental groups on their mortality and cancer events following their participation in the trial.

To support the study in achieving its aims dissemination of the following datasets are requested on a one-off basis.

• Cancer registrations (diagnosis date, tumour site, tumour stage, vital status).

• The addition of Cancer registration mortality data, both deceased status and cause of death where available.

The data requested above should provide the study with the necessary information on disease characteristics of the patient cohort, cancer type, stage at diagnosis and mortality events. There are no alternative less intrusive means of achieving the purpose outlined within this agreement. Participants have consented to the data being accessed and linked to their trial data. Additionally, this application has HRA Ethics approval.

The ASCOT study is run in the Department of Behavioural Science and Health (UCL) from 2015. The participant data collection ran from 2015-2021. Individuals consented to the ASCOT study team accessing their data 10 years after they consented.

To address the GDPR Principle of Data Minimisation, this request is limited to all patients who have consented and been enrolled into the trial who meet the inclusion criteria below.

1. Adults aged ≥18 years (no upper age limit)

2. Individuals diagnosed with non-metastatic (Stage I – III) breast, prostate or colorectal cancer in 2012, 2013, 2014 or 2015

3. Individuals not receiving active anti-cancer treatment (except for those oral anti-cancer treatments taken at home who can be included)

4. Individuals who are able to understand spoken and written English

The recruitment of participants were limited to the following NHS trusts: NHS hospital trusts in London, Bedfordshire or Essex (Bart’s Health NHS Trust, University College London Hospitals NHS Foundation Trust, Southend University Hospital NHS Foundation Trust, Mid Essex Hospital Services NHS Trust, Basildon and Thurrock University Hospitals NHS Trust, Royal Free London NHS Foundation Trust and Barking, Havering and Redbridge University Hospitals NHS Trust, Luton & Dunstable University Hospital NHS Foundation Trust , The Princess Alexandra Hospital NHS Trust and The Whittington Hospital NHS Trust.

UCL aims to conduct research to the highest standards of research integrity. Research is underpinned by policies and procedures that ensure they comply with regulations and legislation that govern the conduct of research; this includes data protection legislation such as the General Data Protection Regulation (GDPR) and the Data Protection Act 2018 (DPA).

UCL rely on Article 6 (1) (e) ‘Public Task’ of the UK General Data Protection Regulation (GDPR) as the lawful basis of processing - "processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller". This is justified as UCL uses personal data to conduct research to improve health, care and services. As a publicly-funded organisation incorporated under a Royal Charter, UCL ensure that it is in the public interest when they use personal data from people who have agreed to take part in research.

UCL rely on Article 9 (2) (j) ‘Research’ as the legal basis for processing under UK GDPR – ‘’the processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject”. This is justified as all UCL’s Health and care research serves the public interest, which means that the data controllers under this agreement have to demonstrate their research serves the interests of society as a whole. Most of UCL’s health and care research follows the UK Policy Framework for Health and Social Care Research.

Eligible patients were identified by searching the electronic cancer databases (e.g. Somerset or Info-flex) of each of the participating NHS trusts cited in objective for processing. The initial list was drawn up by an appropriate member of staff at the trusts (e.g. a nurse or data analyst). The list was carefully cross-checked (via surgical diaries or the multidisciplinary team) to ensure that patients are still alive and have a diagnosis of breast, prostate or colorectal cancer.

Once the list of eligible patients had been finalised the NHS sites only shared a list of finalised ID numbers and cancer types with UCL. The research team at UCL could then issue the staff at the trusts with the appropriate number of copies of the ASCOT initial patient survey (health and lifestyle questionnaire), the letter from the consultant and envelopes. At the back of the survey there was an invitation for participants to leave their details if they would be interested in hearing more about a trial of a lifestyle intervention for cancer patients. The NHS Trusts then sent the letter and survey together in the envelope to the patients to ensure that the research team at UCL did not have initial access to patient identifiable information unless a participant chose to leave their details on the initial survey invitation. All subsequent questionnaires for consented patients for 3/6 and 2 year follow ups were sent by UCL.

Staff at the NHS sites kept a copy of the list of patients with allocated ID numbers that were printed on the initial survey. Completed initial surveys were returned by patients directly to UCL. Staff at UCL then informed the NHS sites of the ID numbers that were returned so reminders could be sent to any patients who hadn’t completed the initial survey. When surveys were returned UCL checked eligibility for the trial for participants who chose to leave their contact details on the survey. If participants were eligible, they were sent an information sheet and consent form for the trial and could consent.

UCL will share trial ID numbers, NHS numbers, names, date of birth, sex and postcode of the enrolled consented participants to allow NHS England to identify the trial participants in the NCRAS data. NHS England will then return the trial ID numbers of data on cancer diagnosis, hospital records, treatment information, health status and mortality for these participants. The patient-level cancer registration data is disseminated from NHS England to UCL’s data safe haven via SEFT (secure electronic file transfer).

UCL will use cancer registration data collected by the National Cancer Registration and Analysis Service (NCRAS; cancer registry in England) to allow them to group the sample population based on the cancer type and stage at diagnosis. The study team will also use the data (on cancer diagnoses and mortality) to compare the experimental groups (those who received the intervention and the control group) to assess if the intervention had a positive impact on these outcomes. The study team are also interested in additional related research questions, for example, exploring whether relationships between health behaviours and survival are moderated by patient variables (like patient reported outcomes of anxiety/depression) to determine if any groups are especially impacted. There is a funded UCL PhD student (who is part of the ASCOT team) who will explore these questions, as well as continuing other planned ASCOT analyses. The disseminated data will be integrated into the trial dataset so that cancer outcomes and survival data become an outcome that can be analysed in relation to various data that the study team collected from participants during the trial.

Data processing will only be carried out by employees of UCL and one enrolled PhD student. All of those carrying out data processing via the safe data haven will complete yearly training on information governance, data protection and confidentiality. The 2nd Principle Investigator (PI) is based from the University of Leeds but will not have access to data disseminated under this agreement. The 2nd PI was involved in the study design plans for analysis and will contribute towards paper writing and will undertake these duties under an honorary contract with UCL.

ASCOT began in 2015 and data collection from participants in the form of the ASCOT patient questionnaires ended in 2021. However, the trial is not considered complete until the study team have received the final NCRAS data from NHS England. UCL plan to examine co-morbidities up to ten years after participants consented. UCL will therefore retain personal identifiers until this time (2028-2029 at the latest). After which UCL will destroy all identifiable data by deleting the database which links patient ID's to patient identifiers, making all the remaining data anonymous to the UCL study team.

All data collected as part of the ASCOT patient questionnaires is pseudonymised using participant ID numbers as soon as the surveys are received. Consented patient identifiable information is stored separately to their ID numbers in a locked filing cabinet in the Department of Epidemiology and Public Health at UCL.

Electronic Data collected as part of the ASCOT patient questionnaire and the ASCOT trial is stored in the data safe haven (a secure environment) by the research team in UCL’s Department of Behavioural Science and Health and accessed remotely by the study team for statistical analysis. The study team can look at the data on the server but are prohibited from moving it to any other machines.

The role of the listed joint data processor VIRTUS data centre is used exclusively for secure server storage purposes only. VIRTUS only supply the physical location for storage. VIRTUS operate 7 layers of physical security on site, including perimeter fencing, access control, CCTV external and internal and restricted pass code access. Staff at VIRTUS data centres will not have access to NHS England data.

The study team will process, store and dispose of patient identifiable information (names and contact details) in accordance with all applicable legal and regulatory requirements, including the Data Protection Act 2018 and any amendments thereto.

All data used in publications and outputs will be completely anonymous using aggregated data with small numbers supressed.

The data on cancer diagnoses disseminated in the first data share will be used to describe the sample in all analyses and subsequent papers, in peer reviewed journals using the trial data. This main trial paper is still in progress.

The data on cancer diagnoses and mortality will be used as part of the trial analysis of the impact of the intervention which will be published as a paper. It will hopefully also be used in further analyses linking variables UCL collected data on in the trial to longer term cancer outcomes and survival. These will all be published as research papers in peer reviewed journals.

A report of the main results will be sent to ASCOT Study participants via a newsletter and shared on the study website once the data has been analysed. As more subsequent studies are published, participants will be signposted to the study website.

These results will be presented at other research institutions, at national and international conferences (e.g. UK Society for Behavioural Medicine, American Society of Clinical Oncology, International Society for Behavioural Nutrition and Physical Activity) and at patient-facing events (e.g. charity hosted), and will be shared with the recruitment sites via email. UCL may seek additional funding to explore the potential to implement their intervention within the NHS over the long-term, which could involve engagement with local and national policymakers.

UCL hope to complete the majority of this dissemination before the end of August 2023 in line with project funding however some research will continue beyond this, in particular the work linking to later cancer events and mortality.

No participants will be identifiable in this data. Data will be aggregated with small numbers supressed such that averages are presented, e.g. number of participants with each type of cancer, average time since diagnosis and the different variables will not be linked.

Breast, prostate and colorectal cancer are 3 of the most common cancers diagnosed each year with approximately 150000 new diagnoses a year in the UK across these three cancer types. This intervention involves a booklet and feedback from a health professional about health behaviours and the potential benefits of making healthy changes to these. If this intervention is shown to benefit cancer patients, it can be incorporated into patient care so that all patients have access to it. This could then improve health behaviours in this group by helping them to move closer to meeting public health recommendations. Such improvements have the potential to improve their cancer outcomes as well as other health conditions. The NHS England data will allow UCL to assess whether the intervention has improved cancer outcomes and survival. If the intervention has resulted in these improvements this could lead to further work focused on how to implement this intervention within the NHS both locally and nationally. The researchers will disseminate the findings widely and discuss with hospitals and funders (including relevant charities such as WCRF, Cancer Research UK and Macmillan). The next steps to work towards providing this intervention as apart of standard care. The aim is that this will happen during 2023.

Data for this study has previously been shared when the data were controlled and managed by Public Health England (PHE). As such there are some yielded benefits to be observed from the access to the data for the study prior to NHS England becoming data controller. These yielded benefits are noted below;

Previously disseminated data has been linked to the trial data so that the participants can be correctly classified based on their cancer diagnosis. This has been used in the analysis for a number of papers which UCL are due to submit to peer reviewed journals in the near future including the main trial results (up to 6 month follow-up) are due to be submitted soon. Preliminary findings suggest the intervention was successful at supporting patients to improve their health behaviours.