MR1407 - Study of Heart and Renal Protection (SHARP) Intrial

The University of Oxford's Clinical Trial Service Unit (CTSU) has extensive experience in developing and running large-scale streamlined randomized trials, many of which have significantly changed clinical practice and majorly influenced national and international guidelines. One of these trials was the Study of Heart and Renal Protection (SHARP).

The SHARP randomized controlled trial took place from 2003-2010. A Data Sharing Agreement is in place with NHS Digital for use of registry data in relation to the SHARP cohort for extended follow-up (reference DARS-NIC-147782-0D7TX)

This Data Sharing Agreement (DARS-NIC-13172-S1S3F) is for an additional SHARP project which is researching the validation of SHARP in-trial adjudicated outcomes versus registry reported outcomes.

The Study of Heart and Renal Protection (SHARP) was the largest randomized controlled trial to be conducted in people with kidney disease to date. It involved almost 9,500 volunteers aged 40 or over with chronic kidney disease recruited from 380 hospitals in 18 countries. Volunteers were randomly allocated to take either daily cholesterol-lowering therapy with a combination tablet containing simvastatin 20mg plus ezetimibe 10mg, or matching dummy "placebo" tablets for an average of 5 years. The SHARP study finished in 2010 and concluded that around a quarter of all heart attacks, strokes, and operations to open blocked arteries could be avoided in people with chronic kidney disease by using the combination of simvastatin and ezetimibe to lower blood cholesterol levels. Further SHARP clinical trial results can be found at: http://www.ctsu.ox.ac.uk/~sharp/ and http://www.sharpinfo.org/.

The SHARP trial was carried out in 18 countries with 1,987 people in the United Kingdom being randomized. It assessed the effect of lowering LDL cholesterol with a combination of simvastatin 20mg plus ezetimibe 10mg versus a matching placebo on serious vascular disease (e.g., heart attacks, strokes) and renal disease (e.g., starting dialysis) events. Participants were followed regularly in study clinics, with all serious adverse events being recorded. Those events which were pre-specified as study outcomes were confirmed by central review of hospital notes by study doctors. This process is referred to as clinical adjudication.

Such outcome measure adjudication is typically seen as the ‘gold standard͛’ for assessment of study outcomes. However, it is very resource intensive and expensive, and took a team of trained clinicians two years to complete for the SHARP trial.

Such increasing cost of research is a contemporary major issue for all researchers conducting trials. If it can be demonstrated that reports from routine healthcare datasets are complete and reliable compared with nurse reported events and/or the final adjudicated outcomes, future trials could be designed which follow-up participants solely through routine healthcare data.

There are examples where routinely collected healthcare data has been compared to study outcomes previously and they have demonstrated that using routinely collected data to record study outcomes can be reliable which could greatly simplify future trial design, and as such benefit patients and the public. However, there are limited such examples in people with renal disease, where complex disease presentations make some researchers doubt the reliability and therefore utility of routinely collected healthcare data. Moreover, people with chronic kidney disease (who have substantially increased morbidity and mortality compared with the general population) are often excluded from large, randomized trials.

Consequently, in renal disease, there is a mismatch between high clinical need and levels of evidence on which to base clinical care. There is, therefore, a particular need to develop novel ways of conducting affordable trials of both old and new treatments in this patient group. An example is the NIHR-funded SIMPLIFIED trial http://www.phpc.cam.ac.uk/pcu/research/researchprojects-list/other-projects/simplified/) which is assessing the effect of high-dose native vitamin D versus standard of care in dialysis patients. The primary outcome is all-cause mortality, however such a trial could become much more informative if the value of routinely collected Hospital Episode Statistics (HES) Admitted Patient Care (APC) data can be confirmed by this proposed study, and so the effect of high dose vitamin D on other non-fatal outcomes such as vascular disease, can be tested. The SHARP trial dataset is in a unique position to validate such methods for both SIMPLIFIED and other future renal trials.

This project seeks to obtain all relevant registry associated outcomes from healthcare data sources at NHS Digital to compare the outcome data collected through nurse-led reporting of adverse events and clinically adjudicated during SHARP with the healthcare information routinely collected by NHS Digital. This is to investigate whether using routinely collected data could be adopted as a more economical alternative to 'clinical adjudication'.

The focus of the study is on non-fatal outcomes (where there is more doubt of validity) derived from registry data including:

• Major cardiac events

• Stroke and its subtypes

• Revascularisation procedures

• End-stage renal disease events

• Admissions with acute kidney injury and

• A range of infections including opportunistic infections unique to immunosuppressed transplant recipients and more common infections

• Heart failure

The primary outcome is all-cause mortality and therefore the SHARP In-Trial study has requested cancer outcome data, and, mortality data including cause of death.

The study involves an evaluation of the specificity and sensitivity of registry data derived outcomes, looking at both the occurrence of events and the time frame in which they occurred. Results of the original trial will be compared against replicated results using registry data derived outcomes.

Efforts were previously made to minimise the data requested which resulted in NHS number, Latest Identifiers and Supplied Identifiers no longer being disseminated. The University of Oxford previously extensively reviewed the Product List and only selected fields believed to be essential to this project, including only the years needed for successful follow-up, and has been limited to a cohort of 1,987 participants from England and Wales. The University of Oxford has taken careful consideration regarding data minimisation efforts and confirm there is no alternative, less intrusive way of achieving the purposes outlined in this application.

The legal basis for processing and storing data under this Agreement is Article 6(1)e (UK GDPR), i.e. it is a task carried out in the public interest. The aim of the study is to provide reliable evidence about whether routinely collected healthcare data is as robust as clinical adjudication in the context of clinical trials, which could result in considerable efficiency savings, and hence facilitate further clinical trials being performed, and hence is in the public interest.

In addition, processing and storage of special category personal data is being done under Article 9(2)(j) (UK GDPR) exemption, i.e. that the processing of the data is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes. The scientific research aim of the study is to provide reliable evidence about the use of routinely collected registry data in clinical trials.

The Study of Heart and Renal protection was supported under section 251 of the NHS Act 2006 to enable the common law duty of confidentiality to be temporarily lifted so that confidential patient information could be transferred to NHS Digital to make the original linkage and also used to flow the identifiers to the applicant without the discloser (in this case NHS Digital) being in breach of the common law duty of confidentiality. Section 251 of the NHS Act 2006 also supports the study to obtain all relevant registry associated outcomes from healthcare data sources, such as NHS Digital and similar organisations, plus other disease specific registries (e.g., the UK Renal Registry), for all UK-based SHARP trial participants from the time of randomisation.

The University of Oxford is the sole Data Controller and also processes data for the purposes described in this Agreement. All processing of NHS Digital data will take place within the University of Oxford.

Funding for this study is provided internally from the University of Oxford.

The number of years requested was to match the follow-up period in SHARP. This will allow the study to have a complete record of HES APC data for UK participants from their date of randomisation to final follow-up.

All geographical areas in England (for HES APC) and England/Wales (for death) were requested in order to identify all clinical events from SHARP participants living in these countries.

No new data will be provided to or by NHS Digital under this Agreement.

The original SHARP clinical trial cohort was flagged with NHS Digital for the purpose of post-trial follow-up under DARS-NIC-147782-0D7TX in 2012. NHS Digital therefore already held the cohort data and after linkage was carried out only the study ID was supplied.

In 2017, the flagged cohort still being held by NHS Digital were extracted and, under the Data Sharing Agreement DARS-NIC-13172-S1S3F (this agreement), used to provide a current status report (i.e. the cohort was not flagged again; a one-off report was produced) which was also linked with HES APC data for the purpose of the SHARP in-trial follow-up.

The data received from NHS Digital is held in an encrypted repository on the servers in the University of Oxford. For the purposes of the research stated in this Data Sharing Agreement, researchers will use such record-level data to derive patient-level and clinical-event-level datasets. Such derived datasets can then be compared to the existing SHARP trial analysis database (i.e. clinical data on clinical events, laboratory results, and study treatment) by using the unique study Patient ID.

Other than the linkages already described within this Agreement, researchers will not link NHS Digital data to other datasets.

No data will be shared with third parties and no attempt will be made to re-identify participants.

Nuffield Department of Population Health (NDPH) researchers (all substantive employees of the University of Oxford) are experienced in handling confidential and participant sensitive data and have appropriate training in information governance. The NDPH servers are protected against unauthorised external access by an appropriate strength firewall.

Access to patient identifiable information is protected by the appropriate authentication procedures (user IDs and passwords). Authentication is only given to personnel with a need to access the required data. Only personnel involved in this project (processing and analysing data) will have access to this data. NDPH has a Corporate Level Security Policy that has been fully adopted by management and will apply fully to this study.

All information is stored securely by the University of Oxford and is kept confidential. Access to the computer database is by unique combinations of usernames and passwords and only authorised study personnel can access information about participants. The building is secure with authorised swipe card access only. No individuals will be identified in any study reports.

All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by ‘Personnel’ (as defined within the Data Sharing Framework Contract i.e. employees, agents and contractors of the Data Recipient who may have access to that data).

The main output will be the results of the analyses. The publicly available outputs will be of aggregate data.

Communication and dissemination of the results will be planned to both enable engagement with the scientific and policymaking communities, and to ensure that knowledge developed by the research can benefit these communities. In particular, given the nature of the project, the results will be targeted to active partners in the clinical trials community, such as the Medical Research Council Hub for Trial Methodology Research and UK Kidney Research Consortium Clinical Trial Network.

Specifically, it is planned for the results to be presented at relevant international scientific meetings, such as the international Clinical Trials Methodology Conference and in peer reviewed journals, e.g., Clinical Trials, CJASN (Clinical Journal of the American Society of Nephrology) or NDT (Nephrology Dialysis Transplantation). In addition to publication in a scientific journal and presentation at meetings, communication channels will include the results of these analyses being posted on the SHARP website (http://www.sharpinfo.org/) in a similar manner to multiple previous SHARP-related publications.

The communication and dissemination activities will target not only an audience of researchers and scientists but also research participants (for example through the NDPH Public Advisory Panel: https://www.ndph.ox.ac.uk/research/participant-panel.

The target time frame for publication of the results of analyses is 2023. Previously stated target dates/deliverables have been delayed due to the Covid pandemic and prolonged serious staff illness.

This project's outputs is hoped to achieve the stated purposes (and thus the benefits of processing) since dissemination of its results are anticipated to clarify whether routinely collected healthcare data is as robust as clinical adjudication in the context of clinical trials.

This will be important and is anticipated to be hugely beneficial because clinical trials are a pivotal part of evidence-based medicine, but their rising costs are problematic and hampering research. Therefore, if this project’s results indicate that routinely collected healthcare data is as robust as clinical adjudication, the results will be presented to clinical trial regulators and policy/decision makers with the aim being to action a change in practice as to how clinical trials are conducted in the future. If the result of the project outputs is accepted by such groups as warranting such change, this would potentially result in very substantial efficiency savings for future trials as:

1, Collection of trial events and trial event adjudication (which requires training and employment of skilled staff) would be greatly streamlined

2. The need for trial participants to attend study clinics (which can be onerous and expensive if travel costs are not reimbursed) would likely be reduced

Given the huge number of clinical trials being conducted in multiple therapeutic areas in the UK and other countries, this could potentially affect multiple research projects and impact upon the ease of clinical trial participation for thousands (if not hundreds of thousands) of clinical trial participants. This in turn would be expected to facilitate participant recruitment into clinical trials, helping to ensure that (i) a wide range of participant types is included, meaning that trial results will be more generalisable; and (ii) that clinical trials are adequately statistically powered to address the questions they intend to answer. In addition, over the subsequent decade, the consequent anticipated cost savings of any such change in practice would be estimated to be in the order of millions of pounds. This in turn will enable such future research to be conducted on a greatly reduced budget, which is vital given the limited funding that national and charity funding bodies can typically offer. This may be particularly important for trials of generic drugs in common conditions (e.g., aspirin in cancer prevention) which do not currently attract industry funding. Such methodological research is becoming increasingly important to the efficiency, design and data collection strategies of future trials and studies, and hence will be of benefit to public health at home and abroad.

In summary, dissemination of the expected measurable outputs are expected to benefit the provision of health care or adult social care or the promotion of health by:

1. Reduced costs, enabling more research as a whole to be conducted for a given overall research budget, with consequent benefits for the provision of health care or adult social care or the promotion of health

2. Greater efficiency leading to increased throughput of research and associated enhancement of evidence-based medicine, with impact upon national and international clinical guidelines

3. Increased awareness of the potential for routinely collected data to augment existing understanding and knowledge of a therapeutic area

In summary, dissemination of the expected measurable outputs is in the public and patient interest because this could result in:

1. A reduction in the demands upon trial participants in terms of time and inconvenience in attending study visits

2. An increased in shared knowledge which will be used to inform provision of healthcare decisions leading to improved quality of patient care

In addition, this project will support a DPhil application for a part-time DPhil student in the University of Oxford, and hence is instrumental in educating and developing the skillset of a future generation of clinical researchers.

Due to the main researcher working on this project having required periods of extended sick leave, this project is still ongoing, so no yielded benefits to date.