EMPA-KIDNEY (The Study of Heart and Kidney Protection With Empagliflozin)

The Study of Heart & Kidney Protection with Empagliflozin (EMPA-KIDNEY) randomized trial is comparing empagliflozin 10mg once daily versus matching placebo in 6609 participants with established chronic kidney disease (CKD) with or without diagnosed diabetes mellitus. The primary aim of the study is to investigate the effect of empagliflozin (a drug belonging to the class of medication called SGLT-2 (sodium glucose co-transport 2) inhibitors which reduce blood sugar levels in people with diabetes) on kidney disease progression or cardiovascular death versus placebo on top of standard of care in patients with pre-existing chronic kidney disease. The main inclusion criterion is participants with chronic kidney disease (CKD) who are at risk of their condition progressing which is defined as blood results showing an estimated glomerular filtration rate (eGFR) of 20-45, or 45-90 mL/min/1.73m2 with urinary albumin:creatinine ratio ≥200 mg/g of albuminuria (or protein:creatinine ratio ≥300mg/g).

During the within-trial period, EMPA-KIDNEY participants are seen in dedicated study clinics at Screening, Randomization (2 to 3 months after screening) and Follow-up Visits (at 2 and 6 months after randomization, and then 6 monthly). Global randomization began on 19th May 2019 and was completed on 16th April 2021 with a total of 6609 participants randomized at sites in the UK, Germany, Italy, Canada, USA, China, Malaysia and Japan. 1133 participants have been recruited from the UK from 53 sites. Participants will be provided with study treatment and followed up as part of this within-trial phase until mid-late 2022.

All UK participants were asked at their Screening visit to provide consent for information about their health to be collected from NHS datasets and registries. Such consent was necessary to enter the trial. The EMPA-KIDNEY study team at the University of Oxford are seeking NHS Digital data to acquire details on deaths (fact of death and date of death) for all UK participants during the within-trial phase. This will enable the EMPA-KIDNEY study team to identify any deaths not reported during this period due to disrupted follow-up (e.g. due to COVID-19 disrupting study visits). Such data will therefore minimize loss to follow-up for UK participants and consequently increase the quality of the trial data.

All assessments will involve intention-to-treat comparisons among all randomized participants of the effects of allocation to empagliflozin versus placebo on the study’s primary outcome of kidney disease progression or cardiovascular death.

The primary research objective will be the time to the first occurrence of:

1. Kidney disease progression - defined as end-stage kidney disease (ESKD; i.e. starting long-term dialysis or receiving a kidney transplant), a sustained decline in eGFR to <10 mL/min/1.73m2, renal death, or a sustained decline of ≥40% in eGFR from randomization; or

2. Cardiovascular death.

Secondary research objectives are:

1. Time to first occurrence of kidney disease progression; and

2. Time to first occurrence of ESKD; and

3. Time to cardiovascular death or ESKD.

Tertiary research objectives are:

1. Time to death from any cause

2. Time to death from cardiovascular causes

3. Time to death from non-cardiovascular causes.

DATA LINKAGE

To achieve these objectives, in particular identifying occurrence of death, the EMPA-KIDNEY study team will need to link study participants to death records i.e. from the Demographics dataset.

The EMPA-KIDNEY study team request linkage of individual participants by providing their personal identifiable information (Study ID, NHS number, date of birth, postcode, full name, sex) to NHS Digital, along with a unique pseudonymised ID number. NHS Digital will flow data back to the Nuffield Department of Population Health at the University of Oxford, via the secure electronic transfer system (SEFT). The flow of data will include patient identifiable information (NHS Number and Date of Birth) for the purposes of linkage validation. The University of Oxford do not perceive there to be an alternative, less intrusive ways of achieving the purpose.

The data subjects will all be UK participants of the EMPA-KIDNEY randomized controlled trial. The EMPA-KIDNEY study team will obtain data for each participant from the time the first participant was consented and screened into the trial on 1st February 2019 until the date of linkage. This will allow the EMPA-KIDNEY study team to identify unreported deaths and provide this information to clinical sites. The sites will then submit updated information via the usual trial processes (i.e. the web-based data entry system). A further data release will allow the EMPA-KIDNEY study team to ensure that all deaths have been captured for all sites.

The EMPA-KIDNEY study team request linkage only to the dataset that contains informal and formal dates of death, for only the English and Welsh participants of the study. The data will be minimised to the cohort recruited during the period of 1st February 2019 to 11th July 2022. Active recruitment (i.e., randomisation) ended on 16-Apr-21, and the final follow-up visits are now being conducted. The EMPA-KIDNEY study team would like to receive data as soon as possible from May 2022. This will allow them to record any missing deaths. A second drop of data will be received in July 2022 to identify any remaining deaths up to the very end of the study.

Both formal and informal date of death are being requested in order to help corroborate all deaths that occur in the cohort. Although only data for deceased members of the cohort is required, a filter to removed living participants cannot be applied due to the automation of the Demographics product. Demographic data for the whole cohort will therefore be disseminated. Filters can not be applied to the Demographic data set to minimise data from 1st February 2019, however the University of Oxford is only requesting details of deaths for participants who consented after that date.

In the future, the EMPA-KIDNEY study team also plan to link participants to equivalent data about deaths received from the Scottish NHS Central Register (NHSCR). University of Oxford will obtain the appropriate permissions before this linkage occurs.

There will be no linkage to other studies or datasets beyond those stated in this Agreement.

A Public and Patient Involvement (PPI) panel was conducted prior to the start of EMPA-KIDNEY. Participants from the UK HARP-III trial (a University of Oxford renal trial that recruited participants with chronic kidney disease (CKD)) were asked to participate and they agreed that the use of data in the way that the EMPA-KIDNEY study team proposed for EMPA-KIDNEY was reasonable. The PPI panel were also asked about linkage of participants to central registries, and this was considered acceptable. There are also plans to have future PPI panel involvement with EMPA-KIDNEY.

The EMPA-KIDNEY trial is overseen by a Steering Committee chaired by members of the University of Oxford, and composed of academics from the contributing regions and employees of Boehringer Ingelheim International GmbH (BI). They are not a Data Controller under this Agreement. The Steering Committee has an advisory and oversight capacity. There are currently 26 voting members of the committee, including 4 from BI. The Steering Committee has the responsibility of finalizing the pre-specified Data Analysis Plans before any unblinding, and drafting, review and approval of the study main publication(s). It is also responsible for the review and approval of proposals for subsequent analyses and publications. However, it is not involved in the detailed conduct of the trial within each country, such as this linkage activity. There have been no steering committee decisions around the use of NHS Digital data and there are no further steering committee meetings planned before the database is locked. This means there will be no situation where BI could vote about the use of this data.

Regarding publications, BI have the contractual right to submit comments to the author(s). The author(s) of publications using unblinded data shall be free to consider such comments at their discretion. There is no right for BI to suppress any publication. The main trial publication Data Analysis Plan (DAP) has been finalized and will be published in due course. The DAP has been approved by the Steering Committee within which there is an overwhelming majority of non-BI members.

Other than the fact of employment by BI, the BI members of the Steering Committee have declared no other Conflict of Interest.

The Steering Committee determine the strategy of the trial and oversee how it is delivered. They have no role in the delivery of the trial, or the methods used to collect information. They had no role in the design of the case report form which collects the vast majority (>99.9%) of the data in this trial.

If the trial results are positive (i.e., participants in the empagliflozin arm have less occurrence of illness & death than those in the placebo arm), and the drug is successfully relabelled and remarketed, it is likely that BI will benefit financially from this.

The EMPA-KIDNEY Data Monitoring Committee (DMC) will advise the Steering Committee if clear evidence of benefit or harm emerges. Unless advised by the DMC, the Steering Committee, collaborators, participants, representatives of the Boehringer Ingelheim, and all study staff will remain blind to results until the end of the study. The DMC is independent of the University of Oxford and Boehringer Ingelheim.

The legal basis for processing and storing data under this Agreement is Article 6(1)(e) (GDPR), i.e., it is a task carried out in the public interest. In addition, processing and storage of special category (sensitive) personal data is being done under Article 9(2)(j) (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. This processing is in the public interest as the results from the trial may potentially inform doctors’ decisions about the management of their patients, and could lead to substantial improvements in the health of people with kidney disease.

The research team have taken the opinion of the South Central – Oxford C Research Ethics Committee, who have granted approval to the study in its current form. Approval has also been granted by the Health Research Authority (HRA), and Health and Care Research Wales (HCRW). Furthermore, data will be limited by data pseudonymisation, data minimisation (by requesting only the data fields that are relevant to the research for the EMPA-KIDNEY cohort only), and technical and organisational safeguards. Analysis will occur in a safe, NHS Data Security Protection (DSP) Toolkit compliant environment and identifiers will not be included in the analysis. All of the data, including identifiers, received from NHSD under this Agreement will be destroyed once any identified deaths have been confirmed.

CONTROLLERSHIP AND FUNDING

The University of Oxford is the sole Data Controller under this Agreement, who also process the data. No other organisations have access to or process the NHS Digital data disseminated under this Agreement. All data processing will take place within the University of Oxford, therefore University of Oxford is listed as sole Data Processor under this Agreement.

The EMPA-KIDNEY trial was envisaged by renal clinician trialists at the University of Oxford and its design was proposed to multiple pharmaceutical companies as the costs of such trials are prohibitive to public funders. BI took up the opportunity to collaborate and reviewed the University of Oxford drafted protocol, but the University of Oxford alone is conducting the trial. The trial is considered a non-commercial trial from the perspective of the National Institute for Health Research (NIHR) Research Portfolio and received support for the Clinical Research Network (CRN). If BI were not available to collaborate on this trial, NDPH would have sought collaboration with a different manufacturer of an SGLT-2inhibitors.

As described within the Protocol, the University of Oxford has responsibility for the conduct, analysis and reporting of the trial.

By working with a commercial sponsor, the EMPA-KIDNEY trial will be able to provide results for this important research more quickly and at less cost than conducting the trial solely within the University of Oxford. Additionally, if the University of Oxford want a positive trial to impact on clinical care, then the results of the trial will need to be submitted to regulators for licence changes. This requires commercial support.

BI requested to be regulatory sponsor of EMPA-KIDNEY and therefore have oversight of the trial (as required by ICH-Good Clinical Practice). BI have provided a supply of study treatment and funding for the trial. BI own the Intellectual Property surrounding empagliflozin, but the EMPA-KIDNEY data is jointly controlled and owned by the University of Oxford and BI (except for the death data provided by NHS Digital that is being requested in this application, for which the University of Oxford will be sole Data Controller).

The study funder is Boehringer Ingelheim International GmbH (BI), who are also taking on the responsibility as regulatory sponsor. For the purposes of this NHS Digital Agreement, BI is not considered a Data Controller (i.e., BI will have no control over what data the EMPA-KIDNEY study team request from NHS Digital or how it is processed, and they will not receive any NHS Digital data). The concept of using linkage data was the University of Oxford’s. The University of Oxford has several decades of experience of using such data to ensure completeness of follow-up in trials. BI do not have this expertise and are supportive (but not directive) of the University of Oxford’s approach.

The University of Oxford also receive support from the UK Medical Research Council and British Heart Foundation (BHF) to run trials, by providing core funding to the Nuffield Department of Population Health (NDPH) at the University of Oxford.

In other aspects of the trial, BI are involved. As described in the Participant Information Leaflet, Consent Form and Privacy Notice, BI will receive a copy of pseudonymised non-NHS Digital data on completion of the within-trial period. However, the data being requested from NHS Digital in this Agreement (or data derived from it) will not be held in this database. No NHS Digital data will be sent to BI. In the event that an unreported death is identified, the site responsible for the participant will be prompted to check their records (which include access to the NHS Spine) and to report the death into the study database via the web-based system, as they would do routinely for deaths identified through other means.

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 list of identifiers that the EMPA-KIDNEY study team at the University of Oxford will send to NHS Digital are as follows:

• Study ID (pseudonymised)

• NHS Number

• Date of birth

• Postcode

• Surname

• Forename

• Sex

The data product the EMPA-KIDNEY study team have requested are as follows:

• Demographics (Study ID, Fact of death, Formal Date of Death, Informal Date of Death, NHS Number, Date of birth)

The data being requested includes historic data starting from 1st February 2019 to present to be received by the EMPA-KIDNEY study team as soon as possible, and then one additional drop of the latest data to be received on 11th July 2022.

NHS Digital will return the study ID and the following identifiers to the EMPA-KIDNEY study team for the purposes of validating linkage. This enables the study team to ensure that the linkage process is robust and accurate. These identifiers will not be used for analysis:

• NHS Number

• Date of Birth

The University of Oxford will retain the data until sites have confirmed any deaths that the EMPA-KIDNEY study team report to them. Identifiers received from NHSD will be destroyed after validation of deaths. NDPH will not be destroying the identifiers held directly from participants.

The data will be transferred from NHS Digital to the Data Controller’s (NDPH, University of Oxford) NHS DSP Toolkit compliant environment via the secure electronic transfer system (SEFT). All data will be transferred, handled and processed in agreement with the NHS Digital Data Sharing Framework Contract, and will be subject to Fair Processing requirements. No NHS Digital data will be transferred to BI.

The data received from NHS Digital will be used as follows:

• Information on deaths will be cross-checked with the trial database. If ‘new’ deaths are identified (i.e., deaths not previously reported by the site), the fact and date of death will be reported to sites to assist them with identifying deceased participants and find medical records relating to their death held locally.

• NHS Digital death data will not be entered into the trial database or used for analysis. In the event that an unreported death is identified, the site responsible for the participant will be prompted to check their records (which include access to the NHS Spine) and to report the death into the study database via the web-based system as they would do routinely for deaths reported in other ways (e.g., via a relative or other healthcare provider).

The EMPA-KIDNEY study team request linkage only to the dataset that contains relevant information on deaths, minimised to the cohort recruited during the period of 1st February 2019 to 11th July 2022. Filters can not be applied to the Demographic data set to minimise data from 1st February 2019, however the University of Oxford is only requesting details of deaths for participants who consented after that date. No other data fields are required for participants who are still alive.

All processing of data will be performed within the Nuffield Department of Population Health at the University of Oxford within an NHS DSP Toolkit compliant environment. No identifiable data will be shared other than with associated researchers working on this project, all of whom are substantive employees of the Nuffield Department of Population Health at the University of Oxford.

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

Researchers will not link the NHS Digital death data to other datasets.

As part of the consent form, participants explicitly agree to the collection, storage, processing, transfer and use of their personal data as explained in the EMPA-KIDNEY Participant Information Leaflet.

NDPH researchers 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.

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. All authorised study personnel are substantive employees of the University of Oxford. The building is secure with authorised swipe card access only. There will be no attempts made to identify participants in any study reports.

EMPA-KIDNEY participants have consented for their personal data (i.e., their name, address, date of birth and medical information) to be accessed by the EMPA-KIDNEY study team, and for these details to be stored securely within the Nuffield Department of Population Health at the University of Oxford.

As detailed in the study documentation, participants give consent for their data to be shared with other parties including central registries, BI, and regulatory authorities. This data sharing will not include the NHS Digital data being requested in this Agreement.

The results of this research would be presented at relevant scientific meetings (e.g., the World Congress in Nephrology), in peer-reviewed journals, and as such should influence clinical practice. The main results paper will be submitted to the world’s leading medical journals (e.g., the Lancet and the New England Journal of Medicine (NEJM)). Lay summaries of important results will be provided on www.empakidney.org following appropriate review by ethics committees (where relevant). The REC committee includes lay members. If time allows (there are very short timelines between getting results and needing to distribute these), the PPI group will be involved in reviewing these.

The NDPH contributes widely to health policy, particularly in the area of vascular risk prevention. It contributes to debate with academic papers, conference participation, lectures to the public and advice to government (including NHS Digital).

The EMPA-KIDNEY study team will share outputs via the following listed channels:

- Study website https://www.empakidney.org/

- Open lectures and talks

- Posters

- Press/media engagement and other public promotion of the research (e.g. via the Nuffield Department of Population Health website (https://www.ndph.ox.ac.uk/), or Twitter account (@oxford_ndph).

The University of Oxford aims to issue the first main publication of results by end 2022.

Participants are kept informed about the study via newsletters, Participant Information Leaflets and the study website https://www.empakidney.org/.

All outputs will only contain results in highly aggregated format and as statistical summaries and measures of association. Reports will be in the form of aggregated outputs with small numbers suppressed in line with the HES Analysis Guide.

Demographic data requested in this Agreement will provide details on deaths (fact of death and date of death) for all UK participants during the within-trial phase and will enable the EMPA-KIDNEY study team to identify any deaths not reported during this period due to disrupted follow-up (e.g. due to COVID-19 disrupting study visits). It is hoped that such data will therefore minimize loss to follow-up for UK participants and consequently increase the quality of the trial data.

This will allow the EMPA-KIDNEY study team to identify unreported deaths and provide this information to clinical sites. The sites will then submit updated information via the usual trial processes (i.e. the web-based data entry system). A further data release will allow the EMPA-KIDNEY study team to ensure that all deaths have been captured for all sites.

This provision of the death data may also mean that the study team does not make contact with and cause any distress to relatives.

Future expected benefits from the EMPA-KIDNEY study are described below:

Chronic kidney disease (CKD) is common. In high-income countries, the prevalence of CKD is about 10% and is likely to increase as average population age rises and diabetes mellitus becomes more prevalent. Cardiovascular risk increases progressively as kidney function declines. There is evidence that lowering low-density lipoprotein cholesterol and blood pressure in people with CKD reduces cardiovascular risk, but substantial residual risk remains, and no other treatments have been shown to reduce cardiovascular risk in this group of patients.

The specific expected benefits of conducting this research are as follows:

1. Impact on kidney patients

If the trial results are positive (i.e., participants in the empagliflozin arm have less occurrence of illness & death than those in the placebo arm), and the drug is successfully relabelled and remarketed, this may translate to improved outcomes for kidney patients worldwide. University of Oxford’s contract with BI prohibits them from preventing any publication that the Steering Committee has approved, even if trial results are negative.

2. Reducing healthcare costs for treating kidney patients

CKD has major effects on the health of those diagnosed, and in its most advanced stages incurs major costs to the NHS. This is because, although only 0.1-0.2% of the UK population receive renal replacement therapy, regular dialysis costs the NHS £30,000 per patient per year. When combined with the costs of managing earlier stage of CKD and its complications, the disease was estimated to cost £1.4 billion each year (2009-2010 estimate: https://pubmed.ncbi.nlm.nih.gov/22815543/). Delaying the progression of CKD could result in significant savings for healthcare providers such as the NHS.

3. Global impact on healthcare for kidney patients

Should results provide important new information, the EMPA-KIDNEY study team hopes that data from the trial will be used to provide evidence to submit to regulators for an extension of the current licenced indication in the UK and globally. Empagliflozin is already available for use, and so the EMPA-KIDNEY study team hopes that results will also provide information for clinicians and guideline groups globally who oversee the care of people with kidney disease (irrespective of any effects on regulatory labelling).

Having complete data for a major clinical trial such as EMPA-KIDNEY is critical to the reliability of data due to public health impact of widespread use of SGLT2i (e.g., empagliflozin). NHS Digital data can contribute to this significantly.

For participants, involvement of BI means that those with CKD may ultimately have access to new drug treatments that could improve their health, or avoid worsening disease and death.

Participants are kept informed about the study via newsletters, Participant Information Leaflets and the study website https://www.empakidney.org/. This would include updates to findings and benefits.

This is a new Data Sharing Agreement. No data has yet been disseminated and there are as yet no yielded benefits.