Real‐world prevalence of the inclusion criteria for the LEADER trial: Data from a national general practice network

Abstract Aims To explore the prevalence and describe the clinical characteristics of people with type 2 diabetes with a similar cardiovascular (CV) profile to that of the LEADER trial participants in a primary care setting in England. Materials and methods In this cross‐sectional analysis, using the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) network database, we identified people with type 2 diabetes meeting the LEADER inclusion criteria. We identified people's CV risk factors using computerized medical records. Additionally, we assessed the prescription pattern of glucagon‐like peptide‐1 receptor agonists (GLP‐1RAs) in this cohort. Results Of 1 275 461 adults, we identified 84 394 with type 2 diabetes, of whom 14 000 (16.6%) met the LEADER inclusion criteria for established or high‐risk CV disease (RCGP RSC‐CVD group). The LEADER cohort was younger than the RCGP RSC‐CVD group (64.2 vs 73.2 years), had higher mean glycated haemoglobin (71.6 vs 67.1 mmol/mol) and blood pressure (BP) values (systolic BP: 135.9 vs 132.9 mmHg; diastolic BP: 77.2 vs 72.7 mmHg), and a higher mean body mass index (32.5 vs 30.9 kg/m2). In the RCGP RSC‐CVD group, only 1215 people (8.7%) had ever been prescribed a GLP‐1RA and 760 (5.4%) had ever received liraglutide. Conclusions In a cohort of English general practice patients, one in six people with type 2 diabetes met the LEADER inclusion criteria, and less than one in 10 of these received liraglutide, a drug which has demonstrated CV benefits amongst others. There is scope to improve the outlook in people with type 2 diabetes and high CV risk through evidence‐based use of specific GLP‐1RAs.

were designed to assess the safety of antidiabetic treatments compared to placebo with respect to major adverse CV events (MACE), including CV death, stroke and non-fatal myocardial infarction (MI).
To date, 17 randomized CVOTs have been completed, and all have confirmed non-inferiority for the therapies assessed in terms of CV safety when compared with placebo. [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] In addition, five of the 17 completed CVOTs have also confirmed significant reductions in the primary MACE composite endpoint for the sodium-glucose cotransporter-2 (SGLT2) inhibitors empagliflozin and canagliflozin, 8,14 and for the glucagon-like peptide-1 receptor agonists (GLP-1RAs) liraglutide, albiglutide and dulaglutide, 16,18,19 suggesting that these agents have direct cardioprotective properties. A sixth study, the SUS-TAIN 6 trial, reported CV superiority compared to placebo for the GLP-1RA semaglutide in a post hoc analysis. 15 In the LEADER trial, the CV effects of the GLP-1RA liraglutide were assessed in patients at high CV risk, and the results confirmed a significant 13% reduction in MACE and 15% reduction in death from any cause for patients treated with once-daily liraglutide compared with placebo, when both were added to standard care. 16 In addition to randomized clinical trials, there is increasing interest in the use of real-world evidence. 20 Whilst CVOT results have undoubtedly provided valuable information about the CV safety separate from glucose-lowering benefits, the inclusion criteria in a specific CVOT, and therefore its results, may apply to only a small percentage of patients with type 2 diabetes, and consequently may not be generalizable to the wider clinical population.
The aims of the present study were, firstly, to evaluate the extent to which the LEADER trial population can be identified in a real-world cohort by assessing the prevalence of patients with type 2 diabetes in an English general practice setting who possess the same inclusion profile as those recruited to the LEADER trial and, secondly, to assess the pattern of GLP-1RA prescription in this cohort.

| MATERIALS AND METHODS
The study was a cross-sectional analysis of all people with type 2 diabetes included in the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) network database, conducted to quantify the proportion of people with CV disease or at high CV risk meeting the eligibility criteria for the LEADER trial.
The RCGP RSC database is a primary care sentinel network. It uploads computerized medical records twice weekly from over 200 primary care practices distributed across England, producing near real-time reports about influenza, vaccine effectiveness and, more recently, other areas of research. 21 Covering, at the time of the present study, a population of >2 000 000 patients, the RCGP RSC provides a representative sample of the national English population in terms of demographics and clinical outcomes. 22,23 As a registrationbased system, where patients are identified with a unique National Health Service (NHS) number, the RCGP RSC database enables capture of a representative population without double counting.
The full methods of the study protocol have been previously published. 24 Using the RCGP RSC database, people with type 2 diabetes who had a similar CV risk profile to those included in the LEADER trial (RCGP RSC-CVD group) were identified and compared with the liraglutide-treated group from the LEADER trial. Within the RCGP RSC-CVD group, the proportion of patients who had ever been prescribed at least one of the following GLP-1RAs was determined: albiglutide, dulaglutide, exenatide, exenatide extended release, liraglutide and lixisenatide. Data for the once-weekly GLP-1RA semaglutide could not be collected, as the treatment was not commercially available in the United Kingdom at the time of the analysis.

| Data analysis
In this analysis, data were extracted for all patients from the RCGP RSC database for information collected up to December 31, 2016, comprising all patients with type 2 diabetes who were aged ≥18 years on or before this date.
People with type 2 diabetes were identified using a two-step ontology-based process. 25 Firstly, we identified individuals with diabetes using a combination of diagnostic codes, glycated haemoglobin (HbA1c) and blood glucose test results, and antihyperglycaemic therapy usage (except metformin). Subsequently, people with diabetes were categorized by diabetes type using a seven-step algorithm that takes into account medication history, diagnosis codes and other key clinical characteristics.
The RCGP RSC-CVD subgroup was identified from the cohort of people with type 2 diabetes by using clinical codes (Tables S1-S12). These coded data included diagnosis and treatment information, prescriptions and laboratory data. 26 Individuals were considered to fulfill the LEADER criteria if they had any of the disease-defining codes at any time in their clinical record. To compare the type 2 diabetes group with the LEADER inclusion criteria for CV disease or risk, we used the closest matching variables available from routine UK primary care data and, when specific conditions were not recorded in primary care data (due to coding limitations or non-specific data entry), broader criteria were used (Table S13).
We reported the clinical characteristics of the RCGP RSC-CVD group. These included age, gender, duration of diabetes, HbA1c, body mass index (BMI), systolic blood pressure (BP) and diastolic BP. Age was reported as per the end of the study period (December 31, 2016).
Duration of diabetes was based on the first indicator of diabetes in the patient's record up to the end of the study period. HbA1c, BMI, systolic and diastolic BP were taken from the latest patient's record.
Clinical characteristics of the cohorts are described using descriptive statistics (percentages, means and SD values).

| Compliance with ethics guidelines
Approval for use of the data was acquired from the RCGP RSC Study Approval Committee. This study did not require ethical approval as it was considered to be a clinical audit when tested against the Health Research Authority/Medical Research Council tool "Is my study research?". 27 Part of our standard way of working is not to process the data of people who have opt-out codes; these affect 2.25% of our population. 28

| RESULTS
At the time of data extraction, the RCGP RSC population comprised 1 275 461 adults. From this population, we identified 84 394 (6.6%) people with type 2 diabetes. Of those with type 2 diabetes, 14 000 (16.6%) met the LEADER trial inclusion criteria (Table 1) Table 2).
Approximately 80% of people in the RCGP RSC-CVD group had established CV disease and 20% were at high CV risk; there was a similar split between established CV disease and those at high CV risk within the liraglutide-treated LEADER cohort (Table 2). A larger proportion of patients from the liraglutide-treated LEADER cohort had prior MI and coronary heart disease compared with the RCGP RSC-CVD group; however, the same subgroup from the RCGP RSC cohort included a larger proportion of patients with prior cerebrovascular events and chronic kidney disease (estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m 2 ; Table 2).
Within the RCGP RSC-CVD group, 1215 patients (8.7%) had a prescription for a GLP-1RA at some point in their medical records, and 760 (5.4%) had previously been treated with liraglutide.

| DISCUSSION
The present study ascertained the prevalence of people with type 2 diabetes possessing a similar CV profile to those included in the LEADER CVOT, and also determined the pattern of prescribing of GLP-1RAs in this cohort. The analysis showed that one in six people (16.6%) with type 2 diabetes in an English primary setting meets the inclusion criteria for the LEADER trial, and that <10% of the RCGP RSC-CVD group had ever been prescribed a GLP-1RA. only to a small proportion of people with type 2 diabetes (15.7% of the total type 2 diabetes population), and that an even smaller proportion of those who are currently treated with SGLT2 inhibitors have the same high CV risk as that of the EMPA-REG trial population (11.1% of the total type 2 diabetes population), thus calling into question whether this class of drug is also being used to its potential, at least in England. 42 In addition, as the pre-approval CVOT SUSTAIN 6 shares identical inclusion criteria to LEADER, 15 the same RCGP RSC-CVD cohort would be identified as in the present study, and may also benefit from the use of the once-weekly GLP-1RA semaglutide, when this is launched in the United Kingdom.
Key strengths and limitations of the data source used in the present study have been reported previously. 22 With the wider acceptance of the results of CVOTs to guide more personalized or tailored therapy for people with type 2 diabetes, as highlighted in the recent ADA/EASD consensus, 33 there is scope to improve the management of people at high CV risk through targeted use of therapies with a proven CV benefit.

ACKNOWLEDGMENTS
The authors would like to thank patients and practices who are members of the RCGP RSC network, and allow their records to be used for this study. The computerized medical record system vendors EMIS, In Practice and TPP. We would also like to thank Apollo Medical systems, the RCGP and University of Surrey colleagues Filipa Ferreira