Effects of Liraglutide in Young Adults With Type 2 DIAbetes (LYDIA) (LYDIA)

Impact of Liraglutide on Cardiac Function and Structure in Young Adults With Type 2 Diabetes: an Open-label, Randomised Active-comparator Trial

There are recent advances in therapies for the treatment of Type 2 Diabetes Mellitus (T2DM) which include the GLP1 analogues and the DPP IV inhibitors. Both of these therapies target the incretin system using different methods to elevate/maintain circulating levels of GLP1 to subsequently achieve improved blood sugar control. Interestingly, GLP1 analogues have been reported not only to improve blood sugar control but to additionally induce weight-loss and emerging experimental evidence has shown it may have beneficial effects on the heart's structure and function. Due to the profile of this condition being a lot worse and younger patients having greater CVD risk, a therapy offering multiple positive effects, in particular the potential cardiometabolic effects, make this line of therapy attractive in this patient population. The aim of this research is to investigate the cardiometabolic effects of Liraglutide (GLP1 analogue) compared to that of its clinically relevant comparator Sitagliptin (DPP IV inhibitor).

T2DM in the young T2DM has traditionally been associated with older age, with type 1 diabetes (T1DM) being the dominant form in younger populations. This traditional profile has dramatically altered over the last couple of decades; the sharp rise in levels of obesity and sedentary lifestyles witnessed in younger age groups in developed countries has resulted in up to a 10 fold increase in the prevalence of T2DM in younger adults and youth [7]. Whereas T2DM was once a rarity in those under 40 years, we have recently shown that T2DM now represents up to 20% of all registered diabetes cases across centres in Leicester and Sheffield in this age group [8]. More worryingly, sub-analysis of the Leicester arm of the international ADDITION study [9], a large population-based screening study that included a small cohort under 40 years of age (n = 445), and an on-going NPRI-MRC study in younger high risk adults, suggest the prevalence of undiagnosed T2DM in those under 40 years is between 2 to 4% depending on the diagnostic criteria used and over 5% in those with a family history or obesity (unreported observations). The focus of health care policy and research has lagged behind this substantive shift in the profile of T2DM towards younger populations. For example, the NHS health checks programme is targeted at those over 40 years of age while previous self-management, lifestyle and pharmaceutical interventions which have been used to inform NICE guidance have been conducted in groups aged between 50 and 70 years. Clinical Burden of T2DM in the young The onset of T2DM in younger adults and youth represents an extreme phenotype that magnifies the disease profile observed in older adults. For example, those diagnosed with T2DM under 40 years are more likely to be classified with Class II or Class III obesity (≥35 kg/m2), have a multigenerational family history of T2DM, lead a sedentary lifestyle and be of minority ethnic origin [7]. Of particular concern is that the onset of T2DM in younger adults is associated with dramatic elevations in the risk of cardiovascular disease, particularly coronary heart disease. For example, the risk of any macrovascular complication in younger adults with T2DM compared to controls has been shown to be twice as high compared to other T2DM patients (HR 7.9 vs. 3.8, respectively)[10]. Myocardial infarction (MI) was found to be the most common macrovascular outcome; the hazard of developing an MI in younger adults with T2DM was 14-fold higher than in control subjects [10]. In contrast, older adults with T2DM had less than four times the risk of developing an MI compared with control subjects. Others have reported that the mortality rate in young people with T2DM was as high as 9% over a 9 year period [11]. Younger people with T2DM therefore have higher complication rates than their peers with Type 1 diabetes, despite a shorter duration of diabetes, highlighting the aggressive nature of the disease. Our group has recently completed a Medical Research Council (MRC) funded study aimed at elucidating the biochemical and cardiac abnormalities associated with T2DM in younger adults (18 to 40 years). The study recruited 20 young adults (18 to 40 years) with diagnosed T2DM and 20 age-matched metabolically healthy obese and lean controls (paper under preparation). Cardiac structure and function were assessed by state-of-the-art tagged cardiac MRI imaging. The most striking finding from this study was evidence of greater diastolic dysfunction in T2DM compared to both the obese and lean controls. Specifically, peak end diastolic strain rate, a highly sensitive measure of left ventricular (LV) diastolic function, had an average value of 1.27 s-1 in the T2DM cohort (the lower limit of normal ranges has been defined as 1.3 s-1) which was 20% lower than the obese controls and 30% lower than the lean controls. This finding is consistent with a larger study that used echocardiography to assess cardiac function and structure in over 150 adolescents and younger adults with T2DM compared to obese and lean controls; this study reported that diastolic dysfunction was reduced from lean to obese and from obese to T2DM. These studies therefore suggest that pre-clinical diastolic dysfunction is already manifest in younger adults with T2DM, despite their relatively young age and short duration of type 2 diabetes. This finding is highly clinically relevant and suggests that the high risk of cardiovascular disease observed in T2DM is predominantly characterised by diastolic dysfunction predisposing these patients to heart failure, in advance of overt systolic compromise [12,13]. For example, 50% of all cases of chronic heart failure have preserved ejection fraction [12,13]. Therefore therapeutic strategies that target both glycaemic control and diastolic cardiac function would be highly desirable in younger adults with T2DM. GLP-1 THERAPY Traditional therapies in the management of T2DM have focused on enhancing the secretion, action and availability of insulin. However, whilst these approaches have been successful in managing blood glucose levels, they have had only modest benefit in reducing rates of myocardial infarction and are associated with a number of deleterious side effects, including hypoglycaemia, bone fractures, congestive heart failure, weight gain, and, in some analyses, increased mortality [14]. Therefore the development of new therapies simultaneously targeting hyperglycaemia and cardiovascular disease is a major priority. Two recently approved classes of incretin based therapies, glucagon like peptide -1 (GLP-1) analogues and dipeptidyl peptidase-4 (DPP-4) inhibitors, have shown promise in these areas. Both classes exert their actions through potentiation of incretin receptor signalling, particularly GLP-1. In vivo GLP-1 is predominately secreted from the L-cells of the distal jejunum, ileum and colon. Low basal levels are secreted in the fasting state, however, levels rapidly and transiently increase in response to the ingestion of food; secretagogues include the major macronutrients, particularly glucose. Endogenous GLP-1 has a short half-life of only 1 - 2 minutes following rapid degradation by the enzyme DPP-4 [15]. GLP-1 receptors are widespread throughout the body including in the pancreas, stomach lining, intestine, brain and heart. The widespread distribution of GLP-1 receptors supports a diversity of pleiotropic effects. Other than the direct effect on appetite regulation, the most widely investigated action of GLP-1 has been around the homeostasis of glucose levels. GLP-1 is known to stimulate insulin secretion in a glucose dependant manor (thus limiting risk of hypoglycaemia), as well as promoting glucose stimulated insulin gene transcription and biosynthesis [14]. It may also have trophic effects on pancreatic beta-cells. Given these glucose dependant effects, GLP-1 therapies have been developed as treatment strategies in the management of T2DM. Due to the short half-life of native GLP-1, therapies have revolved around inhibiting the action of DPP-4 and thus augmenting naturally occurring levels of the incretin hormones, or through the intravenous administration of GLP-1 analogues which are resistant to the action of DPP-4. In contrast to DPP-4 inhibitors, GLP-1 analogues can be administered at supra-physiological levels and thus lead to more profound receptor activation and biological effects. This is particularly true in relation to weight-loss, where GLP-1 analogues have been shown to directly induce substantial weight loss with greater effect seen with higher levels of obesity. In contrast, DPP-4 inhibitors appear weight neutral. Both DPP-4 inhibitors, such as Sitagliptin Vildagliptin and Linagliptin and GLP-1 such as Liraglutide and Exenatide are licensed for use in the management of T2DM within the United Kingdom. Liraglutide is the most recently approved GLP-1 analogue therapy and has the greatest promise in terms of glycaemic efficacy and weight loss. The Liraglutide Effect and Action in Diabetes (LEAD) studies have demonstrated that when used in combination with one or more OADs, doses of 1.2 or 1.8 mg of Liraglutide daily significantly reduce HbA1c (mean HbA1c decrease: 1.0-1.5%) [16-22]. Clinically important decreases in fasting plasma glucose (FPG) and postprandial plasma glucose (PPG) levels have also been well documented following Liraglutide treatment (mean FPG decrease: -1.6 to 2.2 mmol/l and 1.55 to 2.4 mmol/l with 1.2 and 1.8 mg Liraglutide, respectively [16-23]; mean PPG decrease: 2.3-2.6 and 1.81-2.7 mmol/l with 1.2 and 1.8 mg Liraglutide, respectively [16-19]). Liraglutide has also proven to have superior effects on glycaemic control compared to other licensed GLP-1 analogues and DPP-4 inhibitors [20,21]. In addition to improved glycaemic control, Liraglutide is also effective at inducing weight loss and reducing blood pressure. Mean body weight reductions of at least 1 kg were apparent with both doses of Liraglutide (-1.0 to -2.9 kg with 1.2 mg; -1.8 to -3.4 kg with 1.8 mg) [17-21]. Greater decreases in body weight were exhibited by patients with a higher adiposity [23]. Furthermore, when used in daily doses of 2.4 mg or more, Liraglutide has been shown to have a greater effect on body weight than Orlistat [24]. Mean reductions in systolic blood pressure of up to 6 mmHg have also been reported [22] GLP-1 ANALOGUE THERAPY BLOOD PRESSURE AND HEART RATE Observed systolic blood pressure reduction of up to 6 mmHg with Liraglutide would be expected to result in a 15 to 25% reduction in cardiovascular event rate independent of any glucose lowering effect [25]. It is hypothesized a GLP-1 receptor independent nitric oxide mediated vasodilatory pathway is responsible for this, and other potentially beneficial cardiovascular effects on endothelial function, renal sodium excretion and PAI-1 (Plasminogen Activator Inhibitor). GLP-1 receptor activation may also induce sympathetic autonomic activity which stimulates myocyte glucose uptake as well as provoking a positive chronotropic and inotropic response. Although recent reviews suggest these pleiotropic actions have no detrimental effect on cardiovascular events, the clinical significance of a modest elevation in resting heart rate with this treatment is currently unclear [26,27]. Liraglutide is now licensed for use in Europe, Canada, Japan, Mexico and the USA. GLP-1 ANALOGUE THERAPY AND CARDIAC FUNCTION Clearly, substantial clinical benefits accrue from the glucose and weight lowering properties of GLP-1 analogue therapy. However, emerging experimental evidence suggests that GLP-1 therapy has specific cardioprotective effects that are independent of whole body glucose metabolism. In particular, along with improvements in endothelial function, GLP-1 treatment has been shown to have direct effects on cardiac function and structure. For example, rat and canine models of heart failure/cardiomyopathy have demonstrated that GLP-1 administration is associated with improved cardiac output, decreased left ventricular end-diastolic volume and reduced myocyte apoptosis [for recent reviews see 28,29]. Furthermore, mice lacking the GLP-1 receptor were reported to have LV diastolic dysfunction, greater LV wall thickening and impaired LV contractile function [30]. Although limited, human studies have confirmed a link between GLP-1 treatment and cardiac function and structure. For example, after 5 weeks of GLP-1 therapy, 12 patients with chronic heart failure improved their LV function as well as their exercise capacity [31]. Similarly, 72 hours of GLP-1 infusion was associated with improved LV ejection fraction in survivors of acute myocardial infarction [32] and Exenatide administered during percutaneous coronary intervention reduced reperfusion injury and infarct size [33]. Although promising, these studies are limited by their small samples and the non-randomised methodology. However several randomised trials have recently been published. In one study 20 patients were randomised to GLP-1 or saline infusion after percutaneous coronary intervention; GLP-1 was found to ameliorate LV dysfunction [34]. Another randomised cross-over study demonstrated that GLP-1 therapy improved LV function in 14 patients with coronary artery disease [35]. Whilst the above findings are highly promising, data is lacking around the efficacy of GLP-1 therapy at improving cardiac function in high risk populations without overt cardiovascular disease. This limitation is particularly relevant to young individuals with T2DM who are likely to have extreme levels of obesity and present with sub-clinical diastolic dysfunction. Therefore research is needed to investigate to what extent GLP-1 therapy can ameliorate the early stages of cardiac dysfunction. The objective of this study is to determine if Liraglutide, a GLP-1 analogue, leads to improved LV diastolic function in younger adults with T2DM compared to the clinical relevant active comparator Sitagliptin, a DPP-4 inhibitor.

Type 2 Diabetes Mellitus (T2DM), Young Adults, Cardiovascular Disease (CVD), GLP1 Analogue, DPP IV Inhibitor, Obesity, Cardiometabolic Effects, Cardiac MRI

Liraglutide doses will be self-administered by the participant through daily subcutaneous injections. Liraglutide doses will be initiated at 0.6 mg and then increased to 1.2 mg in week two and 1.8mg in week three. The dose will then be maintained at 1.8 mg. Where 1.8 mg doses are not tolerated by the patient, the dose will be lowered to the maximum tolerated dose at the investigators discretion.

Sitagliptin doses will be self-administered by the participant orally at 100mg/day throughout the 26 week period of the study. Sitagliptin is licensed to be used either alone or in combination with other oral antihyperglycemic agents (such as metformin or a sulphonylurea)

Liraglutide (Victoza®, Novo Nordisk) is a stable analogue of the natural hormone glucagon-like peptide-1 (GLP-1). Liraglutide is licensed for use within the United Kingdom and recommended by NICE in combination with metformin, and/or sulphonylurea and/or thiazolidinedione if the following conditions are satisfied. BMI ≥ 35 kg/m2 in those of European descent (with appropriate adjustment for other ethnic groups) and specific psychological or medical problems associated with high body weight, or BMI < 35 kg/m2, and therapy with insulin would have significant occupational implications or weight loss would benefit other significant obesity-related comorbidities.

Sitagliptin (Januvia®, Merck & Co) is an enzyme-inhibiting drug used to inhibit the natural enzyme dipeptidyl peptidase-4 (DPP-4). It is an oral antihyperglycaemic agent used in the treatment of T2DM. Sitagliptin is licensed to be used either alone or in combination with other oral antihyperglycemic agents (such as metformin or a sulphonylurea) and is recommended by NICE as a second line therapy.

It is now well recognised that diastolic dysfunction is the primary characteristic of heart disease in T2DM. Measured by gold-standard tagged cardiac MRI. MRI scans will be anonymised and sent to a stand-alone work station for independent analysis.

(measured blinded to treatment allocation by UHL Pathology labs measured at baseline, 12 weeks to 26 weeks) HbA1c Liver Function Tests Renal Function Tests Lipid profile including total-, LDL- and HDL-cholesterol and triglycerides Thyroid function tests Complete Blood Count (Hematocrit) Vitamin D

(measured blinded to treatment allocation in specialist laboratories at baseline, 12 weeks and 26 weeks): Interleukin-6 C-reactive protein Leptin Adiponectin

Assessed using biological markers. A blood sample will be taken at baseline and 26 weeks and the serum analysed using MSD multiplex panels for the following markers of vascular injury: Panel 1) sICAM-3, e-Selectin, Thrombomodulin, Panel 2) CRP, sICAM, sVCAM, SAA Evaluation of endothelial progenitor cells, stromal derived factor (SDF-1 and GLP-1) and associated biomarkers in a sub-set of participants.

Measures taken at baseline, 12 and 26 weeks BMI Weight Percentage body fat Waist and hip circumferences Systolic and Diastolic Blood pressure (average of 3 measures taken 5 minutes apart) Heart rate (after resting seated for at least 5 minutes

Measured at baseline, 12 and 26 weeks Cardio-respiratory fitness (graded VO2 max test) Total physical activity and time in sedentary behaviour, light-, moderate-, and vigorous-intensity physical activity (ActiGraph GT3X accelerometer worn around the waist during waking hours for 7 consecutive days) Sitting time (thigh mounted ActivPal inclinometer worn for 7 consecutive days)

• Changes to SU, lipid lowering and anti-hypertensive medication usage recorded at baseline, 12 and 26 weeks

HbA1c <7.0%, no weight gain and no minor or major hypoglycemia HbA1c <7.0% and no weight gain HbA1c <7%, SBP <130 mmHg, and no weight gain HbA1c <7% and SBP <130 mmHg Adverse events (see Section 16 for criteria)

Other cardiac measures of function and structure will include: Peak Systolic Strain Left Ventricular Ejection Fraction Stroke volume LV end-diastolic volume LV end-systolic volume LV end-diastolic mass Left Ventricular End Diastolic Mass/volume ratio Pre-and post contrast T1 mapping to calculate volume of distribution, a marker of diffuse cardiac fibrosis Myocardial Perfusion Reserve ( a measure of microvascular function)

Participants will be requested to provide a 7-point glucose profile measured at treatment initiation,12 and 26 week follow-up.

Inclusion Criteria: Capacity to provide informed consent before any trial-related activities Individuals aged 18 - 60 years inclusive Established T2DM BMI ≥ 30 kg/m2 (≥27 kg/m2 for South Asians or other BME populations) On mono or combination oral OAD therapy (sulphonylurea and/or metformin) for ≥ 3months No prescribed thiazolidinediones within the last 3 months An HbA1c value of greater than or equal to 6.5% and less than 10% Exclusion Criteria: < 18 years old Absolute contraindications to MRI Type 1 diabetes (identified through C-peptide analysis) Females of child bearing potential who are pregnant, breast-feeding or intend to become pregnant or are not using adequate contraceptive methods Suffer from terminal illness Have impaired renal function (eGFR < 30 ml/min/1.73m2) ) Impaired liver function (ALAT≥2.5 times upper limit of normal) Known to be Hepatitis B antigen or Hepatitis C antibody positive Clinically significant active cardiovascular disease including history of myocardial infarction within the past 6 months and/or heart failure (NYHA class III and IV) at the discretion of the investigator Recurrent major hypoglycaemia as judged by the investigator Known or suspected allergy to the trial products Known or suspected thyroid disease Receipt of any investigational drug within four weeks prior to this trial Have severe and enduring mental health problems Are not primarily responsible for their own care Are receiving insulin therapy Have taken a thiazolidinedione within the last 3 months Any contraindication to Sitagliptin or Liraglutide Have severe irritable bowel disorder Have pancreatitis or a previous history of pancreatitis

Htike ZZ, Yates T, Brady EM, Webb D, Gray LJ, Swarbrick D, McCann GP, Khunti K, Davies MJ. Rationale and design of the randomised controlled trial to assess the impact of liraglutide on cardiac function and structure in young adults with type 2 diabetes (the LYDIA study). Cardiovasc Diabetol. 2016 Jul 21;15(1):102. doi: 10.1186/s12933-016-0421-6.