Energy Balance Following Islet Transplantation

Background and rationale for the study: Type 1 Diabetes, Hypoglycaemia, IAH and Islet Transplantation

Type 1 diabetes is caused by autoimmune destruction of beta cells within the islets of the pancreas. Those with Type 1 diabetes are dependent on daily insulin replacement for survival and despite major advances in treatment (3), life-expectancy is reduced significantly due to metabolic crises and increased risk of major atherosclerotic vascular events (4). All complications of chronic high glucose can be prevented by tight glycaemic control (5) but this is, however, counter-balanced by a high risk of severe hypoglycaemia leading to a reduced ability to perceive the onset of hypoglycaemia, with disabling confusion and collapse without warning, termed Impaired A Awareness of Hypoglycaemia (AIH). IAH affects up to 25% of people with established Type 1 diabetes every year (10-15 million people world-wide) (6, 7). Approximately 30,000 people in Scotland have Type 1 diabetes and the prevalence is increasing (8). Those with recurrent hypoglycaemia with severe IAH, often experience fluctuations in blood glucose and may be eligible for islet transplantation (5). Islets are transplanted in high numbers via the portal vein under radiological guidance into the liver with glucocorticoid free immunosuppression ("Edmonton protocol"). Two or more transplants are usually necessary to regain insulin secretion to allow adequate control of glucose concentrations and restore awareness of hypoglycaemia.

UK Islet Transplant Consortium (UKITC), Scottish National Islet Programme and anthropometric changes post islet transplantation

Our transplant programme is one of seven in the UKITC; other sites include Newcastle, Oxford, London (Kings, Royal Free), Manchester and Bristol. The primary goal in the UK is to reduce the frequency and restore awareness of hypoglycaemia. Referred patients are assessed in depth by a multidisciplinary team pre-transplant, at the time of transplant and then serially post-transplant. In the UK from April 2008-September 2013 there have been 96 transplants in 76 recipients. Graft survival is >88%, as defined by a stimulated c-peptide of >100pmol/L (90 min following a Fortisip meal), and there have been significant decreases in the frequency of hypoglycaemia: pre- vs. post-transplant, median (interquartile range): 21(7-79) vs. 0(0-1) episodes of hypoglycaemia per annum with significant reductions in HbA1c and insulin dose (all p<0.01)(9). Body weight at 1 year in the UKITC cohort is significantly diminished; pre-transplant: mean (±SEM) 66.3(±1.2) vs. 62.3(±1.1) kg (p=0.001). In common with many programmes (10) peak function is seen at 3 months following the first transplant often with insulin independence but a variable degree of attrition in function may be seen following this with many patients on insulin 1 year following their transplant, although at a much reduced dose versus pre-transplant. In Scotland we have performed 47 transplants in 29 recipients. Those patients with Type 1 diabetes and normal renal function have been assessed previously (≥3<30 months) post-transplant (n=14; 6 males, 8 females; age (range: 30-57 years); all have functioning grafts. Following islet transplantation we have observed significant reductions in body weight, pre versus post-transplant: median(interquartile range) 70.9(64.9-82.5) vs. 66.0(59.8-79.9)kg, p=0.01; BMI: 26.2(24.4-28.6) vs. 24.5(22.2-26.1) kg/m2, p=0.01; %fat mass: 30.0(21.2-35.2) vs. 24.1(18.4-31.2)%, p=0.004; and waist circumference: 85.5(77-89.7) vs. 77.5(73.5-88.3)cm, p=0.01. Total calorific intake assessed via 7 day dietary histories were not different pre- versus post-transplant 1700(1581-1842) vs. 1528(1319-1708) kcal, (p=0.09), although frequency of hypoglycaemia (10(6-18) vs 1(0-3)) episodes per week and calorific intake for hypoglycaemia were reduced. Concurrently HbA1c has improved p<0.001 (11).

Consistent with our observations, significant reductions in body weight of >4kg (10, 12), waist circumference and body fat mass, the latter measured using bioelectrical impedence, within a year of islet transplantation, have been previously reported, despite no significant changes in caloric intake post-transplant (12). In the latter study 30 subjects had all assessments completed and 12 subjects were not on insulin at the time of their last data collection. Insulin use reduced 10 fold and glycaemic control improved. Regression analyses adjusting for confounding variables including exogenous insulin dose and glucagon-like peptide-1 agonist use, confirmed the association of weight loss with islet transplantation. Of note in this cohort of islet transplant recipients, no measurements of energy expenditure were made and the patients including those with islet transplants following kidney transplants were on a variety of immunosuppressant agents which may affect energy balance. More recently the same group published a retrospective study in islet transplant recipients and found a decrease in BMI in their cohort of 33 recipients at 3 years post-transplant versus pre-transplant (13). Dietetic habits and caloric intake and activity, both assessed by open ended questionnaires, were not significantly altered pre- versus post-transplant.

We hypothesise that other components of energy expenditure are increased following islet transplantation secondary to the diminished ratio of systemic to portal insulin concentrations. The energy expenditure components that may be modulated in this way include REE and PPT.

Our proposed study aims to prospectively assess energy intake and food choices using the "gold standard" of serial 7 day weighed food diaries (14, 15) and assess the multiple components of energy expenditure pre- and post-islet transplantation versus subjects commencing insulin pump therapy who have relatively high ratios of systemic to portal insulin concentrations along with control subjects with normal glucose tolerance.

Maintenance of body weight and relationship with energy intake and energy expenditure

Bodyweight reflects the balance between food intake and energy expenditure. Total energy expenditure may be separated into three main components: REE, PPT and physical activity. REE may be defined as energy expenditure at rest in the fasted state measured in a thermoneutral environment. PPT or diet induced thermogenesis is an increase in metabolic rate above REE; PPT demonstrates more variability between people than REE (16, 17). It is the energy generated following food ingestion and has an obligatory component reflecting the energy required to digest, absorb, interconvert and store fuels and a facultative component in which the sympathetic nervous system has a major role (17). The increment in energy expenditure may be accounted for in part by the cost of glucose storage as glycogen but often the measured cost of glucose storage is much higher than this and it is this increment above this "obligatory component" that is termed "facultative thermogenesis" (18). This facultative component can compromise 50% of the thermic effect of food and large inter individual variations may be present (19, 20). With a sedentary lifestyle, REE constitutes 75 to 80%, PPT 10 to 15% and physical activity 10 to 15% of total daily energy expenditure.

Modulators of energy expenditure

The lean body mass, thyroid status and protein turnover of a subject determines the REE (21). We and others have shown that PPT is negatively related to insulin sensitivity in a group at risk for Type 2 diabetes (22). PPT may be affected by beta-adrenergic blockade and other drugs (23, 24). Insulin, a known vasodilator, may also affect energy expenditure (25). Intraportally transplanted islets may become revascularised by tributaries from both the portal vein and the hepatic artery (26) or perhaps by tributaries from just the hepatic artery (27). It is untested if PPT increases as a consequence of islet transplantation secondary to increased insulin concentrations in the portal and splanchnic circulations (28). There has been much progress in the induction agents and immunosuppression therapy used over the last 10 years with improved insulin independence rates (29). Our centre uses the monoclonal antibody alemtuzumab at induction (30) and maintenance immunosuppression is with the calcineurin inhibitor tacrolimus and mycophenolate. Liver transplant patients on tacrolimus have diminished REE, suggesting an inhibition of mitochondrial respiration (31) although in liver transplantation hepatic denervation occurs with autonomic nerve dysfunction which may also decrease energy expenditure (32, 33). It is unlikely therefore that the immunosuppressive therapies increase energy expenditure although there may certainly be negative effects on appetite and energy intake (34).

Hepatic fatty acid oxidation.

Free Fatty Acids (FFA) are oxidised within mitochondria which are present within the whole body with large concentrations in the liver, muscle and heart. Most methodologies measure whole body fatty acid oxidation rates and cannot distinguish between these sources. Hepatic fat oxidation may be altered following islet transplantation and may influence hepatic insulin sensitivity. It is not known whether hepatic fat oxidation is increased following islet transplantation which may explain the weight loss.

Assessment of mitochondrial fatty acid oxidation in the liver:

The assessment of hepatic mitochondrial function has been hampered by invasive and complex techniques, which have in general lacked specificity. Recently, breath testing using stable carbon isotopes has been proposed as a safe, non-invasive and non-radioactive method to assess mitochondrial oxidative capacity. Sodium13C-octanoate can be employed as a substrate used in assessing hepatic mitochondrial oxidative function, particularly, β-oxidative pathway. Sodium octanoate is a medium chain fatty acid, which is metabolised through mitochondrial β-oxidation, producing acetyl Coenzyme A (acetyl Co-A). Acetyl Co-A enters the Krebs cycle to undergo further oxidation leading to the production of CO2 that can be measured during the breath test (35a).

The purpose of this proposal is to understand the mechanisms of the weight loss and fat mass observed. No study has examined energy expenditure prospectively in this detailed way in a cohort of pre- and post-islet transplant and insulin pump therapy recipients.

Detailed plan of investigation:

Hypothesis: Islet transplantation is associated with increased energy expenditure as a consequence of the reduction in the ratio of systemic to portal insulin concentrations.

The aims are to assess:

1. Anthropometric measures: body weight, waist circumference, skin fold thickness, body composition using air displacement plethysmography (Bod Pod).
2. Total energy intake, including the excess energy intake required in the treatment of hypoglycaemia, using 7 day weighed food diaries, The Fear of Hypoglycaemia Survey, Gold and Clarke Score.
3. The activity component of energy expenditure using accelerometry.
4. REE and PPT (using meal tolerance tests) and total metabolic rate (TMR) using doubly labelled water.
5. Hepatic fat oxidation using Sodium 13C octanoate.
6. Liver fat, abdominal subcutaneous and visceral fat using MRS and MRI.

In the islet transplant recipients and insulin pump patients MRS of liver and MRI of liver fat; abdominal subcutaneous and visceral fat; glycaemic lability; frequency of hypoglycaemia assessed using continuous glucose monitoring systems (CGMS); awareness of hypoglycaemia assessed subjectively with the Gold Score and Clarke Scores (35), and mixed meal tolerance tests will continue as per the UKITC protocols, including pre-transplantation. In the glucose tolerant controls all the above will be done except the MR imaging studies, the Fear of Hypoglycaemia Survey (36), the Gold and Clarke scores and CGMS.

STUDY DESIGN In the islet transplant programme recipients are assessed pre-transplant and at 1, 3, 6, 12 months post-transplant and then 6 monthly. The investigations in the transplant or insulin pump subjects, will be based around these timelines. Specifically, we anticipate peak islet function at 3 months.

We aim to recruit 3 groups of participants: those with Type I diabetes that are on the waiting list for islet transplantation, those with Type I diabetes that are on the waiting list for insulin pump therapy, and controls who do not have diabetes (glucose tolerant).

The study contains 3 parts as described below. The participants will be expected to complete all 3 parts at each time point. The 3 parts are completed over 5 separate time points totalling 13 visits:

The following visits are for the patients with type 1 diabetes:

Pre- intervention and then 1,3, 6 and 12 months post intervention.

• Pre-intervention there are 3 study visits over a 4 week period
• Month 1 post intervention there are 2 visits over 2 weeks
• Months 3 post invention there are 3 visits over a 4 week period
• Month 6 post intervention there are 2 visits over 2 weeks
• 12 months post intervention there are 3 visits over 4 weeks

Patients with diabetes normally have 1 routine clinic visit at each time point described as part of their normal follow-up post-intervention. The other visits will be surplus to their usual follow-up following intervention, apart from during their first month post-intervention where they would be followed-up weekly.

Glucose-tolerant controls will only have one set of examinations carried out which will involve 3 visits over 4 weeks.

Pre-intervention, 3 \& 12 months

Part 1:

Blood tests (FBC, LFTS, U\&Es, Coagulation Screen, Lipid Profile, Thyroid function,HbA1c, Glucose, post-transplant Tacrolimus level) this is approx 18ml (3 ½ teaspoons). This is part of the routine follow-up for islet transplant patients.

Anthropometry -body weight, waist, calf and arm circumference, skin fold thickness (including bicep, tricep, scapula, illiac crest and calf), body composition using air displacement plethysmography (BODPOD). These examinations are study specific and not part of the routine follow-up.

A continuous glucose monitoring (CGMS) device and accelerometer will also be fitted to measure energy expenditure over seven days. The participant will be given a food diary and a set of weighing scales to record food intake over this seven day period (including the excess calories required in the treatment of hypoglycaemia) at home. CGMS is frequently used during the follow-up of patients with diabetes. Glucose tolerant controls will not have a CGMS fitted or complete the following surveys.

The participant will complete "The Fear of Hypoglycaemia Survey". This is a questionnaire asking how/whether they modify their behaviour in order to avoid hypoglycaemia and how frequently they worry about their hypoglycaemia in different circumstances. The participants will also be asked about their degree of impaired awareness of hypoglycaemia using specific scoring systems (Gold and Clarke Score).

Subjects within the islet transplant cohort and pump therapy cohort will be asked to go to the clinical research imaging centre and will have magnetic resonance scanning done of their abdomen including their liver and their abdominal fat.

Part 2 (week 2):

The participant will be invited back a week later to assess their resting energy expenditure (REE). At this visit all the equipment from week 1 will be returned. The participant will be weighed on arrival and have a cannula inserted into an arm. Blood samples for glucose, insulin and c-peptide will be taken (15mls). The participants will then be given a "mixed meal" in the form of "Fortisip concentrate" which tastes like a milk shake. Their energy expenditure after this mixed meal will be measured for two hours by a device called an indirect calorimeter which is an open hood ventilation device which measures the amount of oxygen breathed in and the amount of carbon dioxide breathed out. The energy expended after a meal is called post-prandial thermogenesis (PPT) and represents the calories burnt after a meal. Simultaneously blood samples will be taken every 30 minutes for three hours to measure insulin, c-peptide and glucose levels (150 mls). This is the gold standard test for assessing islet function.

"Doubly-labelled Water" studies will be carried out on a proportion of participants. The participant will be given "doubly labelled water" to drink at their second visit, a urine sample will be collected just prior to drinking the water and then they will be instructed to collect further urine samples at home on day 1, 5, 10 and 14 following this, the research nurse will call them regularly during this time period. The samples will be stored in the freezer in a zip locked bag contained in a plastic tub with tight-fitting lid.

Part 3 (week 4):

The participant will be invited back two weeks after part 2 for a study specific visit. The urine samples will be collected. Hepatic Mitochondrial studies will carried out this involves the participant drinking a solution containing Sodium 13C Octanoate which is a fatty acid found in foods. They will then be asked to blow into a bag intermittently over a period of two hours (0, 10, 15, 20, 25, 30, 40, 60, 90 and 120 mins) in order to measure how their liver handles the fatty acid and whether they "burn" it quickly or slowly.

1 month and 6 months post intervention

Only blood test, anthropometry, accelerometer, weighed food diary and CGMS will be carried out. Islet transplant recipients will have a clinical mixed meal tolerance test carried out to assess islet function. These visits will coincide with routine visits.