The Effect of Liraglutide Treatment on Postprandial Chylomicron and VLDL Kinetics, Liver Fat and de Novo Lipogenesis

Phase 4

Completed

Conditions: Type 2 Diabetes

Interventions: Drug: Placebo
Drug: Liraglutide

Registration Number: NCT02765399

Lead Sponsor: Helsinki University Central Hospital

Brief Summary: This study aims to evaluate the mechanisms underlying the effect of incretin therapy on lipoprotein metabolism in subjects with type 2 diabetes and to study the effect of liraglutide on hepatic de novo lipogenesis.

Detailed Description: The well recognized dyslipidemia in people with type 2 diabetes consists of high fasting and non-fasting plasma triglycerides (TG), low high-density lipoprotein (HDL) -cholesterol and preponderance of small dense low-density lipoprotein (LDL) particles nominated as the atherogenic lipid triad. Humans are mostly in a postprandial rather than fasting state and therefore non-fasting TG values reflect more accurately the continuous exposure of arterial wall to triglyceride rich lipoproteins (TRLs) and more importantly, to substantial cholesterol load that these particles deliver.

Postprandial lipemia is highly prevalent even in type 2 diabetes patients with normal fasting TG concentrations. Intestinal overproduction of chylomicrons (CMs) and the structural protein apolipoprotein (apo)-B48 has been identified as an integral feature of postprandial lipemia in type 2 diabetes and insulin resistance. It is clinically important to elucidate the mechanism for delayed postprandial lipemia and the interactions between dysglycemia and dyslipidemia in type 2 diabetes patients.

Recruitment & Eligibility

Status: COMPLETED

Sex: All

Target Recruitment: 23

Inclusion Criteria

Subjects with type 2 diabetes treated with a lifestyle or metformin (any dose)
waist circumference > 88 cm in women and > 92 cm in men
BMI 27-40 kg/m2
triglycerides between 1.0 - 4.0 mmol/L
LDL < 4.5 mmol/l

Exclusion Criteria

Type 1 diabetes
Apo E2/2 phenotype
ALT/AST > 3x ULN
GFR < 60 ml/min, clinically significant TSH outside normal range
Lipid-lowering drugs other than statins within 6 months
Current treatment with pioglitazone, insulin, sulphonylureas, gliptins, glinides, SGLT-2 inhibitors or thiazide diuretics (at a dose of > 25 mg / day)
Blood pressure > 160 mmHg systolic and/or > 105 diastolic
History of pancreatitis or stomach / other major bleeding, thyroid neoplasia, persistent hypothyroidism or persistent hyperthyroidism
Any medical condition that puts the patient in the risk of dehydration
Concurrent medical condition that may interfere with the interpretation of efficacy and safety data during the study.
Females of childbearing potential who are not using adequate contraceptive methods
Subjects who have experienced side-effects previously from GLP-1 agonists
Non-compliance or withdrawal of consent
Any information or clinical event described in liraglutide SPC that is a contraindication for the use of liraglutide

Study & Design

Study Type: INTERVENTIONAL

Study Design: PARALLEL

Arm && Interventions

Group	Intervention	Description
Placebo	Placebo	Placebo subcutaneous injection once daily with following dose escalation: placebo 0.1 ml once daily for one week; placebo 0.2 ml once daily for one week and thereafter placebo 0.3 ml once daily for 3.5 months.
Liraglutide	Liraglutide	Liraglutide subcutaneous injection once daily with following dose escalation: liraglutide 0.6 mg once daily for one week; liraglutide 1.2 mg once daily for one week and thereafter liraglutide 1.8 mg once daily for 3.5 months.

Primary Outcome Measures

Name	Time	Method
Change in HbA1c Level	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in B -Hemoglobiini-A1c level in plasma. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in SAT Area	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): subcutaneous adipose tissue area measured by magnetic resonance imaging (MRI). Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in Liver Fat Content	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): mean liver fat content was measured by magnetic resonance imaging. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Plasma Triglyceride (TG) Area Under Curve (AUC)	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): postprandial plasma TG summary measured using the trapezoidal rule and expressed as AUC (at fasting and at 0.5, 1, 2, 3, 4, 6 and 8 hours) after oral fat tolerance test. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in fP-glucose Level	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): concentration of fasting plasma glucose measured using the hexokinase method. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in Insulin Level	Baseline and after16 weeks	Before vs after intervention (Liraglutide or placebo): Concentration of insulin level in plasma measured using electrochemiluminescence. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in ApoCIII Level	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): apolipoprotein CIII concentration in plasma measured by using turbidimetric immunoassay. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Body Weight	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in body weight. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in Matsuda Index	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Matsuda index was calculated for assessment of insulin sensitivity in plasma at time points 0, 30, 60 and 120 minutes using formula 10,000/square root of \[fasting glucose x fasting insulin\] x \[mean glucose x mean insulin during oral glucose tolerance test\]. The Matsuda index is considered to be the gold standard to determine insulin sensitivity without glucose clamp studies (Matsuda M, DeFronzo RA. Diabetes Care. 22:1462-70). Subjects who don't have insulin resistance have values of Matsuda Index of 2.5 or higher (Kerman WN et al. Stroke 34:1431;2003). Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Change in VAT Area	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): visceral adipose tissue area measured by magnetic resonance imaging (MRI). Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.

Secondary Outcome Measures

Name	Time	Method
Change in Hepatic de Novo Lipogenesis	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Hepatic DNL is calculated from enrichment of deuterated water ingested during the kinetic study at specified time points (0, 4 and 8 hrs.). Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Mean Production Rate of apoB48 in CM	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in mean production rate of ApoB48 in chylomicrons isolated from plasma samples and measured by multicompartmental modeling assay. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (JCI 1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. JIM 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al.
Mean apoB48 FTR to VLDL1 Particles	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in apoB48 chylomicron fractional transfer rate to VLDL1 isolated from plasma by ultracentrifugation and by liquid chromatography/mass spectrometry and calculated with multicompartmental modeling assay. So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. JIM 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. 2021.
Change in Systolic RR	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): systolic blood pressure measurements. Results from Matikainen et al. Diabetes Obes Metab 21:84-94; 2019.
Mean Total Production of apoB48	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): ApoB48 total production in plasma measured by using multicompartmental modeling. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (JCI 63:1262;1979) and have been widely used over 30yrs. So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. JIM 285:562;2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. Diabetes Obes Metab. 23:1191; 2021.
Mean CM FDC of apoB48	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in chylomicron fractional direct clearance rates of apoB48 measured from plasma by liquid chromatography - mass spectrometry with multicompartmental modeling assay. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. 2021.
Change in Direct CM-apoB48 Clearance	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Direct apoB48 clearance rates in isolated chylomicrons and measured by liquid chromatography - mass spectrometry and calculated by multicompartmental modeling assay. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. 2021.
Mean CM-apoB48 Transfer Rates to VLDL1	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in chylomicron-apoB48 transfer rates to VLDL1 isolated from plasma by ultracentrifugation and measured using multicompartmental modeling. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. 2021.
Mean TG Fractional Catabolic Rates in CM	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in triglycerides fractional catabolic rates in isolated chylomicrons from plasma samples measured by multicompartmental modeling assay. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (JCI 1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. JIM 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. DOM 2021.
Mean VLDL1-TG Production Rates	Baseline and after16 weeks	Before vs after intervention (Liraglutide or placebo): Change in VLDL1 production rates measured from isolated VLDL from plasma samples by ultracentrifugation and measured using mathematical modeling. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (JCI 1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. JIM 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. DOM 2021.
Mean Fractional Catabolic Rate of VLDL2-apoB100	Baseline and after 16 weeks	Before vs after intervention (Liraglutide or placebo): Change in VLDL2-apoB100 fractional catabolic rates measured from isolated VLDL2 from plasma by ultracentrifugation and measured using mathematical modeling. The power of mathematical modelling to describe the metabolic pathways of lipid and lipoprotein metabolism was demonstrated by Zech L et al (JCI 1979). So far few studies have focused on the modelling of apo B48 and apo B100 after a meal that is more physiological than the fasting state (Björnson E et al. JIM 2019). Production rates for apo B48, apo B100 and triglycerides in chylomicrons, VLDL1 and VLDL2 were derived from samples taken before and after the tracer injection and after the meal at 0, 30, 45, 60, 75, 90,120, 150 min and at 3, 4, 5, 6, 8, 10, 24 hrs and averages for 24 hrs. Analysis of tracer/ tracee curves of stable isotopes was used to derived the estimates of kinetic parameters using a new mathematical modeling per day. Results from Taskinen et al. DOM 2021.