Results of recent studies using standard long and short acting insulin therapy (Basal - Bolus or BBI) in type 2 diabetes mellitus (T2DM) have not shown benefits to lower risks for heart attacks, strokes, or eye, nerve and kidney problems. Some studies also show a long time between the start of treatment and signs of benefit. This has led to a review of current ways to normalize blood glucose control with basal bolus insulin and how to make blood glucose better. Improving blood sugar with insulin therapy usually causes weight gain, more high sugar levels after meals, and more low blood sugars. Early studies suggest that when people take long-acting insulin and metformin, they have fewer blood sugar extremes when they also take a new type of medicine called glucagon-like polypeptide-1 (GLP-1) agonist named exenatide (Byetta), instead of meal-time insulin. This means there might be a better way to treat Type 2 diabetes.

Participants are asked to take part in an eight month study to find out if middle-aged and older people with Type 2 diabetes who have added risk factors for heart disease can even out their blood sugar levels. They will start on long-acting insulin, mealtime insulin, and metformin, if they are not already on these medications. Their kidney function tests must be normal and they must not be allergic to metformin. Then, after a 2 month run-in phase, they must be willing to be assigned by chance into one of two groups. This means that they will have a 50/50 chance (like flipping a coin) of being in either group. Half of them will be started on the new medicine known as Byetta rather than the meal-time insulin and the other half will remain on the meal-time insulin during the next 6 months (26 weeks) to see which group has more steady blood sugars. They will be asked to use a continuous blood sugar monitoring system called DexCom. A sensor is inserted under the skin in the same areas the insulin is injected. The DexCom can check their blood sugars 24 hours of the day and night and will be worn until 7 days of recordings are collected. In the same 7 day period, they will also be asked to wear a Holter or Telemetry monitor that will record their heart beats and rhythm which will be compared to the blood sugar readings. They will also use home glucose meters to check their glucose levels about 3 to 4 times a day. The study will take place at 12 centers in the United States and enroll about 120-130 people.

Recent medical endpoint studies employing conventional basal bolus insulin therapy (BBI) in type 2 diabetes mellitus (T2DM) have been disappointing, showing either inconsistent or no effect of treatments on risks for micro- or macro-vascular events, or a long interval between treatment initiation and evidence of clinical benefit. In fact, one trial has suggested that treating glycosylated hemoglobin (HbA1C) to lower targets may even lead to harm. This has raised the possibility that more aggressive glucose lowering approaches lead to harm that overwhelms benefit in those with T2DM. Potential explanations for these results include three closely related physiologic processes: glycemic variability, weight gain and hypoglycemia. Too much variability of glucose, especially post-prandial hyperglycemia, poses the dilemma of how to achieve near-normal mean glucose and HbA1C levels without causing insulin-induced hypoglycemia and/or weight gain. All three of these processes have been linked to worsening systemic inflammation and oxidative stress, and to increased renal and cardiovascular risks.

Fortunately, new tools are available that allow us to assess the severity of glycemic variability (continuous glucose monitoring, or CGM), and to investigate the mechanisms through which it may lead to cardiovascular risk (e.g., systemic inflammation and oxidative stress, sensitive measures of diabetic renal disease, and Holter or Telemetry monitoring for hypoglycemia-induced arrhythmias). In addition, preliminary studies have suggested that replacement of rapid-acting analogue (RAA) in traditional BBI with the glucagon-like polypeptide-1 (GLP-1) agonist, exenatide, may substantially reduce glycemic variability without a strong tendency to increase body weight or hypoglycemia.

This research trial, "FLuctuATion reduction with inSUlin and Glp-1 Added togetheR (FLAT-SUGAR)", by using these new methods to optimize glycemic control while limiting unwanted adverse effects, will be a definitive comparative effectiveness trial. This trial is designed to address the following primary hypothesis:

In middle aged and older individuals with T2DM and additional risk factors for cardiovascular disease, and on a background therapy of basal insulin (insulin glargine) and metformin, the addition of the GLP-1 analogue, exenatide, reduces glycemic variability more than the addition of a rapid-acting-analogue (RAA) (insulin aspart, insulin glulisine or insulin lispro) during an active treatment period of 26 weeks.

The primary outcome measure will be the change in the coefficient of variation of continuous glucose readings, as assessed by CGM. Importantly, FLAT-SUGAR will plan, a priori, to assess glycemic variability using CGM. Secondary trial goals will be to explore potential between-group differences in complications that may result from glycemic variability, including hypoglycemia, systemic inflammation and oxidant stress, diabetic renal disease, weight gain and cardiac arrhythmias. If, as we expect, FLAT-SUGAR demonstrates that CGM provides objective verification of reduced glycemic variability in T2DM with the new GLP-1 agonist-based regimen, the main goal of the trial will be accomplished. If reduced variability is associated with lower risks of adverse events of inflammation, albuminuria progression, weight gain, hypoglycemia, and/or cardiac arrhythmia, a long term clinical comparative effectiveness trial powered to evaluate medical outcomes will be justified.

In order to conduct FLAT-SUGAR, a randomized, controlled, multicenter, open-label investigator-initiated trial, the primary funding is supported by Sanofi-Aventis US with donations of other medications and devices by several other companies. The Sponsor-Investigator is the University of Washington, which will also be the Operation Center (OC).The Data Center (DC) is the University of Texas at Houston School of Public Health. There will be 12 clinical sites with diabetes and CGM expertise to screen and enroll qualified participants for approximately 8-10 weeks of a run-in period, then ultimately randomize, and follow 120 total participants for an active treatment period of 26 weeks.

Inclusion Criteria:

1. T2DM for >12 months defined according to current ADA criteria

2. C-peptide >0.5 ng/mL-after informed consent has been signed, samples will be drawn fasting and sent to a central lab

3. Participants must be on insulin therapy. Diabetes, Blood Pressure & Lipid therapy must be stable (in both dose and agent) for ≥3 months (dose of any 1 drug has not changed by more than 2-fold, & new agents not been added within the previous 3 months)

4. HbA1c 7.5-8.5% for enrollment

5. Age at enrollment (screening): 40-75 years (inclusive) when there is a history of cardiovascular disease (defined in 'a'), or 55 to 75 years (inclusive) when there is not a history of cardiovascular disease but 2 or more risk factors (with or without treatment) are present (defined in 'b')

   a) Established cardiovascular disease defined as presence of one of the following: i. Previous myocardial infarction (MI). (most recent must be > 3 months prior enrollment) ii. Previous stroke. (most recent must be >3 months prior enrollment) iii. History of coronary revascularization (e.g., coronary artery bypass graft surgery, stent placement, percutaneous transluminal coronary angioplasty, or laser atherectomy)(most recent must be > 3 months prior enrollment) iv. History of carotid or peripheral revascularization (e.g., carotid endarterectomy, lower extremity atherosclerotic disease atherectomy, repair of abdominal aortic aneurysm, femoral or popliteal bypass). (most recent must be >3 months prior enrollment) v. Angina with either ischemic changes on a resting ECG, or ECG changes on a graded exercise test (GXT), or positive cardiac imaging study vi. Ankle/brachial index <0.9 vii. LVH with strain by ECG or ECHO viii. >50% stenosis of a coronary, carotid, renal or lower extremity artery. ix. Urine albumin to urine creatinine ratio of >30 mg albumin/g creatinine in 2 samples, separated by at least 7 days, within past 12 months) [Target of 50% of study cohort] or b) Increased CVD risk defined as presence of 2 or more of the following: i. Untreated LDL-C >130 mg/dL or on lipid treatment ii. Low HDL-C (<40 mg/dL for men and <50 mg/dL for women) iii. Untreated systolic BP >140 mm Hg, or on antihypertensive treatment iv. Current cigarette smoking v. Body mass index 25-45 (Asian populations 23-45) kg/m2

6. No expectation that participant will move out of clinical center area during the next 8 months, unless move will be to an area served by another trial center

7. Ability to speak & read English

Exclusion Criteria:

1. The presence of a physical disability, significant medical or psychiatric disorder; substance abuse or use of a medication that in the judgment of the investigator will affect the use of CGM, wearing of the sensors, Holter or Telemetry monitor, complex medication regimen, or completion of any aspect of the protocol
2. Cannot have had any cardiovascular event or interventional procedure, (MI, Stroke or revascularization) or been hospitalized for unstable angina within the last 3 months
3. Inability or unwillingness to discontinue use of acetaminophen products during CGM use
4. Inability or unwillingness to discontinue use of all other diabetes agents other than insulin & metformin during trial (including insulin pump participants who will need to convert to BBI)
5. Intolerance of metformin dose <500 mg/day
6. Inability or unwillingness to perform blood glucose testing a minimum of 3 times/per day
7. Creatinine level ≥1.5 for males or 1.4 for females
8. ALT level ≥ 3 times upper limit of normal
9. Current symptomatic heart failure, history of NYHA Class III or IV congestive heart failure at any time, or ejection fraction (by any method) < 25%
10. Inpatient psychiatric treatment in the past 6 months
11. Currently participating in an intervention trial
12. Chronic inflammatory diseases, such as collagen vascular diseases or inflammatory bowel disease
13. History of pancreatitis
14. BMI >45kg/m2
15. For females, pregnant or intending to become pregnant during the next 7 months

Data available from the authors

Starting 6 months after publication

Written request to the investigators