Comparison of the Effectiveness and Tolerability of Exenatide Once-weekly Compared to Basal Insulins

Background: In January 2012, the US Food and Drug Administration approved a once-weekly form
of exenatide, Bydureon, for the treatment of type 2 diabetes mellitus. Evidence from
clinical trials suggested that Bydureon improves glucose control with low risk of
hypoglycemia. Bydureon does not require a dose titration as necessary for other
glucagon-like peptide-1 receptor agonists (GLP-1RAs), and appears to have other advantages,
such as reducing insulin resistance, preventing weight gain, and improving blood pressure
and lipid profiles. However, the degree to which these advantages of Bydureon lead to
improve outcomes in customary clinical care is unknown.

Aims: The aim of this study is to evaluate the effectiveness and tolerability of Bydureon
relative to basal insulin initiated as first-ever injectable therapeutic regimens used in
customary clinical care.

The specific study objectives are as follows:

- To quantify the effectiveness of Bydureon initiation relative to initiation of basal
insulin, on improving:

- Glycated hemoglobin (HbA1c)

- Weight (body mass index (BMI))

- HbA1c simultaneous to reduction in weight

- Blood pressure and lipid profiles

- To monitor the tolerability of Bydureon initiation relative to initiation basal
insulin, on the occurrence of :

- Hypoglycemia

- Gastrointestinal symptoms (nausea, vomiting, diarrhea, and constipation)

- Change in the markers for renal disease (estimated glomerular filtration rate (eGFR),
serum creatinine, and albumin/creatinine ratio (ACR), and the stability of liver
function test (AST, ALT) and standard blood counts (white blood cell, red blood cell,
Hematocrit, Hemoglobin, Platelet)

- To examine these measures of effectiveness and tolerability within potentially
vulnerable subgroups of Bydureon and basal insulin initiators:

- Type 2 diabetes patients with renal impairment

- Elderly Type 2 diabetes patients

- Type 2 diabetes patients who are a minority race

- Type 2 diabetes patients within strata of HBA1c Design: This retrospective cohort study
will be conducted using Optum's electronic health record data from July 2011 through
March 2015 and identified injectable-naive Type 2 diabetes patients who initiated
either Bydureon or basal insulin during the accrual period, January 2012 and January
2015. Injectable-naive Type 2 diabetes patients will be identified. Propensity score
methods will be used to match initiators of Bydureon to basal insulin initiators.

Study Population:

The study population will be selected from the electronic health recrod data included
patients with Type 2 diabetes who initiated Bydureon or basal insulin. Eligible patients
had:

- Received care for at least 6 months prior to the date of study drug initiation (cohort
entry) with at least one outpatient clinic visit in the 6 months prior to cohort entry
(baseline period);

- At least one diagnosis of type 2 diabetes (ICD-9 250.X0 or 250.X2) prior to and
including the date of the qualifying drug initiation, no prior diagnosis (inclusive of
the index date) of type 1 diabetes mellitus (250.X1 or 250.X3), and no diagnosis of
gestational diabetes within the previous 6 months (inclusive of the index date); and

- No evidence of prior injectable antidiabetic treatment, specifically no dispensing of
GLP-1RA or any insulin drug during the 6-month baseline period. Outcomes and Analysis:
To measure the effectiveness of Bydureon relative to basal insulin are changes in HbA1c
and body weight, as well as changes in HbA1c and simultaneous reduction in weight.
Changes in BMI, lipid profiles, and blood pressure will be examined. These variables
are part of the clinical and laboratory data in the electronic health records. Each
outcome will be evaluated for completeness, multiply imputed, and reported across
standardized time intervals. HbA1c, weight and BMI are summarized in baseline and
quarterly (3-month intervals) in the first year following drug initiation. Lipid
measurements and blood pressure are summarized in baseline and bi-annually (6-month
intervals) in the first year following drug initiation.

To assess drug tolerability, the incidence of hypoglycemia, and gastrointestinal symptoms
(nausea, vomiting, diarrhea, and constipation) will be assessed. These outcomes are
ascertained using an ICD-9 algorithm applied to the structured fields and by extracting
mentions of hypoglycemia using a natural language processing (NLP) algorithm developed by
Optum and applied to the free text clinical notes available in the data. In addition the
stability of renal function evaluated by change in eGFR, or albumin/creatinine ratio (ACR);
the change in serum hepatic enzymes [aspartate aminotransferase (AST), alanine
aminotransferase (ALT)], and hematologic measures [red blood cells counts (RBC), white blood
cell counts (WBC), platelets (PLT), hemoglobin (Hgb) and hematocrit (Hct) will be evaluated.
Each of these laboratory values will be evaluated for completeness, multiply imputed, and
reported across standardized time intervals. eGFR and ACR will be summarized in baseline and
quarterly (3-month intervals) in the first year following drug initiation. Hepatic enzymes
and hematologic measures are summarized in baseline and semi-annually (6-month intervals) in
the first year following drug initiation.

Inclusion Criteria:

- At least 18 years old;

- had received care documented in Electronic Health Records (including at least one
out-patient provider visit) for a minimum of 6-months prior index date;

- had at least one diagnosis of type 2 diabetes by The International Classification of
Diseases, Ninth Revision, Clinical Modification (ICD-9-CM: 250.X0 or 250.X2) prior to
and including the date of the study drug initiation, with no prior diagnosis of type1
diabetes (ICD-9-CM: 250.X1 or 250.X3), or gestational diabetes within the 6-months
prior to index date;

- No evidence of prior injectable antidiabetic treatment, specifically no dispensing of
a GLP-1RA or any insulin during the 6-months baseline period prior to study drug
initiation

Exclusion Criteria:

- Prior diagnosis of type1 diabetes (ICD-9-CM: 250.X1 or 250.X3), or gestational
diabetes within the 6-months prior to index date;

- Prior dispensing of a GLP-1RA or any insulin