ORIGINAL

ARTICLE

original article

Diabetes, Obesity and Metabolism 17: 630638, 2015.

2015 John Wiley & Sons Ltd

Efficacy and tolerability of saxagliptin compared with glimepiride

in elderly patients with type 2 diabetes: a randomized, controlled
study (GENERATION)
G. Schernthaner1 , S. Durn-Garcia2 , M. Hanefeld3 , G. Langslet4 , L. Niskanen5 , C. J. stgren6 , E. Malvolti7,
& E. Hardy8
1 Department of Medicine I, Rudolfstiftung Hospital, Vienna, Austria
2 Unidad de Gestion de Endocrinologa y Nutricin, Hospital Universitario de Valme, Sevilla, Spain
3 Study centre Professor Hanefeld, GWT-TUD GmbH, Dresden, Germany
4 Lipid Clinic, Oslo University Hospital, Oslo, Norway
5 Endocrinology and Metabolism, Abdominal Center, Helsinki University Hospital, University of Helsinki, Helsinki, Finland
6 Department of Medical and Health Sciences, Linkping University, Linkping, Sweden
7 AstraZeneca, Istanbul, Turkey
8 AstraZeneca LP, Wilmington, DE, USA

Aims: To assess the efficacy and safety of adjunctive saxagliptin vs glimepiride in elderly patients with type 2 diabetes (T2D) and inadequate glycaemic
control.
Methods: In this multinational, randomized, double-blind, phase IIIb/IV study (GENERATION; NCT01006603), patients aged 65 years were randomized
(1 : 1) to receive saxagliptin 5 mg/day or glimepiride 6 mg/day, added to metformin, during a 52-week treatment period. The primary endpoint was
achievement of glycated haemoglobin (HbA1c) <7.0% at week 52 without confirmed/severe hypoglycaemia. The key secondary endpoint was incidence
of confirmed/severe hypoglycaemia. Safety and tolerability were also assessed.
Results: Of 720 patients randomized (360 in each treatment group; mean age 72.6 years; mean T2D duration 7.6 years), 574 (79.8%) completed the study
(saxagliptin 80.3%; glimepiride 79.2%). Similar proportions of patients achieved the primary endpoint with saxagliptin and glimepiride (37.9 vs 38.2%;
odds ratio 0.99, 95% confidence interval 0.73, 1.34; p = 0.9415); however, a significant treatment-by-age interaction effect was detected (p = 0.0389):
saxagliptin was numerically (but not significantly) superior to glimepiride for patients aged <75 years (39.2 vs 33.3%) and numerically inferior for patients
aged 75 years (35.9 vs 45.5%). The incidence of confirmed/severe hypoglycaemia was lower with saxagliptin vs glimepiride (1.1 vs 15.3%; nominal
p < 0.0001). Saxagliptin was generally well tolerated, with similar incidences of adverse events compared with glimepiride.
Conclusion: As avoiding hypoglycaemia is a key clinical objective in elderly patients, saxagliptin is a suitable alternative to glimepiride in patients with
T2D aged 65 years.
Keywords: dipeptidyl peptidase-4 inhibitor, randomized trial, sulphonylureas, type 2 diabetes
Date submitted 11 November 2014; date of first decision 5 December 2014; date of final acceptance 8 March 2015

Introduction
In elderly patients with diabetes, the risk of hypoglycaemia is
substantially higher compared with younger adults [1,2] and
the harm associated with severe hypoglycaemia may counterbalance the potential benefits of intensive glucose control [3,4].
While for most patients with type 2 diabetes (T2D), the American Diabetes Association (ADA) and the European Association
for the Study of Diabetes (EASD) recommend a target glycated
haemoglobin (HbA1c) concentration of <7.0% [5], glycaemic
targets for elderly patients with limited life expectancy, a history
of severe hypoglycaemia, long-standing or more complicated
Correspondence to: Guntram Schernthaner, MD, Department of Medicine I, Rudolfstiftung
Hospital-Vienna, Juchgasse 25, A1030 Vienna, Austria.
E-mail: guntram.schernthaner@meduniwien.ac.at
Previously

employed by Bristol-Myers Squibb, Paris, France.

T2D are less ambitious than for younger, healthier individuals.

A target HbA1c of 7.58.0% is considered acceptable, according
to recent ADA and EASD guidelines [5].
Saxagliptin is a selective dipeptidyl peptidase-4 (DPP-4)
inhibitor approved in the European Union as additional therapy in adults with T2D with inadequate glycaemic control on
monotherapy with metformin, a sulphonylurea or a thiazolidinedione. Saxagliptin is also approved in the European Union
as a triple oral therapy in combination with metformin plus a
sulphonylurea, when this regimen alone provides inadequate
glycaemic control, and as combination therapy with insulin
(with/without metformin) for adults with T2D [6]. In phase
III studies, saxagliptin improved glycaemic control compared
with placebo and was similar to active comparators (glipizide
or sitagliptin), in combination with metformin [79], metformin plus glimepiride [10], other oral antihyperglycaemic
drugs [11,12] and insulin [13]. In patients aged 65 years, a post

original article

DIABETES, OBESITY AND METABOLISM

hoc analysis of pooled phase III data showed that saxagliptin

(combination therapy or initial monotherapy) provided better
glycaemic control than did placebo, with a low risk of hypoglycaemia [14,15].
Although HbA1c is traditionally used as a primary endpoint for assessing the efficacy of antihyperglycaemic treatments, in elderly patients the risk of hypoglycaemia should also
be considered, as both factors contribute to clinical treatment
decision-making. In the present study, we evaluated whether
saxagliptin was superior to glimepiride as an add-on therapy
to metformin in improving glycaemic control without hypoglycaemia in elderly patients (65 years old) with T2D.

Methods
Patients
Patients with T2D aged 65 years, who were on stable
metformin monotherapy at any dose for 8 weeks before
enrolment and had an HbA1c concentration of 7.09.0%, were
eligible for inclusion in the study. The exclusion criteria were:
type 1 diabetes mellitus; treatment with any antihyperglycaemic therapy other than metformin monotherapy <8 weeks
before enrolment; treatment with systemic glucocorticoids
(except for replacement therapy) or cytochrome P450 3A4
inducers; history of ketoacidosis or hyperosmolar non-ketonic
coma; history of haemoglobinopathies; renal impairment (creatinine clearance <60 ml/min); cognitive function problems;
alcohol or illegal drug abuse for 12 months before enrolment;
and history of hypersensitivity or contraindication to the study
drugs. Additional exclusion criteria were: aspartate aminotransferase levels >3 times the upper limit of normal and/or
alanine aminotransferase levels >3 times the upper limit of normal and/or total bilirubin >34 mol/l; and creatine kinase >10
times the upper limit of normal. All patients abstained from
donating blood, plasma or platelets during the study.
The study was conducted in compliance with the ethical
principles of the Declaration of Helsinki and the International
Conference on Harmonisation notes for Guidance on Good
Clinical Practice. The study protocol was approved by independent ethics committees or institutional review boards. All
patients provided written informed consent before enrolment.

Study Design
This 52-week, multinational, randomized, double-blind,
active-controlled, parallel-arm, phase IIIb/IV study (GENERATION; NCT01006603) was conducted between October
2009 and June 2012 at 152 sites in 12 European countries and
Mexico.
The study consisted of a 2-week screening period, a 2-week
enrolment period, a 2-week single-blinded (to patients only)
placebo lead-in period, and a 52-week double-blinded treatment period, which in turn comprised a 12-week titration
period and a 40-week maintenance period (Figure S1). After
lead-in, patients were randomized (1 : 1) to receive saxagliptin
5 mg/day and placebo or glimepiride 1 mg/day and placebo,
added to metformin. Randomization was carried out via an
interactive web response system and was stratified by age to

Volume 17 No. 7 July 2015

include 60% of patients aged <75 years and 40% of patients

aged 75 years.
During the titration period, glimepiride or placebo was uptitrated every 3 weeks in 1- or 2-mg/day increments to the
optimum dose (fasting plasma glucose 6.1 mmol/l), up to
6 mg/day. During maintenance, no uptitration was performed.
Glimepiride or placebo could be downtitrated if recurrent
hypoglycaemia occurred.

Assessments
The primary efficacy endpoint was the proportion of patients
achieving an HbA1c level of <7.0% at week 52 without confirmed/severe hypoglycaemia. Confirmed hypoglycaemia
was defined as a symptomatic or asymptomatic event with
plasma glucose <3.0 mmol/l, requiring no external assistance.
Severe hypoglycaemia was defined as a symptomatic event
requiring external assistance because of severe impairment in
consciousness or behaviour, with or without plasma glucose
<3.0 mmol/l, but with prompt recovery after glucose/glucagon
administration.
The key secondary endpoint was the proportion of patients
with 1 confirmed/severe hypoglycaemic event over the treatment period. Other secondary endpoints included the proportion of patients achieving HbA1c <7.0 or 6.5% at week 52, and
the change from baseline to week 52 in mean HbA1c.
Post hoc analyses were conducted to determine the number
of confirmed/severe hypoglycaemic events over time, and the
incidence of confirmed/severe hypoglycaemia by HbA1c category at week 52 (<7.0% or 7.0%) stratified by age group, and
by HbA1c category at week 52 (<6.5%; 6.5 to <7.0%; 7.0
to <7.5, and 7.5%).
Safety and tolerability assessments included adverse events
(AEs) and body weight.

Analytical Methods
Blood samples for assessing glycaemic variables were collected
at every visit during the treatment period and were analysed centrally (Quintiles Laboratories Ltd. Europe, Livingstone,
UK). Patients self-monitored glucose levels using a glucometer
(Accu-Chek , Roche Diagnostics, Basel, Switzerland) at least
every second day during lead-in and titration and at least once
a week during maintenance. Blood glucose levels and hypoglycaemic events were recorded in patients diaries.

Statistical Analysis
A sample size of 698 patients (349/treatment arm) was calculated for detecting superiority of saxagliptin in the primary
endpoint, with a two-sided significance level of 0.05 and 80%
power. This assumed a 10% drop-out rate and an odds ratio
(OR) of 1.55 for achieving target HbA1c without hypoglycaemia with saxagliptin compared with glimepiride.
The safety population, which included all randomized
patients who took 1 dose of the study medication, was used
for reporting safety and tolerability results and for primary,
key secondary and post hoc efficacy assessments. The full
analysis population, defined as safety population patients with

doi:10.1111/dom.12461 631

original article
non-missing baseline and 1 post-baseline efficacy data for 1
variable, was used for reporting other secondary endpoints.
Patients who did not complete the 52-week treatment
period were considered to be non-responders, that is, as having not achieved HbA1c target of <7.0% at week 52 without
confirmed/severe hypoglycaemia.
Non-continuous endpoints were analysed using a Cochran
MantelHaenszel test with continuity correction stratified by
age. Treatment-by-age interaction and treatment-by-baseline
HbA1c interaction were determined using a logistic regression
model. To control for multiplicity, primary and key secondary
endpoints were tested in a pre-specified hierarchy. Continuous
endpoints were analysed using an analysis of covariance model,
with treatment group and age as fixed effects and baseline
value of the endpoint as a covariate. Statistical analyses were
performed by Aptiv Solutions GmbH, Cologne, Germany.

Results
Participant Disposition
Of 720 participants randomized, 718 received study medication, and 574 (79.8%) completed the 52-week treatment period
(saxagliptin 80.3%; glimepiride 79.2%; Figure 1). Safety and full
analysis populations comprised 718 and 709 patients, respectively.

Demographics and Baseline Characteristics

Demographics, baseline characteristics and medical history
were balanced across treatment groups and were generally
representative of an elderly population with T2D, inadequate
glycaemic control, and normal renal function or mild renal
impairment (Table 1). The mean dose of glimepiride was
3.3 mg/day.

Efficacy
Primary Efficacy Endpoint. The proportions of patients achieving HbA1c < 7.0% at week 52 without confirmed/severe
hypoglycaemia were similar with saxagliptin and glimepiride:
37.9 vs 38.2% [OR 0.99, 95% confidence interval (CI) 0.73, 1.34;
p = 0.9415 (Table 2)]; however, a significant treatment-by-age
interaction was detected (p = 0.0389). The proportion of
patients aged <75 years achieving the primary endpoint was
numerically (but not significantly) higher with saxagliptin
than with glimepiride (39.2 vs 33.3%), whereas in patients
aged 75 years, the opposite occurred (35.9 vs 45.5%). A
treatment-by-baseline HbA1c interaction was also identified (p = 0.0822). The proportion of patients with baseline
HbA1c levels <7% and 7 to <8% achieving the primary endpoint was numerically higher with saxagliptin, compared
with glimepiride (baseline HbA1c < 7%: 56.1 vs 53.5% and
baseline HbA1c 7 to <8%: 43.5 vs 40.2%). Among patients
with baseline HbA1c 8%, fewer patients in both groups
reached the primary endpoint; a numerically lower proportion of patients treated with saxagliptin reached the primary
endpoint, compared with the glimepiride-treated group
(12.3 vs 25.3%).

632 Schernthaner et al.

DIABETES, OBESITY AND METABOLISM

Key Secondary Endpoint. Fewer patients in the saxagliptin

group experienced 1 confirmed/severe hypoglycaemic event
over the treatment period, compared with the glimepiride
group: 1.1 vs 15.3% (OR 0.06, 95% CI 0.02, 0.17; nominal
p < 0.0001). The difference between saxagliptin vs glimepiride
was numerically greater in patients aged <75 years compared
with patients aged 75 years [ORs 0.04 vs 0.12; nominal
p = 0.3018 (Table 2)]. With glimepiride, fewer patients
aged 75 years experienced hypoglycaemia, compared with
patients aged <75 years (10.5 vs 18.5%).
Secondary Efficacy Endpoints. Overall, a lower proportion
of saxagliptin-treated patients achieved HbA1c < 7.0% at
week 52 than those treated with glimepiride: 44.7 vs 54.7%
(nominal p = 0.0077). The difference between treatment
groups was numerically, but not significantly, greater in
patients aged 75 years than in those aged <75 years (nominal
p = 0.1071). A lower proportion of saxagliptin-treated patients
vs glimepiride also achieved HbA1c 6.5% (18.5 vs 30.9%,
nominal p < 0.0001; Table 2).
The adjusted mean HbA1c change from baseline at week 52
was smaller with saxagliptin than with glimepiride: 0.44 vs
0.64% (nominal p < 0.0001; Table 2), with a greater difference
between treatment groups in patients aged 75 years than
with patients aged <75 years (Table 2). The mean HbA1c
changes from baseline were similar between age groups with
saxagliptin, and numerically greater in patients aged 75 years,
compared with patients aged <75 years, with glimepiride
(nominal p = 0.0839, p = 0.0001; Table 2). Post hoc analyses
assessing potential mechanisms, including baseline estimated
glomerular filtration rate, did not provide any conclusive
results (File S1).
Post hoc Analyses. With glimepiride, confirmed/severe hypoglycaemia occurred throughout the treatment period; events
were more common during weeks 416 (Figure 2A). Furthermore, with glimepiride, hypoglycaemia occurred more frequently in patients who achieved HbA1c < 7.0%, compared
with HbA1c 7.0% at week 52, for the overall population and
for patients aged < 75 years (Figure 2B).
With saxagliptin, confirmed/severe hypoglycaemia incidence was low or absent for all HbA1c intervals; with
glimepiride, hypoglycaemia incidence was highest for the
lower HbA1c intervals (Figure 2C).

Safety and Tolerability

Saxagliptin was generally well tolerated in the present study
population. The incidence of AEs (excluding hypoglycaemia)
was similar for both treatments, and for each age group. The
proportion of patients experiencing any hypoglycaemic event
was lower with saxagliptin than with glimepiride (5.8 vs 34.8%).
The most common AEs with saxagliptin were nasopharyngitis,
arthralgia and diarrhoea (Table 3).
More saxagliptin-treated patients experienced serious AEs
(SAEs), compared with glimepiride (11.4 vs 8.9%). Four
patients experienced SAEs considered to be related to treatment (saxagliptin: cholecystitis, malignant hepatic neoplasm
and bladder calculus; glimepiride: hypoglycaemia). Discontinuations as a result of AEs were higher with saxagliptin

Volume 17 No. 7 July 2015

original article

DIABETES, OBESITY AND METABOLISM

Enrolled (n = 957)
Not randomized (n = 237)
Randomization criteria not met (n = 198)
Voluntary discontinuation (n = 28)
AE (n = 3)
Protocol non-compliance (n = 2)
Other (n = 6)
Randomized (n = 720)

Saxagliptin + metformin
(n = 360)

Glimepiride + metformin
(n = 360)

Discontinued, n = 71 (19.7%)
Voluntary discontinuation (n = 17)
Safety reasons (n = 1)
Protocol non-compliance (n = 1)
Incorrect enrolment (n = 0)
Lost to follow-up (n = 0)
Study-specific criteria (n = 33)
AE (n = 15)
Death (n = 1)
Other (n = 3)

Discontinued, n = 75 (20.8%)
Voluntary discontinuation (n = 19)
Safety reasons (n = 1)
Protocol non-compliance (n = 3)
Incorrect enrolment (n = 2)
Lost to follow-up (n = 1)
Study-specific criteria (n = 34)
AE (n = 7)
Death (n = 1)
Other (n = 7)

Completed 52-week treatment period

n = 289 (80.3%)

Completed 52-week treatment period

n = 285 (79.2%)

Figure 1. Patient disposition. AE, adverse event.

(4.5%) than with glimepiride (3.1%), and were most often

because of neoplasm (saxagliptin, n = 4: bladder transitional
cell carcinoma, colon cancer, malignant hepatic neoplasm
and lymphoma; glimepiride, n = 0) and nervous system
disorders (saxagliptin, n = 3: headache, lethargy and syncope; glimepiride, n = 4: dementia, dizziness, epilepsy and
migraine). There were two deaths in the study (saxagliptin:
myocardial infarction; glimepiride: unknown cause), which
were considered unrelated to treatment.
There were numerically more neoplasm cases on saxagliptin
vs glimepiride (n = 10 vs n = 3). With saxagliptin, events were
in multiple organ systems and occurred mostly in patients with
cancer risk factors and 6 months exposure to saxagliptin in
8/10 cases. One event was considered related to saxagliptin
treatment: a patient with hepatocellular carcinoma risk factors
developed malignant hepatic neoplasm 4.5 months after randomization. More patients in the saxagliptin group reported
fractures than in the glimepiride group (n = 9 vs n = 4). With
saxagliptin, fractures occurred in patients with prior osteoporosis risk factors or pre-existing osteoporosis; these were
associated with 9 months exposure and were not considered
treatment-related.
No differences between treatment groups were noted
for infection, opportunistic infection, skin lesion or localized oedema. In particular, there were no instances of
lymphocytopenia, thrombocytopenia or pancreatitis. More

Volume 17 No. 7 July 2015

heart failure events were observed with glimepiride compared

with saxagliptin [n = 6 (3 SAEs) vs n = 1 (1 SAE)].
Patients treated with saxagliptin showed a mean reduction
in body weight at week 52 vs baseline, compared with a mean
increase in those treated with glimepiride [0.8 kg (95% CI
1.2, 0.4) vs +1.0 kg (95% CI 0.6, 1.3)].

Discussion
Treatment of elderly patients with T2D is a unique challenge,
where the risk of hypoglycaemia and associated complications
must be balanced against the potential benefits of glucose control [16]. Although the risk of microvascular and macrovascular
events has been shown to increase significantly with HbA1c
above thresholds of 6.5 and 7.0%, respectively [17], intensive
glycaemic control may have either no influence or potentially
a detrimental effect on cardiovascular outcomes in patients
with a relatively long duration of diabetes and risk factors for
cardiovascular disease [18]. Severe hypoglycaemia has been
shown to be associated with an increased risk of cardiovascular
disease and severe ventricular arrhythmias [19,20]. As a
consequence, recent guidelines emphasize the need for individualizing HbA1c targets in elderly patients, rather than
achieving strict glycaemic control [5] and recommend that
hypoglycaemia, ability to self-manage, cognitive status, comorbidities and life expectancy are considered when treating

doi:10.1111/dom.12461 633

original article

DIABETES, OBESITY AND METABOLISM

Table 1. Demographic and baseline clinical characteristics.

Men, n (%)
Mean (s.d.) age, years
Age 75 years, n (%)
Race, n (%)
White
Other
Mean (s.d.) BMI*, kg/m2
BMI category*, n (%)
<25 kg/m2
25 and <30 kg/m2
30 kg/m2
Geographic region, n (%)
Central Europe
Latin countries
Nordic countries
Mean (s.d.) duration of T2D, years
HbA1c*
Mean (s.d.), %
Mean, mmol/mol
Mean (s.d.) metformin dose, mg
Abnormal medical history, n (%)
Medical history, most commonly affected body systems, n (%)
Musculoskeletal and connective tissue disorders
Gastrointestinal disorders
Reproductive system and breast disorders
Neoplasms
History of vascular diseases, n (%)
Hypertension
Coronary artery disease
Previous myocardial infarction
Cardiovascular accident
Stable angina
History of lipid disorder, n (%)

Saxagliptin + metformin
(n = 360)

Glimepiride + metformin
(n = 360)

All patients
(n = 720)

217 (60.3)
72.5 (5.7)
143 (39.7)

228 (63.3)
72.7 (5.4)
144 (40.0)

445 (61.8)
72.6 (5.6)
287 (39.9)

352 (97.8)
8 (2.2)
29.9 (5.0)

355 (98.6)
5 (1.4)
29.3 (4.7)

707 (98.2)
13 (1.8)
29.6 (4.9)

51 (14.2)
147 (40.8)
161 (44.7)

66 (18.3)
137 (38.1)
156 (43.3)

117 (16.3)
284 (39.4)
317 (44.0)

133 (36.9)
74 (20.6)
153 (42.5)
7.6 (6.4)

141 (39.2)
62 (17.2)
157 (43.6)
7.6 (6.0)

274 (38.1)
136 (18.9)
310 (43.1)
7.6 (6.2)

7.58 (0.67)
59
1,647 (705)
278 (77.2)

7.62 (0.65)
60
1,572 (671)
290 (80.6)

7.60 (0.66)
60
1,609 (689)
568 (78.9)

120 (33.3)
85 (23.6)
52 (14.4)
53 (14.7)

121 (33.6)
82 (22.8)
60 (16.7)
49 (13.6)

241 (33.5)
167 (23.2)
112 (15.6)
102 (14.2)

276 (76.7)
31 (8.6)
34 (9.4)
19 (5.3)
17 (4.7)
220 (61.1)

279 (77.5)
36 (10.0)
20 (5.6)
21 (5.8)
21 (5.8)
213 (59.2)

555 (77.1)
67 (9.3)
54 (7.5)
40 (5.6)
38 (5.3)
433 (60.1)

BMI, body mass index; HbA1c, glycated haemoglobin; s.d., standard deviation; T2D, type 2 diabetes.
*n = 718 (n = 359, saxagliptin; n = 359, glimepiride).
n = 715 (n = 357, saxagliptin; n = 358, glimepiride).
n = 720 (n = 360, saxagliptin; n = 360, glimepiride).

this population [21]. The evidence base for using individualized glycaemic targets, however, is limited. A recent study in
elderly patients with T2D showed vildagliptin to be superior
to placebo in achieving individualized glycaemic targets [22].
Mean HbA1c targets set by investigators were 7.0%, the
lowest target recommended in this population [21].
In the present GENERATION study, the superiority of
saxagliptin compared with glimepiride was not demonstrated
for the primary endpoint. Saxagliptin was numerically superior to glimepiride for the primary endpoint in patients aged
<75 years and numerically inferior in patients aged
75 years, with a statistically significant interaction of treatment with age. Whereas the incidence of confirmed/severe
hypoglycaemia was substantially higher with glimepiride
than with saxagliptin, more patients treated with glimepiride
achieved the HbA1c target than those treated with saxagliptin.
The incidence of hypoglycaemia with saxagliptin was low, even
for patients achieving the lowest HbA1c values. Furthermore,

634 Schernthaner et al.

with glimepiride, hypoglycaemia occurred throughout the

treatment period and across all levels of glycaemic control.
An important factor contributing to fact that the primary
endpoint was not achieved in the present study was the unexpected difference in the efficacy of glimepiride between age
groups. With glimepiride, glycaemic control was greater and
hypoglycaemia incidence was lower in patients aged 75 years,
compared with <75 years, whereas with saxagliptin the same
variables were consistent across the age groups. Although better glycaemic control was observed in patients aged 75 years
on glimepiride, this may not constitute a benefit if it is accompanied by a higher rate of hypoglycaemia.
There is a tendency for glimepiride clearance to increase with
decreasing renal function [23], which may affect glycaemic efficacy. Furthermore, hypoglycaemia incidence with glimepiride
is higher in patients with mild renal impairment than in those
with normal renal function [24]; however, in the present study,
the increased glimepiride efficacy in patients aged 75 years

Volume 17 No. 7 July 2015

original article

DIABETES, OBESITY AND METABOLISM

Table 2. Efficacy results at week 52.

HbA1c < 7.0% without confirmed/severe hypoglycaemic events, n

All patients, n (%)
<75 years, n (%)
75 years, n (%)
<7% baseline HbA1c, n (%)
7% to <8% baseline HbA1c**, n (%)
8% baseline HbA1c, n (%)
1 confirmed/severe hypoglycaemic events, n
All patients, n (%)
<75 years, n (%)
75 years, n (%)
HbA1c < 7.0%, n
All patients, n (%)
<75 years, n (%)
75 years, n (%)
HbA1c 6.5%, n
All patients, n (%)
HbA1c, adjusted mean change from baseline, n
All patients, % (95% CI)
<75 years, % (95% CI)
75 years, % (95% CI)

Saxagliptin +
metformin

Glimepiride +
metformin

359
136 (37.9)
85 (39.2)
51 (35.9)
23 (56.1)
103 (43.5)
10 (12.3)
359
4 (1.1)
2 (0.9)
2 (1.4)
356
159 (44.7)
99 (46.3)
60 (42.3)
356
66 (18.5)
353
0.44 (0.51, 0.37)
0.46 (0.54, 0.37)
0.40 (0.52, 0.28)

359
137 (38.2)
72 (33.3)
65 (45.5)
23 (53.5)
92 (40.2)
22 (25.3)
359
55 (15.3)
40 (18.5)
15 (10.5)
353
193 (54.7)
110 (51.4)
83 (59.7)
353
109 (30.9)
345
0.64 (0.71, 0.57)
0.57 (0.65, 0.48)
0.74 (0.86, 0.62)

OR*or difference(95% CI)

0.99 (0.73, 1.34); p = 0.9415*
1.29 (0.87, 1.91)*
0.67 (0.42, 1.08)*
1.15 (0.48, 2.75)*
1.14 (0.79, 1.65)*
0.41 (0.18, 0.94)*
0.06 (0.02, 0.17); p < 0.0001*
0.04 (<0.01; 0.17)*
0.12 (0.03; 0.54)*
0.67 (0.50, 0.90); p = 0.0077*
0.81 (0.56, 1.19)*
0.49 (0.31, 0.79)*
0.51 (0.36, 0.72); p < 0.0001*
0.20 (0.10, 0.30); p < 0.0001
0.11 (0.01, 0.23); p = 0.0839
0.34 (0.18, 0.51); p < 0.0001

CI, confidence interval; HbA1c, glycated haemoglobin; OR, odds ratio.

*OR for saxagliptin vs glimepiride.
Difference in adjusted mean change from baseline for saxagliptin vs glimepiride.
n = 217, saxagliptin; n = 216, glimepiride.
n = 142, saxagliptin; n = 143, glimepiride.
n = 41, saxagliptin; n = 43, glimepiride.
**n = 237, saxagliptin; n = 229, glimepiride.
n = 81, saxagliptin; n = 87, glimepiride.
n = 214, saxagliptin; n = 214, glimepiride.
n = 142, saxagliptin; n = 139, glimepiride.

could not be attributed to differences in baseline renal

function between populations, and patients with creatinine
clearance <60 ml/min had been excluded from the study.
Lack of awareness of hypoglycaemic symptoms may explain
the unexpectedly lower incidence of hypoglycaemia in patients
aged 75 years [25]. In elderly patients, fewer individual
hypoglycaemic symptoms of lower intensity are generally
reported, compared with younger adults [26], and symptoms are often non-specific in nature, including weakness,
unsteadiness, sleepiness, faintness and poor concentration
[26]; therefore, in the present study, hypoglycaemia could
have been under-reported, which may account for the reduced
hypoglycaemia incidence observed in the older age group of
glimepiride-treated patients. With saxagliptin, the number of
patients reporting hypoglycaemia may have been too low for
this effect to become apparent.
As we are not aware of other studies comparing saxagliptin
with glimepiride in elderly patients with T2D, the results of
the present study should ideally be interpreted in the context of other head-to-head studies between DPP-4 inhibitors
and sulphonylureas in elderly patients with T2D. A recent
meta-analysis [27] identified 12 randomized studies between
DPP-4 inhibitors and sulphonylureas; however, only one study
enrolled elderly patients only and none used an endpoint

Volume 17 No. 7 July 2015

combining glucose-lowering and hypoglycaemia as in the

present GENERATION study. The meta-analysis confirmed
that patients on DPP-4 inhibitors were less likely to achieve an
HbA1c target of <7% compared with those on sulphonylureas
(OR 0.91; 95% CI 0.84, 0.99), but had a lower risk of hypoglycaemia on DPP-4 inhibitors compared with sulphonylureas
(OR 0.13; 95% CI 0.11, 0.16) [27]. In the randomized study
that enrolled elderly patients with T2D, glycaemic control
with alogliptin was found to be non-inferior to glipizide
and alogliptin had a lower risk of hypoglycaemia compared
with glipizide (5.4 vs 26.0%) [28]; however, results from
the alogliptinglipizide head-to-head study were not analysed by age group, so it is difficult to relate them to those
obtained in the present study. Also recently reported were
two placebo-controlled studies in elderly patients (70 years)
with T2D treated with other DPP-4 inhibitors (vildagliptin
[22] and linagliptin [29]). The definition of hypoglycaemia
used was in line with the GENERATION study for vildagliptin
(<3.1 mmol/l) [22] and a less stringent definition was used
for linagliptin (<3.9 mmol/l) [29]. While it is not appropriate to compare results with those of the GENERATION
study directly, HbA1c reductions from baseline and low rates
of hypoglycaemia were demonstrated in both studies and
GENERATION, in elderly patients [22,29], although it

doi:10.1111/dom.12461 635

original article

DIABETES, OBESITY AND METABOLISM

Saxagliptin + metformin

A
19

20
16

16
Number of events

Glimepiride + metformin

18
15

13
12

11

10

11

10

7
5

4
0

2
0

04

48

812 1214 1416 1618 1824 2428 2832 3236 3640 4044 4448
Time to event (weeks)

30
25.0

25
18.5

Patients (%)

20
15

13.3

12.6

11.3

9.7

10
5
1.4

1.2

0.9

1.9

0.8

1.1

0
HbA1c < 7% HbA1c 7% HbA1c < 7% HbA1c 7% HbA1c < 7% HbA1c 7%
All patients
C

Age < 75 years

Age 75 years

30
25

Patients (%)

21.0
20

17.0
14.7

15
9.1

10
5

2.7
0

0
Patients (%)*: 13.3%
24.4%
HbA1c category:
< 6.5%

1.0

0
30.8%

29.2%

6.5< 7.0%

25.3%

21.4%

7.0< 7.5%

28.6%

20.8%

7.5%

Figure 2. (A) Confirmed/severe hypoglycaemic events over the 52-week treatment period (safety population); (B) incidence of confirmed/severe
hypoglycaemia by glycated haemoglobin (HbA1c) category at week 52 and age group; (C) incidence of confirmed/severe hypoglycaemic events by HbA1c
category at week 52 (safety population). *Patients achieving indicated HbA1c categories.

should be noted that the percentage of patients with hypoglycaemia was not reported by age subgroups in the two
placebo-controlled studies.
Saxagliptin add-on to metformin was generally well tolerated in patients aged 65 years, with an overall incidence of
AEs similar to glimepiride. The safety profile of saxagliptin was
consistent with that reported in previous studies in older
patients [14,15]. No pancreatitis was observed. An imbalance

636 Schernthaner et al.

in AEs of neoplasm and fracture was observed with saxagliptin

vs glimepiride; however, events occurred in patients with prior
risk factors, and the exposure to saxagliptin was short. Furthermore, the size and duration of the present study was insufficient
for evaluating rare events. Safety findings should be considered
in the context of the overall saxagliptin clinical programme,
where no imbalances in AEs of neoplasm and fracture have
been observed with saxagliptin in this age group [14,15].

Volume 17 No. 7 July 2015

original article

DIABETES, OBESITY AND METABOLISM

Table 3. Summary of adverse events (safety population).

1 AE (excluding hypoglycaemia), n (%)

1 related AE, n (%)
Deaths, n (%)
1 SAE, n (%)
1 related SAE, n (%)
Discontinuation because of AEs, n (%)
Discontinuation because of SAEs, n (%)
1 hypoglycaemic event*, n (%)
Most common AEs (3%), n (%)
Nasopharyngitis
Arthralgia
Diarrhoea
Bronchitis
Upper respiratory tract infection
Urinary tract infection
Back pain
Cough
Dizziness
Headache

Saxagliptin +
metformin
(n = 359)

Glimepiride +
metformin
(n = 359)

213 (59.3)
26 (7.2)
1 (0.3)
41 (11.4)
3 (0.8)
16 (4.5)
9 (2.5)
21 (5.8)

213 (59.3)
28 (7.8)
1 (0.3)
32 (8.9)
1 (0.3)
11 (3.1)
5 (1.4)
125 (34.8)

23 (6.4)
17 (4.7)
15 (4.2)
14 (3.9)
12 (3.3)
11 (3.1)
10 (2.8)
9 (2.5)
6 (1.7)
6 (1.7)

35 (9.7)
9 (2.5)
19 (5.3)
7 (1.9)
14 (3.9)
19 (5.3)
18 (5.0)
13 (3.6)
17 (4.7)
12 (3.3)

AE, adverse event; SAE, serious adverse event.

*Any hypoglycaemic event.

The overall cardiovascular safety with saxagliptin was previously assessed in the Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus (SAVOR)
study of 16 500 patients with T2D with a history of, or at
risk of, cardiovascular events [30]. Saxagliptin was non-inferior
to placebo for cardiovascular safety in the overall population
and in patients aged >75 years [30]. Furthermore, in the overall population, there were no imbalances in malignancies, fractures and events of pancreatitis between saxagliptin and placebo
groups. In the GENERATION study, cardiovascular events
were balanced across the two groups, although their incidence
was too low for a meaningful comparison.
Several limitations should be considered when interpreting
the results of the present study. The primary endpoint included
only patient-reported confirmed/severe hypoglycaemia; this
was deemed the most rigorous method, although the rate of
hypoglycaemia may have been underestimated; as noted above,
elderly patients may under-report hypoglycaemic symptoms.
Furthermore, a more stringent definition of hypoglycaemia was
used (blood glucose <3.0 mmol/l) than the <3.9 mmol/l value
recommended by some professional diabetes societies [31].
Patients with chronic kidney disease were excluded from the
present study because, at the time of planning, guidelines recommended that renal dysfunction was a contraindication for
metformin. Finally, glycaemic targets were based on guidelines
valid when the study protocol was developed [32] and were not
individualized. A target HbA1c of <7.0% may no longer be considered appropriate for certain elderly patients [5,21].
Overall, the GENERATION study showed that saxagliptin
provided glycaemic control with a low risk of hypoglycaemia
and was well tolerated in patients aged 65 years, inadequately
controlled on metformin. As avoidance of hypoglycaemia is a

Volume 17 No. 7 July 2015

key clinical objective in elderly patients, saxagliptin is a suitable

alternative to glimepiride in this population. Given the paucity
of data in this age group, the present study adds significant
value to the evidence available for treating elderly patients
with T2D.

Acknowledgements
Medical writing services were provided by Ioana Dumitrescu,
PhD, of QXV Communications, Macclesfield, UK and were
funded by AstraZeneca and Bristol-Myers Squibb. This study
was funded by AstraZeneca and Bristol-Myers Squibb.

Conflict of Interest
G. S. reports personal fees from AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, Janssen Pharmaceuticals,
Novo Nordisk, Sanofi, Servier Laboratories, and Takeda, outside the submitted work. M. H. reports personal fees from
Bayer, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Roche,
Sanofi-Aventis, and Takeda, outside the submitted work. C. J.
. reports personal fees from AstraZeneca during the conduct of the study and personal fees from AstraZeneca and
Bristol-Myers Squibb, outside the submitted work. L. N. reports
a research grant from AstraZeneca during the conduct of the
study and personal fees from Novo Nordisk, GlaxoSmithKline,
and Janssen-Cilag, a research grant from Sanofi, and personal
fees and research grants from AstraZeneca, Boehringer Ingelheim, and Eli Lilly, outside the submitted work. G. L. reports
personal fees from Bristol-Myers Squibb and Janssen Pharmaceuticals, outside the submitted work. S. D.-G. reports personal
fees from AstraZeneca during the conduct of the study and personal fees from AstraZeneca and Bristol-Myers Squibb, outside
the submitted work. E. H. is an employee of AstraZeneca. E. M.
is currently an employee of AstraZeneca and was employed by
Bristol-Myers Squib at the time the study took place.
G. S. was involved in planning and designing the study,
and was responsible for data interpretation and manuscript
review. M. H. was responsible for the study conduct, data
interpretation and manuscript review. C. J. . was responsible for data collection and interpretation, and manuscript
review. L. N. was responsible for data acquisition and interpretation, and manuscript review. G. L. was responsible for the
study conduct, data acquisition, analysis and interpretation and
manuscript review. S. D.-G. contributed to the conduct of the
study, the acquisition, analysis and interpretation of the data,
and reviewed the manuscript. E. H. was responsible for data
analysis, data interpretation and manuscript development. E.
M. was responsible for data analysis, data interpretation and
manuscript development.

Supporting Information
Additional Supporting Information may be found in the online
version of this article:
Figure S1. Study design.
File S1. HbA1c analyses by estimated glomerular filtration
rate sub-groups.

doi:10.1111/dom.12461 637

original article
References
1. Lecomte P. Diabetes in the elderly: considerations for clinical practice. Diabetes
Metab 2005; 31: 5S105111.
2. American Diabetes Association. Standards of medical care in diabetes-2011.
Diabetes Care 2011; 34: S1161.
3. Boussageon R, Bejan-Angoulvant T, Saadatian-Elahi M et al. Effect of intensive glucose lowering treatment on all cause mortality, cardiovascular death,
and microvascular events in type 2 diabetes: meta-analysis of randomised controlled trials. BMJ 2011; 343: d4169.
4. Seshasai SR, Kaptoge S, Thompson A et al. Diabetes mellitus, fasting glucose,
and risk of cause-specific death. N Engl J Med 2011; 364: 829841.
5. Inzucchi SE, Bergenstal RM, Buse JB et al. Management of hyperglycemia in type
2 diabetes, 2015: a patient-centered approach: update to a position statement
of the American Diabetes Association and the European Association for the
Study of Diabetes. Diabetes Care 2015; 38: 140149.
6. Bristol-Myers Squibb. Saxagliptin summary of product characteristics. Available
from URL: http://www.ema.europa.eu/docs/en_GB/document_library/Summary
_of_opinion/human/001039/WC500044316.pdf. Accessed 21 January 2015.
7. DeFronzo RA, Hissa MN, Garber AJ et al. The efficacy and safety of saxagliptin
when added to metformin therapy in patients with inadequately controlled type
2 diabetes with metformin alone. Diabetes Care 2009; 32: 16491655.
8. Goke B, Gallwitz B, Eriksson J, Hellqvist A, Gause-Nilsson I. Saxagliptin is
non-inferior to glipizide in patients with type 2 diabetes mellitus inadequately
controlled on metformin alone: a 52-week randomised controlled trial. Int J
Clin Pract 2010; 64: 16191631.
9. Scheen AJ, Charpentier G, Ostgren CJ, Hellqvist A, Gause-Nilsson I. Efficacy and
safety of saxagliptin in combination with metformin compared with sitagliptin
in combination with metformin in adult patients with type 2 diabetes mellitus.
Diabetes Metab Res Rev 2010; 26: 540549.
10. Moses RG, Kalra S, Brook D, Sockler J, Visvanathan J, Fisher SA. Saxagliptin
(SAXA) effectively reduces hba1c and is well tolerated when added to a
combination of metformin (MET) and sulfonylurea (SU). Diabetes 2012; 61:
A282.
11. Chacra AR, Tan GH, Ravichandran S, List J, Chen R. Safety and efficacy of
saxagliptin in combination with submaximal sulphonylurea versus up-titrated
sulphonylurea over 76 weeks. Diab Vasc Dis Res 2011; 8: 150159.
12. Hollander PL, Li J, Frederich R, Allen E, Chen R. Safety and efficacy of saxagliptin
added to thiazolidinedione over 76 weeks in patients with type 2 diabetes
mellitus. Diab Vasc Dis Res 2011; 8: 125135.
13. Barnett AH, Charbonnel B, Li J, Donovan M, Fleming D, Iqbal N. Saxagliptin
add-on therapy to insulin with or without metformin for type 2 diabetes
mellitus: 52-week safety and efficacy. Clin Drug Investig 2013; 33: 707717.
14. Karyekar CS, Ravichandran S, Allen E, Fleming D, Frederich R. Tolerability and
efficacy of glycemic control with saxagliptin in older patients (aged 65 years)
with inadequately controlled type 2 diabetes mellitus. Clin Interv Aging 2013;
8: 419430.
15. Doucet J, Chacra A, Maheux P, Lu J, Harris S, Rosenstock J. Efficacy and safety
of saxagliptin in older patients with type 2 diabetes mellitus. Curr Med Res Opin
2011; 27: 863869.
16. McLaren LA, Quinn TJ, McKay GA. Diabetes control in older people. BMJ 2013;
346: f2625.
17. Zoungas S, Chalmers J, Ninomiya T et al. Association of HbA1c levels with
vascular complications and death in patients with type 2 diabetes: evidence
of glycaemic thresholds. Diabetologia 2012; 55: 636643.

638 Schernthaner et al.

DIABETES, OBESITY AND METABOLISM

18. Skyler JS, Bergenstal R, Bonow RO et al. Intensive glycemic control and the
prevention of cardiovascular events: implications of the ACCORD, ADVANCE,
and VA diabetes trials: a position statement of the American Diabetes Association and a scientific statement of the American College of Cardiology
Foundation and the American Heart Association. Diabetes Care 2009; 32:
187192.
19. Goto A, Arah OA, Goto M, Terauchi Y, Noda M. Severe hypoglycaemia and
cardiovascular disease: systematic review and meta-analysis with bias analysis.
BMJ 2013; 347: f4533.
20. Stahn A, Pistrosch F, Ganz X et al. Relationship between hypoglycemic episodes
and ventricular arrhythmias in patients with type 2 diabetes and cardiovascular
diseases. Diabetes Care 2013; 37: 516520.
21. Sinclair A, Morley JE, Rodriguez-Manas L et al. Diabetes mellitus in older people:
position statement on behalf of the International Association of Gerontology
and Geriatrics (IAGG), the European Diabetes Working Party for Older People
(EDWPOP), and the International Task Force of Experts in Diabetes. J Am Med
Dir Assoc 2012; 13: 497502.
22. Strain WD, Lukashevich V, Kothny W, Hoellinger MJ, Paldanius PM. Individualised treatment targets for elderly patients with type 2 diabetes using
vildagliptin add-on or lone therapy (INTERVAL): a 24 week, randomised,
double-blind, placebo-controlled study. Lancet 2013; 382: 409416.
23. Accord Healthcare. Glimepiride 4 mg tablets: summary of product characteristics. Available from URL: http://www.medicines.org.uk/EMC/history/25845/SPC/
Glimepiride+4+mg+Tablets. Accessed 21 January 2015.
24. Schernthaner G, Grimaldi A, Di MU et al. GUIDE study: double-blind comparison
of once-daily gliclazide MR and glimepiride in type 2 diabetic patients. Eur J Clin
Invest 2004; 34: 535542.
25. Bremer JP, Jauch-Chara K, Hallschmid M, Schmid S, Schultes B. Hypoglycemia
unawareness in older compared with middle-aged patients with type 2 diabetes. Diabetes Care 2009; 32: 15131517.
26. McAulay V, Deary IJ, Frier BM. Symptoms of hypoglycaemia in people with
diabetes. Diabet Med 2001; 18: 690705.
27. Zhang Y, Hong J, Chi J, Gu W, Ning G, Wang W. Head-to-head comparison
of dipeptidyl peptidase-IV inhibitors and sulfonylureas a meta-analysis from
randomized clinical trials. Diabetes Metab Res Rev 2014; 30: 241256.
28. Rosenstock J, Wilson C, Fleck P. Alogliptin versus glipizide monotherapy in
elderly type 2 diabetes mellitus patients with mild hyperglycaemia: a prospective, double-blind, randomized, 1-year study. Diabetes Obes Metab 2013; 15:
906914.
29. Barnett AH, Huisman H, Jones R, von Eynatten M, Patel S, Woerle HJ. Linagliptin
for patients aged 70 years or older with type 2 diabetes inadequately controlled with common antidiabetes treatments: a randomised, double-blind,
placebo-controlled trial. Lancet 2013; 382: 14131423.
30. Scirica BM, Bhatt DL, Braunwald E et al. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes mellitus. N Engl J Med 2013; 369:
13171326.
31. American Diabetes Association Workgroup on Hypoglycemia. Defining and
reporting hypoglycemia in diabetes: a report from the American Diabetes
Association Workgroup on Hypoglycemia. Diabetes Care 2005; 28: 12451249.
32. Nathan DM, Buse JB, Davidson MB et al. Medical management of hyperglycemia
in type 2 diabetes: a consensus algorithm for the initiation and adjustment
of therapy: a consensus statement of the American Diabetes Association and
the European Association for the Study of Diabetes. Diabetes Care 2009; 32:
193203.

Volume 17 No. 7 July 2015