Liraglutide improves treatment satisfaction in people with type2 diabetes compared with sitagliptin, each as an add on to metformin
Liraglutide improves treatment satisfaction in peoplewith Type 2 diabetes compared with sitagliptin, eachas an add on to metformin
M. Davies, R. Pratley*, M. Hammer†, A. B. Thomsen† and R. Cuddihy‡
University of Leicester, Leicester, UK, *University of Vermont, Burlington, VT, USA, †Novo Nordisk A ⁄ S, Bagsvaerd, Denmark and ‡International Diabetes Center,Minneapolis, MN, USA
Patient-reported outcomes from clinical trials offer insight into the impact of disease on health-related quality of life,
including treatment satisfaction. This patient-reported outcomes evaluation was a substudy of a 26-week randomized, open-label trial comparing the once-daily injectable human GLP-1 analogue liraglutide with once-daily oral sitagliptin, both added tometformin. The patient reported outcomes substudy aimed to evaluate treatment satisfaction using the Diabetes TreatmentSatisfaction Questionnaire (DTSQ) at baseline and 26 weeks.
In the main 26-week randomized, open-label study (n = 658), liraglutide, 1.2 or 1.8 mg, injected with a pen, led to
greater HbA1c reduction than oral sitagliptin, 100 mg once daily, both added to metformin = 1500 mg daily: mean HbA1creduction was 1.5, 1.2 and 0.9% (7, 10 and 14 mmol ⁄ mol) for liraglutide 1.8 mg, 1.2 mg and sitagliptin, respectively(P < 0.0001 for both liraglutide doses vs. sitagliptin) and liraglutide patients lost more weight (3 vs.1 kg; P < 0.0001). In thispatient-reported outcomes substudy (liraglutide 1.8 mg, n = 171; 1.2 mg, n = 164; sitagliptin, n = 170) DTSQ scores wereanalyzed by ANCOVA with treatment and country as fixed effects and baseline value as covariate.
Overall treatment satisfaction, calculated by adding satisfaction scores for ‘current treatment’, ‘convenience’,
‘flexibility’, ‘understanding’, ‘recommend’, and ‘continue’, improved in all groups at 26 weeks; greater improvement withliraglutide (4.35 and 3.51 vs. 2.96; P = 0.03 for liraglutide 1.8 mg vs. sitagliptin) may reflect greater HbA1c reduction andweight loss. Patients perceived themselves to be hyperglycaemic significantly less frequently with liraglutide 1.8 mg(difference = )0.88; P < 0.0001) and 1.2 mg (difference = )0.49; P = 0.01). Perceived frequency of hypoglycaemia was similaracross all groups.
Injectable liraglutide may lead to greater treatment satisfaction than oral sitagliptin, potentially by facilitating
greater improvement in glycaemic control, weight loss and ⁄ or perception of greater treatment efficacy.
GLP-1, glucagon-like peptide-1; HRQoL, health-related quality of life
disease course [1,2] as well as influencing patients’ health-related
quality of life (HRQoL), including treatment satisfaction [3,4].
In addition to the multiple physical sequelae of diabetes mellitus
Patient-reported outcomes from clinical trials provide
and its treatment, psychosocial factors significantly affect the
information on the impact of a disease on HRQoL and mayidentify the extent to which treatment meets patients’ needs andexpectations. Patient-reported outcomes data typically reflect
Correspondence to: Melanie Davies, Professor of Diabetes Medicine,
perceptions of the efficacy and tolerability of treatment, as well as
University Hospitals of Leicester, Leicester Royal Infirmary, Infirmary Square,Leicester LE1 5WW, UK. E-mail: [email protected]
treatment preferences and, as such, complement physician
ª 2011 The Authors.
Diabetic Medicine ª 2011 Diabetes UK
Satisfaction with liraglutide vs. sitagliptin • M. Davies et al.
appraisals of the clinical value of specific therapies. As greater
plasma glucose, indices of B-cell function, body weight, fasting
treatment satisfaction may be associated with improved
lipids and treatment satisfaction. Safety assessments included
adherence to treatment and self-management behaviour ,
adverse events and self-reported hypoglycaemia.
patient-reported measures of treatment satisfaction offer a
Treatment satisfaction was assessed at baseline and 26 weeks,
clinically valuable indication of the likelihood that patients will
or on withdrawal if this occurred prior to study completion, using
choose, and adhere to, a given treatment. In the case of incretin-
the status version of the validated Diabetes Treatment
based therapies, a relatively new class of anti-diabetic agents,
Satisfaction Questionnaire (DTSQs) in a subgroup that
healthcare practitioners require quantitative and qualitative
included all exposed subjects from the 10 countries where
patient feedback to guide prescribing decisions. This is
patient-reported outcomes data were gathered (n = 505 ⁄ 658;
particularly important given the advantages some of these
77% of exposed subjects). The remaining three countries were
agents appear to offer over traditional oral anti-diabetic drugs
excluded from the patient-reported outcomes substudy because
and insulin, namely significantly lower risks of hypoglycaemia
of lack of local, linguistically validated patient-reported
and weight gain [6,7], both of which may affect treatment
outcomes measures. As the DTSQ was completed either at
satisfaction [8,9]. The present patient-reported outcomes
26 weeks or on withdrawal, no subjects were excluded from the
evaluation was conducted in a predefined subpopulation of a
patient-reported outcomes analysis. DTSQs consisted of eight
randomized controlled trial that compared the efficacy and safety
items, each analysed individually, while overall treatment
of approved doses of two incretin-based therapies, liraglutide, an
satisfaction was calculated by summing individual scores from
injectable once-daily human glucagon-like peptide-1 (GLP-1)
six items: satisfaction with ‘current treatment’, ‘convenience’,
analogue and the orally administered dipeptidyl peptidase 4
‘flexibility’, ‘understanding’, ‘recommend’ and ‘continue’, where
(DPP-4) inhibitor sitagliptin, each added on to metformin in
‘recommend’ and ‘continue’ indicate the likelihood that a
patients with Type 2 diabetes poorly controlled on metformin
patient would recommend or continue treatment after study
completion. Each item was scored on a scale from 0 (‘very
In patients with Type 2 diabetes, adding liraglutide to failing
dissatisfied ⁄ inconvenient’) to 6 (‘very satisfied ⁄ convenient’); a
metformin monotherapy improves glycaemic control and lowers
higher score indicates greater treatment satisfaction. Perceived
weight with low hypoglycaemic risk, while reducing systolic
frequency of hyperglycaemia and hypoglycaemia was measured
blood pressure . Synergism between sitagliptin and
separately using one item each, also scored on a scale from 0
metformin also seems logical, as metformin stimulates GLP-1
(‘none of the time’) to 6 (‘most of the time’), where lower scores
secretion , while sitagliptin increases the half-life of
indicate lower frequency of perceived hypo- or hyperglycaemia
endogenous GLP-1. In clinical trials, adding sitagliptin to
and therefore better perceived glycaemic control. Subjects did not
metformin therapy decreased key glycaemic control parameters
receive guidance on how to determine whether their blood
with minimal hypoglycaemia [12–14]. However, the magnitude
glucose level was high or low; they simply answered the questions
of HbA1c reduction appears significantly lower with DPP-4
‘How often have you felt that your blood sugars have been
inhibitors than with GLP-1 receptor agonists [10,15]. Other key
unacceptably high ⁄ low recently?’ Patients were not specifically
differentiators between these two therapy classes are their effects
asked about their appetite or about symptoms of nausea.
on weight and mode of administration: GLP-1 receptor agonists
Treatment satisfaction scores were analysed using an
facilitate weight loss whereas DPP-4 inhibitor monotherapy is
ANCOVA model, with treatment and country as fixed effects
usually weight neutral  and, while GLP-1 receptor agonists are
and baseline value as covariate, with no imputation for missing
injected, DPP-4 inhibitors are taken orally. Although it is often
stated that patients resist injectable therapies, published datasuggest this is not by any means a universal finding [4,5,16] and it
is possible that this stance may at times be overvalued,representing a form of physician-driven clinical inertia rather
Baseline characteristics of the full analysis set for the main study
than an evidence-based concern .
population and patient-reported outcomes subpopulation werecomparable and treatment group demographics well balanced.
Baseline mean HbA
1c values were 8.4% (68 mmol ⁄ mol) in both
liraglutide groups and 8.5% (69 mmol ⁄ mol) in the sitagliptin
Liraglutide, 1.2 or 1.8 mg, injected once daily using a pen device,
group and baseline weights were 93.7, 94.6 and 93.1 kg in
was compared with oral sitagliptin, 100 mg once daily, both
the 1.2 mg, 1.8 mg and sitagliptin groups, respectively.
added to a stable dose of metformin (‡ 1500 mg daily), in a
Corresponding body mass indices were 32.6, 33.1 and
26-week randomized, open-label study carried out in 13
32.6 kg ⁄ m2. The main study randomized 665 individuals, of
countries. Study design, methods, efficacy and safety results are
which 658 were exposed to at least one dose of trial product,
reported elsewhere . The primary efficacy endpoint was
and 554 (83%) completed the trial. A total of 505 subjects
change in HbA1c from baseline to week 26, with secondary
(liraglutide 1.2 mg n = 164; liraglutide 1.8 mg n = 171;
endpoints including proportion of subjects reaching HbA1c
sitagliptin n = 170) were included in the current patient-
< 7% (53 mmol ⁄ mol) and £ 6.5% (48 mmol ⁄ mol), fasting
In this trial, liraglutide led to significantly greater reduction in
1.8 mg than sitagliptin (difference = )0.88; P < 0.0001) and
HbA1c than sitagliptin (mean HbA1c reduction: 1.50, 1.24 and
the same was found when comparing the 1.2-mg dose of
0.90% (7, 10 and 14 mmol ⁄ mol) for liraglutide 1.8 mg, 1.2 mg
liraglutide with sitagliptin (difference = )0.49; P = 0.01). The
and sitagliptin, respectively; P < 0.0001 for both liraglutide doses
perceived frequency of hypoglycaemia was similar across all
vs. sitagliptin). The proportion of patients reaching HbA1c < 7%
(53 mmol ⁄ mol) was 54.6, 43.4 and 22.4%, respectively, andliraglutide-treated subjects lost significantly more weight (3 vs.
1 kg). Treatment-emergent adverse events occurred in 66.1%(1.2 mg) and 72.9% (1.8 mg) of liraglutide-treated subjects vs.
While all groups reported an increase in treatment satisfaction,
58.0% of sitagliptin-treated patients. For liraglutide, the majority
subjects receiving liraglutide 1.8 mg reported significantly
of excess adverse events were early gastrointestinal side effects,
greater improvement in overall treatment satisfaction than
typically nausea that was mostly mild and transient ;
those taking sitagliptin, despite the fact that liraglutide was
although nausea occurred more frequently during the first few
injected while sitagliptin was oral, and that treatment-emergent
weeks with liraglutide (21–27%) than with sitagliptin (5%),
adverse events occurred in more liraglutide patients. This could
symptoms had decreased to levels observed with sitagliptin
reflect patients’ recognition that the GLP-1 analogue offered
(< 3%) by the end of the trial. The most common adverse events
better control of hyperglycaemia and the potential for weight
in sitagliptin-treated patients were nasopharyngitis and
loss, although further data are needed to confirm this. Our data
headache, reported in 11.9 and 10.0% of patients, respectively.
also highlight the positive impact of improved glycaemic control
The proportion of subjects experiencing hypoglycaemia (mostly
on treatment satisfaction. Of interest, there was no difference
minor) was low and comparable in all groups .
between liraglutide and sitagliptin on DTSQ items relating to
Table 1 presents the patient-reported outcomes results.
treatment convenience and flexibility, indicating that patients
Overall, treatment satisfaction was comparable between
were no less satisfied with the injectable than the oral agent.
groups at baseline and improved in all groups after 26 weeks.
These findings concur with much of the published literature in
However, improvement in overall treatment satisfaction was
suggesting that patients may prefer an injected to an oral therapy
significantly greater with liraglutide 1.8 mg (4.35) than
if it leads to greater improvement in glycaemic control,
sitagliptin (2.96) [between-group difference = 1.39 (95% CI
perception of greater treatment efficacy and ⁄ or facilitates
0.13; 2.64); P = 0.03]; differences in overall treatment
weight loss [4,5,16]. As obese subjects with Type 2 diabetes
satisfaction between liraglutide 1.2 mg and sitagliptin, and
report poorer health status and greater symptom impact than
between the two liraglutide doses, were not significant. Patients
non-obese patients [18,19], the improvement in HRQoL
reported significantly greater improvement in treatment
afforded by weight-lowering therapies may be particularly
satisfaction with liraglutide 1.8 mg than sitagliptin on three
welcome. Our finding of reduced perceived frequency of
hyperglycaemia with liraglutide is also in accordance with data
‘recommend’ (difference = 0.41; P = 0.003) and ‘continue’
showing that the association between HbA1c and HRQoL may
(difference = 0.44; P = 0.01). Patients perceived themselves to
be mediated by the perceived frequency of hyper- and
be hyperglycaemic significantly less frequently with liraglutide
Table 1 Patient-reported outcomes summary
)0.49 ()0.86; )0.12) P = 0.01 )0.88 ()1.25; )0.51) P < 0.0001
Data are differences in Diabetes Treatment Satisfaction Questionnaire score least square means between weeks 0 and 26.
*Columns on the right show estimated treatment difference with 95% confidence intervals.
P-values indicate statistical significance at the 5% level.
ª 2011 The Authors.
Diabetic Medicine ª 2011 Diabetes UK
Satisfaction with liraglutide vs. sitagliptin • M. Davies et al.
Despite the potential for greater glucose-lowering efficacy
and weight loss with GLP-1 receptor agonists compared withDPP-4 inhibitors, and the reported effects on patient
The authors accept direct responsibility for this paper and are
satisfaction and adherence, some clinicians remain hesitant
grateful for the contribution made by Watermeadow Medical
to use GLP-1 receptor agonists, perhaps perceiving injectable
(supported by Novo Nordisk A ⁄ S, Bagsvaerd, Denmark) in
therapies to be more complex and less desirable than a once-
developing the draft manuscript from an agreed outline and for
daily oral medication. However, this reluctance to prescribe
GLP-1 receptor agonists fails to take into account the fact thatmany commonly used anti-diabetic therapies are associated
with weight gain and hypoglycaemia, both of whichnegatively affect patient quality of life, and that weight gain
1 Davis WK, Hess GE, Hiss RG. Psychosocial correlates of survival in
itself may exacerbate other components of the metabolic
diabetes. Diabetes Care 1988; 11: 538–545.
2 Lustman PJ, Anderson RJ, Freedland KE, De Groot M, Carney RM,
Clouse RE. Depression and poor glycemic control: a meta-analytic
Patient-reported outcomes and treatment satisfaction are
review of the literature. Diabetes Care 2000; 23: 934–942.
important factors to consider when choosing a glucose-
3 Bradley C, Speight J. Patient perceptions of diabetes and diabetes
lowering therapy for patients with Type 2 diabetes. This study
therapy: assessing quality of life. Diabetes Metab Res Rev 2002; 18:
provides evidence of greater improvement in treatment
4 Nicolucci A, Cucinotta D, Squatrito S, Lapolla A, Musacchio N,
satisfaction with an injectable GLP-1 agent, liraglutide 1.8 mg,
Leotta S et al.; QuoLITy Study Group. Clinical and socio-eco-
than an oral DPP-4 inhibitor, sitagliptin, potentially by
nomic correlates of quality of life and treatment satisfaction in
facilitating greater improvement in glycaemic control and
patients with type 2 diabetes. Nutr Metab Cardiovasc Dis 2009;
weight loss. These results challenge the perception that patients
‘prefer’ oral to injected glucose-lowering therapies.
5 Peyrot M, Rubin RR. How does treatment satisfaction work?:
modeling determinants of treatment satisfaction and preference.
Diabetes Care 2009; 32: 1411–1417.
6 Ahren B. Emerging dipeptidyl peptidase-4 inhibitors for the
treatment of diabetes. Expert Opin Emerg Drugs 2008; 13: 593–
MD has acted as consultant, advisory board member and
speaker for Novartis, Novo Nordisk, Sanofi-Aventis, Lilly,
7 Lovshin J, Drucker D. Incretin-based therapies for type 2 diabetes
mellitus. J Nat Rev Endocrinol 2009; 5: 262–269.
Merck Sharp & Dohme, Roche, BMS and speaker for Servier.
8 Marrett E, Stargardt T, Mavros P, Alexander C. Patient-reported
She has received grants in support of investigator and
outcomes in a survey of patients treated with oral antihypergly-
investigator-initiated trials from Novartis, Novo Nordisk,
caemic medications: associations with hypoglycaemia and weight
Sanofi-Aventis, Lilly, Pfizer, Merck Sharp & Dohme,
gain. Diabetes Obes Metab 2009; 11: 1138–1144.
GlaxoSmithKline and Servier. RP has received research
9 Pollack MF, Purayidathil FW, Bolge SC, Williams SA. Patient-
grants, funds for speaking, honoraria and consulting fees
associations with adherence; treatment satisfaction and health-
from Novartis and has ownership interests in Novartis;
related quality of life. Diabetes Res Clin Pract 2010; 87: 204–
research grants, speakers fees, honoraria and consulting fees
from Takeda; research grants speakers fees, honoraria and
10 Pratley RE, Nauck M, Bailey T, Montanya E, Cuddihy E, Filetti S
consulting fees from Novartis fees from Merck; research
et al., for the 1860-LIRA-DPP-4 Study Group. Liraglutide versussitagliptin for patients with type 2 diabetes who did not have
grants, honoraria and consulting fees from Roche; research
adequate glycaemic control with metformin: a 26-week, rando-
grants, honoraria and consulting fees from GlaxoSmithKline;
mised, parallel-group, open-label trial. Lancet 2010; 375: 1447–
research grants from Eli Lilly; research grants, speakers fees,
honoraria and consulting fees from Novo Nordisk; research
11 Nauck M, Frid A, Hermansen K, Shah NS, Tankova T, Mitha IH
grants from Mannkind, Sanofi-Aventis and Pfizer, and
et al.; LEAD-2 Study Group Efficacy and safety comparisonof liraglutide, glimepiride, and placebo, all in combination
honoraria and consulting fees from Eisai, Glenmark and
with metformin, in type 2 diabetes: the LEAD (liraglutide effect and
AstraZeneca ⁄ BMS. MH is an employee and stakeholder of
action in diabetes)-2 study. Diabetes Care 2009; 32: 84–90.
Novo Nordisk A ⁄ S. ABT is an employee of Novo Nordisk
12 Charbonnel B, Karasik A, Liu J, Wu M, Meininger G. Efficacy and
and has shares in the company. RC serves as PI or co-
safety of the dipeptidyl peptidase-4 inhibitor sitagliptin added to
investigator for sponsored clinical trials research for Amylin,
ongoing metformin therapy in patients with type 2 diabetes inade-quately controlled with metformin alone. Diabetes Care 2006; 29:
13 Raz I, Chen Y, Wu M, Hussain S, Kaufman KD, Amatruda JM et al.
Johnson ⁄ Lifescan, Mannkind, Medtronic, Novo Nordisk,
Efficacy and safety of sitagliptin added to ongoing metformin
Quotient Diagnostics, ResMed, Roche, Sanofi-Aventis and
therapy in patients with type 2 diabetes. Curr Med Res Opin 2008;
Takeda; is an advisory board member for Abbott, Bayer,
14 Scott R, Loeys T, Davies MJ, Engel SS. Efficacy and safety of
CeQur, Eli Lilly, Novo Nordisk and Roche; receives support
sitagliptin when added to ongoing metformin therapy in patients
for educational activities from Lifescan, Eli Lilly, Merck,
with type 2 diabetes. Diabetes Obes Metab 2008; 10: 959–
15 Gilbert MP, Pratley RE. Efficacy and safety of incretin-based ther-
18 Matza L, Yurgin N, Boye KS, Malley K, Shorr JM. Obese versus
apies in patients with type 2 diabetes mellitus. Eur J Intern Med
non-obese patients with type 2 diabetes: patient-reported outcomes
and utility of weight change. Curr Med Res Opin 2007; 23: 2051–
16 Houlden R, Ross S, Harris S, Yale JF, Sauriol L, Gerstein HC.
Treatment satisfaction and quality of life using an early insuliniza-
19 Sundaram M, Kavookjian J, Patrick J, Miller LA, Madhavan SS,
tion strategy with insulin glargine compared to an adjusted oral
Scott VG. Quality of life, health status and clinical outcomes in
therapy in the management of Type 2 diabetes: The Canadian
type 2 diabetes patients. Qual Life Res 2007; 16: 165–167.
INSIGHT Study. Diabetes Res Clin Pract 2007; 78: 254–258.
20 Kleefstra N, Ubink-Veltmaat L, Houweling S, Groenier KH,
17 Kruger D, Spollett G. Addressing barriers to timely intensification
Meyboom-de Jong B, Bilo HJ. Cross-sectional relationship
of diabetes care: the relationship between clinical inertia and patient
between glycaemic control, hyperglycaemic symptoms and quality of
behavior. Consultant 2009; 49: S20–25.
life in type 2 diabetes (ZODIAC-2). Neth J Med 2005; 63: 215–221.
ª 2011 The Authors.
Diabetic Medicine ª 2011 Diabetes UK
ACKNOWLEDGMENT OF ORDER AND TERMS AND CONDITIONS OF SALE This is to acknowledge receipt of your ("Buyer") purchase order for the specified products (the "Products") of ELANTAS PDG,INC. ("Seller") and to set forth the terms and conditions ("Terms and Conditions")of sale for this order. If the Terms and Conditions of this acknowledgment differ in any way from the t
Status: Site of Special Scientific Interest (SSSI) notified under Section 28 of theWildlife and Countryside Act 1981Local Planning Authorities: WEST WILTSHIRE DISTRICT COUNCIL, WiltshireCounty CouncilDescription and Reasons for Notification:Inwood is a structurally varied and botanically rich example of southern calcareousash-wych elm and dry ash-maple woodland. It supports an extremely rich gro