Original Investigation
Kidney Disease End Points in a Pooled Analysis of Individual
Patient–Level Data From a Large Clinical Trials Program of the
Dipeptidyl Peptidase 4 Inhibitor Linagliptin in Type 2 Diabetes
Mark E. Cooper, MBBS, PhD,1 Vlado Perkovic, MBBS, PhD,2 Janet B. McGill, MD,3
Per-Henrik Groop, MD, DMSC,1,4,5 Christoph Wanner, MD,6 Julio Rosenstock, MD,7
Uwe Hehnke, MSc,8 Hans-Juergen Woerle, MD,8 and Maximilian von Eynatten, MD8
Background: Although assessment of cardiovascular safety is mandated by regulatory agencies for the
development of new drugs to treat type 2 diabetes, evaluation of their renal safety has been relatively neglected.
Study Design: Individual patient–level data pooled analysis of 13 phase 2 or 3 randomized, double-blind,
placebo-controlled, clinical trials of the dipeptidyl peptidase 4 inhibitor linagliptin.
Setting & Participants: Participants who participated in any of 13 randomized clinical trials and fulfilled predefined inclusion/exclusion criteria, such as being drug-naive (hemoglobin A1c, 7.0%-11.0% [53-97 mmol/mol])
or being on background glucose-lowering therapy (hemoglobin A1c, 6.5%-10.5% [48-91 mmol/mol]).
Intervention: Of 5,466 consenting individuals with inadequately controlled type 2 diabetes, 3,505 received
linagliptin, 5 mg/d, and 1,961 received placebo.
Outcomes: The primary kidney disease outcome was defined as first occurrence during the study of
6 predefined safety end points: new onset of moderate elevation of albuminuria (urinary albumin-creatinine
ratio [ACR] .30 mg/g with baseline values # 30 mg/g), new onset of severe elevation of albuminuria
(ACR . 300 mg/g with baseline values # 300 mg/g), reduction in kidney function (serum creatinine increase
to $250 mmol/L from a baseline value , 250 mmol/L), halving of estimated glomerular filtration rate (loss of
baseline eGFR . 50%), acute renal failure (ascertained from diagnostic codes), or death from any cause.
Measurements: Albuminuria was assessed using ACR. GFR was estimated using the CKD-EPI (Chronic
Kidney Disease Epidemiology Collaboration) equation.
Results: Cumulative exposure (person-years) was 1,751 for linagliptin and 1,055 for placebo. The primary
composite outcome occurred in 448 (12.8%) and 306 (15.6%) participants in the linagliptin and placebo
groups, respectively. Linagliptin treatment significantly reduced the hazard of kidney disease events by 16%
compared with placebo (HR, 0.84; 95% CI, 0.72-0.97; P 5 0.02).
Limitations: Retrospective and hypothesis-generating study involving short- to midterm clinical trials.
Conclusions: Linagliptin was not associated with increased kidney disease risk in patients with type 2
diabetes. The potential of this drug to improve kidney disease outcomes warrants further investigation.
Am J Kidney Dis. 66(3):441-449. ª 2015 The Authors. Published by Elsevier Inc. on behalf of the National
Kidney Foundation, Inc. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
INDEX WORDS: Dipeptidyl peptidase 4 (DPP-4) inhibition; linagliptin; kidney disease end points; albuminuria;
renal function; renal risk; type 2 diabetes mellitus (T2DM); hyperglycemia; glucose control; glucose-lowering
therapy; pooled analysis.
T
he number of new medications available for the
treatment of type 2 diabetes has substantially
increased over the past 2 decades.1 The primary goal
of these agents is to effectively lower elevated blood
glucose concentrations in order to reduce the risk of
long-term complications that result from chronic
hyperglycemia.2,3 However, it is equally critical for
any new agent to also be proved to be safe. Although
specific aspects of toxicology, including renal
handling of drugs likely to be administered for years,
are routinely evaluated in the very early stages of drug
development, it is only after sufficient clinical exposure that potential safety concerns might become
apparent. As an example, systematic reviews and
From the 1Baker IDI Heart and Diabetes Institute, Melbourne,
Victoria; 2George Institute for Global Health, University of Sydney, New South Wales, Australia; 3Washington University in St
Louis, St Louis, MO; 4Folkhälsan Institute of Genetics, Folkhälsan
Research Center, Biomedicum Helsinki; 5Division of Nephrology,
Department of Medicine, Helsinki University Central Hospital,
Helsinki, Finland; 6University of Würzburg, Würzburg, Germany;
7
Dallas Diabetes and Endocrine Center at Medical City, Dallas,
TX; and 8Boehringer Ingelheim, Ingelheim, Germany.
Received May 16, 2014. Accepted in revised form March 5,
2015. Originally published online May 7, 2015.
Address correspondence to Mark E. Cooper, MBBS, PhD,
Baker IDI Heart and Diabetes Institute, PO Box 6492, Melbourne,
VIC, Australia. E-mail: [email protected]
2015 The Authors. Published by Elsevier Inc. on behalf of the
National Kidney Foundation, Inc. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).
0272-6386
http://dx.doi.org/10.1053/j.ajkd.2015.03.024
Am J Kidney Dis. 2015;66(3):441-449
441
Cooper et al
meta-analyses of clinical trials pointed out the potential for increased risks of myocardial infarction
associated with the thiazolidinedione rosiglitazone.4,5
Largely as a result of these types of concerns, regulatory requirements issued by the US Food and Drug
Administration (FDA) in 2008 stipulate that new type
2 diabetes drugs should rule out an unacceptable increase in cardiovascular risk in phase 2 and phase 3
clinical trials prior to drug approval.6-8
Microvascular complications, such as retinopathy,
nephropathy, and neuropathy, are widespread in type
2 diabetes and account for significant morbidity and
mortality.9 Therefore, current treatment goals in
diabetes include prevention of microvascular complications such as preservation of kidney function.2,3
However, an approach to assess kidney disease end
points for novel diabetes drugs is not generally
applied. More commonly, specific renal data have
been obtained from secondary analyses of cardiovascular outcome trials with several studies identifying effects on glomerular filtration rate (GFR),
although this was not the primary end point of those
trials.10-12 Despite the value of acquiring these
important data, such long-term renal safety evidence
has often not been a priority of the clinical trial program, but has only become available several years
after drug approval.
Linagliptin, a novel member of the dipeptidyl
peptidase 4 (DPP-4) inhibitor class, has previously
been shown to significantly improve glucose control
without causing weight gain or increasing hypoglycemia risk.13-15 In accordance with the guidance by
the FDA, an early cardiovascular meta-analysis was
performed that ruled out an unacceptable increase in
cardiovascular risk for this drug and linagliptin was
consequently approved in the United States in May
2011.16 Unlike other members of the DPP-4 inhibitor
class, linagliptin is not primarily cleared by the kidney
and can be prescribed to patients with type 2 diabetes
at one single dose irrespective of kidney function.15,17
Such pharmacologic qualities support the use of linagliptin in a broad range of patients with type 2 diabetes,
including those with increased prevalence and risk of
renal microvascular complications, as well as in patients with declining kidney function.18 The objective
of this study was to explore kidney disease end points
in a large set of patients with type 2 diabetes treated
with linagliptin. Comprehensive assessments of safety
events were performed by developing a systematic and
innovative approach based on individual-patient data
from a large clinical trials program.
METHODS
Study Design and Data Source
This pooled kidney disease analysis included all randomized,
double-blind, and placebo-controlled clinical trials of linagliptin of
442
12-week or longer duration for which database lock of either
predefined interim or final analysis was completed before February
13, 2011. Individual-patient data from 2 phase 2 and 11 phase 3
trials were included in the data set only if they received either
linagliptin at a dose of 5 mg once daily or placebo. Studies were
conducted globally over periods of 12 to 76 weeks as monotherapy
or in combination with other glucose-lowering agents in patients
with inadequately controlled type 2 diabetes. Open-label extension
periods of primary double-blind randomized clinical trials were not
considered (Table S1, available as online supplementary material).
Detailed study designs and primary and secondary efficacy and
safety results of the individual studies have been published previously (Table S1).19
All patients provided written informed consent. Local ethics
committees/institutional review boards reviewed and approved all
study protocols. All studies were conducted in accordance with
Good Clinical Practice guidelines and the principles of the
Declaration of Helsinki.
Setting and Participants
Eligibility criteria for each of the 13 trials were similar. Most
relevant common inclusion criteria were age 18 years or older or
21 years or older, diagnosis of type 2 diabetes, and body mass
index # 40 or #45 kg/m2. At screening, hemoglobin A1c levels
ranged either from 7.0% to 11.0% (53-97 mmol/mol) in treatmentnaive participants or from 6.5% to 10.5% (48-91 mmol/mol) in
participants previously treated with one or more glucose-lowering
therapy. In the majority of trials, previous glucose-lowering therapies, if any, had to be unchanged for at least 8 weeks prior to
informed consent. Most relevant common exclusion criteria at
screening included the following: decreased hepatic function
defined by serum levels of either alanine transaminase, aspartate
transaminase, or alkaline phosphatase more than 3 times the upper
limit of normal; myocardial infarction, stroke, or transient
ischemic attack within the previous 3 or 6 months; any requirement for hemodialysis within the previous 3 months; and kidney
transplantation.
Individuals whose blood glucose levels were not adequately
controlled during each of the 13 trials received additional rescue
therapy to ensure overall safety, as appropriate.
The pooled population consisted of all randomly assigned individuals (n 5 5,466) who received at least one dose of study drug
(treated set: linagliptin group, n 5 3,505; placebo group, n 5 1,961).
Kidney Disease End Points
Based on clinical guidelines or recommendations made by
medical associations and regulatory bodies,2,3,7,20 we defined the
primary composite outcome as first occurrence of 6 individual and
clinically relevant kidney disease end points: (1) new onset of
moderate elevation of albuminuria (urinary albumin-creatinine ratio
[ACR] . 30 mg/g at any time during study conduct with baseline
values # 30 mg/g), (2) new onset of severe elevation of albuminuria
(ACR . 300 mg/g at any time during study conduct with baseline
values # 300 mg/g), (3) reduction in kidney function (serum
creatinine increase to $250 mmol/L [$2.8 mg/dL] from a baseline
value , 250 mmol/L as defined by European Medicines Agency;
increase observed at a minimum of 2 consecutive visits during study
conduct with a between-visit time window of at least 4 weeks), (4)
halving of estimated GFR (eGFR; loss of baseline eGFR . 50% as
defined by the FDA, and observed at a minimum of 2 consecutive
visits during study conduct with a between-visit time window of at
least 4 weeks), (5) incidence of acute renal failure, and (6) death from
any cause. Although similar end points have previously been
interpreted as efficacy parameters in renal studies,10,11 they were
assessed as safety outcomes in the linagliptin program.
Individual components of the primary composite outcome were
defined as secondary outcomes.
Am J Kidney Dis. 2015;66(3):441-449
Kidney Disease End Points and Linagliptin
Kidney Disease Assessments
Kidney function was determined with the CKD-EPI (Chronic
Kidney Disease Epidemiology Collaboration) equation.21 Because
the majority of the linagliptin development program was conducted in
the era before creatinine assay standardization, 1 of the 2 utilized
central laboratories did not perform standardization to isotopedilution mass spectrometry (IDMS) values for the measurement of
serum creatinine. Therefore, serum creatinine values from this laboratory were standardized for appropriate use in the CKD-EPI equation
(ie, reduction by 5%). Albuminuria was determined at each study visit
by ACR from a spot urine sample at baseline and during the randomized treatment period. Normoalbuminuria was defined as baseline ACR # 30 mg/g, moderate elevation of albuminuria was defined
as baseline ACR . 30 to 300 mg/g, and severe elevation of albuminuria was defined as baseline ACR . 300 mg/g. All assessments of
urine and blood were performed at a central laboratory (Quintiles
Laboratories, Clearstone Central Laboratories [formerly MDS
Pharma Services], and Covance Laboratories). Acute renal failure was
defined as ID 20000003 of the standardized MedDRA query (Medical
Dictionary for Regulatory Activities, version 14.0) broad search.
Overall death was based on fatal adverse event outcome.
Overall safety assessments included adverse events, vital signs,
and clinical laboratory variables. Based on relevant safety signals
of recent cardiorenal outcomes trials in patients with type 2 diabetes and chronic kidney disease (CKD),11 additional adverse
events of special interest for this pooled kidney disease analysis
included hypotension (identified by investigator from individual
patient report form evaluation) and hyperkalemia (identified by
central laboratory serum potassium levels . 5.0 mmol/L).
Statistical Analysis
Statistical analyses were performed using SAS, version 9.2,
software (SAS Institute Inc). Analyses were based on individualpatient data from the treated set. Descriptive statistics were used
for demographic and baseline characteristics, as well as to determine
primary and secondary outcomes (such as incidence, incidence rates
per 1,000 person-years, and time at risk). Comparability between
treatment groups for demographic and baseline characteristics was
explored by applying stratified 2-sample t tests and CochranMantel-Haenszel tests, respectively.
The hazard ratio (HR) for the time to first event of interest of the
primary composite outcome was calculated using the Cox proportional hazards model with the primary exposure treatment
group. Adjustments were performed for individual studies. Subgroup analyses of the primary composite outcome were performed
by age, race, and concomitant antihypertensive treatment (with or
without angiotensin-converting enzyme [ACE] inhibitor and/or
angiotensin receptor blocker [ARB] therapy at baseline). Interaction tests were conducted in order to exclude the heterogeneity of
the treatment effect regarding age, race, antihypertensive treatment, kidney function at baseline, and individual studies. These
interaction analyses were performed using an additional Cox
regression model including the respective factor, as well as the
interaction term of the respective factor with treatment.
Statistical analysis of the secondary outcomes was based on the
individual components of the primary composite outcome.
Cox proportional hazards model analyses were also performed
to calculate HRs of adverse events of special interest (ie, hypotension and hyperkalemia). Renal and safety parameters were
reported using descriptive statistics.
RESULTS
Demographic and Baseline Characteristics
This pooled kidney disease analysis included a total of
5,466 treated individuals: 3,505 received linagliptin,
Am J Kidney Dis. 2015;66(3):441-449
5 mg, once daily and 1,961 received placebo. Median
treatment exposure was 171 (range, 1-531) days for
linagliptin and 172 (range, 1-531) days for placebo.
Cumulative exposure was 1,751 person-years for linagliptin and 1,055 person-years for placebo. Overall,
participants in the linagliptin group (8.3%) were less
likely to discontinue treatment compared with those
receiving placebo (12.6%).
Demographic and baseline clinical and biochemical
characteristics (Table 1) and background glucoselowering and cardiovascular therapies (Table 2) were
well balanced between the 2 treatment groups.
Kidney Disease End Points
Primary Composite Outcome
The primary composite outcome occurred in 448
(12.8%) individuals in the linagliptin group and 306
(15.6%) individuals in the placebo group (Table 3).
Linagliptin treatment significantly reduced the hazard
of the first occurrence during the study of the primary
composite outcome by 16% as compared with placebo (HR, 0.84; 95% confidence interval [CI],
0.72-0.97; P 5 0.02; Fig 1). Sensitivity analyses
ruled out a significant study by treatment interaction
(P for heterogeneity 5 0.6).
Kidney function at baseline was identified as a
potential confounder for the primary composite outcome. Sensitivity analysis showed that adjustment for
kidney function at baseline did not influence the association between reduced renal risk and linagliptin
treatment (HR, 0.83; 95% CI, 0.72-0.97; P 5 0.02).
Moreover, the observed risk reduction for kidney
disease end points with linagliptin was consistent
across examined subgroups, including those defined by
race and concomitant use of renin-angiotensin system
(RAS) inhibitors, with borderline evidence of heterogeneity in the age subgroup (Table 3; Fig 1B).
Secondary Outcomes
Secondary analyses of the individual components
of the primary composite outcome showed that the
majority of kidney disease end points occurred more
frequently in the placebo group (Table 3). The
overall most frequent event reported was new onset
of moderate elevation of albuminuria, with an incidence rate of 9.3% and 10.9% in the linagliptin and
placebo groups, respectively. Linagliptin treatment
significantly reduced the hazard of new onset of
moderate elevation of albuminuria by 18% (HR,
0.82; 95% CI, 0.69-0.98; P 5 0.03; Fig 2). The
second most reported event was new onset of severe
elevation of albuminuria, occurring in 2.4% and
3.2% in the linagliptin and placebo groups, respectively (HR, 0.86; 95% CI, 0.61-1.20; P 5 0.4; Fig 2).
Rare kidney disease end points such as reduction in
kidney function, halving of eGFR, or acute renal
443
Cooper et al
Table 1. Demographic and Baseline Characteristics
Linagliptin
(n 5 3,505)
Placebo
(n 5 1,961)
Table 2. Background Medication at Enrollment
Age (y)
,65 y
58.8 6 10.5 59.1 6 10.5 0.7b
2,425 (69.2) 1,320 (67.3) 0.7
Male sex
Race
White
Asian
Black
1,844 (52.6) 1,060 (54.1) 0.5
0.5
2,250 (64.2) 1,330 (67.8)
1,179 (33.6)
564 (28.8)
76 (2.2)
67 (3.4)
Body weight (kg)
Body mass index (kg/m2)
eGFR
$90 mL/min/1.73 m2
60-,90 mL/min/1.73 m2
30-,60 mL/min/1.73 m2
,30 mL/min/1.73 m2
79.9 6 17.8
29.2 6 5.1
82.1 6 17.5 0.1b
29.8 6 5.1 0.2b
0.5
1,820 (51.9)
982 (50.1)
1,355 (38.7)
748 (38.1)
255 (7.3)
167 (8.5)
75 (2.1)
64 (3.3)
Urinary ACR
#30 mg/g
.30-300 mg/g
.300 mg/g
Missing
2,307
813
192
193
(65.8) 1,274 (65.0)
(23.2)
447 (22.8)
(5.5)
132 (6.7)
(5.5)
108 (5.5)
Time since T2DM diagnosis
#1 y
.1-5 y
.5 y
Missing
422
975
2,108
0
(12.0)
201 (10.2)
(27.8)
503 (25.7)
(60.1) 1,256 (64.0)
(0.0)
1 (0.1)
HbA1cc (%)
HbA1cc (mmol/mol)
FPGd (mg/dL)
Systolic BPe
Normotensive: ,140
mm Hg
Grade 1: 140-159 mm Hg
Grade 2: 160-179 mm Hg
Grade 3: $180 mm Hg
0.6
0.5
8.2 (0.9)
8.2 (0.9)
66.0 6 9.8
66.0 6 9.8
165.8 6 45.5 165.2 6 45.9
131.4 6 15.8 132.2 6 15.5
2,438 (69.6) 1,348 (68.7)
880 (25.1)
160 (4.6)
27 (0.8)
0.2b
0.3b
0.7b
0.3
501 (25.5)
98 (5.0)
14 (0.7)
Diastolic BPe
78.3 6 9.1
78.3 6 9.3 0.7b
Normotensive: ,90 mm Hg 3,060 (87.3) 1,712 (87.3) 0.7
Grade 1: 90-99 mm Hg
383 (10.9)
215 (11.0)
Grade 2: 100-109 mm Hg
56 (1.6)
26 (1.3)
Grade 3: $110 mm Hg
6 (0.2)
8 (0.4)
Medical history of
hypertension
Medical history of CAD
2,231 (63.7) 1,265 (64.5) 0.7
447 (12.8)
295 (15.0) 0.6
Note: Values for categorical variables are given as number
(percentage); for continuous variables, as mean 6 standard
deviation.
Abbreviations: ACR, albumin-creatinine ratio; BP, blood
pressure; CAD, coronary artery disease; eGFR, estimated
glomerular filtration rate; FPG, fasting plasma glucose; HbA1c,
hemoglobin A1c; T2DM, type 2 diabetes mellitus.
a
P for difference between groups calculated by the CochranMantel-Haenszel test, unless otherwise indicated.
b
P calculated by the 2-sample t test.
c
Baseline HbA1c data were available for 5,459 (linagliptin,
3,503; placebo, 1,956) participants.
d
Baseline FPG data were available for 5,379 (linagliptin,
3,448; placebo, 1,931) participants.
e
Baseline systolic and diastolic BP data were available for
5,466 (linagliptin, 3,505; placebo, 1,961) participants.
444
Linagliptin
(n 5 3,505)
Pa
No. of glucose-lowering
background drugs
0
1
$2
Placebo
(n 5 1,961)
Pa
0.9
583 (16.6)
1,171 (33.4)
1,751 (50.0)
Glucose-lowering background
therapy
Monotherapy
Metformin
719 (20.5)
Sulfonylurea
278 (7.9)
Insulin
146 (4.2)
Other
28 (0.8)
Combination therapy
Metformin 1 sulfonylurea 1,093 (31.2)
Metformin 1 insulin
489 (14.0)
Other
169 (4.8)
Antihypertensive background
therapyb
ACE inhibitors
ARBs
b-Blockers
Diuretics
Calcium antagonists
Fixed-dose combinations
Other
307 (15.7)
655 (33.4)
999 (50.9)
334
158
152
11
(17.0)
(8.1)
(7.8)
(0.6)
0.8
0.4
0.4
0.4
407 (20.8) 0.9
483 (24.6) 0.8
109 (5.6) 0.4
2,229 (63.6) 1,314 (67.0) 0.9
1,028
530
751
605
595
381
1
(29.3)
(15.1)
(21.4)
(17.3)
(17.0)
(10.9)
(0.0)
622
295
465
392
357
230
3
(31.7)
(15.0)
(23.7)
(20.0)
(18.2)
(11.7)
(0.2)
0.7
0.3
0.7
0.9
0.7
0.9
0.2
Lipid-lowering background
therapyb
Statins
Fibrates
Niacin
Other
1,483 (42.3)
929 (47.4) 0.1
1,293
221
33
119
818
133
23
83
Acetylsalicylic acid
1,231 (35.1)
(36.9)
(6.3)
(0.9)
(3.4)
(41.7)
(6.8)
(1.2)
(4.2)
0.1
0.6
0.5
0.9
763 (38.9) 0.5
Note: Values are given as number (percentage).
Abbreviations: ACE, angiotensin-converting enzyme; ARB,
angiotensin receptor blocker.
a
P for difference between groups calculated by the CochranMantel-Haenszel test.
b
Patients might have had more than one medication.
failure were comparable between the linagliptin and
placebo groups (Table 3).
Five (0.1%) and 4 (0.2%) deaths occurred during
the study periods in the linagliptin and placebo
groups, respectively. None of these was considered
drug related by the investigators.
Safety and Tolerability
Overall, clinical adverse events occurred in 2,151
(61.4%) participants in the linagliptin group and
1,242 (63.3%) in the placebo group. The proportion
of participants having drug-related adverse events was
12.9% in the linagliptin group and 13.4% in the
placebo group. Hypotensive episodes, as reported by
investigators, occurred in 11 (0.3%) participants in the
linagliptin group and 7 (0.4%) in the placebo group
Am J Kidney Dis. 2015;66(3):441-449
Kidney Disease End Points and Linagliptin
Table 3. Incidence, Incidence Rates, and Time at Risk of Primary and Secondary Outcomes
Linagliptin (n 5 3,505)
Incidence
Overall
Age
,65 y
$65 y
448 (12.8)
IR
Placebo (n 5 1,961)
Time at Risk
Primary Composite Outcomea
268.4
1,669.4
Incidence
IR
Time at Risk
306 (15.6)
311.7
981.6
297 (12.2)
151 (14.0)
259.4
287.9
1,145.0
524.4
213 (16.1)
93 (14.5)
331.0
275.1
643.5
338.1
Race
White
Asian
Black
280 (12.4)
161 (13.7)
7 (9.2)
247.1
323.6
180.3
1,133.0
497.6
38.8
207 (15.6)
90 (16.0)
9 (13.4)
296.4
359.7
272.2
698.3
250.2
33.1
Antihypertensive treatment
Without ACEi/ARB
With ACEi/ARB
224 (11.3)
224 (14.7)
247.5
904.9
293.0
764.4
Secondary Outcomes
149 (13.9)
157 (17.7)
289.4
336.3
514.8
466.8
Albuminuria
Moderate elevation
Severe elevation
326 (9.3)
84 (2.4)
191.1
48.0
1,705.7
1,751.4
213 (10.9)
62 (3.2)
210.9
59.0
1,009.8
1,051.7
2.3
1.1
29.5
2.8
1,774.6
1,776.7
1,764.9
1,777.3
10.3
1.9
34.9
3.7
1,065.1
1,069.3
1,060.5
1,069.9
Reduction in kidney function
Halving of eGFR
Acute renal failure
Death
4
2
52
5
(0.1)
(0.1)
(1.5)
(0.1)
11
2
37
4
(0.6)
(0.1)
(1.9)
(0.2)
Note: Incidence values are given as number (percentage). IRs expressed per 1,000 person-years; time at risk, in person-years.
Abbreviations: ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; eGFR, estimated glomerular
filtration rate; IR, incidence rate.
a
The primary outcome is a composite of 6 kidney disease end points, listed under the secondary outcomes.
(HR, 1.12; 95% CI, 0.43-2.95; P 5 0.8). Elevated
serum potassium levels were documented in 28 (0.8%)
participants in the linagliptin group and 24 (1.2%) in the
placebo group (HR, 1.02; 95% CI, 0.59-1.77; P 5 0.9).
Mean changes in eGFR, ACR, blood pressure, and
body weight from baseline to week 24 in the linagliptin
and placebo groups are presented in Table 4. Table S2
provides additional information on blood pressure
changes during interventions as categorized by the International Society of Hypertension recommendations.
No significant differences were observed for changes in
blood pressure among normo- and hypertensive patients when treated with either linagliptin or placebo.
DISCUSSION
In this pooled kidney disease analysis, a comprehensive assessment of kidney disease end points for a
glucose-lowering drug was performed based on
available data from an early clinical development
program. Such safety evaluations have previously
been requested to improve informed treatment decision making and to weigh overall risks and benefits of
novel glucose-lowering agents as early as possible in
drug development. A recent example of early drug
safety evaluation was the FDA guidance issued in
2008 requesting that all new oral glucose-lowering
drugs for treating type 2 diabetes have to explore
cardiovascular safety prior to regulatory approval.7
Am J Kidney Dis. 2015;66(3):441-449
However, similar guidance on renal safety has not
generally been requested, although type 2 diabetes is
associated with considerable renal burden. This is
noteworthy because there is a large body of evidence
that supports the view that macro- and microvascular
changes often occur simultaneously and share common
risk factors and underlying mechanisms.22 Therefore, it
would be of relevance for new diabetes drugs to provide
early safety information on renal complications in
addition to assessing cardiovascular safety. This
consideration is further supported because individuals
with diabetes per se are known to be at significant
increased risk of developing CKD. As the leading cause
of kidney failure in the United States, diabetes accounts
for nearly 45% of new cases.23
To further define the potential impact of a glucoselowering drug on renal risk and outcomes, a novel and
systematic approach was developed to assess kidney
disease end points in patients with type 2 diabetes
treated with the DPP-4 inhibitor linagliptin.24
Although several studies have assessed the safety
and tolerability of DPP-4 inhibitors in patients with
type 2 diabetes and decreased kidney function,15,25-27
clinical trials with primary kidney disease outcomes
are currently not available.28 In order to address this
important gap, linagliptin was chosen as an appropriate candidate because it does not require dose
adjustment due to nonrenal clearance and thus can be
445
Cooper et al
Figure 1. Comparison of risk estimates between the linagliptin and placebo groups using a Cox proportional hazards model for the
primary composite outcome defined as first occurrence of new onset of moderate elevation of albuminuria, new onset of severe elevation of albuminuria, reduction in kidney function, halving of estimated glomerular filtration rate, acute renal failure, or death from any
cause. Adjustments were performed for individual studies and for treatment groups. Primary analysis in all individuals depicted on top
and in selected subgroups below: analyses by age (,65 and $65 years; P for heterogeneity 5 0.04), race (white, Asian, and black;
P for heterogeneity 5 0.7), and concomitant use of renin-angiotensin system inhibitors (with and without; P for heterogeneity 5 0.9).
Abbreviations: ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; CI, confidence interval.
given as a single oral dose in patients with type 2 diabetes at any stage of decreased kidney function.17,18
It was fortunate that large data sets from phase 2 and
phase 3 clinical trials were available16 and, importantly,
included individuals at increased risk of kidney disease
end points, such as patients with eGFRs , 30 mL/min/
1.73 m2,15 and elderly patients (aged $ 70 years).13
Overall, our pooled kidney disease analysis included
Figure 2. Comparison of risk estimates between the linagliptin and placebo groups using a Cox proportional hazards model for
secondary outcomes (any first occurrence of the individual components of the composite kidney disease outcome). Abbreviations:
CI, confidence interval; eGFR, estimated glomerular filtration rate.
446
Am J Kidney Dis. 2015;66(3):441-449
Kidney Disease End Points and Linagliptin
Table 4. Change in Renal and Selected Safety Parameters From Baseline to Week 24
Linagliptin (n 5 3,505)
Baseline
Placebo (n 5 1,961)
Week 24: Difference From
Baseline
Baseline
Week 24: Difference From
Baseline
eGFR (mL/min/1.73 m2)
Urinary ACR (mg/g)a
Blood pressure (mm Hg)
Systolic
Diastolic
n 5 3,505 87.2 6 0.4 n 5 2,661 20.8 6 0.2 n 5 1,961 85.6 6 0.5 n 5 1,413 20.3 6 0.2
n 5 1,005 311.7 6 26.2 n 5 707
217.0 6 21.3 n 5 580
366.8 6 33.1 n 5 406
19.6 6 25.8
n 5 3,505 131.4 6 0.3
n 5 3,505 78.3 6 0.2
n 5 2,751
n 5 2,751
20.3 6 0.3
20.2 6 0.2
Body weight (kg)
n 5 3,074
n 5 2,488
0.2 6 0.1
80.4 6 0.3
n 5 1,961 132.2 6 0.3
n 5 1,961 78.3 6 0.2
n 5 1,463
0.0 6 0.4
n 5 1,463 20.5 6 0.2
n 5 1,667
n 5 1,210
82.2 6 0.4
0.0 6 0.1
Note: Numbers of patients indicate participants with measurements. Unless otherwise indicated, values given as mean 6 standard
error.
Abbreviations: ACR, albumin-creatinine ratio; eGFR, estimated glomerular filtration rate.
a
In participants with urinary ACRs $ 30 mg/g at baseline.
13 randomized controlled clinical trials made up of
more than 5,000 individuals with type 2 diabetes. We
found that linagliptin treatment was not associated with
an increase in renal risk. Moreover, linagliptin treatment reduced significantly the hazard of kidney disease
end points by 16% when compared with placebo. This
is an intriguing finding and generates the novel hypothesis that this agent may have potential to slow CKD
progression in type 2 diabetes, as recently reported
through its effects on the “surrogate” albuminuria.29
The primary aim of our pooled kidney disease
analysis was to rule out an increased renal risk with
linagliptin. It is important to exclude potential deleterious effects as a result of drug-drug interactions,
particularly as they pertain to the kidney. For
example, a recent trial targeting dual blockade of the
RAS by combining the novel renin inhibitor aliskiren
with the established classes of either ACE inhibitors
or ARBs in patients with type 2 diabetes supports
such a possibility.11 This trial had to be terminated
prematurely, partially as a result of excess risk for
adverse events (eg, hypotension and hyperkalemia) in
the aliskiren group.30 Of note, in our analysis, linagliptin was not associated with an increased risk of
hypotension or hyperkalemia. Furthermore, this lack
of a deleterious effect of linagliptin was observed in
this cohort, in which 45% of participants were
concomitantly treated with a RAS inhibitor at baseline. Hence, our analysis indicates that there is no
increased renal or hemodynamic risk with the combined blockade of the RAS and DPP-4 system.
As outlined, the data are consistent with a potential beneficial effect of linagliptin on the onset
and progression of kidney disease in type 2 diabetes.
In line with findings from this pooled kidney disease
analysis and a previous analysis that has identified
an albuminuria-lowering effect of linagliptin, which
appears to be at least partially independent of its
glucose-lowering action,29 preclinical studies have
Am J Kidney Dis. 2015;66(3):441-449
shown that linagliptin may reduce urinary albumin
excretion, renal fibrosis, oxidative stress, and
inflammation in different experimental models.31-35
In the present study, more than 500 individuals
with eGFRs , 60 mL/min/1.73 m2 were included.
Kidney function at baseline did not influence the
association between reduced renal risk and linagliptin
treatment, indicating a possible renal effect of linagliptin even at advanced stages of CKD. Moreover,
the risk reduction for kidney disease end points with
linagliptin was seen in participants with or without
concomitant treatment with RAS inhibitors at baseline. Although treatments based on blockade of the
RAS have generally been associated with a reduction
in albuminuria and improved kidney disease outcome
in patients with type 2 diabetes,36,37 there was no
evidence that background ACE-inhibitor and/or ARB
therapy attenuated the renal effects of linagliptin.
Based on positive findings from the previous
pooled analysis with respect to a potential renal
benefit of linagliptin,29 further research is now
underway to explore glucose-dependent and -independent effects of linagliptin in the kidney. Specifically, the MARLINA-T2D trial (Efficacy, Safety
and Modification of Albuminuria in Type 2 Diabetes Subjects With Renal Disease With Linagliptin;
ClincalTrials.gov study number NCT01792518) is
designed and powered to explore the albuminurialowering potential of linagliptin in patients with
type 2 diabetes and prevalent albuminuria already
treated with ACE inhibitors or ARBs. This 24-week,
randomized, controlled study was initiated in early
2013 and results are expected in 2015.
A major strength of the present study is the large
sample size, which provides greater power to detect
significant differences than individual randomized
controlled trials. Nonetheless, we acknowledge that
this pooled kidney disease analysis was of clinical
trials not primarily designed to investigate kidney
447
Cooper et al
disease end points. However, safety assessments were
standardized across trials and laboratory assessments
of albuminuria and kidney function were conducted
by a central laboratory. Another limitation is the use
of a primary composite outcome that combines
different kidney disease end points. However, all of
our selected kidney disease end points are likely part
of broader pathologic pathways to kidney failure in
patients with diabetes. Finally, average durations of
the included trials were relatively short, which could
explain the overall low observation rate of significant
kidney disease progression and hard kidney disease
outcome events. To address the high unmet medical
need in patients with type 2 diabetes and CKD, the
recently initiated CARMELINA trial (Cardiovascular
and Renal Microvascular Outcome Study With
Linagliptin in Patients With Type 2 Diabetes Mellitus;
ClinicalTrials.gov study number NCT01897532) will
investigate the long-term effect of linagliptin on major
cardiovascular and kidney disease outcomes in adults
with type 2 diabetes at risk of macro- and microvascular events.
In conclusion, it is suggested that emerging data
from clinical development programs should be systematically used to assess kidney disease end points in
patients with type 2 diabetes treated with new glucoselowering agents. Analogous to the current efforts to
better assess cardiovascular safety, renal safety analyses could provide reliable and valid information on
kidney disease in the early stages of drug development. In our analysis, a novel approach for kidney
disease assessments has been developed using the
large clinical trials program of the novel DPP-4 inhibitor linagliptin. Linagliptin was not associated with
an increased renal risk but was associated with a
significant reduction in clinically relevant kidney
disease end points in patients with type 2 diabetes.
ACKNOWLEDGEMENTS
Part of this work was presented as an abstract at the American
Society of Nephrology’s Kidney Week 2012, San Diego, CA,
October 30 to November 4, 2012 (poster TH-PO530).
Support: Audrey Koïtka-Weber, PhD, provided scientific
consulting and medical writing services for this study; this assistance was funded by Boehringer Ingelheim, the manufacturer of
linagliptin. Boehringer Ingelheim was the funding source and
sponsor for all clinical trials. The sponsor was involved in study
design, data collection, data analysis, and interpretation of data.
Each author had full access to the data. The manuscript was
prepared by the authors under the supervision of Dr Cooper, with
the sponsor-funded scientific writing assistance of Dr KoïtkaWeber. The decision to submit the manuscript was the responsibility of Dr Cooper.
Financial Disclosure: Dr Cooper has received speaker honoraria from Boehringer Ingelheim, Servier, and Eli Lilly and is an
advisory member for Boehringer Ingelheim, AstraZeneca, BristolMyers Squibb, Merck Sharp and Dohme, and Novo-Nordisk. Dr
Perkovic is supported by a Heart Foundation of Australia Cardiovascular Research Network Fellowship; has received speaker
448
honoraria from AstraZeneca, Boehringer Ingelheim, Merck,
Roche, and Servier; serves on Steering Committees for trials
supported by Abbott, Baxter, Boehringer Ingelheim, Janssen, and
Pfizer; and is a consultant for Abbott, Astellas, Baxter, and Vitae
Pharmaceuticals. Dr McGill has received grants from Sanofi,
Novartis, Takeda, Mannkind, and Andromeda; has served as a
speaker and consultant for Merck and Janssen; and is a consultant
for Boehringer Ingelheim, Sanofi, Mannkind, and Novo Nordisk.
Dr Groop has received speaker honorariums from Boehringer
Ingelheim, Cebix, Eli Lilly, Genzyme, Merck Sharp and Dohme,
Novartis, and Novo Nordisk and research grants from Eli Lilly and
Roche and is an advisory member for Boehringer Ingelheim and
Novartis. Dr Rosenstock has served on scientific advisory boards
and received honorarium or consulting fees from Pfizer, Roche,
Sanofi, Novo Nordisk, Eli Lilly, MannKind, GlaxoSmithKline,
Takeda, Daiichi Sankyo, Johnson & Johnson, Novartis, Boehringer Ingelheim, and Lexicon and grants/research support from
Merck, Pfizer, Sanofi, Novo Nordisk, Roche, Bristol-Myers
Squibb, Eli Lilly, Forest, GlaxoSmithKline, Takeda, Novartis,
AstraZeneca, Amylin, Johnson & Johnson, Daiichi Sankyo,
MannKind, Lexicon, and Boehringer Ingelheim. Dr Wanner has
received speaker honoraria from Abbvie, AstraZeneca, BristolMyers Squibb, MSD, Mitsubishi, Roche, and Sanofi-Genzyme;
serves on Steering Committees for trials supported by Abbvie and
Boehringer Ingelheim; and is a consultant for Abbvie, Baxter, Keryx
and Mitsubishi. Mr Hehnke and Drs Woerle and von Eynatten are
employees of Boehringer Ingelheim.
Contributions: research idea and study design: MEC, MvE;
statistical analysis: UH; data analysis/interpretation: MEC, VP,
JBM, P-HG, CW, JR, UH, H-JW, MvE; supervision or mentorship: MEC. Each author contributed important intellectual content
during manuscript drafting or revision and accepts accountability
for the overall work by ensuring that questions pertaining to the
accuracy or integrity of any portion of the work are appropriately
investigated and resolved. MEC takes responsibility that this study
has been reported honestly, accurately, and transparently; that
no important aspects of the study have been omitted; and that any
discrepancies from the study as planned have been explained.
SUPPLEMENTARY MATERIAL
Table S1: Clinical trials included in pooled kidney disease
analysis.
Table S2: Change in blood pressure from baseline to week 24.
Note: The supplementary material accompanying this article
(http://dx.doi.org/10.1053/j.ajkd.2015.03.024) is available at
www.ajkd.org
REFERENCES
1. Guthrie RM. Evolving therapeutic options for type 2 diabetes mellitus: an overview. Postgrad Med. 2012;124(6):82-89.
2. American Diabetes Association. Standards of medical care in
diabetes—2014. Diabetes Care. 2012;37(suppl 1):S14-S80.
3. National Kidney Foundation. KDOQI clinical practice
guideline for diabetes and CKD: 2012 update. Am J Kidney Dis.
2012;60(5):850-886.
4. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of
myocardial infarction and death from cardiovascular causes.
N Engl J Med. 2007;356(24):2457-2471.
5. Singh S, Loke YK, Furberg CD. Long-term risk of cardiovascular events with rosiglitazone: a meta-analysis. JAMA.
2007;298(10):1189-1195.
6. Drucker DJ, Goldfine AB. Cardiovascular safety and diabetes drug development. Lancet. 2011;377(9770):977-979.
7. Food and Drug Administration Center for Drug Evaluation
and Research. Guidance for Industry: Diabetes Mellitus Evaluating Cardiovascular Risk in New Antidiabetic Therapies
Am J Kidney Dis. 2015;66(3):441-449
Kidney Disease End Points and Linagliptin
to Treat Type 2 Diabetes. 2008. http://www.fda.gov/downloads/
Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/
UCM071627.pdf. Accessed April 9, 2014.
8. Goldfine AB. Assessing the cardiovascular safety of diabetes
therapies. N Engl J Med. 2008;359(11):1092-1095.
9. Cumbie BC, Hermayer KL. Current concepts in targeted
therapies for the pathophysiology of diabetic microvascular complications. Vasc Health Risk Manag. 2007;3(6):823-832.
10. Mann JF, Anderson C, Gao P, et al. Dual inhibition of the
renin-angiotensin system in high-risk diabetes and risk for stroke
and other outcomes: results of the ONTARGET trial. J Hypertens.
2013;31(2):414-421.
11. Parving HH, Brenner BM, McMurray JJ, et al. Cardiorenal
end points in a trial of aliskiren for type 2 diabetes. N Engl J Med.
2012;367(23):2204-2213.
12. Perkovic V, Heerspink HL, Chalmers J, et al. Intensive
glucose control improves kidney outcomes in patients with type 2
diabetes. Kidney Int. 2013;83(3):517-523.
13. 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(9902):1413-1423.
14. Gallwitz B, Rosenstock J, Rauch T, et al. 2-Year efficacy and
safety of linagliptin compared with glimepiride in patients with type 2
diabetes inadequately controlled on metformin: a randomised,
double-blind, non-inferiority trial. Lancet. 2012;380(9840):475-483.
15. McGill JB, Sloan L, Newman J, et al. Long-term efficacy
and safety of linagliptin in patients with type 2 diabetes and severe
renal impairment: a 1-year, randomized, double-blind, placebocontrolled study. Diabetes Care. 2013;36(2):237-244.
16. Johansen OE, Neubacher D, von Eynatten M, Patel S,
Woerle HJ. Cardiovascular safety with linagliptin in patients with
type 2 diabetes mellitus: a pre-specified, prospective, and adjudicated meta-analysis of a phase 3 programme. Cardiovasc Diabetol.
2012;11:3.
17. Graefe-Mody U, Friedrich C, Port A, et al. Effect of renal
impairment on the pharmacokinetics of the dipeptidyl peptidase4 inhibitor linagliptin(*). Diabetes Obes Metab. 2011;13(10):
939-946.
18. Friedrich C, Emser A, Woerle HJ, Graefe-Mody U. Renal
impairment has no clinically relevant effect on the long-term
exposure of linagliptin in patients with type 2 diabetes. Am J
Ther. 2013;20(6):618-621.
19. Doupis J. Linagliptin: from bench to bedside. Drug Des
Devel Ther. 2014;8:431-446.
20. European Medicines Agency. Guideline on Clinical Investigation of Medicinal Products in the Treatment or Prevention of
Diabetes Mellitus. 2012. http://www.ema.europa.eu/docs/en_GB/
document_library/Scientific_guideline/2012/06/WC500129256.pdf.
Accessed April 9, 2014.
21. Levey AS, Stevens LA, Schmid CH, et al. A new equation to
estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):
604-612.
22. Orasanu G, Plutzky J. The pathologic continuum of diabetic vascular disease. J Am Coll Cardiol. 2009;53(5)(suppl):
S35-S42.
Am J Kidney Dis. 2015;66(3):441-449
23. Collins AJ, Foley RN, Herzog C, et al. US Renal Data System 2012 annual data report. Am J Kidney Dis. 2013;61(1)(suppl 1):
e1-e480.
24. Gallwitz B. Linagliptin—a novel dipeptidyl peptidase inhibitor for type 2 diabetes therapy. Clin Med Insights Endocrinol
Diabetes. 2012;5:1-11.
25. Arjona Ferreira JC, Marre M, Barzilai N, et al. Efficacy
and safety of sitagliptin versus glipizide in patients with type 2
diabetes and moderate-to-severe chronic renal insufficiency. Diabetes Care. 2013;36(5):1067-1073.
26. Lukashevich V, Schweizer A, Shao Q, Groop PH,
Kothny W. Safety and efficacy of vildagliptin versus placebo in
patients with type 2 diabetes and moderate or severe renal
impairment: a prospective 24-week randomized placebocontrolled trial. Diabetes Obes Metab. 2011;13(10):947-954.
27. Nowicki M, Rychlik I, Haller H, et al. Long-term treatment
with the dipeptidyl peptidase-4 inhibitor saxagliptin in patients
with type 2 diabetes mellitus and renal impairment: a randomised
controlled 52-week efficacy and safety study. Int J Clin Pract.
2011;65(12):1230-1239.
28. Panchapakesan U, Mather A, Pollock C. Role of GLP-1
and DPP-4 in diabetic nephropathy and cardiovascular disease.
Clin Sci (Lond). 2013;124(1):17-26.
29. Groop PH, Cooper ME, Perkovic V, Emser A, Woerle HJ,
von Eynatten M. Linagliptin lowers albuminuria on top of recommended standard treatment in patients with type 2 diabetes and
renal dysfunction. Diabetes Care. 2013;36(11):3460-3468.
30. de Leeuw PW. ACP Journal Club. Aliskiren increased adverse
events in patients with diabetes and kidney disease who were receiving ACE inhibitors or ARBs. Ann Intern Med. 2013;158(6):JC7.
31. Alter ML, Ott IM, von Websky K, et al. DPP-4 inhibition
on top of angiotensin receptor blockade offers a new therapeutic
approach for diabetic nephropathy. Kidney Blood Press Res.
2012;36(1):119-130.
32. Ishibashi Y, Matsui T, Maeda S, Higashimoto Y,
Yamagishi S. Advanced glycation end products evoke endothelial
cell damage by stimulating soluble dipeptidyl peptidase-4 production and its interaction with mannose 6-phosphate/insulin-like
growth factor II receptor. Cardiovasc Diabetol. 2013;12:125.
33. Kanasaki K, Shi S, Kanasaki M, et al. Linagliptin-mediated
DPP-4 inhibition ameliorates kidney fibrosis in streptozotocininduced diabetic mice by inhibiting endothelial-to-mesenchymal
transition in a therapeutic regimen. Diabetes. 2014;63(6):2120-2131.
34. Nakashima S, Matsui T, Takeuchi M, Yamagishi SI.
Linagliptin blocks renal damage in type 1 diabetic rats by suppressing advanced glycation end products-receptor axis. Horm
Metab Res. 2014;46(10):717-721.
35. Nistala R, Habibi J, Aroor A, et al. DPP4 inhibition attenuates filtration barrier injury and oxidant stress in the Zucker
obese rat. Obesity. 2014;22(10):2172-2179.
36. Brenner BM, Cooper ME, de Zeeuw D, et al. Effects of
losartan on renal and cardiovascular outcomes in patients with type 2
diabetes and nephropathy. N Engl J Med. 2001;345(12):861-869.
37. Lewis EJ, Hunsicker LG, Clarke WR, et al. Renoprotective
effect of the angiotensin-receptor antagonist irbesartan in patients
with nephropathy due to type 2 diabetes. N Engl J Med. 2001;345(12):
851-860.
449