Professional Documents
Culture Documents
Diabetes 2
Cardiovascular outcome trials of glucose-lowering drugs
or strategies in type 2 diabetes
Rury R Holman, Harald Sourij, Robert M Cali
Lancet 2014; 383: 200817
This online publication has
been corrected. The corrected
version rst appeared at
thelancet.com on June 27, 2014
See Editorial page 1945
This is the second in a Series of
two papers about diabetes
Diabetes Trials Unit, University
of Oxford, Oxford, UK
(Prof R R Holman FRCP,
H Sourij MD); Division of
Endocrinology and
Metabolism, Medical
University of Graz, Graz,
Austria (H Sourij); and Duke
Clinical Research Institute,
Duke University School of
Medicine, Durham, NC, USA
(Prof R M Cali MD)
Correspondence to:
Prof Rury R Holman, Diabetes
Trials Unit, OCDEM, Churchill
Hospital, Old Road, Headington,
Oxford OX3 7LJ, UK
rury.holman@dtu.ox.ac.uk
Few trials of glucose-lowering drugs or strategies in people with type 2 diabetes have investigated cardiovascular
outcomes, even though most patients die from cardiovascular causes despite the benecial eects of lipid-reducing
and blood pressure-lowering treatments. The evidence-based reduction in risk of microvascular disease with glucose
lowering has resulted in guidelines worldwide recommending optimisation of glycosylated haemoglobin, but no trial
results have shown unequivocal cardiovascular risk reduction with glucose lowering. However, results of the post-trial
follow-up of the UK Prospective Diabetes Study, and of a meta-analysis of the four glucose-lowering outcome trials
completed to date, suggest about a 15% cardiovascular relative risk reduction per 1% decrement in HbA1c. The 2008
US Food and Drug Administration industry guidance for licensing of antidiabetic drugs greatly increased the number
of cardiovascular outcome trials in diabetes, but most trials opted for non-inferiority designs aiming primarily to
show absence of cardiovascular toxicity in the shortest possible time. This unintended consequence of the new
regulations has meant that the potential long-term benets, and the possible risks of new therapies, are not being
assessed eectively. Also, essential head-to-head trials of therapies for this complex progressive disease, to answer
issues such as how best to achieve and maintain optimum glycaemia without promoting weight gain or hypoglycaemia,
are not being undertaken. In this Series paper, we summarise randomised controlled cardiovascular outcome trials in
type 2 diabetes, provide an overview of ongoing trials and their limitations, and speculate on how future trials could
be made more ecient and eective.
Introduction
The rapidly emerging diabetes pandemic is one of the
most challenging noncommunicable disease threats to
public health in the 21st century. More than 380 million
people worldwide have type 2 diabetes,1 most of whom
will die from cardiovascular disease, the second most
common cause of death being their increased risk of
cancer.2 People with type 2 diabetes have a roughly
2008
Series
Year of
Population
publication
Follow up
Intervention
Outcome
UKPDS3,12
1998
4209
Median
100 years
DIGAMI 213
2005
1253
Median
21 years
All-cause mortality
Group 1 versus 2: HR 103 (95% CI 079134)
Group 2 versus 3: HR 123 (95% CI 089169)
PROactiv14
2005
T2DM with
macrovascular disease
5238
Mean
345 months
Addition of pioglitazone or
placebo to usual diabetes care
ADVANCE15
2008
11 140
Median
5 years
ACCORD16
2008
10 251
Mean
35 years
HEART2D17
2009
1115
Mean
26 years
VADT18
2009
T2DM
1791
Median
56 years
RECORD19
2009
T2DM on maximum
tolerated dose of
metformin or SFU
4447
Mean
55 years
Addition of rosiglitazone or
combination of metformin
and sulfonylurea
BARI 2D20
2009
2368
Mean
53 years
Insulin-sensitisation versus
insulin-provision treatment
strategy
ADDITION21
2011
Screen-detected T2DM
3055
Mean
53 years
ORIGIN22
2012
12 537
Median
62 years
SAVOR-TIMI5323
2013
16 492
Median
21 years
EXAMINE24
2013
5380
Median
15 years
LOOK-AHEAD25
2013
Overweight or obese
T2DM with or without
CVD history
5145
Median
96 years
T2DM=type 2 diabetes mellitus. SFU=sulfonylurea. RRR=relative risk reduction. MI=myocardial infarction. CV=cardiovascular. MACE=cardiovascular mortality, myocardial infarction, stroke. CVD=cardiovascular
disease. PAD=peripheral artery disease. CAD=coronary artery disease. IFG=impaired fasting glucose. IGT=impaired glucose tolerance. ACS=acute coronary syndrome. HR=hazard ratio.
Table 1: Randomised, controlled, cardiovascular outcome trials (>1000 subjects, >1 year of follow up) of glucose-lowering drugs or strategies in people with type 2 diabetes
Series
Series
Series
TECOS
Intervention
Population
Primary outcome
Study status
ClinicalTrials.
gov identier
14 000
Ongoing,
not recruiting
12/2008Q3/2014
NCT00790205
TOSCA IT
3371
Pioglitazone versus
sulfonylurea
Recruiting
09/200812/2018
NCT00700856
CANVAS
4330
Canagliozin 100 mg
versus canagliozin
300 mg versus placebo
Ongoing,
not recruiting
12/200903/2017
NCT01032629
ELIXA
6000
Lixisenatide versus
placebo
Ongoing,
not recruiting
06/201001/2015
NCT01147250
14 000
Recruiting
06/201012/2017
NCT01144338
BI 10773 trial
7000
Empagliozin 10 mg
versus empagliozin
25 mg versus placebo
Ongoing,
not recruiting
07/201003/2018
NCT01131676
LEADER
9340
Ongoing,
not recruiting
08/201010/2015
NCT01179048
CAROLINA
6000
Linagliptin versus
glimepiride
Ongoing,
not recruiting
10/201009/2018
NCT01243424
REWIND
9622
Dulaglutide versus
placebo
Ongoing,
not recruiting
07/201104/2019
NCT01394952
MK-3102 trial
4000
Recruiting
10/201210/2017
NCT01703208
SUSTAIN6
3260
Semaglutide 05 mg
versus semaglutide
10 mg versus placebo
Ongoing,
not recruiting
02/201301/2016
NCT01720446
ITCA650 trial
2000
Recruiting
03/201307/2018
NCT01455896
Dapagliozin 10 mg
versus placebo
Recruiting
04/201304/2019
NCT01730534
Recruiting
07/201301/2018
NCT01897532
EXSCEL
DECLARE-TIMI 58
17 150
CARMELINA
8300
DEVOTE
7500
CV death, MI or stroke
Recruiting
10/201311/2018
NCT01959529
Ertugliozin trial
3900
Ertugliozin 5 mg versus
ertugliozin 15 mg versus
placebo
Recruiting
11/201306/2020
NCT01986881
T2DM=type 2 diabetes. CVD=cardiovascular disease. CV=cardiovascular. MI=myocardial infarction. UA=unstable angina. ACS=acute coronary syndrome.
Table 2: Ongoing cardiovascular outcome trials of glucose-lowering drugs or strategies (in order of starting date)
2012
Series
200
150
100
50
0
1996
2000
2004
2008
Year
2012
2016
2020
2013
Series
Series
Conclusions
A critical view of the large cardiovascular trials done over
the last four decades shows that, until recently, few had
adequate statistical power to inform practice, and that the
designs of the trials constrained the range of questions
that could be addressed. The UKPDS taught us that
improved glycaemic control in newly-diagnosed type 2
diabetes patients reduces the risk of microvascular
www.thelancet.com Vol 383 June 7, 2014
2015
Series
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
2016
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Series
47
48
49
50
51
52
53
54
55
56
57
58
Hess CN, Rao SV, Kong DF, et al. Embedding a randomized clinical
trial into an ongoing registry infrastructure: unique opportunities
for eciency in design of the Study of Access site For Enhancement
of Percutaneous Coronary Intervention for Women (SAFE-PCI for
Women). Am Heart J 2013; 166: 42128.
Lauer MS, DAgostino RB Sr. The randomized registry trialthe
next disruptive technology in clinical research? N Engl J Med 2013;
369: 157981.
Barker AD, Sigman CC, Kello GJ, Hylton NM, Berry DA,
Esserman LJ. I-SPY 2: an adaptive breast cancer trial design in the
setting of neoadjuvant chemotherapy. Clin Pharmacol Ther 2009;
86: 97100.
Topol EJ. Individualized medicine from prewomb to tomb. Cell
2014; 157: 24153.
Gallin EK, Bond E, Cali RM, et al. Forging stronger partnerships
between academic health centers and patient-driven organizations.
Acad Med 2013; 88: 122024.
Kent DM, Rothwell PM, Ioannidis JP, Altman DG, Hayward RA.
Assessing and reporting heterogeneity in treatment eects in
clinical trials: a proposal. Trials 2010; 11: 85.
2017