Embed Size (px)
Citation preview
Cardiovascular Outcomes Trials inType 2 Diabetes: Where Do We GoFrom Here? Ref lections From aDiabetes Care Editors’ ExpertForumDiabetes Care 2018;41:14–31 | https://doi.org/10.2337/dci17-0057
In December 2008, the U.S. Food and Drug Administration issued guidance tothe pharmaceutical industry setting new expectations for the development ofantidiabetes drugs for type 2 diabetes. This guidance expanded the scope and costof research necessary for approval of such drugs by mandating long-termcardiovascular outcomes trials (CVOTs) for safety. Since 2008, 9 CVOTs have beenreported, 13 are under way, and 4 have been terminated. Reassuringly, each ofthe completed trials demonstrated the noninferiority of their respective drugs toplacebo for their primary cardiovascular (CV) composite end point. Notably, fouradditionally provided evidence of CV benefit in the form of significant decreases inthe primary CV composite end point, two suggested reductions in CV death, andthree suggested reductions in all-cause mortality. Although these trials haveyielded much valuable information, whether that information justifies theinvestment of time and resources is controversial. In June 2016, a Diabetes CareEditors’ Expert Forum convened to review the processes and challenges of CVOTs,discuss the benefits and limitations of their current designs, and weigh the merits ofmodifications that might improve the efficiency and clinical value of future trials.Discussion and analysis continuedwith the CVOT trial results released in June 2017 attheAmericanDiabetes Association’s Scientific Sessions and in September 2017 at theEuropean Association for the Study of Diabetes scientific meeting. This article sum-marizes the discussion and findings to date.
In December 2008, theU.S. Food andDrugAdministration (FDA) issued guidance to thepharmaceutical industry setting out new expectations for the development of antidia-betes drugs for type 2 diabetes (1). This guidance focused specifically on cardiovascular(CV) safety, largely in recognition of the excess burden of cardiovascular disease (CVD)in type 2 diabetes (2). The FDA was responding to prevailing concerns about thepotential for increased CVD risk associated with certain antidiabetes drugs, notablythe thiazolidinedione (TZD) rosiglitazone (3,4). The guidancedand subsequent similarrequirements from the European Medicines Agency (5)deffectively expanded thescope and cost of research necessary to secure approval of new antidiabetes drugsby mandating long-term safety trials.In the 9 years since the guidance was issued, 9 long-term, prospective CV outcomes
trials (CVOTs) have been completed, 13 are under way, and 4 were started but termi-nated early. Collectively, these studies have involved more than 190,000 participants.
1American Diabetes Association, Arlington, VA2Cedars-Sinai Medical Center, Los Angeles, CA3McMaster University and Hamilton Health Sci-ences, Hamilton, Ontario, Canada4Diabetes Trial Unit, Oxford Centre for Diabetes,Endocrinology and Metabolism, Oxford, U.K.5Lunenfeld-Tanenbaum Research Institute,Mount Sinai Hospital, University of Toronto, Tor-onto, Ontario, Canada6Diabetes Research Institute, University of Mi-ami Leonard M. Miller School of Medicine, Mi-ami, FL7Duke University Medical Center, Durham, NC8University of North Carolina School of Medicine,Chapel Hill, NC9Yale School of Medicine and Yale New HavenHospital, New Haven, CT10Keenan Research Centre for Biomedical Science,Li Ka Shing Knowledge Institute, St. Michael’s Hos-pital, and Departments ofMedicine andNutritionalSciences, University of Toronto, Toronto, Ontario,Canada11Diabetes Unit, Department of Internal Medi-cine, Hadassah Hebrew University Hospital, Jer-usalem, Israel12Dallas Diabetes Research Center at MedicalCity and The University of Texas SouthwesternMedical Center, Dallas, TX13Oregon Health & Science University, Portland,OR
Corresponding author:William T. Cefalu, [email protected].
Received 24 October 2017 and accepted 24October 2017.
This article contains Supplementary Data onlineat http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dci17-0057/-/DC1.
This article is featured in a podcast available athttp://www.diabetesjournals.org/content/diabetes-core-update-podcasts.
© 2017 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered. More infor-mation is available at http://www.diabetesjournals.org/content/license.
See accompanying articles, pp. 11,136, 143, 150, and 156.
William T. Cefalu,1 Sanjay Kaul,2
Hertzel C. Gerstein,3 Rury R. Holman,4
Bernard Zinman,5 Jay S. Skyler,6
Jennifer B. Green,7 John B. Buse,8
Silvio E. Inzucchi,9 Lawrence A. Leiter,10
Itamar Raz,11 Julio Rosenstock,12 and
Matthew C. Riddle13
14 Diabetes Care Volume 41, January 2018
DIABETES
CARE
EXPER
TFO
RUM
The nine completed trials examined threedipeptidyl peptidase 4 (DPP-4) inhibitors(6–8), four glucagon-like peptide 1 (GLP-1) receptor agonists (9–12), and twosodium–glucose cotransporter 2 (SGLT2)inhibitors (13,14). Each demonstrated thenoninferiority of their respective drugs toplacebo in their major adverse cardiacevent (MACE) primary composite endpoint. Four additionally provided evi-dence of CV benefit in the form of signif-icant decreases in the MACE primarycomposite end point (10,11,13,14), twofound reductions in CV death (10,13),and three showed reductions in all-causemortality (10,12,14)dalthough the sta-tistical robustness of findings for thesesecondary end points in some cases maybe controversial.The completed CVOTs have provided
much valuable information, and becausenumerous antidiabetes drugs have beendeveloped and tested in the past decade,it appears that the FDA mandate has notdiscouraged pharmaceutical companiesfrom pursuing approval for potentiallysuccessful drugs. Still, questions remainregarding whether the information ob-tained throughCVOTs designed accordingto the FDAmandate justifies the time andresources needed, especially in light ofthe neutral results ofmany of the studies.In June2016,aDiabetesCareEditors’ ExpertForum convened and began to review theprocesses and challenges of CVOTs, dis-cuss the benefits and drawbacks of theircurrent designs, and weigh the merits ofpotential modifications that might im-prove the efficiency and clinical value offuture trials. This process continuedthrough September 2017 to include addi-tional CVOTdata reported at that time. This
article summarizes the proceedings of theExpert Forum and CVOT findings to date.
PRE-GUIDANCE RESEARCH ANDCONTROVERSY
Recent data indicate that, because of ad-vances in research and clinical care, morbid-ityandmortalityhavedecreasedsignificantlyinboth type1and type2diabetes.However,individualswithdiabetes still havegreaterCVrisk than those without diabetes (15). Thus,additional strategies to reduce this riskcontinue to be evaluated.
Well before 2008, diabetes was knowntomore than double the risk for CV eventsand was considered by some a coronaryartery disease equivalent (16). Althoughseveral major clinical trials ofmore versusless stringent glycemic control initiallyfailed to demonstrate that intensive glu-cose lowering significantly reduces CV risk(17–23), meta-analyses indicated a mod-estly reduced risk of nonfatal myocardialinfarction (MI) (24,25). In addition, long-term noninterventional follow-up of theDCCT (Diabetes Control and Complica-tions Trial) (26,27) in type 1 diabetesand the UKPDS (UK Prospective DiabetesStudy) (28) and VADT (Veterans AffairsDiabetes Trial) (29) in type 2 diabetes sug-gested an eventual “legacy” CV benefit ofinitially tighter glycemic control (Table 1)(17–22,26–30), but one that was not asapparent as the effect on microvascularcomplications. However, increased CVmortality in patients assigned to a moreintensive glucose-lowering strategy in theACCORD (Action to Control Cardiovascu-lar Risk in Diabetes) trial (20,30) and noreduction in CVmortality risk in the VADT(22,29) tempered enthusiasm for strin-gent glucose lowering, particularly inolder patients at high CV risk.
Less well understood was the CV safetyprofile of available antidiabetes drugs. Be-fore the FDA issued its 2008 guidance, ap-proval decisions for such medicationsfocused largely on short-term glycemic ef-ficacy (i.e., A1C reduction), along withsafety data from 6- to 12-month phase2 and 3 randomized clinical trials. Thesetrials typically enrolled younger cohortswith relatively recent onset of diabetesand low CV risk. Indeed, individuals withestablished CVD were usually excluded.Thus, these studies had low CV eventrates, and the events that did occur wereneither prespecified as end points nor in-dependently adjudicated. As a result, esti-mates of theCV safety of thesedrugswereinconsistent and unreliable (31,32).
Concerns about this gap in assessmentof CV safety grew after two highly contro-versialmeta-analysesofCV risk (3,33)werepublished in the mid-2000sdthe first onmuraglitazar, an investigational dual per-oxisome proliferator–activated receptor(PPAR)-a and -g agonist, and the secondon the FDA-approved TZD (and PPAR-gagonist) rosiglitazone. The muraglitazaranalysis showed more than twice the in-cidence of CV complications, includingcongestive heart failure (HF) and death,with the drug compared with standardtherapy (33). Although this drug had re-ceived a vote for approval from the FDA’sEndocrinologic and Metabolic Drugs Ad-visory Committee (34), these findings ef-fectively ended its development. Themeta-analysis of rosiglitazone, which hadalready been marketed for a decade,suggested a statistically significant 43%increased risk of MI and a statisticallynonsignificant 64% increased risk of CVdeath versus comparators (3). In thewake of these publications and the heated
Table 1—Early major trials evaluating the effects of intensive glycemic control of diabetes
Study Diabetes type CV composite MI CV mortality All-cause mortality
DCCT/EDIC (17,26,27) Type 1 ↔ ↓ d d d d ↔ ↓
UKPDS Type 2Main randomization (SUor insulin vs. conventional therapy)(18,28) d d ↔ ↓ d d ↔ ↓Additional randomization of overweight patients(metformin vs. SU vs. conventional therapy) (19,28) d d ↓* ↓* d d ↓* ↓*
ACCORD (20,30) Type 2 ↔ ↔ ↓ ↔ ↑ ↑ ↑ ↔
ADVANCE (21) Type 2 ↔† ↔ ↔ ↔
VADT (22,29) Type 2 ↔ ↓ ↔ ↔ ↔ ↔ ↔ ↔
Left columns show initial results; right columns show long-term follow-up.↔, Neutral effect; ↓, decrease; ↑, increase;d, not assessed/reported;ADVANCE, Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation; SU, sulfonylurea. Adapted from Bergenstalet al. (97). *Metformin group only. †A decrease was reported in a combined CV/microvascular composite but was found to be mostly attributable tonephropathy.
care.diabetesjournals.org Cefalu and Associates 15
controversy they ignited, the FDA issuedits CV safety guidance.After a subsequent study specifically
designed to test the CV safety of rosiglitazonedid not confirm the earlier findings (35,36),the FDA lifted its prescription restrictionsfor rosiglitazone (37). Even so, use of ro-siglitazone remains negligible. Trials ofanother TZD, pioglitazone, have shownCV benefit in high-risk patients. The PRO-active (Prospective Pioglitazone ClinicalTrial in Macrovascular Events) study (38)found that pioglitazone reduced the sec-ondary composite end point of all-causemortality, nonfatal MI, and stroke in pa-tients with type 2 diabetes who were athigh risk for macrovascular events. Themore recent IRIS (Insulin Resistance Inter-vention After Stroke) trial (39,40) foundthat, in insulin-resistant patients (manywith prediabetes but none with diabetes)with previous ischemic stroke or transientischemic attack, treatment with this drugsignificantly reduced fatal and nonfatalMI and stroke.
REVIEW OF THE GUIDANCE
The FDA, following its usual process, castits expectations as recommendationsrather than requirements, but also notedthat theywere “for immediate implemen-tation to ensure that relevant issues re-lated tominimizing cardiovascular risk areconsidered in ongoing drug developmentprograms” (1). The guidance outlined adetailed process for evaluating the CVsafety of type 2 diabetes drugs in devel-opment (Table 2) (1,41).
It was suggested that separate CVOTswould be necessary only if meta-analysescould not rule out an unacceptable levelof risk. However, the number ofCVeventsaccrued during a typical phase 2 or 3 de-velopment program is usually insufficienttoprovidehigh statistical confidence. Thus,as noted in a review by Regier et al. (42),every new antidiabetes drug approvedsince 2008 has had a dedicated CVOT toevaluate whether the upper limit of atwo-sided 95% CI of the estimated hazardratio (HR) is ,1.8, the requirement nec-essary to gain initial approval. Althoughthe FDA’s guidance specifically excludedinsulin from its CV safety evaluation, not-ing that its necessity as a lifesaving therapyfor type 1 diabetes makes such evaluationimpractical, some newer insulin formula-tions have been studied in CVOT trialsbased on findings from meta-analysis ofphase 2 and 3 studies (43).
SUMMARY OF CVOTS TO DATE
Trial DesignsFigure 1 displays a timeline of completedand ongoing diabetes CVOTs, as well asseveral related trials that assessed CVoutcomes but were not initiated as a di-rect result of the 2008 FDA guidance(6–14,39,44–58). These trials weremainlydesigned to rule out unacceptable CV risk,but some were powered to estimate su-periority after noninferiority was dem-onstrated. They typically have studiedpopulations in which some or all patientshad advanced atherosclerotic CV risk orestablished CVD to ensure accrual of
sufficient events in a timely manner.Some had active run-in periods to en-hance adherence to the study drug. Theirparticipants have a long duration of dia-betes (mean 7.1–16.4 years), with base-line mean A1C ranging from 7.2 to 8.7%.The characteristics of the completedCVOTs and their primary findings aresummarized in Table 3 (6–14,59,60). On-goingCVOTsaresummarized inSupplementaryTable1 (44–56), and SupplementaryTable2summarizes additional trials that assessedCV outcomes but were not initiated as adirect result of the 2008 FDA guidance(39,57,58,61).
Most of the CVOTs were designed asphase4event-driven trials requiring.600primary end point events to rule out withstatistical confidence anHRwith anupperconfidence limit$1.3 for a 3-point MACE(CV death, nonfatal MI, and nonfatalstroke). In three trials (TECOS [Trial Eval-uating Cardiovascular Outcomes withSitagliptin] [8], ELIXA [Evaluation ofLixisenatide inAcute Coronary Syndrome][9]), and FREEDOM-CVO [A Study to Eval-uate Cardiovascular Outcomes in PatientsWith Type 2 Diabetes Treated With ITCA650] [53]) a 4-point MACE, adding hos-pitalization for unstable angina to thecomponents of the 3-point MACE, is theprimary end point. In the ACE (AcarboseCardiovascular Evaluation) trial (58) of thea-glucosidase inhibitoracarbose,conductedin China in patients with coronary heartdisease and impaired glucose tolerance(IGT), the primary end point is a 5-pointMACE, adding hospitalization for unstable
Table 2—Summary of the 2008 FDA guidance on CVOTs (1)
To adequately evaluate the CV safety of type 2 diabetes drugs in development, future development programs should include:
c Phase 2 and 3 trials that include patients at higher risk for CV events, are of sufficient size and duration to enable enough CV events to allow for ameaningful evaluation of CV risk, and are designed to facilitate latermeta-analysis; the CV events should include CVmortality,MI, and stroke and canalso include hospitalization for ACS, urgent revascularization procedures, and other end points such as HF hospitalization*
c Independent adjudication of CV eventsc Meta-analysis of the phase 2 and 3 trials to be conducted at the end of the research program, following a protocol developed in advance thatprespecifies the end points to be assessed and the statistical methods to be employed
cAnalysis of premarketing data comparing the CVevents occurringwith the agent to those occurringwith the control group and demonstrating that theupper limit of a two-sided 95% CI of the estimated risk ratio is,1.8; if this cannot be done through the meta-analysis described above, it should beaccomplished in a separate, large CV safety trial
c For agents whose 95% CI upper limit falls between 1.3 and 1.8 in premarketing analysis, completion of a postmarketing trial or continuation of apremarketing trial after approval may be needed to conclusively show that the upper limit of the two-sided 95% CI is,1.3 with a “reassuring” pointestimate of overall CV risk
*TheFDAhasgainedvaluable experience fromCVOTs andCV risk assessments intended tomeet theDecember2008 guidance. TheFDAhas acknowledgedthat allowing MACE+ (a composite of CV death, nonfatal MI, or nonfatal stroke plus the less-specific end point of unstable angina requiring eitherhospitalization or revascularization) in the premarket CV risk assessment to exclude a 95% CI upper limit$1.8 is acceptable because it allows for a morereasonable sample size for evaluation. Provided no countervailing safety finding is observed, MACE+ in the premarket risk assessment strikes anappropriate balance of safety assessment without an undue burden to companies bringing new therapies to market. However, to provide longer-term,more reassuring CV safety data in the postmarketing setting, the exclusion of a 95% CI upper limit $1.3 should rely on the MACE composite(CV death, nonfatal MI, and nonfatal stroke). Furthermore, these data should come from a dedicated CV trial and not from the meta-analysis ofmultiple phase 2 or 3 trials (41).
16 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
angina or HF to the components of the3-point MACE. Additionally, most analysisplans permitted testing for superiority toplacebo if the noninferiority HR thresholdof,1.3 were achieved, providing an op-portunity to show potential CV benefit.All of the CVOTs in populations with
diabetes were designed to promote “gly-cemic equipoise” between the treatmentgroups to minimize the potentially con-founding effect of differences in glycemiccontrol. Accordingly, treatment intensifi-cation with other oral antidiabetes drugsor insulin was more prevalent in the con-trol groups. Even so, modest between-group differences in A1C changes wereobserved at the end of most trials, withhigher values in the placebo arms.All primary resultswere analyzed in the
intention-to-treat (ITT) (as randomized)or modified ITT (as treated with at leastone dose) population, with evaluations inthe on-treatment or per-protocol popula-tions reported as sensitivity analyses.Most study patients were receivingnear-optimal CV risk management atbaseline, as shown by a high proportionof patients receiving antihypertensive,lipid-lowering, and antiplatelet medica-tions. Differences in study populations,
baseline patient characteristics, and de-signs make it difficult to compare resultsamong these trials.
Effects on Cardiometabolic RiskFactorsThe impact on cardiometabolic risk fac-tors with DPP-4 inhibitors was minimalexcept for small reductions in A1C of0.3% at study end (6–8).
The EMPA-REG OUTCOME (BI 10773[Empagliflozin]CardiovascularOutcomeEventTrial in Type 2 Diabetes Mellitus Patients)trial of the SGLT2 inhibitor empagliflozinshowed a small but statistically significant0.3% A1C reduction at study end, simi-lar reductions in systolic (4–5 mmHg)and diastolic (2 mmHg) blood pressureand weight (;2 kg), and minimal in-creases in LDL and HDL cholesterol, withno changes observed in heart rate (13). Inthe CANVAS (Canagliflozin Cardiovascu-lar Assessment Study) Program (14),there were statistically significant reduc-tions in mean A1C updated over time(0.58%), systolic (3.93mmHg) and diastolic(1.39 mmHg) blood pressure, and weight(1.6 kg) and increases in LDL and HDL cho-lesterol (4.68 and 2.05 mg/dL, respec-tively) with canagliflozin treatment (14).
In the ELIXA trial of the GLP-1 receptoragonist lixisenatide (9), there were statis-tically significant reductions in A1C atstudy end (0.3%), systolic blood pressure(0.8 mmHg), and weight (0.7 kg), with aslightly increased heart rate (0.4 bpm). Inthe LEADER (Liraglutide Effect and Actionin Diabetes: Evaluation of CardiovascularOutcome Results) trial (10), treatmentwith liraglutide was associated with asmall but statistically significant increasein heart rate (3.0 bpm), with reductions inmean updated A1C (0.4%), systolic bloodpressure (1.2 mmHg), and weight (2.3 kg)and a slight increase in diastolic bloodpressure (0.6 mmHg). Similar resultswere found in SUSTAIN-6 (Trial to Evalu-ate Cardiovascular and Other Long-termOutcomes With Semaglutide in SubjectsWith Type 2 Diabetes) (11) with 0.5 and1.0 mg doses of semaglutide; heart rateincreased by 2.0 and 2.5 bpm, respec-tively, with robust reductions in A1C atstudy end (0.7 and 1.0%, respectively), sys-tolic blood pressure (1.3 and 2.6 mmHg,respectively), and weight (2.9 and 4.3 kg,respectively) and no difference in diastolicblood pressure. Finally, in the EXSCEL(Exenatide Study of Cardiovascular EventLowering) trial (12) with once-weekly
Figure 1—Completed and ongoing CVOTs (6–14,39,44–58). 3-P, 3-point; 4-P, 4-point; 5-P, 5-point. DECLARE-TIMI 58, Multicenter Trial to Evaluate theEffect of Dapagliflozin on the Incidence of Cardiovascular Events; ESRD, end-stage renal disease; HARMONYOutcomes, Effect of Albiglutide,When Addedto Standard Blood Glucose Lowering Therapies, onMajor Cardiovascular Events in SubjectsWith Type 2 DiabetesMellitus; PIONEER 6, A Trial Investigatingthe Cardiovascular Safety of Oral Semaglutide in Subjects With Type 2 Diabetes; REWIND, Researching Cardiovascular Events With a Weekly Incretin inDiabetes; VERTIS CV, Cardiovascular Outcomes Following Ertugliflozin Treatment in Type 2 Diabetes Mellitus Participants With Vascular Disease.
care.diabetesjournals.org Cefalu and Associates 17
Tab
le3—
CVOTsco
mpletedafterissu
ance
oftheFD
A2008guidan
ce
DPP-4
inhibitors
GLP-1
receptor
agon
ists
SGLT2inhibitors
SAVOR-TIM
I53
(6)
(n=16,492)
EXAMINE(7,59)
(n=5,380)
TECOS(8)
(n=14,671)
ELIXA(9)
(n=6,068)
LEADER
(10)
(n=9,340)
SUSTAIN-6
(11)*
(n=3,297)
EXSC
EL(12)
(n=14,752)
EMPA
-REG
OUTC
OME(13,60)
(n=7,020)
CANVAS
Program
(14)
(n=10,142)
Intervention
Saxagliptin/
placebo
Alogliptin/
placebo
Sitagliptin/
placebo
Lixisenatide/
placebo
Liraglutide/
placebo
Semaglutide/
placebo
Exen
atideQW/
placebo
Empaglifl
ozin/
placebo
Can
aglifl
ozin/
placebo
Maininclusion
criteria
Type2diabetes
andhistory
of
ormultiplerisk
factors
forCVD
Type2diabetes
andACSwithin
15–90days
before
rand
omization
Type2diabetes
andpreexisting
CVD
Type2diabetes
andan
acute
coronary
even
twithin
180days
before
screen
ing
Type2diabetes
andpreexisting
CVD,kidney
disease,o
rHF
at$50
yearsof
ageor
$1CV
risk
factor
at$60
years
ofage
Type2diabetes
andpreexisting
CVD,HF,orCKD
at$50
years
ofageor
$1CVrisk
factor
at$60
years
ofage
Type2diabetes
withor
without
preexisting
CVD
Type2diabetes
andpreexisting
CVD,w
ith
BMI#45
kg/m
2
andeG
FR$
30mL/min/
1.73
m2
Type2diabetes
andpreexisting
CVDat
$30
years
ofageor
$2CVrisk
factorsat
$50
years
ofage
A1C
inclusion
criterion(%)
$6.5
6.5–11.0
6.5–8.0
5.5–11.0
$7.0
$7.0
6.5–10.0
7.0–10.0
7.0–10.5
Age
(years)†
65.1
61.0
65.4
60.3
64.3
64.6
6263
.163.3
BMI(kg/m
2)†
31.1
28.7
30.2
30.2
32.5
32.8
31.8
30.7
32
Diabetes
duration
(years)†
10.3
7.1
11.6
9.3
12.8
13.9
1257
%.10
13.5
Eventsplanned/
observed
(n)
1,040/1,222
650/621
1,300
/1,690
844/805
$611/1,302
$122/254
1,36
0/1,74
469
1/772
688/1,011
Medianfollow-up
(years)
2.1
1.5
3.0
2.1
3.8
2.1
3.2
3.1
2.4
Statinuse(%)
7891
8093
7273
7477
75
PriorCVD/CHF(%)
78/13
100/28
74/18
100/22
81/18
60/24
73.1/16.2
99/10
65.6/14.4
A1C
/A1C
change
(%)‡
8.0/–0.3
8.0/–0.3
7.2/–0.3
7.7/–0.3
8.7/–0.4
8.7/–0.7or
–1.0§
8.0/–0.53
8.1/–0.3|
8.2/–0.58
Year
started/repo
rted
2010/201
320
09/201
32008/201
520
10/201
52010/201
620
13/201
62010/2017
2010/2015
2009/201
7
Primaryou
tcom
e¶3-pointMACE
3-pointMACE
4-pointMACE
4-pointMACE
3-pointMACE
3-pointMACE
3-pointMACE
3-pointMACE
3-pointMACE
1.00
(0.89–1.12)
0.96(95%
UL#1.16)
0.98
(0.89–1.08)
1.02(0.89–1.17)
0.87
(0.78–0.97)
0.74(0.58–0.95)
0.91(0.83–1.00)
0.86
(0.74–0.99)
0.86(0.75–0.97)#
Keysecond
ary
outcom
e¶Expan
ded
MACE
4-pointMACE
3-pointMACE
ExpandedMACE
Expanded
MACE
ExpandedMACE
Individual
componen
tsofMACE
(see
below)
4-pointMACE
All-cause
and
CVmortality
(see
below)
1.02
(0.94–1.11)
0.95(95%
UL#1.14)
0.99
(0.89–1.10)
1.00(0.90–1.11)
0.88
(0.81–0.96)
0.74(0.62–0.89)
0.89
(0.78–1.01)
CVdeath¶
1.03(0.87–1.22)
0.85(0.66–1.10)
1.03
(0.89–1.19)
0.98(0.78–1.22)
0.78
(0.66–0.93)
0.98(0.65– 1.48)
0.88(0.76–1.02)
0.62
(0.49–0.77)
0.96(0.77–1.18)**
0.87
(0.72–1.06)#
MI¶††
0.95(0.80–1.12)
1.08(0.88–1.33)
0.95
(0.81–1.11)
1.03(0.87–1.22)
0.86
(0.73–1.00)
0.74(0.51–1.08)
0.97(0.85–1.10)
0.87
(0.70–1.09)
0.89(0.73–1.09)#
Stroke¶††
1.11(0.88–1.39)
0.91(0.55–1.50)
0.97
(0.79–1.19)
1.12(0.79–1.58)
0.86
(0.71–1.06)
0.61(0.38–0.99)
0.85(0.70–1.03)
1.18
(0.89–1.56)
0.87(0.69–1.09)#
Continuedon
p.19
18 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
exenatide, mean updated A1C was 0.53%lower, weight was 1.3 kg lower, and sys-tolic blood pressurewas 1.6mmHg lower,but heart rate was 2.5 bpm higher.
Overall, the impact of treatment inter-ventions on cardiometabolic risk factorswas small, and the resultingeffects of thesechanges on CV outcomes remain unclear.
CV Outcomes
DPP-4 Inhibitors
Five trials enrolling 49,618 patients havebeen designed to evaluate the CV safetyof DPP-4 inhibitors. Three have reportedoutcome results (6–8), and two are ongo-ing (44,45).One additional trial, OMNEON(A Study to Assess Cardiovascular Out-comes Following Treatment With Omari-gliptin [MK-3102] in Participants WithType 2 Diabetes Mellitus [MK-3102-018])(62), evaluating the investigational once-weekly omarigliptin in 4,202 patientswithtype 2 diabetes and CVD, was terminatedas a business decision (63). Although allthree completed trials (6–8) met the pri-mary objective of excluding an HR upperlimit $1.3, none were associated withany suggestion of CV benefit. Likewise,at its termination after amedian follow-upof 96 weeks, the OMNEON trial reportedan HR of 1.00 (95% CI 0.77–1.29) for itsprimary 3-point MACE end point (62).
The SAVOR-TIMI 53 (Saxagliptin As-sessment of Vascular Outcomes Recorded inPatients with DiabetesMellitus–Thrombolysisin Myocardial Infarction) trial (6) with sax-agliptin suggested an increased risk forincident HF (HR 1.27 [95% CI 1.07–1.51],P = 0.007 [6]), with some support pro-vided by a trend noted in the EXAMINE(Examination of Cardiovascular Outcomeswith Alogliptin versus Standard of Care)trial with alogliptin (HR 1.19 [95% CI0.90–1.58], P = 0.220 [7,59]), earning awarning by the FDA, especially in patientswith underlying heart and kidney disease(64).
CAROLINA (Cardiovascular OutcomeStudy of Linagliptin Versus Glimepiridein Patients With Type 2 Diabetes) (44)was designed to establish noninferiorityof linagliptin relative to the sulfonylureaglimepiride, thus including an active com-parator. However, because the CV safetyof glimepiride has not been established, aplacebo-controlled trial (CARMELINA [Car-diovascular and Renal Microvascular Out-come Study With Linagliptin in PatientsWith Type 2 Diabetes Mellitus] [45]) isalso being conducted to establish the CV
Tab
le3—
Continued
DPP-4
inhibitors
GLP-1
receptor
agon
ists
SGLT2inhibitors
SAVOR-TIM
I53
(6)
(n=16,492)
EXAMINE(7,59)
(n=5,380)
TECOS(8)
(n=14,671)
ELIXA(9)
(n=6,068)
LEADER
(10)
(n=9,340)
SUSTAIN-6
(11)*
(n=3,297)
EXSC
EL(12)
(n=14,752)
EMPA
-REG
OUTC
OME(13,60)
(n=7,020)
CANVAS
Program
(14)
(n=10,142)
HFho
spitalization¶
1.27(1.07–1.51)
1.19(0.90–1.58)
1.00
(0.83–1.20)
0.96(0.75–1.23)
0.87
(0.73–1.05)
1.11(0.77–1.61)
0.94(0.78–1.13)
0.65
(0.50–0.85)
0.67(0.52–0.87)#
Unstableangina
hospitalization¶
1.19(0.89–1.60)
0.90(0.60–1.37)
0.90
(0.70–1.16)
1.11(0.47–2.62)
0.98
(0.76–1.26)
0.82(0.47–1.44)
1.05(0.94–1.18)
0.99
(0.74–1.34)
d
All-causemortality¶
1.11(0.96–1.27)
0.88(0.71–1.09)
1.01
(0.90–1.14)
0.94(0.78–1.13)
0.85
(0.74–0.97)
1.05(0.74–1.50)
0.86(0.77–0.97)
0.68
(0.57–0.82)
0.87(0.74–1.01)**
0.90
(0.76–1.07)#
Worsening
neph
ropathy¶
‡‡
1.08(0.88–1.32)
dd
d0.78
(0.67–0.92)
0.64(0.46–0.88)
d0.61
(0.53–0.70)
0.60(0.47–0.77)#
d,not
assessed/reported;95%UL,up
perlim
itof95%CI;CHF,congestive
heartfailure.*Po
wered
toruleou
tanHRup
permargin$1.8;superiorityhypo
thesisno
tprespecified.†Age
andBMIw
ererepo
rted
asmeans
inalltrialsexcept
EXAMINE,which
repo
rted
medians;diabetesdu
ration
was
repo
rted
asmeans
inallbut
four
trials,w
ithSA
VOR-TIM
I58,EXAMINE,andEXSCEL
repo
rtingmedians
andEM
PA-REG
OUTCOMErepo
rtingas
percentage
ofpo
pulation
withdiabetes
duration
.10
years.‡ForTECOS,LEADER,EXSCEL,and
theCA
NVASProgram,A1C
difference
was
updatedovertime;forSAVOR-TIM
I53,EXAMINE,ELIXA,SUSTAIN-6,and
EMPA
-REG
OUTCOME,A1C
difference
was
atstud
yend.§A
1Cchange
of0.66%with0.5mgand1.05%with1mgdo
seofsemaglutide.|A1C
change
of0.30
inEM
PA-REG
OUTCOMEisbasedon
pooled
resultsforb
othdo
ses(i.e.,
0.24%for10
mgand0.36%for25
mgof
empagliflozin).¶Outcomes
repo
rted
asHR(95%
CI)un
lessotherw
iseno
ted.#N
ontrun
catedintegrated
data(refersto
pooled
datafrom
CANVAS,includ
ingbefore
20Novem
ber
2012
plus
CANVAS-R).**Trun
catedintegrated
dataset(referstopo
oled
datafrom
CANVASafter20Novem
ber2012plus
CANVAS-R;prespecified
intreating
hierarchyastheprincipaldatasetfor
analysisforsup
eriorityof
all-cause
mortalityandCVdeathintheCANVASProgram).††Reportedforfatalandno
nfataleventsinalltrialsexcept
EXAMINE,ELIXA,and
SUSTAIN-6,w
hich
repo
rted
forn
onfataleventson
ly.‡‡Worsening
neph
ropathy
was
definedas
doub
lingof
creatinine
level,initiation
ofdialysis,renaltransplantation,or
creatinine
.6.0mg/dL
(530
mmol/L)inSA
VOR-TIM
I53;as
thenewon
setof
macroalbu
minuria(urine
albu
mincreatinine
ratio.300mg/g)or
ado
ublingoftheserumcreatinine
leveland
aneG
FRof#45
mL/min/1.73m
2,the
need
forcon
tinu
ousrenal-replacementtherapy,ord
eath
from
renaldisease
inLEADER,SUSTAIN-6,and
EMPA
-REG
OUTCOME;andas
40%redu
ctionineG
FR,renal-replacementtherapy,or
deathfrom
renalcausesinCANVAS.Worseningneph
ropathywas
aprespecified
exploratoryadjudicatedou
tcom
einSA
VOR-TIM
I53,LEADER,
SUSTAIN-6,and
CANVASbu
tno
tinEM
PA-REG
OUTCOME.
care.diabetesjournals.org Cefalu and Associates 19
and renal safety of linagliptin by includingan enriched population with CV risk andevidence of renal compromise. These tri-als may help to answer the question of CVsafety with sulfonylureas, which has beenlingering for nearly 50 years (65).Overall, the DPP-4 inhibitors have es-
tablished MACE safety.SGLT2 Inhibitors
The CV and renal safety of SGLT2 in-hibitors is being evaluated in nine trialsenrolling 62,378 patients. Of these, twodEMPA-REGOUTCOME(13) and theCANVASProgram (14)dhave been completed,and each reported positive CV and renaloutcomes. The other seven trials are sched-uled for completion within the next 2–3years (46–52).In four SGLT2 inhibitor trials, a compos-
ite 3-pointMACE is the primary end point;in two, a composite renal outcome is theprimary end point; and in three, a com-posite ofHFoutcomes andCVdeath is theprimary end point in people with estab-lished HF. Although primarily renal stud-ies, CANVAS-R (CANVAS-Renal) (66) (datafrom which were included in the com-bined report on the overall CANVAS Pro-gram [14]) and the ongoing CREDENCE(Evaluation of the Effects of Canagliflozinon Renal and Cardiovascular Outcomes inParticipants With Diabetic Nephropathy)trial (46) also prospectively collected/arecollecting adjudicated CV outcomes, asrequired by the FDA mandate. Dapa-CKD (A Study to Evaluate the Effect ofDapagliflozin onRenalOutcomes andCar-diovascular Mortality in Patients WithChronic Kidney Disease) (50) will alsoevaluate CV death or HF hospitalizationin addition to the primary composite re-nal outcome. Plans for a dedicated studyto evaluate empagliflozin on the progres-sion of CKD were recently announced(67). Although primarily HF studies,Dapa-HF (Study to Evaluate the Effect ofDapagliflozin on the Incidence ofWorsen-ing Heart Failure or Cardiovascular Deathin Patients With Chronic Heart Failure[49]) and EMPEROR-Reduced (Empagli-flozin Outcome Trial in Patients WithChronic Heart Failure with Reduced Ejec-tion Fraction [51]), bothofwhich are in HFpatients with reduced ejection fraction,and EMPEROR-Preserved (EmpagliflozinOutcome Trial in Patients With ChronicHeart Failure With Preserved EjectionFraction [52]), in HF patients with pre-served ejection fraction, will also prospec-tively evaluate adjudicated renal outcomes.
EMPA-REGOUTCOME(13)wasanevent-driven trial in patients at high CV riskrandomly treated with 10 or 25 mgof empagliflozin or placebo, with a me-dian observation time of 3.1 years. Theprimary hypothesis was noninferiorityfor the primary 3-point MACE endpoint with empagliflozin (pooled dosesof 10 and 25 mg) versus placebo, withan HR upper limit of ,1.3. The primaryend point was significantly reduced withempagliflozin treatment (HR 0.86 [95% CI0.74–0.99], P = 0.038). This reductionwasdriven by a significant decrease in CVdeath (HR 0.62 [95% CI 0.49–0.77], P ,0.001), with numerical, but not statisti-cally significant, differences in nonfatalMI (favoring empagliflozin) and nonfatalstroke (favoring placebo). There was nodifference in the key secondary end pointof 3-point MACE plus hospitalization forunstable angina (HR 0.89 [95% CI 0.78–1.01], P = 0.08). Significant reductions inprespecified end points of all-cause mor-tality (32%) and HF hospitalization (35%)were also observed. Superiority in termsof CV death was large and clinically im-portant (38% risk reduction), statisticallyrobust (P, 0.001), based on a large num-ber of events (n = 309), occurring early(within 3 months), consistently seenwith both doses across subgroups, andestablished across sensitivity analysesthat accounted for silent MIs (whichwere excluded from event adjudication),nonassessable deaths, and missing data.There was an increased rate of treatment-emergent genital infection but no increasein other adverse events, including diabeticketoacidosis, bone fractures, venous throm-boembolism,hypoglycemicevents, amputa-tions, and acute kidney injury. Interestingly,reductions in CV outcomes appeared to beunexplained by empagliflozin’s glucose-lowering effect or by its effects on bloodpressure, weight, or uric acid, suggestingthat other mechanisms activated bySGLT2 inhibition may contribute to thisdrug’s cardioprotective effects (68–70).
Basedon this favorable benefit-risk pro-file, the FDAapproved anew indication forempagliflozin on 2 December 2016 “to re-duce the risk of cardiovascular death inadult patients with type 2 diabetes melli-tus and cardiovascular disease” (71), aclaim that represented a clinical break-through in treating patients with type 2diabetes. It was also a regulatory break-through because CV death was a second-ary outcome but, in view of the positive
results of the primary outcome and therobustness of the CV death data, this in-dication was approved. On 15 December2016, the American Diabetes Associa-tion (ADA) endorsed this recommenda-tion in its Standards of Medical Care inDiabetesd2017 (72). Although EMPA-REGOUTCOMEwas thefirst trial to demon-strate the CV superiority of an antidiabetesdrug, more research is needed to elucidatetheunderlyingmechanisms, confirmwhetherthe CV benefits are a class effect, and deter-mine whether empagliflozin conveys a CVbenefit to patients without established CVDor diabetes.
The CANVAS Program was originallydesigned in two phases to establish theCV protective effects of canagliflozin inhigh-risk patients with type 2 diabetesand established CVD or risk factors forCVD (73). The first phase was includedin the meta-analysis presented to theFDA for regulatory approval (74), whichruled out an HR upper limit $1.8. How-ever, these resultswere publicly disclosedat an advisory committee meeting in2013 after partial (group-level) unblindingdue to an observed dose-dependent in-crease in LDL cholesterol. The FDA feltthat the lipid data necessitated public dis-closure, and with it, the CV event datawere also disclosed, which had the poten-tial to compromise the postmarketingphaseof the trial to fulfill the requirementof ruling out an HR upper limit $1.3. Be-causeof theseextenuatingcircumstances, theFDA accepted an integrated analysis plan,whereby data from extended follow-up ofthe CANVAS first phase before and after thedata disclosure (20 November 2012) wouldbe combined with new data from CANVAS-R(66) to address CV safety as assessed byexclusion of a postmarketing risk ratio of1.3. A sequential hypothesis testing planwas used, whereby a truncated integrateddata set of pooled data fromCANVAS after20 November 2012 plus CANVAS-R wasprespecified as the principal data set foranalysis for superiority of all-cause mortal-ity and CV death (75).
Theprimary3-pointMACEendpointwassignificantly reduced with canagliflozintreatment (HR 0.86 [95% CI 0.75–0.97],P = 0.02), thereby meeting the crite-rion for noninferiority using the 1.3 riskmargin (14). It is debatable, from a rigor-ous statistical perspective, whether an in-ference of superiority is justified becauseit was apparently not prespecified in thetesting sequence. Because the all-cause
20 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
mortality difference was not statisticallysignificant in the truncated integrateddata set (HR 0.90 [95% CI 0.76–1.07], P =0.24), given the statistical rules of the pre-specified hierarchical testing strategy, allsubsequent results, including the favor-able effects on HF hospitalization, pro-gression of albuminuria, and worseningnephropathy, were deemed exploratory.The point estimate for nonfatal or totalstroke was favorable for canagliflozin.Therewasan increasedrateof treatment-
emergent genital infection as expected,but no increase in venous thromboem-bolism or hypoglycemic events. Diabeticketoacidosis was very low but slightly nu-merically increased with canagliflozin, andthere was a nearly twofold increase in riskfor lower-leg amputation (6.3 and 3.4events/1,000 patient-years for canagliflozinand placebo, respectively) and a 26% rela-tive increase in risk of bone fracture withcanagliflozin treatment. The underlyingmechanisms for these findings remain un-clear. There are some potential biologicalexplanations for bone fractures, but nonefor the small but significant andunexpectedincrease in lower-leg amputations. Neitherbone fracture nor lower-limb amputationhas been documented with the otherSGLT2 inhibitors, so additional evaluationis need to clarify whether this should beviewed as a class effect.GLP-1 Receptor Agonists
The CV safety of GLP-1 receptor agonistshas been assessed in eight trials enrolling60,090 patients, of which four have re-ported outcomes (9–12). A fifth trial hasbeen completed but not reported (53), andthe other three are scheduled for comple-tion within the next 1–2 years (54–56).The first reported, ELIXA (9), which was
conducted in patients with a history ofrecent acute coronary syndrome (ACS),was CV neutral, confirming the noninfer-iority of lixisenatide with respect to a4-point MACE but showing no beneficialeffect on any CV outcome.The second trial, LEADER (10), demon-
strated the CV noninferiority, as well asthe statistical superiority, of once-dailytreatment with liraglutide. The reductionin 3-point MACE (HR 0.87 [95% CI 0.78–0.97]) with liraglutide was driven by a sig-nificant reduction in CV death (HR 0.78[95% CI 0.66–0.93]) with numerical, butnot statistically significant, differences innonfatal MI and nonfatal stroke favoringliraglutide. The P value for the 3-pointMACE met superiority in all sensitivity
analyses. All-cause mortality, but not HFhospitalization, was also significantly re-duced with liraglutide. An interesting ob-servation is the delayed separation of theKaplan-Meier curves (indicating time tobenefit) in LEADER (.12 months for CVdeath and .18 months for all-causedeaths and HF hospitalization) (10), sug-gesting that the CV benefit may be con-veyed via an impact on atheroscleroticdisease, which contrasts with the earliertime tobenefit (,3months) in the EMPA-REG OUTCOME trial (13). On 25 August2017, based on the results of the LEADERtrial, the FDA approved a new indicationfor liraglutide “to reduce the risk of majoradverse cardiovascular events in adultswith type 2 diabetes mellitus and estab-lished cardiovascular disease” (76).
SUSTAIN-6 confirmed the noninferior-ity of once-weekly treatment with 0.5or 1 mg of the long-acting semaglutide,which is currently under FDA review (11).This phase 3 trial was designed to ruleout an HR upper limit$1.8 and achievedstatistical superiority (HR 0.74 [95% CI0.58–0.95], P , 0.001), but superiorityanalysis was not prespecified. The favor-able effect on the 3-point MACE was ac-companied by a significant decrease innonfatal stroke (HR 0.61 [95% CI 0.38–0.99], P = 0.04) and a nonsignificant de-crease in nonfatal MI (HR 0.74 [95% CI0.51–1.08], P = 0.12), with no trend forCV death or all-cause mortality. The riskreduction for the primary outcome wasseen despite an increase in pulse rate,an effect seen in all CVOTs in this drugclass to date. On 18 October 2017, theFDA’s Endocrinologic and MetabolicDrugs Advisory Committee voted in favorof approval for semaglutide (77).
The fourth trial, EXSCEL (12), was per-formed in a usual-care setting among pa-tientswith type2diabeteswith orwithoutprevious CVD. It confirmed the noninfer-iority, but not superiority, of once-weeklytreatment with 2 mg of the long-actingextended-release exenatide (HR 0.91[95% CI 0.83–1.00], P = 0.06). The risk ofdeath from any cause was 6.9% in theexenatide group and 7.9% in the placebogroup (HR 0.86 [95% CI, 0.77–0.97]). Thisdifference was considered only explor-atory on the basis of the hierarchical test-ing plan. The rates of CV death, fatal ornonfatal MI, fatal or nonfatal stroke, HFhospitalization, and ACS hospitalization;the incidences of acute pancreatitis, pan-creatic cancer, and medullary thyroid
carcinoma; and serious adverse eventsincluding severe hypoglycemia did notdiffer significantly between the two treat-ment groups.
Of interest, the treatment persistencewith weekly exenatide was low, with 43%drug discontinuation. The pragmatic na-ture of the study design, with visits every6 months and limited study support, mayexplain the low treatment adherence andpersistence. It is remarkable that despitethis limited drug exposure and a hetero-geneous population of whom27%had nohistory of CVD, the 3-point MACE reduc-tion of 9% came so close to reaching sta-tistical significance, with HRs of almost allmeasured parameters in the direction ofbenefit.
Although full results have not yet beenpublished, the phase3 FREEDOM-CVO tri-al, which evaluated continuous deliveryof exenatide and was designed for safetypurposes to accrue a limited number ofCVevents, reported in a press release (53)that the primary objective of achieving anHR upper limit,1.8 had been met, and anew drug application has been submittedto the FDA (78). Treatment persistencecould not have been an issue in this trialgiven the continuous delivery of exena-tide via a subcutaneously implantedmini-osmotic pump.However, the limitednumber of participants prevents any de-finitive conclusions beyond meeting theHR requirement for regulatory approval.
The heterogeneity in CVoutcomes (i.e.,null effects with lixisenatide and exena-tide and favorable effects with liraglutideand semaglutide) might be mainly the re-sult of differences in study populations,trial designs, and treatment persistence.However, it could also reflect differencesin pharmacokinetic andpharmacodynamicproperties (i.e., short-acting lixisenatideis a once-daily prandial drug that acts pri-marily on postprandial glucose, whereasliraglutide and semaglutide are longer-acting drugs that act mainly on fastingglucose with a carryover effect on post-prandial glucose) or other trial or drugdifferences, including in structural simi-larity to human GLP-1. What is clear sofar is that drugs in this class, if tolerated,are safe, and, if affordable, should be con-sidered more often for type 2 diabetesmanagement.Other Drugs
PPARAgonists. Two trials to determine theCV effects of aleglitazar, a dual activatorof PPAR-a and -g, were started but
care.diabetesjournals.org Cefalu and Associates 21
terminated early due to safety concernsand futility for efficacy. The AleCardio trial(79) enrolled 7,226 patients with a recentACS event. At its termination after a me-dian follow-up of 104 weeks, several CVand non-CV safety concerns were identi-fied, including increased risks of HF andbone fractures, decreased estimated glo-merular filtration rate (eGFR), and evi-dence of increased risk of gastrointestinalbleeding, which was unanticipated. TheALEPREVENT trial (80) enrolled 1,999 pa-tients with prediabetes or type 2 diabetesand established stable CVD. At its termina-tion after a brief treatment period (58 638 days), aleglitazar was associated withincreased incidences of hypoglycemiaand muscular events.As previously mentioned, the IRIS trial
of the TZD pioglitazone (39,40) reported asignificant 24% reduction in its compositeend point of fatal or nonfatal stroke orMIcompared with placebo after 4.8 years offollow-up in 3,876 insulin-resistant sub-jects with a recent ischemic stroke ortransient ischemic attack but without di-abetes. No between-group difference inall-cause mortality was found.TOSCA.IT (Thiazolidinediones or Sulfo-
nylureas Cardiovascular Accidents Inter-vention Trial) (81), a trial comparing theCV safety of pioglitazone to a sulfonyl-urea in 3,028 patients, was terminatedon the basis of a futility analysis after amedian follow-up of 57months. A total of213 events were adjudicated and in-cludedintheCVanalysis.Theprimarycom-posite end point (nonfatal MI, nonfatalstroke, all-cause death, and urgent coro-nary revascularization) showed no signif-icant difference between pioglitazoneand sulfonylurea treatment (HR 0.96[95% CI 0.74–1.26], P = 0.790). Pioglita-zone was associated with a reduced riskof severe hypoglycemia (,0.2 vs. 2.0%,P, 0.0001) and moderate hypoglycemia(10 vs. 32%, P , 0.001). Despite the factthat pioglitazone showed evidence of CVbenefit in both the PROactive (38) andIRIS (39,40) trials, this study failed to rep-licate the treatment benefit, albeit in alower-risk population. It should be notedthat TOSCA.IT was ultimately underpow-ered, with a substantial percentage of pa-tients not taking their assigned medica-tion because of concerns that emergedduring the trial about a possible linkbetween pioglitazone and bladder can-cer. A longer and larger active-controlledstudy (e.g., CAROLINA [44]) should shed
more light on the relative CV effects ofsulfonylureas.Insulin. Although undertaken 5 years be-fore the 2008 FDA guidance and thus out-side of the primary focus of this article,the ORIGIN Trial (Outcome Reductionwith an Initial Glargine Intervention)(82,83), established the neutral CV effectsof insulin glargine compared with stan-dard glycemic care in 12,537 participantswith CV risk factors plus impaired fastingglucose, IGT, or type 2 diabeteswhowerefollowed for a median of 6.2 years.
More recently, DEVOTE (A Trial Com-paring Cardiovascular Safety of InsulinDegludec Versus Insulin Glargine in Pa-tients With Type 2 Diabetes at High Riskof Cardiovascular Events) (57), involving7,637 patients with established CVDand amean diabetes duration of 16 years,confirmed the CV safety of insulin deglu-dec compared with insulin glargine (HR0.91 [95% CI 0.78–1.06], P = 0.21). Thisstudy was conducted after the FDArequired a preapproval degludec CVOTbased on the results of its phase 2 and3meta-analysis (43). Degludecwas statis-tically superior with regard to hypoglyce-mia risk, with a lower rate of both severeand nocturnal severe hypoglycemia (by40 and 53%, respectively; P , 0.001 forboth comparisons). These findings corre-sponded to absolute incidence rates forsevere hypoglycemia of 4.9 and 6.6% andfor severe nocturnal hypoglycemia of 1.0and 1.9% for degludec and glargine, re-spectively. Interestingly, despite the dif-ferences in severe hypoglycemia, nodifferences in CV mortality were found.a-Glucosidase Inhibitors. The ACE trial ofacarbose (58) in 6,522 Chinese patientswith coronary heart disease and IGTshowed no relative risk reduction for theprimaryoutcomeof 5-pointMACE (3-pointMACE plus hospitalization for unstable an-gina or HF) (HR 0.98 [95% CI 0.86–1.11],P = 0.73) or for any secondary end pointscompared with placebo. Although acar-bose treatment did not reduce CV risk,a modest reduction in incident type 2diabetes was observed (HR 0.82 [95%CI 0.71–0.94], P = 0.005) over a medianfollow-up of 5 years with a dose of 50 mgthree times daily.
Although ORIGIN, IRIS, DEVOTE, andACE were not initiated as a direct resultof the 2008 FDA guidance and OMNEON,AleCardio, ALEPREVENT, and TOSCA.IT were terminated early, these trials none-theless contributed important insights
to our overall understanding of the CVeffects of pharmacotherapies for type 2diabetes.
Benefits and Challenges of the CurrentCVOT Design
Benefits
Trials have shown that insulins glargine anddegludec, sitagliptin, alogliptin, saxagliptin,lixisenatide, and once-weekly exenatide(6–9,12,57,82,83) have neutral effectson MACE outcomes, thereby supportingthe use of these drugs when neededto improve glycemic control with the goalof limiting microvascular complicationswithout increasing CV risk. Well-designedand adequately powered CVOTs for otherantidiabetes drugs have also provided valu-able information beyond their primarysafety purpose. The evidence for CV benefitfrom empagliflozin, canagliflozin, liraglutide,and possibly semaglutide versus placebo(10,11,13,14), supplemented by datafrom other studies, already has promptedreconsideration of treatment guidelines(70,84) and is likely to alter clinical prac-tice and reduce CV events and deathsamong people similar to those enrolledin these trials.
Other insights from these trials havederived from secondary end points, whichare in some cases less statistically robustbut nonetheless of potential importance.
Notably, recent CVOTs have focusedattention on the pressing problem ofHF, which affects older people with dia-betes more frequently than MI (85).Empagliflozin caused an early, large, andsomewhat unexpected reduction in hos-pitalization for or death from HF (HR0.61 [95% CI 0.47–0.79], P , 0.001) (86).Canagliflozin also had favorable effects onHF, although perhaps of lessermagnitude(14). Less certain, but also of interest, isthe incidental observation of increasedHF with saxagliptin and a consistent non-significant trend with alogliptin, despiteno effect on other CV end points (6,59).If this harm is confirmed, it will draw at-tention to previously unsuspected effectsof and differences amongDPP-4 inhibitors,which block cleavage of many circulatingpeptides and thusmayhave various down-stream effects (either benign or harmful)other than reducing clearance and increas-ing blood levels of GLP-1.
Another clinically important benefit ofthese trials has been assessment of theeffect of the tested drugs on kidney dis-ease, itself a CV risk factor. For example,
22 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
empagliflozin (HR 0.61 [95% CI 0.53–0.70], P , 0.001) (60), canagliflozin (HR0.60 [95% CI, 0.47–0.77], P , 0.0001)(14), liraglutide (HR 0.78 [95% CI 0.67–0.92], P = 0.003) (10), and semaglutide(HR 0.64 [95% CI 0.46–0.88], P = 0.005)(11) all reduced progression of renal dis-ease more than would be expected fromthe improvements in glucose and bloodpressure provided by these drugs. Impor-tantly, the renal benefits of the SGLT2 in-hibitors extended to actual deterioration ofrenal function as denoted by a doubling ofserum creatinine or major reduction ineGFR, whereas those of the GLP-1 receptoragonists appear to be mainly on the pro-gression to macroalbuminuria. However,semaglutide appears to have a deleteriouseffect of diabetic retinopathy, possibly re-lated to rapid anddramatic improvement inglycemic control inpatientswithpreexistingretinopathy (11). A lesser trend for ad-verse eye outcomes with liraglutide re-quires further assessment (10).Together, these findings have ener-
gized the medical community. Many CVexperts appear to have revised their pre-vious skepticism about the potential forCV benefits from diabetes-specific thera-pies. Diabetes researchers are exploringmechanisms that may explain the clinicaleffects first noted in these trials. The find-ings with empagliflozin and canagliflozin,which alter renal clearance of sodium aswell as glucose and thus cause hemody-namic changes, have led to reconsidera-tion of abnormalities in salt and waterbalance that are intrinsic to diabetes, anarea long inactive in physiological re-search (87). Another provocative pro-posal is that injured myocardial andrenal tissues may benefit from access tothe energy supplied by modestly highercirculating ketone levels during treatmentwith SGLT2 inhibitors (68,69). These hy-potheses open new areas of investigationfor both clinical and basic research.Challenges
Aside from obvious questions about com-peting priorities for allocation of financial,clinical, and professional resources and, insomecases, delays in regulatory approval,the currently advised structure for CVOTshas several practical limitations (Table 4).The most serious of these is the lack ofgeneralizability of findings to the entirepopulation of people with diabetes.As currently performed, randomized
controlled trials (RCTs) requireparticipantsto have certain inclusion characteristics.
They must be able to provide informedconsent, agree to attend the study sitefor an extended period of time, and bephysically able to do so. To answer ques-tions of safety and efficacy, they must beat high risk to experience the outcome inquestion (in this case, a CV event or death)in order to provide an adequate number ofevents in an acceptable period of time.They cannot be entirely representative ofthe general population, and therefore thefindings of such trials can be extrapolatedto a wider population only with consider-able caution. As an example, the EMPA-REG OUTCOME trial provides reasonableconfidence that people with type 2 diabe-tes and a history of previous CVD are likelyto benefit from empagliflozin therapy inthe same manner as did the patients inthe trial itself (13). However, there is lesscertainty that empagliflozin will similarlyimprove outcomes for individuals with di-abetes and no history of CVD or for thosewith CVD but without diabetes. In theCANVAS Program, one-third of partici-pants had only CV risk factors and no evi-dence of CVD. Perhaps the less impressivebut still positive outcomes with canagliflozinin this research program may be relatedto this population subset that mightrequire longer drug exposure in a longer-term trial (14). Notably, a subgroup anal-ysis of the LEADER trial suggested thehypothesis that liraglutide effectivelylowered CV risk only in peoplewith estab-lished CVD (HR 0.83 [95% CI 0.74–0.93]compared with HR 1.20 [95% CI 0.86–1.67] in those without CVD, with a signif-icant unadjusted interaction P value of0.04) (10). Due to the large number ofstatistical tests that were performed forsubgroups, the possibility that this was achance observation cannot be excluded.Nonetheless, the FDA-approved indica-tion for liraglutide for CV risk reductionwas restricted only to those with estab-lished CVD.
In addition, the present CVOT designhas serious limitations related to thenatural histories of diabetes and CVD.Because the concerns of regulatoryagencies, pharmaceutical companies,health systems, and in many cases, indi-viduals with diabetes have relativelyshort timelines, current CVOTs addressonly short-term outcomesdnot thoseoccurring .5 years after the onset oftreatment. However, several lines of evi-dence argue for the importance of longer-term follow-up.
Experience from the UKPDS and DCCT/EDIC (Epidemiology of Diabetes Interven-tions and Complications) cohorts, whichwere treated early in the course of diabe-tes, shows that importantmedical compli-cations of diabetes, including blindness,renal failure, amputation, CV death, andothers, do not usually appear in the first5 years. Rather, each has a characteristicand much longer development timeline(26–28). In the case of CVD, significantnumbers of CV events in the UKPDS andDCCT/EDIC did not accumulate until af-ter .10 years of follow-up. In part be-cause of the slow accrual of events, theapparent CV benefits of early randomiza-tion to intensive glycemic control werenot observed until long after initiationof treatment. The CV benefits likewise oc-curred long after the change in A1C dif-ference disappeared after cessation ofrandomized treatment, leading to theconcept of a “legacy effect,” for whichthere is no proven explanation. Theselong-term benefits were shown throughobservational and possibly incompletedata within which residual confoundingcannot be excluded, so they cannot con-fidently be attributed only to glycemiccontrol. Nevertheless, a mechanism plau-sibly underlying the legacy effect is a re-duction of glucose-related changes ofvascular tissue structure, occurring duringintensive glycemic therapy but persistingthereafter, thatmayhave slowed theearlyprogression of atherosclerosis and led to amuch later reduction in overt CV events.Whether any nonglycemic effect of treat-ments in these trials contributed to thisbenefit cannot be determined.
This view is consistentwith the findingsof other large trials of short-term glucose-lowering strategies using similar treatmentsthatwereapplied later in thenatural historyof type 2 diabetes. In ACCORD (20,88) andVADT (22,29), there were only modest re-ductions in some secondary CV end pointsin patients with a long duration of diabetesand high CV risk, but in ACCORD, therewas an unexplained early increase in mor-tality. Specifically, very intensive glucose-lowering treatment goals achieved mainlywith the use of metformin, sulfonylureas,TZDs, and intensive insulin regimens in thesetrials were unable to diminish the risk ofevents in patientswhoalreadyhadadvancedCV pathology.
The failure of targeting very intensiveglucose lowering in ACCORD and VADThighlights the benefits of treatment with
care.diabetesjournals.org Cefalu and Associates 23
Table 4—Limitations of current CVOT structure and opportunities for improvement
Current limitations
Lack of generalizability Current CVOTs include participants who are at high risk for a CV event or death and thusare not representative of the larger population.
Short timeline for assessing potential benefits CV benefit may not become apparent until long after initiation of treatment. CurrentCVOTs do not assess outcomes occurring .5 years after the onset of treatment.
Short timeline for assessing potential harm CVOTs lasting,5 years are not likely to detect risks that may become apparent only afteryears of treatment. This may be especially concerning for agents with complexmechanisms of action.
Placebo-controlled design Nearly all CVOTs to date have tested one drug against placebo, with both groupsattempting to attain comparable glycemic control using regimens that also includeother medications. Problems with this design include 1) comparable glycemic controlgenerally has not been achieved and the lower A1Cs found with the study drug maycontribute to CV benefits, 2) some drugs used in the comparison groupsmay themselveshave an adverse effect on CV events, and 3) the CV benefits found with four agents todate make the future use of placebo ethically challenging.
Opportunities for improvementLower-risk, more diverse populations Primary intervention trials in lower-risk populations could determine whether diabetes
medications offer CVprotection for thosewhodonot yet haveCVD. Thiswould requirelarger and/or longer studies but would yield valuable information with regard to CVDprevention.
Longer-term follow-up Trial designs that prespecify longer-term follow-up could better identify longer-termsafety issues and late beneficial effects, produce better cost-effectiveness data, andimprove understanding of changing treatment requirements over time. Such a designwould require new consent procedures to permit lifelong follow-up, strategies toincrease therapy adherence and persistence, expanded use of EMRs, and innovativestatistical approaches to permit serial reporting of key clinical outcomes over time.
Active comparators Using an active comparator instead of placebo could address the drawbacks of placebo-controlled trials but will require sufficient knowledge of the CV impact of thecomparator to avoid confounding interpretation of the results. Although challenging,thismay become feasible as understanding of the CV safety of newer agents increases.Across-trials consistency in enrollment criteria and the capturing of baseline patientcharacteristics will facilitate such efforts.
Innovative designs Adoptionof factorial or adaptivedesigns, superiority trials, trials embeddedwithinhealthcare systems or networks, and/or employment of “big data” to dissect the effects ofnew diabetes medications may provide practical opportunities for further investigation.
Standardized definitions Standard definitions of important safety and microvascular outcomes would facilitatebetter comparisons among agents. Collaborative efforts should be made tostandardize definitions of high-priority safety and microvascular outcomes, akin tothose established for CV outcomes.
Modification of end points and analyses Incorporating weighted composite end points that include estimation of the severity ofevents, aswell asmultiple events in the samepatient,may yieldmorenuancedfindingsin future studies. The design of trials for new agents should be informed by data frompreviousCVOTswithinthesamedrugclass. Important secondaryoutcomeswith robuststatistical findings that are biologically plausible and supportable by external evidenceshould be independently considered even if primary composite outcomes are notachieved. Such approaches would require buy-in from regulatory bodies andmechanismstoensureequity fordevelopersoffirst-in-class interventionsand/or thosewho willingly adopt more complex trial designs. How best to incorporate predefinedsafety concerns into primary analyses should also be considered.
Establishment of biorepositories Future trials should obtain informed consent to store participants’ biological samples incase unexpected results warrant further investigation. Such biorepositories couldincrease opportunities to investigate various mechanisms contributing to CV events andkey subgroups and will become increasingly important for subsequent biomarker, gutmicrobiota, and genomic analyses to facilitate precision medicine opportunities.
Enhanced efficiency and cost-sharing options Conducting a CVOT for each newdiabetes drug is cumbersome and expensive. Strategiesto enhance trial efficiency, such as collecting CVoutcomesdata fromtrials designed forother purposes, should be strongly considered, as should newmodels for cost-sharingamong pharmaceutical, governmental, and other organizations.
Involvement of patients and advocacy organizations Involving patients and their advocates in designing future trials will help to ensure thatpatients’ views and wishes are taken into account and that patient-related outcomemeasures are fully integrated. Such efforts would likely increase patient buy-in and helpto minimize discontinuation, improve treatment adherence and persistence, and avoidmissing data.
24 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
empagliflozin, canagliflozin, liraglutide,and semaglutide in their respective CVOTs.Favorable CV effects with these drugs pre-sumably resulted fromeffects beyond sim-ply lowering glucose in patients withadvanced diabetes, except perhaps inthe trial testing semaglutide, during whicha pronounced A1C reduction occurredin the group receiving active therapy.Whether the beneficial effects of thesedrugs would be similar or different if theywere initiated closer to the time of diabetesdiagnosis and earlier in the course of CVDdevelopment remains to be seen.The current CVOT design also provides
only limited objective information aboutharmful outcomes that occur .5 yearsafter an intervention. Whether the newantidiabetes drugs currently being evalu-ated have risks thatmay appear only afteryears of therapy is unknown, but CVOTswith a duration of,5 years are not likelyto detect them. Concerns may be greaterfor therapeutic drugswith complexmech-anisms of action that are not currentlywell understood, including TZDs, DPP-4 in-hibitors, and (despite their strong short-term protective effects) SGLT2 inhibitors.
CVOTS OF THE FUTURE:OPPORTUNITIES FORIMPROVEMENT
As previously noted, the completedCVOTs have yielded some important in-formation about both safety concernsand potential medical benefits. Regardingsafety, the increased risk of HF hospitali-zation with saxagliptin in SAVOR-TIMI53 (6) and a nonsignificant trend withaloglipitin in EXAMINE (7,59) resulted inthe addition of new safety warnings tothe labels of those medications (64).Given these safety signals for HF, whichwere unexpected from the registrationtrials, it is likely that studies of individualantidiabetes drugs will continue to berequired.Further, four of the completed tri-
als (LEADER, SUSTAIN-6, EMPA-REGOUTCOME, and the CANVAS Program)reported clinically relevant relative riskreductions ofMACE outcomes in patientswithhighCVrisk (Table 5) (10,11,13,14,60).Although it is inherently inappropriate tocompare results among trials becausethey are conducted in different popula-tions, with different protocols, and at dif-ferent sites, the numbers needed to treat(NNTs) to prevent a MACE occurrence inEMPA-REGOUTCOME (63over 3.1 years),
LEADER (53 over 3.8 years), and SUSTAIN-6 (44 over 2 years) were similar to thoseobserved for widely recommended thera-pies to prevent CVD such as lipid loweringwith statins, antihypertensive therapy,and aspirin (89). These three trials, to-gether with similar but more controver-sial findings in the CANVAS Program,support an early and evolving case forthe use of these antihyperglycemic ther-apies in patients with type 2 diabetes andevidence of clinical CVD who are .50years of age.
Although the trials conducted thus farappear to both satisfy the FDA guidancerecommendations and provide additionaluseful information, there are clearly op-portunities for improvement (Table 4).The next generation of diabetes trialsshould be smarter, simpler, and innova-tively designed to make more efficientuse of resources and produce more gen-eralizable results while still addressingsafety issues. The design and conduct offuture trials should be standardizedenough to permit reliable between-trialcomparisons but also nimble enough toexplore unanticipated safety concernsthat may arise. In pursuing these goals,some modifications to the current CVOTnoninferiority trial design, outlined be-low, are worthy of consideration.
Lower-Risk and More DiversePopulationsIncorporation of recent trial findings intodiabetes care guidelines has been limitedto date, in part because the populationsstudied were selected for high CV risk.Although this practice is consistent withFDA recommendations and essential todemonstrate safety, it remains unclearwhether results seen in high-risk patientsare translatable to patients with a shorterduration of diabetes or without estab-lished CV complications.
Enrollment of very high-risk subjectsalso may hamper the ability of trials todetect a benefit from drug therapy. Asexamples, the ELIXA trial of lixisenatide(9) and the EXAMINE trial of alogliptin(7) exclusively recruited patients with arecent hospitalization for an acute coro-nary event. Such patients are easy toidentify based on hospital records andare likely to have a high number of fu-ture MACE events. Thus, such trials maybe completed more quickly than thoseenrolling patients with more stable CVD.However, this strategymay fail to detect a
beneficial effect of the intervention stud-ied because the CVD is too advanced, thedrug exposure is too short, and/or theearly events may be less amenable tothe intervention.
The best way to demonstrate benefitmight be to select lower-risk patientswhohave not yet manifested CVD to find outwhether their likelihood of developingCVD can be reduced. To achieve a statisti-cally significant number of MACE events,this will require larger and/or longer stud-ies but would yield valuable informationabout CVD prevention.
Longer-Term Follow-UpTrial designs that prespecify longer-termfollow-upmaybedesirable to identify anylonger-term safety issues and beneficialeffects that are either slowly evolving orresulting from a legacy effect of earliertreatment. Toward that end, innovativeconsent procedures that permit lifelongfollow-up should be explored, as shouldstrategies to increase patient adherenceto and persistence with assigned therapy.The ability to capture medication adher-ence and relevant clinical and laboratorydata will be crucial and should be facili-tated by use of more sophisticated elec-tronic medical records (EMRs).
Statistical approaches would need tobedeveloped (andapprovedby regulatorybodies) to permit robust assessment ofkey clinical outcomes over an extendedtimescale. For example, a trial could pro-vide definitive early information on the de-gree of glucose lowering achieved (e.g., at1 year) and also be powered to assess theimpact of an intervention on a primaryclinical outcome at a later time (e.g., at3 years). Continued follow-up thereaftercould examine recurrent or less frequentclinical outcomes, maximize collection ofsafety data, and better assess the durabil-ity of the glucose-lowering effect.
Even a few such trials could yield a richharvest of clinically relevant information tobetter inform clinical guidelines and ther-apeutic choices for patients and producemore robust cost-effectiveness data forhealth care providers. Longer trials wouldalso provide a better understanding of theevolution of treatment requirements overtime, especially if participants are random-ized to specific therapeutic sequences.
Active ComparatorsAnother difficulty in interpreting and ap-plying the results of most CVOTs com-pleted to date concerns comparing the
care.diabetesjournals.org Cefalu and Associates 25
drug in question to placebo, with bothgroups attempting to attain comparableglycemic control using regimens that in-clude other medications. There are sev-eral problems with this design. First, ingeneral, comparable glycemic control isnot achieved, with the study drug groupusually having lower A1C values. Such dif-ferences might contribute to any CV ben-efits seen. Second, some drugs usedmorefrequently in the comparison groups maythemselves have CV effects. Third, mov-ing forward, it may not, for ethical rea-sons, be possible to restrict backgrounduse of drugs that have recently beenshown to have beneficial effects.However, using an active comparator
as a control will require sufficient knowl-edge of the CV impact of the comparatorto avoid confounding interpretation ofthe results. Although challenging, thismay become feasible as our understand-ing of the CV safety of newer antidiabetesdrugs increases. Future trialsmay success-fully assess the impact of a new drug com-pared with a previously tested drug if thenewer trial enrolls a patient populationand adopts a design that is similar tothat with which the comparator drug wasstudied. Combinations of drugs known tobe cardioprotective may also need to be
tested to explore whether the CV benefitsare compounded. Between-trial consis-tency in study design, enrollment criteria,and ascertainment of clinical data will fa-cilitate these efforts.Innovative DesignsCreative trial designs likely will be neededto fully address current concerns. In addi-tion to traditional RCTs, factorial or adap-tive designs, superiority trials, trialsembedded within health care systems ornetworks, and/or employment of “bigdata” to dissect the effects of new diabe-tes medications may provide practicalopportunities for further investigation.Factorial designswould increase efficiencyby allowing simultaneous testing of multi-ple interventions, and superiority trialswould put greater emphasis on potentialbenefits of therapy and likely make re-cruitment easier by increasing the valueproposition forparticipants. Pragmatic de-signs that embed studies within healthcare systems’ EMR platforms would pro-vide a natural conduit for conducting long-term posttrial observational follow-upstudies. However, lessons from EXSCEL(12) must be considered to avoid havinglow treatment persistence in suchdesigns.
EmbeddingRCTs into routineclinical carewould also enable the multidimensional
collection of additional clinical outcomesrelevant to diabetes that are not collectedroutinely in current trials because of costand complexity. Such outcomes includerates of cancer, bone fractures, depression,and dementia, as well as unexpected off-target side effects undetected during con-ventional drugdevelopmentprograms. Thiswould extend assessment of safety beyondCV safety and include long-term durabilityof effects. Sample sizes could be muchlarger using the embedded trial approachbecauseof lower costs perpatient, enablingmore powerful assessments of the hetero-geneity of treatment effects. Trials of thisnature also could be crucial in realizing thegoalofprecisionpharmacotherapy in type2diabetes, should different classes of drugsor specific drugs within classes be found tobe better for patients with different clinicalcharacteristics (31). However, defining therandomization process to avoid con-founders and any imbalances betweenthe groups studied will be crucial.
Although registries or EMRs may af-ford the opportunity for large, pragmatic,and relatively inexpensive trials, the suc-cess of such trials likely will depend onhaving a simple intervention and an ap-propriate health systemwith an adequateEMR, as is the case with the ongoing
Table 5—Risk reduction in four completed trials showing evidence of CV benefit
LEADER (10) SUSTAIN-6 (11) EMPA-REG OUTCOME (13,60) CANVAS Program (14)
Subjects (n) 9,340 3,297 7,020 10,142
Mean age (years) 64.3 64.6 63.1 63.3
Diabetes duration (years)* 12.8 13.9 57% .10 13.5
Mean baseline A1C (%) 8.7 8.7 8.1 8.2
Mean placebo-corrected A1C difference (%)† 20.4 20.7 (0.5 mg dose) 20.24 (10 mg dose) 20.5821.0 (1.0 mg dose) 20.36 (25 mg dose)
Median follow-up duration (years) 3.8 2.1 3.1 2.4
3-point MACE RRR (%) 13 26 14 14
3-point MACE ARR (%) 1.9 2.3 1.6 d‡
CV death RRR (%) 22 2 38 4§; 13|
Nonfatal MI RRR (%) 12 26 13 15
Nonfatal stroke RRR (%) 11 39 +24 10
All-cause mortality RRR (%) 15 +5 32 13§; 10|
HF hospitalization RRR (%) 13 +11 35 33
Worsening nephropathy RRR (%)¶ 22 36 39 40
Boldface type indicates statistical significance. +, increased relative risk;d, not reported; ARR, absolute risk reduction; RRR, relative risk reduction.*Reported as mean in all trials except EMPA-REG OUTCOME, which reported percentage of population with diabetes duration.10 years. †For LEADERand the CANVAS Program, difference was updated over time; for SUSTAIN-6 and EMPA-REG OUTCOME, difference was at study end. ‡Data needed tocalculate ARR in 3-point MACE (specifically the NNT) were not reported for the CANVAS Program; instead this trial reported an incidence rate differenceof 4.6 per 1,000 patient-years for canagliflozin vs. placebo. §Truncated integrated data set (refers to pooled data from CANVAS after 20 November2012 plus CANVAS-R; prespecified in treating hierarchy as the principal data set for analysis for superiority of all-cause mortality and CV death in theCANVAS Program). |Nontruncated integrated data (refers to pooled data from CANVAS, including before 20 November 2012 plus CANVAS-R. ¶Worseningnephropathy is defined as the new onset of macroalbuminuria (urine albumin creatinine ratio.300 mg/g) or a doubling of the serum creatinine leveland an eGFR rate of#45 mL/min/1.73 m2, the need for continuous renal-replacement therapy, or death from renal disease in LEADER, SUSTAIN,and EMPA-REG OUTCOME and as 40% reduction in eGFR, renal-replacement therapy, or death from renal causes in CANVAS. This outcome was aprespecified exploratory adjudicated outcome in LEADER, SUSTAIN-6, and CANVAS but not in EMPA-REG OUTCOME.
26 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
ADAPTABLE (Aspirin Dosing: A Patient-Centric Trial Assessing Benefits and Long-term Effectiveness) trial of aspirin therapyfor atherosclerotic CVD (90). As men-tioned above, irrespective of trial design,treatment randomization will remain cru-cial to determining a medication’s effectwithout undue bias.
Standardized DefinitionsSafety and efficacy comparisonswould begreatly enhanced by standardizing thedefinitions of important outcomes. Thestrategies and definitions used to identifyevents such as pancreatitis, pancreaticmalignancy, progression of retinopathy,andworsening of renal function have var-ied considerably among the CVOTs con-ducted to date. Collaborative effortsshould be made to generate definitionsfor high-priority safety and microvascularoutcomes, akin to those established forCV outcomes.
Modification of End Points andAnalysesIn light of the noted differences in effectsof drugs on components of MACE com-posite end points, it may be desirable toreassess the end points and primary anal-yses to be adopted in future studies toprovide a more nuanced look at CV out-comes. Trials likely will still be poweredbased on the time to the first primary endpoint, despite disadvantages that includepossible heterogeneity of the compositecomponents, results drivenby componentsof lesser importance, and insufficientpower to draw definitive conclusions forall components of the composite (91). Toaddress these limitations, alternative an-alytic strategies such as weighted com-posite outcomes (92) and the win ratioapproach (93) should be considered. Tobetter capture the full impact of treat-ments, subsequent and recurrent eventsalso need to be evaluated for both bene-fits and risks using validated statistical ap-proaches (94,95). When testing drugswithin the same class, data from previoustrials should be used to inform both thedesign and sample size requirements,with a Bayesian approach considered topotentially limit sample sizes to moremanageable and less costly proportionswithout compromising study power. Im-portant secondary outcomes, especiallymortality or serious irreversiblemorbidityevents, should not be dismissed even ifthe primary composite outcomes are notachieved; they should be independently
considered, especiallywhen the statisticalfindings are robust, biologically plausible,and supportable by external evidence.This way forward would require buy-infrom regulatory bodies, as well as mech-anisms to ensure equity for developers offirst-in-class interventions and/or thosewho willingly adopt more complex trialdesigns.
Another challenge is how best to incor-porate predefined safety concerns intothe primary analysis. One possibility isto add safety issues as components ofthe primary composite outcome if the in-cluded safety and efficacy componentscan be weighted appropriately to reflecttheir clinical importance. Such a com-bined end point would then reflect theintervention’s overall net clinical benefitto the patient. If there is adequate power,it may be more desirable to evaluate keyefficacy and key safety outcomes sepa-rately as predefined coprimary end points,with superiority and noninferiority analy-ses respectively, as was done in EXSCEL(12). In general, however, the FDA typicallyprefers trials to establish efficacy andsafety separately, with approval basedon whether efficacy (benefit) outweighssafety (harm).
Establishment of BiorepositoriesMany differentmechanisms contribute tothe occurrence of CV events, includingatherosclerosis, hemodynamic changes,arrhythmias, and, for strokes, hemorrhagicmechanisms. To gain additional insightinto both mechanisms and subgroups, itwould be desirable for future trials tohave informed consent to store biological(i.e., serum, urine, fecal, andDNA) samplesin case unexpected results warrant furtherinvestigation or for additional hypothesis-generating studies. Greater use of patientphenotyping, genotyping, andmetabolomicdata, similar towhat isnowbeingdone inthefield of oncology,may help to better identifypredictors of benefit or harm that so farhave escaped detection. Biorepositorieswill be of key importance in this regard,as well, by allowing for subsequent bio-marker, gutmicrobiota, and genomic anal-yses to facilitate future precision medicineopportunities.
Enhanced Efficiency and Cost-SharingOptionsThe current system for evaluating the CVsafety of new diabetes drugs, in whicheach is assessed individually in a large,long-termCVOTfundedbyapharmaceutical
sponsor, is cumbersome and expensive.Strategies to enhance the efficiency oftrials, such as collecting CV outcomesdata from trials designed for other pur-poses, should be strongly considered. Al-though evidence that the expense of suchstudies is suppressing innovation in newdiabetes drug development is lacking, it isconceivable that, if the cost of CVOTswere reduced, more innovative researchcould be supported. Thus, models forcost-sharing among pharmaceutical, gov-ernmental, and other organizationsshould be explored.
Involvement of Patients and AdvocacyOrganizationsWhen designing future trials, all stake-holders should be involved. The directinvolvement of patients and their advo-cates will be vitally important to ensurethat patients’ views and wishes are takeninto account and that patient-related out-come measures are fully integrated.There is some concern, however, thatpatient-reported outcome measures arenot standardized, limiting comparisonsbetween studies. Efforts to improvepatients’ understanding of clinical trialsand to enlist the help of patients in de-veloping trial information literature andconsent processes might improve pa-tients’ willingness to participate as sub-jects. The importance of minimizingparticipant discontinuation, improvingstudy drug adherence and treatmentpersistence, and avoiding missing datacannot be overstated when seeking toobtain definitive trial outcomes. The rapidgrowth of digital health and monitoringdevices with Internet connectivity may of-fer further opportunities to access patient-specific outcomes or provide mechanisticinsights (29,96).
CONCLUSIONS
The 2008 FDA guidance for diabetes drugdevelopment greatly increased the collec-tion of data on conventional CVoutcomesof treatment. Studies completed to dateand those now under way are providingstrong evidence for CV benefit for severaldrugs and reassurance about the lack ofCV risk for many others. In addition, theyhave yielded further information onnon-CV benefits, such as reduction of re-nal disease, and potential harms, such asHF and lower-limb amputations. Thisinformation has built a strong founda-tion of outcomes evidence that is truly
care.diabetesjournals.org Cefalu and Associates 27
transforming diabetes care. The informa-tion on improvedCVoutcomes for specificantidiabetes drugs should be consideredin revised clinical treatment recommenda-tions, given that cardioprotection is anadded benefit.However, after nearly a decade of ex-
perience with CVOTs designed on the ba-sis of the 2008 guidance, it is time toreevaluate how best to conduct such tri-als. Future trialsmight benefit from focus-ing on populations more typical of thoseseen in routine care, with longer durationof follow-up, greater consistency in theascertainment of outcomes, and improvedstatistical methods for analysis. In addi-tion, greater efficiency and relevance toclinical practice might be obtained bycollaboration of various stakeholders inhealth caredincluding pharmaceuticalcompanies, regulatory groups, insurers,health systems, consumer advocates,and peoplewith diabetes themselvesdinthe conduct of trials, perhaps by embed-ding clinical protocols within health sys-tems. It is to be expected that futuretrials, like those already completed, willprovide both answers to the primary re-search questions posed and new insightsregarding previously unexpected benefitsand risks.
Acknowledgments. Writing and editing sup-port services for this article were provided byDebbie Kendall of Kendall Editorial in Richmond,Virginia. The authors thank Christian S. Kohler,the ADA’s Associate Publisher, Scholarly Journals,and his staff for their assistance, guidance, andexpertise in convening the Expert Forum.Duality of Interest. W.T.C., prior to joining theADA, received grant support awarded to hisinstitution from Novo Nordisk and Sanofi. Heserved as a consultant for Adocia, Intarcia Ther-apeutics, and Sanofi. He has no active disclo-sures. S.K. is a consultant to and advisor forBoehringer Ingelheim, Eli Lilly, andNovoNordiskand is a stock shareholder in Johnson & Johnson.H.C.G. holds the McMaster-Sanofi PopulationHealth Institute Chair in Diabetes Research andCare. He has received research grant supportfrom AstraZeneca, Eli Lilly, Merck, and Sanofi;honoraria for speaking engagements fromAstraZeneca, Boehringer Ingelheim, Novo Nordisk,and Sanofi; and consulting fees fromAbbott, Amgen,AstraZeneca, Boehringer Ingelheim, Eli Lilly, Merck,Novo Nordisk, and Sanofi. R.R.H. reports receivinggrants fromAstraZeneca; grants andpersonal feesfrom Bayer, Boehringer Ingelheim, and Merck;personal fees from Amgen, Novartis, and Servier;and other support from Elcelyx Therapeutics,GlaxoSmithKline, Janssen, andTakeda.B.Z. is a con-sultant to and has received honoraria fromAstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen,Merck, Novo Nordisk, and Sanofi and has received
grant support from AstraZeneca, Boehringer Ingel-heim, and Novo Nordisk. J.S.S. has been an advisorto Adocia, AstraZeneca, BD Technologies, BoehringerIngelheim, Dance Biopharm, Diavacs, Elcelyx Ther-apeutics, Eli Lilly, Ideal Life, ImmunoMolecularTherapeutics, Intarcia Therapeutics, Intrexon,Merck, Orgenesis, Sanofi, Servier, vTv Therapeu-tics, Valeritas, and Viacyte. He has received re-search funding from Mesoblast and Viacyte. Heis a member of the board of directors of Dexcom,Intarcia Therapeutics, Moerae Matrix, and Vaso-Prep Surgical. He is a stock shareholder in DanceBiopharm, Dexcom, Ideal Life, Intarcia Therapeu-tics, Intrexon, Moerae Matrix, and VasoPrep Sur-gical. J.B.G. is a consultant to Boehringer IngelheimandDaiichi Sankyo and has received research fund-ing fromAstraZeneca, GlaxoSmithKline, andMerck.J.B.B. has received contracted consulting fees, paidto his institution, and travel support from Adocia,AstraZeneca, Dance Biopharm, Dexcom, ElcelyxTherapeutics, Eli Lilly, Fractyl, GI Dynamics, Intar-cia Therapeutics, Lexicon, Merck, Metavention,NovaTarg, Novo Nordisk, Orexigen, PhaseBio,Sanofi, Shenzhen HighTide, and vTv Thera-peutics and grant support from AstraZeneca,Bayer, Boehringer Ingelheim, Eli Lilly, GI Dy-namics, GlaxoSmithKline, Intarcia Therapeutics,Johnson & Johnson, Lexicon, Medtronic, Merck,Novo Nordisk, Orexigen, Sanofi, Scion NeuroStim,Takeda, and Theracos. He has received fees andholds stock options in Insulin Algorithms andPhaseBio and serves on the board of theAstraZenecaHealthCareFoundation. S.E.I. has servedas a consultant to Alere, Janssen, and vTv Thera-peutics. He has participated on clinical trial steer-ing or executive committees for AstraZeneca,Boehringer Ingelheim, Daiichi Sankyo, Eisai,Novo Nordisk, and Sanofi/Lexicon. He has servedas a member of a data monitoring committee forIntarcia Therapeutics. L.A.L. has received researchfunding from, has provided continuing medica-tion education on behalf of, and/or has acted asan advisor to AstraZeneca, Boehringer Ingelheim,Eli Lilly, GlaxoSmithKline, Janssen, Merck, NovoNordisk, Sanofi, and Servier. I.R. has received feesfor serving on advisory boards for AstraZeneca,Boehringer Ingelheim, Concenter BioPharma/Silkim, Eli Lilly, LabStyle Innovations, Merck Sharp&Dohme,NovoNordisk, Orgenesis, Pfizer, Sanofi,and SmartZyme Innovation; fees for consultantservices from AstraZeneca/Bristol-Myers Squibb,Diabetes Medical Center, FutuRx, Gili Medical,Insuline Medical, Kamada, and Medial EarlySign;and fees for serving on speakers’ bureaus fromAstraZeneca/Bristol-Myers Squibb, Boehringer In-gelheim, Eli Lilly, Johnson& Johnson,Merck Sharp& Dohme, Novartis Pharma AG, Novo Nordisk,Sanofi, and Teva. He is a stock shareholder inGlucoMe, Insuline Medical, LabStyle Innovations,and Orgenesis. J.R. has served on scientific advi-sory boards and received honoraria or consultingfees from AstraZeneca, Boehringer Ingelheim, EliLilly, Intarcia Therapeutics, Janssen, NovoNordisk,and Sanofi. He has received grants/research sup-port from AstraZeneca, Boehringer Ingelheim,Bristol-Myers Squibb, Eli Lilly, Genentech,GlaxoSmithKline, Intarcia Therapeutics, Janssen,Lexicon, Merck, Novo Nordisk, Pfizer, and Sanofi.M.C.R. has received honoraria for consulting,speaking, or serving on committees for Adocia,AstraZeneca, Elcelyx Therapeutics, Eli Lilly,GlaxoSmithKline, Sanofi, Theracos, and Valeritas
and has received research support through hisinstitution from AstraZeneca, Eli Lilly, and NovoNordisk; these potential dualities have been re-viewed andmanaged by Oregon Health& ScienceUniversity. No other potential conflicts of interestrelevant to this article were reported.
References1. U.S. Food and Drug Administration. Guidancefor industry: diabetes mellitusdevaluatingcardiovascular risk in new antidiabetic thera-pies to treat type 2 diabetes [Internet]. Avail-able from www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm071627.pdf. Accessed 31 October20162. Laakso M. Cardiovascular disease in type 2 di-abetes from population to man to mechanisms:the KellyWest Award Lecture 2008. Diabetes Care2010;33:442–4493. Nissen SE, Wolski K. Effect of rosiglitazone onthe risk of myocardial infarction and death fromcardiovascular causes. N Engl J Med 2007;356:2457–24714. Caveney E, Turner JR.White paper. A reviewofFDA guidance: understanding the FDA guidanceon assessing cardiovascular risks for new antidia-betic therapies [Internet]. Available from http://www.quintiles.com/library/white-papers/new-fda-guidance-on-antidiabetic-therapies. Accessed31 October 20165. EuropeanMedicinesAgency. Guideline on clin-ical investigation of medicinal products in thetreatment or prevention of diabetes mellitus [In-ternet], 2012. Available from http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/06/WC500129256.pdf.Accessed 3 November 20166. Scirica BM, Bhatt DL, Braunwald E, et al.;SAVOR-TIMI 53 Steering Committee and Investi-gators. Saxagliptin and cardiovascular outcomesin patients with type 2 diabetes mellitus. N Engl JMed 2013;369:1317–13267. White WB, Cannon CP, Heller SR, et al.;EXAMINE Investigators. Alogliptin after acute cor-onary syndrome in patients with type 2 diabetes.N Engl J Med 2013;369:1327–13358. Green JB, Bethel MA, Armstrong PW, et al.;TECOS Study Group. Effect of sitagliptin on cardio-vascular outcomes in type 2 diabetes. N Engl JMed 2015;373:232–2429. Pfeffer MA, Claggett B, Diaz R, et al.; ELIXAInvestigators. Lixisenatide in patients with type 2diabetes and acute coronary syndrome. N Engl JMed 2015;373:2247–225710. Marso SP, Daniels GH, Brown-Frandsen K,et al.; LEADER Steering Committee; LEADER TrialInvestigators. Liraglutide and cardiovascular out-comes in type 2 diabetes. N Engl J Med 2016;375:311–32211. Marso SP, Bain SC, Consoli A, et al.; SUSTAIN-6 Investigators. Semaglutide and cardiovascularoutcomes in patients with type 2 diabetes. N EnglJ Med 2016;375:1834–184412. Holman RR, Bethel MA, Mentz RJ, et al.;EXSCEL Study Group. Effects of once-weekly exe-natide on cardiovascular outcomes in type 2 di-abetes. N Engl J Med 2017;377:1228–123913. Zinman B,Wanner C, Lachin JM, et al.; EMPA-REG OUTCOME Investigators. Empagliflozin, car-diovascular outcomes, and mortality in type 2diabetes. N Engl J Med 2015;373:2117–2128
28 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
14. Neal B, Perkovic V, Mahaffey KW, et al.;CANVASProgramCollaborativeGroup. Canagliflozinand cardiovascular and renal events in type 2 di-abetes. N Engl J Med 2017;377:644–65715. Rawshani A, Rawshani A, Franzen S, et al.Mortality and cardiovascular disease in type 1and type 2 diabetes. N Engl J Med 2017;376:1407–141816. National Cholesterol Education Program(NCEP) Expert Panel on Detection, Evaluation,andTreatment of HighBloodCholesterol inAdults(Adult Treatment Panel III). Third Report of theNational Cholesterol Education Program (NCEP)Expert Panel on Detection, Evaluation, and Treat-ment of High Blood Cholesterol in Adults (AdultTreatment Panel III) final report. Circulation 2002;106:3143–342117. Nathan DM, Genuth S, Lachin J, et al.; Diabe-tes Control and Complications Trial ResearchGroup. The effect of intensive treatment of dia-betes on the development and progression oflong-term complications in insulin-dependent di-abetes mellitus. N Engl J Med 1993;329:977–98618. UK Prospective Diabetes Study (UKPDS)Group. Intensive blood-glucose control with sul-phonylureas or insulin compared with conven-tional treatment and risk of complications inpatients with type 2 diabetes (UKPDS 33). Lancet1998;352:837–85319. UK Prospective Diabetes Study (UKPDS)Group. Effect of intensive blood-glucose controlwith metformin on complications in overweightpatients with type 2 diabetes (UKPDS 34). Lancet1998;352:854–86520. Gerstein HC, Miller ME, Byington RP, et al.;Action to Control Cardiovascular Risk in DiabetesStudy Group. Effects of intensive glucose lowering intype2diabetes. N Engl JMed2008;358:2545–255921. Patel A, MacMahon S, Chalmers J, et al.;ADVANCE Collaborative Group. Intensive bloodglucose control and vascular outcomes in patientswith type 2 diabetes. N Engl J Med 2008;358:2560–257222. Duckworth W, Abraira C, Moritz T, et al.;VADT Investigators. Glucose control and vascularcomplications in veterans with type 2 diabetes.N Engl J Med 2009;360:129–13923. Ohkubo Y, Kishikawa H, Araki E, et al. Inten-sive insulin therapy prevents the progression ofdiabetic microvascular complications in Japanesepatients with non-insulin-dependent diabetesmellitus: a randomized prospective 6-year study.Diabetes Res Clin Pract 1995;28:103–11724. Ray KK, Seshasai SR, Wijesuriya S, et al. Effectof intensive control of glucose on cardiovascularoutcomes and death in patients with diabetesmellitus: a meta-analysis of randomised con-trolled trials. Lancet 2009;373:1765–177225. Turnbull FM, Abraira C, Anderson RJ, et al.;Control Group. Intensive glucose control andmac-rovascular outcomes in type 2 diabetes. Diabeto-logia 2009;52:2288–229826. Nathan DM, Cleary PA, Backlund JY, et al.;Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Com-plications (DCCT/EDIC) Study Research Group. In-tensive diabetes treatment and cardiovasculardisease in patients with type 1 diabetes. N Engl JMed 2005;353:2643–265327. Orchard TJ, Nathan DM, Zinman B, et al.;Writing Group for the DCCT/EDIC ResearchGroup. Association between 7 years of intensive
treatment of type 1 diabetes and long-term mor-tality. JAMA 2015;313:45–5328. Holman RR, Paul SK, Bethel MA, MatthewsDR, Neil HAW. 10-year follow-up of intensive glu-cose control in type 2diabetes.N Engl JMed2008;359:1577–158929. Hayward RA, Reaven PD, Wiitala WL, et al.;VADT Investigators. Follow-up of glycemic controland cardiovascular outcomes in type 2 diabetes.N Engl J Med 2015;372:2197–220630. ACCORD Study Group. Nine-year effects of3.7 years of intensive glycemic control on cardio-vascular outcomes. Diabetes Care 2016;39:701–70831. Holman RR, Sourij H, Califf RM. Cardiovascu-lar outcome trials of glucose-lowering drugs orstrategies in type 2 diabetes. Lancet 2014;383:2008–201732. John M, Gopalakrishnan Unnikrishnan A,Kalra S, Nair T. Cardiovascular outcome trialsfor anti-diabetes medication: a holy grail ofdrug development? Indian Heart J 2016;68:564–57133. Nissen SE, Wolski K, Topol EJ. Effect of mur-aglitazar on death and major adverse cardiovas-cular events in patients with type 2 diabetesmellitus. JAMA 2005;294:2581–258634. Drugs.com. Bristol-Myers Squibb and Merckjoint statement on FDA advisory committee voteon Pargluva (muraglitazar), an investigationaloral treatment for type 2 diabetes [Internet].Available from https://www.drugs.com/nda/pargluva_050909.html. Accessed 3 November201635. Home PD, Pocock SJ, Beck-Nielsen H, et al.;RECORD Study Team. Rosiglitazone evaluated forcardiovascular outcomes in oral agent combina-tion therapy for type 2 diabetes (RECORD): a mul-ticentre, randomised, open-label trial. Lancet2009;373:2125–213536. Mahaffey KW, Hafley G, Dickerson S, et al.Results of a reevaluation of cardiovascular out-comes in the RECORD trial. Am Heart J 2013;166:240–249.e137. U.S. Food and Drug Administration. FDA drugsafety communication: FDA requires removal ofsome prescribing and dispensing restrictions forrosiglitazone-containing diabetes medicines [In-ternet]. Available from https://www.fda.gov/drugs/drugsafety/ucm376389.htm. Accessed 9November 201738. Dormandy JA, Charbonnel B, Eckland DJ,et al.; PROactive Investigators. Secondary preven-tion of macrovascular events in patients withtype 2 diabetes in the PROactive Study (PROspec-tive pioglitAzone Clinical Trial In macroVascularEvents): a randomised controlled trial. Lancet2005;366:1279–128939. Kernan WN, Viscoli CM, Furie KL, et al.; IRISTrial Investigators. Pioglitazone after ischemicstroke or transient ischemic attack. N Engl J Med2016;374:1321–133140. Young LH, Viscoli CM, Curtis JP, et al. IRISInvestigators. Cardiac outcomes after ischemicstroke or TIA: effects of pioglitazone in patientswith insulin resistance without diabetes. Circula-tion 2017;135:1882–189341. U.S. Food and Drug Administration Centerfor Drug Evaluation and Research. ApplicationNumber 204042Orig1s000: summary re-view [Internet]. Available from https://www.accessdata.fda.gov/drugsatfda_docs/nda/2013/
204042Orig1s000SumR.pdf. Accessed 5 May201742. Regier EE, Venkat MV, Close KL. More thanseven years of hindsight: revisiting the FDA’s2008 guidance on cardiovascular outcomes trialsfor type 2 diabetes medications. Clin Diabetes2016;34:173–18043. Novo Nordisk. Novo Nordisk receives com-plete response letter in the US for Tresiba andRyzodeg [Internet]. Available from https://www.novonordisk.com/bin/getPDF.1676900.pdf.Accessed 4 November 201644. Boehringer Ingelheim. CAROLINA: cardio-vascular outcome study of linagliptin versusglimepiride in patients with type 2 diabetes. In:ClinicalTrials.gov [Internet]. Bethesda, MD, Na-tional Library of Medicine. Available from https://clinicaltrials.gov/show/NCT01243424. NLM Identifier:NCT01243424. Accessed 6 April 201745. Boehringer Ingelheim. Cardiovascular and re-nal microvascular outcome study with linagliptinin patients with type 2 diabetes mellitus(CARMELINA). In: ClinicalTrials.gov [Internet]. Be-thesda,MD,National Library ofMedicine. Availablefrom https://clinicaltrials.gov/show/NCT01897532.NLM Identifier: NCT01897532. Accessed 10February 201746. JanssenResearch&Development. Evaluationof the effects of canagliflozin on renal and cardio-vascular outcomes in participants with diabeticnephropathy (CREDENCE). In: ClinicalTrials.gov[Internet]. Bethesda, MD, National Library ofMedicine. Available from https://clinicaltrials.gov/show/NCT02065791. NLM Identifier:NCT02065791. Accessed 10 February 201747. AstraZeneca.Multicenter trial to evaluate theeffect of dapagliflozin on the incidence of cardio-vascular events (DECLARE-TIMI 58). In: Clinical-Trials.gov [Internet]. Bethesda, MD, NationalLibrary of Medicine. Available from https://clinicaltrials.gov/show/NCT01730534. NLM Iden-tifier: NCT01730534. Accessed 10 February 201748. Merck Sharp & Dohme. Cardiovascular out-comes following ertugliflozin treatment in type 2 di-abetesmellitus participantswith vascular disease, theVERTIS CV study (MK-8835-004). In: ClinicalTrials.gov[Internet]. Bethesda, MD, National Library of Medi-cine.Available fromhttps://clinicaltrials.gov/show/NCT01986881. NLM Identifier: NCT01986881.Accessed 14 February 201749. AstraZeneca. Study to evaluate the effect ofdapagliflozin on the incidence of worsening heartfailure or cardiovascular death in patients withchronic heart failure (Dapa-HF). In: ClinicalTrials.gov [Internet]. Bethesda, MD, National Libraryof Medicine. Available from https://clinicaltrials.gov/show/NCT03036124. NLM Identifier:NCT03036124. Accessed 6 April 201750. AstraZeneca. A study to evaluate the effect ofdapagliflozin on renal outcomes and cardiovascu-lar mortality in patients with chronic kidney dis-ease (Dapa-CKD). In: ClinicalTrials.gov [Internet].Bethesda, MD, National Library of Medicine.Available from https://clinicaltrials.gov/show/NCT03036150. NLM Identifier: NCT03036150.Accessed 4 April 201751. Boehringer Ingelheim. Empagliflozin out-come trial in patients with chronic heart failurewith reduced ejection fraction (EMPEROR-Reduced). In: ClinicalTrials.gov [Internet]. Bethesda,MD, National Library of Medicine. Available fromhttps://clinicaltrials.gov/show/NCT03057977.
care.diabetesjournals.org Cefalu and Associates 29
NLM Identifier: NCT03057977. Accessed 6 April201752. Boehringer Ingelheim. Empagliflozin out-come trial in patients with chronic heart failurewith preserved ejection fraction (EMPEROR-Preserved). In: ClinicalTrials.gov [Internet].Bethesda, MD, National Library of Medicine.Available from https://clinicaltrials.gov/show/NCT03057951. NLM Identifier: NCT03057951.Accessed 6 April 201753. Intarcia Therapeutics. Intarcia announcessuccessful cardiovascular safety results in phase3 FREEDOM-CVO trial for ITCA 650, an investiga-tional therapy for type 2 diabetes: company alsoreports new $75million financing formanufactur-ing scale-up and inventory build for anticipatedglobal launch of ITCA 650 [Internet]. Availablefrom https://www.intarcia.com/media/press-releases/2016-may-6-cardiovascular-safety.html.Accessed 14 February 201754. Gerstein HC, Colhoun HM, Dagenais GR,et al.; REWIND Trial Investigators. Design andbaseline characteristics of participants in the Re-searching cardiovascular Events with a WeeklyINcretin in Diabetes (REWIND) trial on the cardio-vascular effects of dulaglutide. Diabetes ObesMetab 14 July 2017 [Epub ahead of print].https://doi.org/10.1111/dom.1302855. GlaxoSmithKline. Effect of albiglutide,when added to standard blood glucose lower-ing therapies, on major cardiovascular eventsin subjects with type 2 diabetes mellitus. In:ClinicalTrials.gov [Internet]. Available from https://clinicaltrials.gov/show/NCT02465515. NLMIdentifier: NCT02465515. Accessed 14 February201756. Novo Nordisk A/S. A trial investigating thecardiovascular safety of oral semaglutide in subjectswith type 2 diabetes (PIONEER 6). In: ClinicalTrials.gov [Internet]. Available fromhttps://clinicaltrials.gov/show/NCT02692716. NLM Identifier:NCT02692716. Accessed 29 June 201757. Marso SP, McGuire DK, Zinman B, et al.;DEVOTE Study Group. Efficacy and safety of de-gludec versus glargine in type 2 diabetes. N Engl JMed 2017;377:723–73258. HolmanRR, ColemanRL, Chan JCN, et al.; ACEStudy Group. Effects of acarbose on cardiovascu-lar and diabetes outcomes in patients with coro-nary heart disease and impairedglucose tolerance(ACE): a randomised, double-blind, placebo-controlled trial. Lancet Diabetes Endocrinol2017;5:877–88659. Zannad F, Cannon CP, Cushman WC, et al.;EXAMINE Investigators. Heart failure and mortal-ity outcomes in patients with type 2 diabetes tak-ing alogliptin versus placebo in EXAMINE: amulticentre, randomised, double-blind trial. Lan-cet 2015;385:2067–207660. Wanner C, Inzucchi SE, Lachin JM, et al.;EMPA-REG OUTCOME Investigators. Empagliflozinandprogression of kidney disease in type 2 diabetes.N Engl J Med 2016;375:323–33461. Theodorakis MJ, Coleman RL, Feng H, et al.Baseline characteristics and temporal differencesin Acarbose Cardiovascular Evaluation (ACE) trialparticipants. AmHeart J. 1 September 2017 [Epubahead of print]. https://doi.org/10/1016/j.ahj.2017.09.00162. Gantz I, Chen M, Suryawanshi S, et al. A ran-domized, placebo-controlled study of the cardio-vascular safety of the once-weekly DPP-4 inhibitor
omarigliptin in patients with type 2 diabetes mel-litus. Cardiovasc Diabetol 2017;16:11263. Merck. Merck provides update on filing plansfor omarigliptin, an investigationalDPP-4 inhibitorfor type 2 diabetes [Internet]. Available fromhttp://www.mrknewsroom.com/news/company-statements/merck-provides-update-filing-plans-omarigliptin-investigational-dpp-4-inhibi.Accessed 7 April 201764. U.S. Food and Drug Administration. FDA drugsafety communication: FDA adds warnings aboutheart failure risk to labels of type 2 diabetesmed-icines containing saxagliptin and alogliptin [Inter-net]. Available from http://www.fda.gov/Drugs/DrugSafety/ucm486096.htm. Accessed 1 January201765. Rosenstock J, Marx N, Kahn SE, et al. Cardio-vascular outcome trials in type 2 diabetes and thesulphonylurea controversy: rationale for theactive-comparator CAROLINA trial. Diab Vasc DisRes 2013;10:289–30166. Neal B, Perkovic V, Matthews DR, et al.;CANVAS-R Trial Collaborative Group. Rationale, de-sign andbaseline characteristics of theCANagliflozincardioVascular Assessment Study-Renal (CANVAS-R): a randomized, placebo-controlled trial. DiabetesObes Metab 2017;19:387–39367. Eli Lilly. Jardiance (empagliflozin) tablets tobestudied in chronic kidney disease [Internet]. Avail-able from https://investor.lilly.com/releasedetail.cfm?releaseid=1029854. Accessed 11 October201768. Ferrannini E, Mark M, Mayoux E. CV protec-tion in the EMPA-REG OUTCOME trial: a “thriftysubstrate” hypothesis. Diabetes Care 2016;39:1108–111469. Mudaliar S, Alloju S, Henry RR. Can shift infuel energetics explain the beneficial cardiorenaloutcomes in the EMPA-REG OUTCOME study? Aunifying hypothesis. Diabetes Care 2016;39:1115–112270. Abdul-Ghani M, DeFronzo RA, Del Prato S,Chilton R, Singh R, Ryder REJ. Cardiovascular dis-ease and type 2 diabetes: has the dawn of a newera arrived? Diabetes Care 2017;40:813–82071. U.S. Food and Drug Administration. FDA ap-proves Jardiance to reduce cardiovascular deathin adults with type 2 diabetes [Internet]. Availablefrom http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm531517.htm. Accessed14 February 201772. American Diabetes Association. Pharmaco-logic approaches to glycemic control. Sec. 8. InStandards of Medical Care in Diabetesd2017.Diabetes Care 2017;40(Suppl. 1):S64–S7473. Neal B, Perkovic V, de Zeeuw D, et al. Ratio-nale, design, and baseline characteristics of theCanagliflozin Cardiovascular Assessment Study(CANVAS)–a randomized placebo-controlled trial.Am Heart J 2013;166:217–223.e1174. Janssen Pharmaceuticals. FDA briefing docu-ment: NDA 204042 Invokana (canagliflozin) tab-lets [Internet]. Endocrinologic and MetabolicDrugs Advisory Committee Meeting. 10 January2013. Available from https://wayback.archive-it.org/7993/20170405220010/https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/UCM334550.pdf. Accessed 10 October 201775. Neal B, Perkovic V, Mahaffey KW, et al.;CANVAS Program collaborative group. Optimizing
the analysis strategy for the CANVAS Program: aprespecified plan for the integrated analyses ofthe CANVAS and CANVAS-R trials. Diabetes ObesMetab 2017;19:926–93576. Victoza liraglutide injection 1.2 mg/1.8 mg[prescribing information]. Plainsboro, NJ, NovoNordisk, 201777. Novo Nordisk. Semaglutide receives positive16-0 vote in favour of approval from FDAAdvisoryCommittee [Internet]. Available from https://www.novonordisk.com/bin/getPDF.2142854.pdf.Accessed 19 October 201778. Intarcia Therapeutics. Intarcia submits newdrug application (NDA) to FDA for U.S. marketingapproval of ITCA 650 in type 2 diabetes [Internet].Available from https://www.intarcia.com/media/press-releases/2016-nov-21-nda-submission.html.Accessed 10 April 201779. Lincoff AM, Tardif J-C, Schwartz GG, et al.;AleCardio Investigators. Effect of aleglitazar oncardiovascular outcomes after acute coronarysyndrome in patients with type 2 diabetes melli-tus: the AleCardio randomized clinical trial. JAMA2014;311:1515–152580. Erdmann E, Califf R, Gerstein HC, et al. Effectsof the dual peroxisome proliferator-activated re-ceptor activator aleglitazar in patients with type 2diabetes mellitus or prediabetes. Am Heart J2015;170:117–12281. VaccaroO,MasulliM, Nicolucci A, et al.; Thia-zolidinediones Or Sulfonylureas CardiovascularAccidents Intervention Trial (TOSCA.IT) studygroup; Italian Diabetes Society. Effects on theincidence of cardiovascular events of the additionof pioglitazone versus sulfonylureas in patientswith type 2 diabetes inadequately controlledwith metformin (TOSCA.IT): a randomised, multi-centre trial. Lancet Diabetes Endocrinol 2017;5:887–89782. Gerstein HC, Bosch J, Dagenais GR, et al.;ORIGIN Trial Investigators. Basal insulin and car-diovascular and other outcomes in dysglycemia.N Engl J Med 2012;367:319–32883. ORIGIN Trial Investigators. Cardiovascularand other outcomes postintervention with insulinglargine and omega-3 fatty acids (ORIGINALE). Di-abetes Care 2016;39:709–71684. Canadian Diabetes Association ClinicalPractice Guidelines Expert Committee. Phar-macologic management of type 2 diabetes:2016 interim update. Can J Diabetes 2016;40:484–48685. McMurray JJ, Gerstein HC, Holman RR,Pfeffer MA. Heart failure: a cardiovascular out-come in diabetes that can no longer be ignored.Lancet Diabetes Endocrinol 2014;2:843–85186. Fitchett D, Zinman B,Wanner C, et al.; EMPA-REG OUTCOME� trial investigators. Heart failureoutcomes with empagliflozin in patients withtype 2 diabetes at high cardiovascular risk: resultsof the EMPA-REG OUTCOME� trial. Eur Heart J2016;37:1526–153487. Weidmann P, Ferrari P. Central role of sodiumin hypertension in diabetic subjects. DiabetesCare 1991;14:220–23288. Gerstein HC, Miller ME, Genuth S, et al.;ACCORD Study Group. Long-term effects of inten-sive glucose lowering on cardiovascular outcomes.N Engl J Med 2011;364:818–82889. TheNNTGroup.Quick summaries of evidence-based medicine [Internet]. Available from http://www.thennt.com. Accessed 5 February 2017
30 Cardiovascular Outcomes Trials in Type 2 Diabetes Diabetes Care Volume 41, January 2018
90. Jones WS, Roe MT, Antman EM, et al. Thechanging landscape of randomized clinical trialsin cardiovascular disease. J Am Coll Cardiol2016;68:1898–190791. Bethel MA, Holman R, Haffner SM, et al. De-termining themost appropriate components for acomposite clinical trial outcome. AmHeart J 2008;156:633–64092. Stolker JM, Spertus JA, Cohen DJ, et al. Rethink-ing composite end points in clinical trials: insights frompatientsandtrialists. Circulation2014;130:1254–1261
93. Pocock SJ, Clayton TC, Stone GW. Design ofmajor randomized trials: part 3 of a 4-part serieson statistics for clinical trials. J Am Coll Cardiol2015;66:2757–276694. Rogers JK, Jhund PS, Perez AC, et al. Effect ofrosuvastatin on repeat heart failure hospitaliza-tions: the CORONA Trial (Controlled RosuvastatinMultinational Trial in Heart Failure). JACC HeartFail 2014;2:289–29795. Rogers JK, Pocock SJ, McMurray JJ, et al. Ana-lysing recurrent hospitalizations in heart failure: a
review of statistical methodology, with applicationto CHARM-Preserved. Eur J Heart Fail 2014;16:33–4096. Gallin EK, Bond E, Califf RM, et al. Forgingstronger partnerships between academic healthcenters and patient-driven organizations. AcadMed 2013;88:1220–122497. Bergenstal RM, Bailey CJ, Kendall DM. Type 2diabetes: assessing the relative risks and benefitsof glucose-lowering medications. Am JMed 2010;123:374.e9–e18
care.diabetesjournals.org Cefalu and Associates 31
