Dipeptidyl-peptidase-4 Inhibitors and Heart Failure: Class Effect, Substance-Specific Effect, or Chance Effect?

Abstract

Opinion statement:
The increased risk of heart failure hospitalizations related to treatment with the DPP-4 inhibitor saxagliptin observed in the SAVOR TIMI 53 trial, is likely not to be a chance effect, but rather a previously unrecognized side effect of this drug, as this risk was very consistently apparent across all subgroups of this large multicenter, prospective, randomized trial. Whether this side effect might represent a class effect of all DPP-4 inhibitors remains to be seen. Results of randomized prospective multicenter trials with the DPP-4 inhibitors alogliptin and vildagliptin have in fact generated new uncertainties and clearly not totally excluded the possibility of a class side effect. A meta-analysis of 59 randomized controlled trials with various DPP-4 inhibitors evaluating data from 36,620 patients with diabetes and a minimal observation period of 24 weeks, confirmed a 21 % increase of heart failure events compared to placebo treatment, however, not in comparison to treatment with other blood glucose lowering drugs. German registry data also did not show an increased risk for heart failure for the latter comparison. Potential interactions of DPP-4 inhibitors with other drugs, e.g. ACE inhibitors, have been discussed in relation to the increased heart failure risk, as well as interactions with peptides regulating cardiovascular functions that are also split by DPP-4 enzymes such as BNP, substance P, and NPY. Results from ongoing large multicenter trials with the DPP-4 inhibitors sitagliptin and linagliptin are expected to clarify the potential heart failure issue related to treatment with DPP-4 inhibitors.

Full text

Metadata of the article that will be visualized in OnlineFirst
1 Article T itle Dipeptidyl-peptidase-4 Inhibitors and Heart Failure: Class Effect,
Substance-Specific Effect, or Chance Effect?
2 Article Sub- T i tl e
3 Article Copyright -
Year
Springer Science+Business Media New York 2 014
(This will be the copyright line in the final PDF)
4 Journal Nam e Current Treatm ent Op ti ons in Cardiovascular Medi cine
5
Correspondi ng
Author
Family Name Standl
6 Particl e
7 Given Name Eberhard
8 Suffix
9 Organizati on Muni ch Diabetes Research Group e.V. at
Helmhol tz Centre
10 Division
11 Address Ingol staedter Landstrasse 1, Neuherberg 85764,
Germany
12 e-mai l eberhard.standl@lrz.uni-muenchen.de
13
Author
Family Name Erbach
14 Particle
15 Gi ven Name Michael
16 Suffi x
17 Organi zation Sciarc Institute
18 Divisi on
19 Address Baierbrunn, Ge rmany
20 e-mai l
21
Author
Family Name Schnell
22 Particle
23 Gi ven Name Ol iv er
24 Suffi x
25 Organi zation Munich Diabetes Research Group e.V. at
Helmhol tz Centre
26 Divisi on
27 Address Ingol staedter Landstrasse 1, Neuherberg 85764,
Germany

28 e-mai l
29
Schedule
Recei ved
30 Revi sed
31 Accepted
32 Abstract
33 Keywords
separated by ' - '
Heart fai lure - DPP-4 i nhibitors - Diabetes therapy - Side effects -
Cardiovascular safety of diabetes drugs
34 Foot note
information
This arti cle i s part of the Topi cal Collection on Prevention

UNCORRECTED PROOF
1Curr Treat Options Cardio Med (2014) 16:353
2DOI 10.1007/s11936-014-0353-y
3
4
5Prevention (L Sperling and D Gaita, Section Editors)
6Dipeptidyl-peptidase-4
7Inhibitors and Heart Failure:
8Class Effect, Substance-
9Specific Effect, or Chance
10 Effect?
11Q2 Eberhard Standl
1,*
12 Michael Erbach
2
13 Oliver Schnell
1
14 Address
15
*,1
Munich Diabetes Research Group e.V. at Helmholtz Centre, Ingolstaedter
16 Landstrasse 1, 85764, Neuherberg, Germany
17 Email: eberhard.standl@lrz.uni-muenchen.de
18
2
Sciarc Institute, Baierbrunn, Germany
19
20
21 *Springer Science+Business Media New York 2014
22
23 This article is part of the Topical Collection on Prevention
24
25 Keywords Heart failure IDPP-4 inhibitors IDiabetes therapy ISide effects ICardiovascular safety of diabetes
26 drugs
27
28 Opinion statement
29 The increased risk of heart failure hospitalizations related to treatment with the DPP-4
30 inhibitor saxagliptin observed in the SAVOR TIMI 53 trial, is likely not to be a chance
31 effect, but rather a previously unrecognized side effect of this drug, as this risk was
32 very consistently apparent across all subgroups of this large multicenter, prospective,
33 randomized trial. Whether this side effect might represent a class effect of all DPP-4
34 inhibitors remains to be seen. Results of randomized prospective multicenter trials with
35 the DPP-4 inhibitors alogliptin and vildagliptin have in fact generated new uncer-
36 tainties and clearly not totally excluded the possibility of a class side effect. A
37 meta-analysis of 59 randomized controlled trials with various DPP-4 inhibitors evaluat-
38 ing data from 36,620 patients with diabetes and a minimal observation period of
39 24 weeks, confirmed a 21 % increase of heart failure events compared to placebo treat-
40 ment, however, not in comparison to treatment with other blood glucose lowering
41 drugs. German registry data also did not show an increased risk for heart failure for
42 the latter comparison. Potential interactions of DPP-4 inhibitors with other drugs,
43 e.g. ACE inhibitors, have been discussed in relation to the increased heart failure risk,
44 as well as interactions with peptides regulating cardiovascular functions that are also

UNCORRECTED PROOF
45 split by DPP-4 enzymes such as BNP, substance P, and NPY. Results from ongoing large
46 multicenter trials with the DPP-4 inhibitors sitagliptin and linagliptin are expected to
47 clarify the potential heart failure issue related to treatment with DPP-4 inhibitors.
48
49 Introduction
50 Chronic heart failure develops rather frequently in pa-
51 tients with diabetes mellitus, i.e. some 30 % more of-
52 ten compared to non-diabetic subjects [1,2,3••,4].
53 Thus, heart failure has emerged as a clinically impor-
54 tant issue in the context of diabetes-associated cardiac
55 complications [1,2,3••,4]. Heart failure, however,
56 has also attracted wide attention in relation to the de-
57 velopment of innovative blood glucose-lowering
58 drugs for people with diabetes, e.g. dual PPAR
59 alpha/gamma agonists (“glitazars”)andthe
60 thiazolidindiones or PPAR gamma agonists
61 (“glitazones”). Because of the negative experience with
62 muraglitazar and a doubling of the rate of heart failure
63 and cardiovascular complications in the published la-
64 belling studies [5], the American Food and Drug Ad-
65 ministration and the European Medicines Agency
66 have updated their labelling rules for new diabetes
67 drugs, particularly focusing on cardiovascular safety
68 [6]. This is further augmented by similar adverse re-
69 sults obtained from randomized, placebo-controlled,
70 prospective cardiovascular outcome studies with
71 rosiglitazone [7–9], but also with pioglitazone [10,
72 11]. The disadvantageous outcome results with
73 aleglitazar in the recently stopped Aleglitazar to Re-
74 duce Cardiovascular Risk in Coronary Heart Disease
75Patients with a Recent Acute Coronary Syndrome
76Event and Type 2 Diabetes Mellitus (ALECARDIO)
77Study seem to underpin the validity of the new regula-
78tory measures [12]. The debate over an increased heart
79failure risk connected with the use of blood glucose-
80lowering drugs, however, has entered a new phase, af-
81ter the results of the randomized, placebo-controlled
82Saxagliptin Assessment of Vascular Outcomes Recorded
83in Patients with Diabetes Mellitus (SAVOR-TIMI 53)
84Study became available. Unexpectedly, the authors
85found a 27 % increased risk for heart failure hospitaliza-
86tions in the group on treatment with saxagliptin, a
87dipeptidyl-peptidase-4 (DPP-4) inhibitor, compared to
88the placebo group [13•]. Furthermore, results of ran-
89domized, prospective, multicenter trials with other
90DPP-4 inhibitors, i.e. with alogliptin and vildagliptin
91have in fact generated new uncertainties and clearly
92not excluded the possibility that an increased heart fail-
93ure risk may comprise a previously unrecognized side ef-
94fect of the whole class of DPP-4 inhibitors [14,15•,16].
95The mini-review presented here attempts to evalu-
96ate whether the increased risk of heart failure hospital-
97izations seen in SAVOR-TIMI 53 represents a class side
98effect of DPP-4 inhibitors, a saxagliptin specific side ef-
99fect, or a “chance”effect.
100 Evidence from randomized cardiovascular outcome trials
101
102 SAVOR-TIMI 53 randomized 16,492 patients with type 2 diabetes and a
103 history of cardiovascular (CV) events or at high CV risk either to saxagliptin
104 or placebo and followed them for a median of 2.1 years [13•]. A primary
105 endpoint event (i.e. a composite of CV death, myocardial infarction, or
106 stroke) occurred in 613 patients of the saxagliptin group and 609 patients of
107 the placebo group. In other words, the CV event rate was identical in both
108 groups [hazard ratio (HR) for saxagliptin 1.00; 95 % confidence interval (CI)
109 0.89 –1.12; p=0.99 for superiority and pG0.001 for non-inferiority]. An in-
110 creased CV risk from saxagliptin treatment, therefore, appeared to be ex-
111 cluded. On the other hand, an advantage of saxagliptin treatment in terms of
112 CV complications could not be substantiated either. Looking at the
353, Page 2 of 10 Curr Treat Options Cardio Med (2014) 16:353

UNCORRECTED PROOF
113 predefined and adjudicated secondary endpoints, however, an imbalance in
114 heart failure events requiring hospitalization became apparent. More patients
115 in the saxagliptin group had been hospitalized for heart failure compared to
116 the placebo group (3.5 % vs. 2.8 %; HR 1.27; 95 % CI 1.07 –1.51; p= 0,007).
117 This result was unexpected and could have also reflected a chance finding in
118 the context of multiple statistical comparisons [13•]. Meanwhile, however, it
119 is clear [15•] that this imbalance in heart failure hospitalizations had oc-
120 curred both in the 2,105 patients with a prior history of heart failure, i.e. a
121 high risk group for heart failure, as well as in the remaining patients without
122 a prior history of heart failure (11.7 vs. 10.2 % in patients with prior heart
123 failure, HR 1.21; 95 % CI 0.93 –1.58 vs. 2.3 and 1.7 in patients without prior
124 heart failure, HR 1.32; 95 % CI 1.04 –1.65; p=0.68 for interaction).
125 Conversely, looking at the concentrations of NT-pro-BNP measured at
126 baseline, it was revealed that the excess of heart failure hospitalizations
127 hadhappenedmoreorlessexclusively in the highest NT-pro-BNP
128 quartile (10.9 vs. 9.0 %; HR 1.31; 95 % CI 1.0 –1.6; p=0.021), sug-
129 gestingaroleof“subclinical”heart failure as a risk factor of the unto-
130 ward effect of saxagliptin. Beyond that, the increased heart failure risk in
131 the saxagliptin group was consistently seen in all of subgroups. The re-
132 sult was independent of demographic or biochemical variables such as
133 age, gender, BMI, renal function, glycemic, and lipid parameters, as well
134 as a concurrent therapy with diabetes drugs like insulin, metformin, or
135 sulphonylureas or with the various classes of antihypertensive drugs,
136 including ACE inhibitors or angiotensin receptor blockers, or with aspi-
137 rin or statins. Thus, the increased heart failure risk seems likely not to be
138 a chance effect, but rather a previously unrecognized side effect of
139 saxagliptin, although it is important to note that it had no impact on the
140 overall primary CV outcome in SAVOR-TIMI 53.
141 In parallel to SAVOR-TIMI 53, the Examination of Cardiovascular Out-
142 comes with Alogliptin versus Standard of Care (EXAMINE) Trial was pub-
143 lished, another randomized and placebo-controlled study evaluating the
144 DPP-4 inhibitor alogliptin in 5,380 patients with diabetes and recent myo-
145 cardial infarction or unstable angina, respectively, with a mean follow-up of
146 18 months [14]. Similar to SAVOR-TIMI 53, EXAMINE was focused on CV
147 safety in a cohort of diabetic patients at high risk for CV complications.
148 Again, the primary outcome was defined as a composite of CV death, non-
149 fatal myocardial infarction, or stroke and showed no difference between the
150 two treatment groups. Three hundred and five patients in the alogliptin
151 group (11.3 %) and 316 patients in the placebo group (11.8 %) had
152 developed a primary endpoint event (HR 0.96; 95 % CI upper limit
153 1.16; pG0.001 for non-inferiority, p=0.32 for superiority). Surprisingly,
154 no results with regard to heart failure events were released in the primary
155 publication despite the fact that some 28 % of patients in both study
156 arms had a prior history of heart failure [14]. In published responses to
157 enquiries about heart failure, a statement was made by the investigators
158 of EXAMINE that no significant differences had been observed for heart
159 failure between the two treatment groups of the trial [15•]; this state-
160 ment was not qualified further. According to presentations of the
161 EXAMINE data at large international meetings such as EASD or ACC,
162 however, the situation seems to be less clear [17]. Looking at a newly
Curr Treat Options Cardio Med (2014) 16:353 Page 3 of 10, 353

UNCORRECTED PROOF
163 defined explorative composite endpoint (all-cause mortality, non-fatal
164 myocardial infarction or stroke, emergency revascularization for instable
165 angina, or hospitalization for heart failure) no significant differences
166 were found (HR 0.98; 95 % CI 0.86 –1.12) and the frequency of heart
167 failure within this composite endpoint amounted to 3.1 % in the
168 alogliptingroupcomparedto2.9%intheplacebogroup(HR1.07;
169 95 % CI 0.79 –1.46). However, this contrasts with the tendency that all
170 patients requiring hospitalization for heart failure in the trial were con-
171 sidered irrespective of other prior events. One hundred and six patients
172 were contained in the alogliptin group compared to 89 patients in the
173 placebo group (HR 1.19; 95 % CI 0.89 –1.58). A full publication of
174 these data is urgently needed. For the time being, a preliminary meta-
175 analysis looking at all heart failure in SAVOR-TIMI 53 and EXAMINE
176 (Table 1) fosters the suspicion that an increased heart failure risk might
177 be an emerging side effect of the whole class of DPP-4 inhibitors, not
178 just of saxagliptin (a total of 395 patients with hospitalization requiring
179 heart failure on DPP-4 inhibitors in comparison to 317 patients on
180 placebo, HR 1.24; 95 % CI 1.07 –1.44).
181 Another DPP-4 inhibitor, vildagliptin, has been investigated in terms of
182 cardiac safety in diabetic heart failure patients in the Vildagliptin in Ven-
183 tricular Dysfunction Diabetes (VIVIDD) Trial [16]. Again, only information
184 derived from congressional presentations is currently available and a full
185 publication is eagerly awaited. The VIVIDD Trial had enrolled 254 patients
186 with diabetes and chronic heart failure NYHA class I-III and randomized
187 them to either vildagliptin or placebo therapy. The echocardiographic ejec-
188 tion fraction was determined as primary endpoint and measured at baseline
189 and at 1-year follow-up. Both groups showed significant improvement of
190 some 4 % at 1 year with no significant difference existing between the two
191 groups. Unexpected, however, was the finding of a significant increase of
192 the left-ventricular end-diastolic and end-systolic volume as well as of
193 the stroke volume in the vildagliptin arm compared to placebo. Con-
194 versely, plasma BNP concentrations had fallen by 28 % in the
195 vildagliptin group, and by 14 % in the placebo group. The difference,
196 however, was not statistically significant. Likewise, no significant differ-
197 ence was seen in terms of all-cause mortality (four patients on placebo
198 vs. 11 patients on vildagliptin). The trial, however, was not powered to
199 detect differences in clinical endpoints and conclusions regarding cardiac
200 safety of vildagliptin are, therefore, limited. Hence, uncertainties remain
201 which would need to be addressed in a much larger trial with a much
t1:1Table 1. Heart failure requiring hospitalization in prospective, randomized studies evaluating DPP-4 in-
hibitors and cardiovascular outcomes –a Meta-Analysis
Study DPP-4 inhibitor Placebo OR (95 % CI)
t1:2
SAVOR-TIMI 53 289 228 1.27 (1.06–1.52)t1:3
EXAMINE 106 89 1.19 (0.89–1.58)t1:4
Combined 395 317 1.24 (1.07–1.44)t1:5
353, Page 4 of 10 Curr Treat Options Cardio Med (2014) 16:353

UNCORRECTED PROOF
202 longer follow-up. The finding of larger hearts after a 1 year of therapy
203 with vildagliptin does not exclude an increased heart failure risk, but
204 might be rather suggestive for it.
205 Potential pathophysiologic links
206
207 DPP-4 inhibitors inhibit more or less specifically the enzyme DPP-4 that
208 exists as trans-membranous exo-peptidase in many cells of the body and
209 splits off dipeptides at the N-terminal end of proteins or peptides immedi-
210 ately after a proline of alanine residue (sometimes also after other amino
211 acid residue) in position 2 [18]. Among a wide scope of aspects, DPP-4 has
212 turned out to be an important regulator of the incretin effect, as it spits the
213 two main incretin hormones released from the L- or K-cells, respectively,
214 in the small intestines, i.e. glucagon-like peptide 1 (GLP-1) and glucose-
215 dependent insulinotropic polypeptide (GIP). In the case of GLP-1, the
216 hormonelosesitscapacitytobindtotheGLP-1receptorofcellsandto
217 induce signal transduction by the removal of the dipeptide. Inhibition of
218 GLP-1 degradation by DPP-4 inhibitors increases the GLP-1 receptor-
219 dependent effects and is the basis of the mode of action of the class of
220 blood glucose-lowering drugs called DPP-4 inhibitors. On the other
221 hand, the GLP-1 molecule shortened by the dipeptide seems to have
222 physiologic albeit GLP-1 receptor-independent effects, e.g. at the heart
223 [18]. DPP-4 inhibitors, therefore, shift the effects of the original peptide
224 and the shortened peptide in favor of the original peptide. Whether this
225 shift also impacts the biological action of the shortened peptide is un-
226 clear,butmight,however,beofimportanceinthecontextofapotential
227 link between the use of DPP-4 inhibitors and the occurrence of heart
228 failure.
229 Meanwhile, it has become apparent that the DPP-4 enzyme splits a mul-
230 titude of biologically important peptides that exert effects on the heart [18,
231 19]. An incomplete list of these peptides (the number of the amino acid
232 residues is shown in brackets) summarizes their cardiac effects (other effects
233 are again shown in brackets):
234 –GLP-1 (GLP-1 7–36): increased cardiac function, glucose uptake,
235 decreased contractility, apoptosis (blood vessels: increased NO-
236 production, decreased inflammation)
237 –B-Type Natriuretic Peptide (BNP 1–32): decreased LV-remodeling
238 (blood vessels: increased vasodilatation; kidney: increased natriuresis)
239 –Substance P (SP 1–11): decreased chronotropy and inotropy (brain:
240 altered cardiac adrenergic tone)
241 –Neuropeptide Y (NPY 1–36): increased Ca2+i-voltage
242 –Peptide YY (PYY 1–36): (blood vessels: increased collateral blood flow)
243 –GLP2 (GLP-1 1–33): (blood vessels: increased blood flow, blood
244 pressure and heart frequency)
245 –Stromal-derived factor 1 alpha (SDF1 alpha 1–68): (progenitor cells:
246 increased homing of progenitor cells in ischemic myocardium, in-
247 creased angiogenesis)
Curr Treat Options Cardio Med (2014) 16:353 Page 5 of 10, 353

UNCORRECTED PROOF
248 –GIP (GIP 1–42): in rodents, receptors detected in the atrium and
249 ventricle (lipogenesis?, but also effects on signal transduction path-
250 ways of endothelial cells)
251 An enhancement of these effects, therefore, could exhibit substantial
252 modulations of heart function, and a connection with the increased oc-
253 currence of heart failure cannot primarily be excluded. Our current
254 knowledge of this area of research, however, is rather limited. Studies
255 are certainly warranted about whether circulating concentrations of
256 BNP and their measurement are affected by DPP-4 inhibition, especially
257 in patients with heart failure. In addition, the DPP-4 enzyme splits the ba-
258 sic fibroblast growth factor (bFGF), perhaps anchoring proteins for cyto-
259 kines within the extracellular matrix (18). Conversely, some of the
260 cleavage products of DPP-4 induced degradation exert profound cardiac
261 effects. They are summarized below:
262 –GLP-1 9–36: increased cardiac function, glucose uptake, de-
263 creased apoptosis (blood vessels: increased vasodilatation)
264 –BNP 3–32: (kidney: increased natriuresis)
265 –NPY 3–36: (blood vessels: increased angiogenesis)
266 Whether these effects might be modified in the context of therapeutic
267 DPP-4 inhibition is again unclear and merits scientific attention. It is
268 also noteworthy in this connection that PYY 3–36 in comparison to
269 PYY 1–36 penetrates the blood brain barrier much more easily and
270 induces anorexic effects in the brain. Moreover, PYY 3–36 represents
271 an anti-secretory and pro-absorptive hormone and regulates the post-
272 prandial water and sodium influx into the gut, especially in the ileum
273 and colon [19]. Obviously, therapeutic usage of DPP-4 inhibition
274 might influence a multitude of biological processes in a very complex
275 way, not to mention the various affinities and specificities in regulat-
276 ing signal transduction pathways and partial inhibitory effects of the
277 enzymes DPP-8 and DPP-9. Hence, not only do potential connections
278 withtheoccurrenceofheartfailureneedfurtherclarification,butalso
279 aspects beyond.
280 Discussion
281
282 In aggregate, weighing the evidence in relation to the increased occurrence of
283 heart failure requiring hospitalization while on therapy with saxagliptin, one
284 seems to deal with a previously unknown side effect of a blood glucose
285 lowering DPP-4 inhibitor, as this signal appears to be very consistent and
286 robust in the huge data base of SAVOR-TIMI 53 [13•,15•]. It is important in
287 terms of clinical relevance, however, that this increased risk for heart failure
288 had no effect on the primary endpoint and by an independent committee
289 adjudicated endpoint which comprised a composite of cardiovascular death,
290 non-fatal myocardial infarction, and non-fatal stroke [13•]. The NT-pro-BNP
291 concentrations measured in SAVOR should have been unaffected by therapy
292 with the DPP-4 inhibitor and suggest a role of “subclinical”heart failure as a
293 risk factor for the observed side effect. NT-pro-BNP is produced in equimolar
353, Page 6 of 10 Curr Treat Options Cardio Med (2014) 16:353

UNCORRECTED PROOF
294 amounts when BNP is generated from pro-BNP. Conversely, it remains to be
295 determined whether the BNP concentrations measured in the VIVIDD trial
296 might have been influenced by the administration of the DPP-4 inhibitor
297 vildagliptin, and if so, whether such an effect might have had an effect on
298 cardiac function [16]. Among a number of effects, BNP is involved in left-
299 ventricular remodeling [18], and an increase of left-ventricular end-dia-
300 stolic and end-systolic volume was seen in VIVIDD connected with the
301 use of the DPP-4 inhibitor vildagliptin, although the primary endpoint,
302 i.e. the ejection fraction, showed averysimilarimprovementasinthe
303 placebo arm [16].
304 Potential interactions with ACE inhibitors have also been discussed in the
305 context of possible pathogenic links [20–22]. Paradoxical increases of blood
306 pressure and heart rate have been described at higher doses of ACE inhibitors
307 which might have been induced by interactions with SP1-11 or NPY1-36
308 [19–22]. Increased concentrations of norepinephrine and signs of enhanced
309 vaso-constrictory effects of angiotensin II have been observed [21,23,24].
310 Vildagliptin has been associated with cardiac arrhythmias in experiments
311 with dogs and an increased number of first degree AV blocks has been noted
312 in humans [25].
313 At present, however, it has neither been confirmed nor excluded by the
314 results of the VIVIDD as well as the EXAMINE study whether the heart fail-
315 ure findings in SAVOR may represent a class side effect of all DPP-4 inhibi-
316 tors. A recent meta-analysis has raised new serious concerns. Fifty-nine
317 available prospective and randomized studies evaluating DPP-4 inhibitors
318 for at least 24 weeks have gathered a data base of a total of 36,620 patients
319 with a mean follow-up time of 46.7 weeks [26]. A highly significant increase
320 of heart failure hospitalizations were reported on treatment with DPP-4 in-
321 hibitors compared to placebo therapy (n = 24,111, RR 1.21, 1.03 –1.42),
322 whereas no significant differences were found in terms of all-cause mortality,
323 CV mortality, myocardial infarction, or stroke. In comparison to other blood
324 glucose-lowering agents, however, DPP-4 inhibitors showed comparable
325 clinical outcomes including heart failure [26]. Trends in favor of DPP-4 in-
326 hibitors regarding all-cause mortality, myocardial infarction, and stroke did
327 not reach the level of significance. These observations are in agreement with
328 1-year follow-up data of a large German registry, called DiaRegis, enrolling
329 non-insulin requiring patients with type 2 diabetes [27]. Although add-on
330 treatments were at the discretion of the individual physician and were
331 not allocated randomly, patients on new DPP-4 inhibitor therapy as
332 compared to new sulphonylurea therapy exhibited some non-significant
333 trends for lower rates of stroke and unstable angina, like in the meta-
334 analysis, but no difference was noted in terms of heart failure. Con-
335 flicting data were also recently reported at the Joint Meeting of the In-
336 ternational Society of Endocrinology and the American Endocrine Society
337 [28]. In a retrospective cohort study using the Cleveland Clinic electronic
338 health record system, patients with type 2 diabetes who received a pre-
339 scription for metformin plus a DPP-4 inhibitor had a significant, albeit
340 small, increased risk for heart failure compared with those who received
341 metformin and other oral antidiabetic agents [28]. In contrast, a large
342 data base of “real world”type 2 diabetic patients obtained at the Joslin
343 Diabetes Center, Boston, did not find an adverse heart failure signal in
Curr Treat Options Cardio Med (2014) 16:353 Page 7 of 10, 353

UNCORRECTED PROOF
344 patients starting DPP-4 inhibitor therapy, but rather the opposite, i.e. a
345 less frequent rate of heart failure [28].
346 Thus, in all, important arguments exist at present, not to take the heart
347 failure findings seen in SAVOR-TIMI 53 as unequivocally granted as being
348 indicative for a class side effect of all DPP-4 inhibitors.
349 Perspectives
350
351 Regarding DPP-4 inhibitors and heart failure, the book is not closed yet. The
352 much larger and longer ongoing randomized controlled trials evaluating the
353 DPP-4 inhibitors sitagliptin and linagliptin and looking at hard CV outcomes
354 will probably be key [29,30]. The Trial Evaluating Cardiovascular Outcomes
355 with Sitagliptin (TECOS) and the Cardiovascular Outcome Study of
356 Linagliptin versus Glimepiride in Patients with Type 2 Diabetes (CAROLI-
357 NA), in which more than 20,000 patients with diabetes have been enrolled,
358 are expected to clarify the issue [29,30]. Finally, likewise, still ongoing large
359 multicenter studies with GLP-1 receptor agonists might contribute new no-
360 tions on the topic.
361
362 Acknowledgments
363
364 This activity was supported by an Educational Grant of the Association for the Support of International
365 Scientific Communication in Diabetology e.V., Munich, Germany
366 Compliance with Ethics Guidelines
367
368
369 Conflict of Interest
370 Dr. Eberhard Standl, Dr. Michael Erbach, and Dr. Oliver Schnell each declare no potential conflicts of
371 interest.
372
373 Human and Animal Rights and Informed Consent
374 This article does not contain any studies with human or animal subjects performed by any of the authors.
375376
377 References
378
379 Papers of particular interest, published recently, have been
380 highlighted as:
381 •Of importance
382 •• Of major importance
383 1. Standl E, Schnell O. A new look at the heart in dia-
384 betes mellitus: from ailing to failing. Diabetologia.
385 2000;43(12):1455–69.
386 2. Ryden L, Standl E, Bartnik M, Van den Berghe G,
387 Betteridge J, de Boer MJ, et al. Guidelines on diabetes,
388 pre-diabetes, and cardiovascular diseases: executive
389summary. The Task Force on Diabetes and Car-
390diovascular Diseases of the European Society of
391Cardiology (ESC) and of the European Associa-
392tion for the Study of Diabetes (EASD). Eur Heart
393J. 2007;28(1):88–136. doi:10.1093/eurheartj/
394ehl260.
353, Page 8 of 10 Curr Treat Options Cardio Med (2014) 16:353

UNCORRECTED PROOF
395 3.•• Ryden L, Grant PJ, Anker SD, Berne C, Cosentino F,
396 Danchin N, et al. ESC Guidelines on diabetes, pre-
397 diabetes, and cardiovascular diseases developed in
398 collaboration with the EASD: the Task Force on dia-
399 betes, pre-diabetes, and cardiovascular diseases of the
400 European Society of Cardiology (ESC) and devel-
401 oped in collaboration with the European Association
402 for the Study of Diabetes (EASD). Eur Heart J.
403 2013;34(39):3035–87. doi:10.1093/eurheartj/
404 eht108.
405 most recently updated comprehensive guideline on heart
406 failure & diabetes mellitus.
407 4. McMurray JJ, Gerstein HC, Holman RR, Pfeffer MA.
408 Heart failure: a cardiovascular outcome in diabetes
409 that can no longer be ignored. Early online. Lancet
410 Diabetes Endocrinol. 2014. doi:10.1016/S2213-
411 8587(14)70031-2.
412 5. Nissen SE, Wolski K, Topol EJ. Effect of muraglitazar
413 on death and major adverse cardiovascular events in
414 patients with type 2 diabetes mellitus. JAMA.
415 2005;294(20):2581–6. doi:10.1001/
416 jama.294.20.joc50147.
417 6. Center for Drug Evaluation and Research. Guidance for
418 industry diabetes mellitus: evaluating cardiovascular
419 risk in new antidiabetic therapies to treat type 2 diabe-
420 tes. 2008. Available at http://www.fda.gov/downloads/
421 Drugs/GuidanceComplianceRegulatoryInformation/
422 Guidances/ucm071627.pdf (accessed May 1, 2014).
423 7. Nissen SE, Wolski K. Effect of rosiglitazone on the risk
424 of myocardial infarction and death from cardiovascular
425 causes. N Engl J Med. 2007;356(24):2457–71.
426 doi:10.1056/NEJMoa072761.
427 8. Kahn SE, Haffner SM, Heise MA, Herman WH,
428 Holman RR, Jones NP, et al. Glycemic durability of
429 rosiglitazone, metformin, or glyburide monotherapy.
430 N Engl J Med. 2006;355(23):2427–43. doi:10.1056/
431 NEJMoa066224.
432 9. Home PD, Pocock SJ, Beck-Nielsen H, Curtis PS,
433 Gomis R, Hanefeld M, et al. Rosiglitazone evaluated
434 for cardiovascular outcomes in oral agent combina-
435 tion therapy for type 2 diabetes (RECORD): a
436 multicentre, randomised, open-label trial. Lancet.
437 2009;373(9681):2125–35. doi:10.1016/S0140-
438 6736(09)60953-3.
439 10. Dormandy JA, Charbonnel B, Eckland DJ, Erdmann
440 E, Massi-Benedetti M, Moules IK, et al. Secondary
441 prevention of macrovascular events in patients with
442 type 2 diabetes in the PROactive Study (PROspective
443 pioglitAzone Clinical Trial In macroVascular Events):
444 a randomised controlled trial. Lancet.
445 2005;366(9493):1279–89. doi:10.1016/S0140-
446 6736(05)67528-9.
447 11. Erdmann E, Charbonnel B, Wilcox RG, Skene AM,
448 Massi-Benedetti M, Yates J, et al. Pioglitazone use and
449 heart failure in patients with type 2 diabetes and
450 preexisting cardiovascular disease: data from the
451 PROactive study (PROactive 08). Diabetes Care.
452 2007;30(11):2773–8. doi:10.2337/dc07-0717.
45312. Lincoff AM, Tardif JC, Schwartz GG, Nicholls SJ,
454Rydén L, Neal B, et al. Effect of Aleglitazar on Car-
455diovascular Outcomes After Acute Coronary Syn-
456drome in Patients With Type 2 Diabetes Mellitus. The
457AleCardio Randomized Clinical Trial. JAMA.
4582014;311(15):1515–25.
45913.•Scirica BM, Bhatt DL, Braunwald E, Steg PG, David-
460son J, Hirshberg B, et al. Saxagliptin and cardiovas-
461cular outcomes in patients with type 2 diabetes
462mellitus. N Engl J Med. 2013;369(14):1317–26.
463doi:10.1056/NEJMoa1307684.
464detailed information on fully published randomized con-
465trolled trials on DPP4 inhibitors & heart failure.
46614. White WB, Cannon CP, Heller SR, Nissen SE,
467Bergenstal RM, Bakris GL, et al. Alogliptin after acute
468coronary syndrome in patients with type 2 diabetes.
469N Engl J Med. 2013;369(14):1327–35. doi:10.1056/
470NEJMoa1305889.
47115.•Standl E. Saxagliptin, alogliptin, and cardiovascular
472outcomes. N Engl J Med. 2014;370(5):483. 10.1056/
473NEJMc1313880#SA1.
474detailed information on fully published randomized con-
475trolled trials on DPP4 inhibitors & heart failure.
47616. McMurray J. The Vildagliptin in Ventricular Dys-
477function Diabetes (VIVIDD) trial. Presented at the
478Heart Failure Congress 2013, Lissabon, Portugal.
4792013;99. abstract.
48017. Nainggolan L. Heart-failure data for alogliptin
481prompt more debate at ACC. Medscape. March 31,
4822014. Available at http://www.medscape.com/
483viewarticle/822849 (accessed 14-05-05).
48418. Ussher JR, Drucker DJ. Cardiovascular biology of the
485incretin system. Endocr Rev. 2012;33(2):187–215.
486doi:10.1210/er.2011-1052.
48719. Parker SL, Balasubramaniam A. Neuropeptide Y Y2
488receptor in health and disease. Br J Pharmacol.
4892008;153(3):420–31. doi:10.1038/sj.bjp.0707445.
49020. Fadini GP, Avogaro A. Cardiovascular effects of DPP-
4914 inhibition: beyond GLP-1. Vasc Pharmacol.
4922011;55(1–3):10–6. doi:10.1016/j.vph.2011.05.001.
49321. Marney A, Kunchakarra S, Byrne L, Brown NJ. Inter-
494active hemodynamic effects of dipeptidyl peptidase-
495IV inhibition and angiotensin-converting enzyme
496inhibition in humans. Hypertension.
4972010;56(4):728–33. doi:10.1161/
498HYPERTENSIONAHA.110.156554.
49922. Jackson EK, Dubinion JH, Mi Z. Effects of dipeptidyl
500peptidase iv inhibition on arterial blood pressure.
501Clin Exp Pharmacol Physiol. 2008;35(1):29–34.
502doi:10.1111/j.1440-1681.2007.04737.x.
50323. Jackson EK, Mi Z. Sitagliptin augments sympathetic
504enhancement of the renovascular effects of angio-
505tensin II in genetic hypertension. Hypertension.
5062008;51(6):1637–42. doi:10.1161/
507HYPERTENSIONAHA.108.112532.
50824. Boschmann M, Engeli S, Dobberstein K, Budziarek P,
509Strauss A, Boehnke J, et al. Dipeptidyl-peptidase-IV
Curr Treat Options Cardio Med (2014) 16:353 Page 9 of 10, 353

UNCORRECTED PROOF
510 inhibition augments postprandial lipid mobilization
511 and oxidation in type 2 diabetic patients. J Clin
512 Endocrinol Metab. 2009;94(3):846–52.
513 doi:10.1210/jc.2008-1400.
514 25.Q3 Vildagliptin. Arznei-Telegramm 2008; Jg. 39, Nr. 6,
515 66–67.
516 26. Krum H, Skiba M, Wu S, Hopper I. Heart failure and
517 dipeptidyl peptidase-4 inhibitors. Eur J Heart Fail.
518 2014. doi:10.1002/ejhf.90.
519 27. Gitt AK, Bramlage P, Binz C, Krekler M, Deeg E,
520 Tschope D. Prognostic implications of DPP-4 inhib-
521 itor vs. sulfonylurea use on top of metformin in a
522 real world setting - results of the 1 year follow-up of
523 the prospective DiaRegis registry. Int J Clin Pract.
524 2013;67(10):1005–14.
52528. Naingolan L. Conflicting data on sitagliptin and
526heart failure in diabetes. Medscape 2014, July 1.
527Available at http://www.medscape.com/viewarticle/
528827653 (accessed 14-07-02).
52929. Q4Green JB, Bethel MA, Paul SK, Ring A, Kaufman KD,
530Shapiro DR, et al. Rationale, design, and organiza-
531tion of a randomized, controlled Trial Evaluating
532Cardiovascular Outcomes with Sitagliptin (TECOS)
533in patients with type 2 diabetes and established car-
534diovascular disease. Am Heart J. 2013;166(6):983–9
535e7. doi:10.1016/j.ahj.2013.09.003.
53630. CAROLINA: Cardiovascular Outcome Study of
537Linagliptin Versus Glimepiride in Patients With Type
5382 Diabetes. ClinicalTrials.gov Identifier:
539NCT01243424 (accessed May 1, 2014)
540
353, Page 10 of 10 Curr Treat Options Cardio Med (2014) 16:353

UNCORRECTED PROOF
AUTHOR QUERIES
AUTHOR PLEASE ANSWER ALL QUERIES.
Q1. Please check the suggested running page title if appropriate. Otherwise, please provide
short running title with maximum of 65 characters including spaces.
Q2. Degree is mandatory. Please provide.
Q3. Please provide complete bibliographic details of this reference.
Q4. Please check if the page range here is correct.