J O U R N A L O F T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y
`ª 2 0 1 7 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N
`
`V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0
`
`P U B L I S H E D B Y E L S E V I E R
`
`h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 7 . 0 4 . 0 1 4
`
`THE PRESENT AND FUTURE
`
`STATE-OF-THE-ART REVIEW
`
`Novel Diabetes Drugs and the
`Cardiovascular Specialist
`
`Naveed Sattar, MD, PHD,a Mark C. Petrie, MD,a Bernard Zinman, MD,b James L. Januzzi, JR, MDc,d
`
`ABSTRACT
`
`Recently, treatment with 2 newer classes of type 2 diabetes drugs were found to reduce events in patients with diabetes
`and cardiovascular (CV) disease, a group common in cardiology clinics. The sodium-glucose cotransporter 2 inhibitor,
`empagliflozin, markedly and rapidly reduced CV death and heart failure hospitalization, likely with hemodynamic/
`metabolic-driven mechanisms of action. More recently, the glucagon-like peptide–1 receptor agonists liraglutide and
`semaglutide also reduced CV death and/or major adverse CV events, but did so more slowly and did not influence heart
`failure risks, suggesting alternative mechanisms of benefit. We will discuss drug therapy for diabetes relative to CV
`risk, briefly summarize key findings of CV benefit from recent trials, discuss potential mechanisms for benefits of
`sodium-glucose cotransporter 2 inhibitors and glucagon-like peptide–1 agonists, and suggest how such drugs might be
`embraced by CV specialists to reduce CV events and mortality in their patients. (J Am Coll Cardiol 2017;69:2646–56)
`© 2017 by the American College of Cardiology Foundation.
`
`M ost
`
`focused their
`cardiologists have
`traditional
`risk
`efforts on managing
`factors, and have paid less attention to
`type 2 diabetes (T2D) therapies whose primary role
`is to lower glucose. This may be because, until
`recently, T2D therapies other than metformin had
`little obvious favorable effect on cardiovascular (CV)
`outcomes,
`the principal cause of morbidity and
`mortality in T2D. Indeed, for cardiologists, the most
`common diabetes drug intervention was to stop drugs
`that may cause heart
`failure (e.g., glitazones);
`
`initiation or titration of drugs for diabetes care was
`most commonly referred to primary caregivers or
`diabetes specialists. If anything, concerns about CV
`safety were more prevalent than reassurance as to
`the potential benefits of these agents.
`
`THE RISE OF CV SAFETY AND
`OUTCOME TRIALS IN DIABETES CARE
`
`In light of concerns regarding CV safety of new
`glucose-lowering drugs being developed, the U.S.
`
`Listen to this manuscript’s
`audio summary by
`JACC Editor-in-Chief
`Dr. Valentin Fuster.
`
`From the aInstitute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, United Kingdom; bLunenfeld
`Tanenbaum Research Institute, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada; cCardiology Division,
`Massachusetts General Hospital, Boston, Massachusetts; and the dBaim Institute for Clinical Research, Boston, Massachusetts.
`Dr. Sattar has received speaker fees or consulting honoraria from Amgen, Sanofi, Boehringer Ingelheim, Eli Lilly, AstraZeneca
`(on Operations Committee for EXSCEL trial), Novo Nordisk, and Janssen; and has received research funds from AstraZeneca. Dr.
`Petrie has received speaker fees or consulting honoraria from Takeda, Novartis, AstraZeneca, Maquet, Boehringer Ingelheim, Pfizer,
`Daiichi-Sankyo, Servier, and Eli Lilly; has served as a consultant for Novo Nordisk; and has served on clinical events committees for
`Roche, Bayer, Stealth Biotherapeutics, AstraZeneca, GlaxoSmithKline, Astellas, Cardiorentis, Reservlogix, and Boehringer Ingelheim
`(including for the EMPA-REG OUTCOME trial). Dr. Zinman has received grant support from Boehringer Ingelheim, Novo Nordisk, and
`AstraZeneca; has received consulting and speaking honoraria from Boehringer Ingelheim, Eli Lilly, Merck, Novo Nordisk, Astra-
`Zeneca, Sanofi, and Janssen; and is an investigator in EMPA-REG OUTCOME, LEADER, and EXSCEL. Dr. Januzzi is supported by the
`Hutter Family Professorship; has received grant support from Siemens, Singulex, and Prevencio; has received consulting income
`from Roche Diagnostics, Critical Diagnostics, Sphingotec, Phillips, Novartis, Janssen and Boehringer Ingelheim; and participates in
`clinical endpoint committees/data safety monitoring boards for Pfizer, Novartis, Amgen, Janssen, and Boehringer Ingelheim
`(including for the EMPA-REG OUTCOME Trial). Drs. Sattar and Petrie contributed equally to this work. Deepak L. Bhatt, MD, MPH,
`served as Guest Editor-in-Chief for this paper. Bejamin M. Scirica, MD, served as Guest Editor for this paper.
`
`Manuscript received December 21, 2016; revised manuscript received April 3, 2017, accepted April 3, 2017.
`
`Novo Nordisk Exhibit 2109
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00001
`
`

`

`J A C C V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`M A Y 3 0 , 2 0 1 7 : 2 6 4 6 – 5 6
`
`Sattar et al.
`The Cardiovascular Importance of New Diabetes Drugs
`
`2647
`
`Food and Drug Administration (FDA) and European
`Medicines Agency mandated that new therapies for
`diabetes had to demonstrate CV safety in prospective,
`randomized controlled outcome trials. Current rec-
`ommendations for trial design of new therapies for
`T2D have been recently reviewed (1) and include
`iterative assessment of drug safety, with initially lib-
`eral pre-approval statistical boundaries to exclude
`unacceptable CV risk, followed by more restrictive
`boundaries
`post-approval. For
`phase
`4
`post-
`marketing outcome trials, ultimately,
`the upper
`bound of the 95% confidence interval (CI) for any T2D
`treatment should not exceed 1.30 for major adverse
`cardiovascular events (MACE), whereas a 1.80 upper
`limit applies to phase 3 trials. Additionally,
`the
`recommendation was made that
`trials evaluating
`novel T2D therapies should focus on high-risk pop-
`ulations (such as those with vascular disease, with
`renal impairment, or at advanced age) and should
`include long-term data, and that all MACE events
`measured in such trials should be adjudicated by an
`independent committee.
`Although designed to detect a risk signal, remark-
`ably, results from recent “cardiovascular outcomes
`trials” (CVOTs) may lead to a meaningful change in
`how cardiologists might approach the patient with
`T2D, as these CVOTs have shown not only CV safety,
`but also reduced CV and all-cause mortality in some
`studies (2–4). These trials include patients who are
`common to cardiologists’ practices, and the magni-
`tude of the results compares favorably with the
`landmark cardiology trials that have shaped our
`international cardiology guidelines (5,6).
`Clearly, cardiologists would do well to keep up
`with this evolving area of T2D CVOTs to ensure that
`their patients potentially benefit from newer thera-
`pies for diabetes care. In addition, a good under-
`standing of the potential risks of diabetes drugs in
`treating patients with CV disease is also important.
`Before discussing newer therapies, reviewing expe-
`rience of the CV effects of older drugs is helpful.
`
`DIABETES DRUGS THAT HAVE LESS
`FAVORABLE OR UNCERTAIN CV OR
`MORTALITY RISK BENEFITS
`
`glucose-
`landmark
`of
`Although meta-analyses
`lowering trials suggest that intensive glycemic con-
`trol does reduce risk for CV disease events (7),
`improved CV outcomes as a function of intensive
`glucose control appear modest in comparison to the
`calculated CV benefits from lipid and blood pressure
`management (8). In addition, some concerning sig-
`nals for risk of CV events have been associated with
`
`certain widely-used diabetes medications,
`including sulfonylureas, thiazolidinediones,
`dipeptidylpeptidase 4 inhibitors, and insulin.
`
`A B B R E V I A T I O N S
`
`A N D A C R O N Y M S
`
`CI = confidence interval
`
`CV = cardiovascular
`
`CVOT = cardiovascular
`outcomes trial
`
`GLP = glucagon-like peptide
`
`SULFONYLUREAS. Although widely used for
`care of T2D, drugs from the sulfonylurea class
`of drugs (although perhaps less so for glicla-
`zide) (9) have been associated with a higher
`risk for CV events, notably including a higher
`risk for nonfatal myocardial infarction (MI) or
`CV death, relative to other diabetes drugs
`(10). For example, a meta-analysis of 72 small
`or modest-sized randomized controlled trials
`found that all-cause mortality; CV mortality;
`and a composite of MI, stroke, and CV mor-
`tality were all increased in patients treated
`with glibenclamide, glipizide, and tolbuta-
`mide compared with metformin (11). Based
`on these and other data, sulfonylurea medications
`carry a “black box” CV warning from the FDA
`regarding heightened risk for CV events, although the
`same is not true in many non-U.S. countries.
`
`HbA1c = glycosylated
`hemoglobin
`
`HR = hazard ratio
`
`MACE = major adverse
`cardiovascular event(s)
`
`MI = myocardial infarction
`
`SGLT = sodium-glucose
`cotransporter
`
`T2D = type 2 diabetes
`
`(TZDs)
`THIAZOLIDINEDIONES. Thiazolidinediones
`are agonists for the peroxisome proliferator-activated
`receptors that regulate gene expression, resulting in
`improved glucose utilization and reduced glucose
`production. TZDs improve a number of CV risk factors
`and became widely used at one point; however,
`reports of potential CV risk (including reports of fluid
`retention with incident heart failure as well as a
`possible increased risk for incident MI [12]), and
`earlier reports of excess bladder cancer risk (now
`debated [13])
`led to reduction in their use. For
`example, the RECORD (Rosiglitazone Evaluated for
`Cardiac Outcomes and Regulation of Glycemia in
`Diabetes) trial reported an adjusted risk for incident
`heart failure (hazard ratio [HR]: 2.25; 95% CI: 1.27 to
`3.97) (14), similar to findings in a meta-analysis (15).
`The MI risks for rosiglitazone have now been largely
`dispelled (16), whereas pioglitazone does have trial
`evidence to show net CV benefit (17), but the height-
`ened heart failure risk, as well as weight gain and
`potential risks for fractures with this class of drugs,
`has led to a reduction in their use (18).
`
`DIPEPTIDYL PEPTIDASE-4 INHIBITORS. Dipeptidyl
`peptidase-4 (DPP-4) is an enzyme that degrades many
`peptides,
`including glucagon-like peptide (GLP)-1;
`thus, pharmacological inhibition of DPP-4 prolongs
`the half-life
`and biological
`activity of GLP-1.
`Inhibitors of DPP-4 have modest glucose-lowering
`effects, but although 3 recent CVOTs did show
`evidence of CV safety according to FDA criteria, they
`did not demonstrate net CV benefits (19–21) contra-
`dicting an earlier meta-analysis (22). Furthermore,
`
`Novo Nordisk Exhibit 2109
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00002
`
`

`

`2648
`
`Sattar et al.
`The Cardiovascular Importance of New Diabetes Drugs
`
`J A C C V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`M A Y 3 0 , 2 0 1 7 : 2 6 4 6 – 5 6
`
`T A B L E 1 Summary of Key Findings of the 3 Positive CVOTs in T2D, Detailing Adverse Effects and Broad Beneficial Mechanisms
`Implicated in CV Benefits
`
`EMPA-REG OUTCOME (2)
`
`LEADER (3)
`
`SUSTAIN-6 (4)
`
`Agent
`
`Empagliflozin (SGLT2 inhibitor)
`
`Inclusion criteria
`
`All with T2D and CVD
`HbA1c 7%–10%
`
`Duration of trial
`Baseline HbA1c
`Primary endpoint
`CV death
`MI
`Stroke
`HF hospitalization
`Noteworthy adverse
`effects
`Likely broad mechanisms
`of benefit
`
`3.1 yrs
`8.1%
`Y 14% (1% to 26%)
`Y 38% (23% to 51%)
`Y 13% (9% to 30%)
`[ 24% (8% to 67%)
`Y 35% (15% to 50%)
`Genitourinary infections,
`no excess DKA
`Rapid effects suggest a
`hemodynamic or metabolic
`benefit, although a vascular
`benefit may also occur
`
`Liraglutide (once-daily
`GLP-1 agonist)
`Age >50 yrs with CVD or >60 yrs
`with $1 CV risk factor
`HbA1c >7%
`
`3.8 yrs
`8.7%
`Y 13% (3% to 22%)
`Y 22% (7% to 34%)
`Y 12% (3% to 25%)
`Y 11% (11% to 28%)
`Y 13% (5% to 27%)
`More gallstones, GI side effects
`
`Semaglutide (once-weekly
`GLP-1 agonist)
`Age >50 yrs with CVD or >60 yrs
`with $1 CV risk factor
`HbA1c >7%
`
`2.05 yrs
`8.7%
`Y 26% (5% to 42%)
`Y 2% (48% to 35%)
`Y 26% (8% to 49%)
`Y 39% (1% to 72%)
`[11% (23% to 61%)
`Higher retinopathy rates
`
`Slower effects suggest benefits via
`less atherothrombosis and/or
`avoidance of hypoglycemia
`
`Slower effects suggest benefits via
`less atherothrombosis
`
`Y ¼ decrease; [ ¼ increase; CV ¼ cardiovascular; CVOT ¼ cardiovascular outcomes trials; DKA ¼ diabetic ketoacidosis; EMPA-REG OUTCOME ¼ Empagliflozin Cardiovascular
`Outcome Event Trial in Type 2 Diabetes Mellitus Patients; GI ¼ gastrointestinal; GLP-1 ¼ glucagon-like peptide-1; HbA1c ¼ glycosylated hemoglobin; HF ¼ heart failure;
`LEADER ¼ Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results; MI ¼ myocardial infarction; SGLT2 ¼ sodium-glucose cotransporter-2;
`SUSTAIN-6 ¼ Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes-6.
`
`because of recent data suggesting a higher risk for
`incident heart failure associated with use of sax-
`agliptin and alogliptin, recent regulatory warnings
`have been put in place for these 2 agents. Although
`meta-analyses suggest the risk for incident heart
`failure to be significant with this class of drug (rela-
`tive risk: 1.13; 95% CI: 1.01 to 1.26) (23), not all DPP-4
`inhibitors have been linked to heart failure risk; for
`example, recent data suggest no increased risk for
`incident heart failure related to sitagliptin use (24).
`
`INSULIN. Insulin is effective for glucose lowering and
`is very widely used for the treatment of advanced
`T2D. Therapy with insulin commonly leads
`to
`increased body weight and is associated with greater
`hypoglycemia risks. Thus, although insulin might
`improve glycemic control,
`its other effects may
`theoretically attenuate its clear glucose-lowering
`benefits in subgroups with particular susceptibility
`to hypoglycemia or the adverse effects of hypogly-
`cemia. There was also some expectation that
`exogenous insulin administration early in the course
`of T2D may have beneficial effects on CV outcomes;
`however,
`the results of
`the ORIGIN (Outcome
`Reduction with an Initial Glargine Intervention) trial
`failed to demonstrate any CV benefit (25).
`
`DIABETES DRUGS RECENTLY REPORTED TO
`REDUCE CV AND CV MORTALITY RISK
`
`Although numerous therapies for T2D have been
`associated with an increased risk of CV events, 3
`
`recent CVOTs have shown benefit in terms of hard
`clinical endpoints (Table 1) (2–4). We first review the
`results for the sodium-glucose cotransporter (SGLT) 2
`inhibitor, empagliflozin, before discussing results for
`2 GLP-1 receptor agonists.
`Of course, it should be noted that up until these
`recent trials, metformin was the only drug with
`possible evidence for CV benefit, albeit
`in very
`modest numbers of patients and with low event
`numbers. In the UKPDS (UK Prospective Diabetes
`Study), metformin-treated patients had a 30% lower
`risk for macrovascular disease than did patients not
`given metformin (26). Importantly, metformin does
`not cause weight gain or increased risk for hypogly-
`cemia, has many years of safety evidence, and is
`inexpensive; thus, it is widely used as a first-line
`therapy for the patient with CV disease.
`
`SGLT2 INHIBITORS. SGLT2 is a low-affinity, high-ca-
`pacity glucose transporter located in the proximal
`tubule of the nephron; SGLT2 is responsible for 90%
`of glucose reabsorption. Inhibition of SGLT2 results in
`decrease of blood glucose due to glycosuria. Second-
`ary effects of SGLT2 inhibition include a modest
`diuretic effect (sodium loss is also promoted), weight
`loss, and lowering of blood pressure.
`The only available CVOT for SGLT2 inhibitors
`recently reported reduction in CV events following
`treatment with empagliflozin compared with placebo.
`The EMPA-REG OUTCOME (Empagliflozin Cardiovas-
`cular Outcome Event Trial in Type 2 Diabetes Mellitus
`Patients)
`trial
`included
`7,020
`patients with
`
`Novo Nordisk Exhibit 2109
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00003
`
`

`

`J A C C V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`M A Y 3 0 , 2 0 1 7 : 2 6 4 6 – 5 6
`
`Sattar et al.
`The Cardiovascular Importance of New Diabetes Drugs
`
`2649
`
`F I G U R E 1 Comparing and Contrasting the Outcome Benefits in the EMPA-REG OUTCOME, LEADER, and SUSTAIN-6 Trials
`
`A
`
`%
`
`80
`
`60
`
`40
`
`20
`
`0
`
`-20
`
`-40
`
`-60
`
`-80
`
`B
`
`%
`
`80
`
`60
`
`40
`
`20
`
`0
`
`-20
`
`-40
`
`-60
`
`-80
`
`EMPA-REG
`
`LEADER
`
`SUSTAIN-6
`
`EMPA-REG
`
`LEADER
`
`SUSTAIN-6
`
`CV death
`
`HF hospitalization
`
`MI
`
`Stroke
`
`(A) Cardiovascular death and heart failure hospitalization. In the EMPA-REG OUTCOME trial, the most marked benefit was seen on reduction in CV death and heart
`failure hospitalization, with a more modest benefit in LEADER on cardiovascular death, and little evidence for such benefits in SUSTAIN-6. One must be careful to note
`that event numbers in the latter trial were small, in keeping with much larger confidence intervals. (B) Acute myocardial infarction and stroke. Greatest reductions in
`these events were seen in SUSTAIN-6 (although, once again, small event numbers suggest caution in interpretation), whereas such benefits were more modest in
`LEADER (although directionally similar). By contrast, such effects in EMPA-REG OUTCOME were mixed, although once again caution is needed, given overlapping
`confidence intervals. CV ¼ cardiovascular; EMPA-REG OUTCOME ¼ Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients; HF ¼ heart
`failure; LEADER ¼ Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results; MI ¼ myocardial infarction; SUSTAIN-6 ¼ Trial to Evaluate
`Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes-6.
`
`established CV disease, and randomized them to pla-
`cebo, empagliflozin 10 mg, or empagliflozin 25 mg. All
`study participants had established CV disease.
`The primary endpoint of EMPA-REG OUTCOME was
`3-point MACE (CV mortality, nonfatal MI, and
`nonfatal stroke). Patients randomized to empagli-
`flozin had a modest reduction in the primary endpoint
`(HR: 0.86; 95% CI: 0.74 to 0.99; p ¼ 0.04 for superi-
`ority; absolute risk reduction [ARR]: 1.6%). The
`reduction in the primary endpoint was driven pre-
`dominately by a substantial reduction in CV death
`(HR: 0.62; 95% CI: 0.49 to 0.77; p < 0.001; ARR: 2.2%),
`whereas nonfatal MI and stroke were not significantly
`altered; a 32% reduction in all-cause mortality was
`also observed (Figures 1A and 1B).
`Interestingly,
`benefit from empagliflozin in EMPA-REG OUTCOME
`was similar between the 2 doses tested. In recognition
`of the statistically robust effect on CV mortality, the
`FDA recently granted an indication to empagliflozin to
`reduce risk for CV death (27).
`Notably,
`in EMPA-REG OUTCOME, heart failure
`hospitalization was reduced by 35% (HR: 0.65; 95%
`CI: 0.50 to 0.85; p ¼ 0.002; ARR 1.4%), with a rapid
`separation in the survival curves suggesting acute
`benefit of the drug. The reduction in heart failure
`
`events was particularly clinically relevant, as drugs
`from other classes of glucose-lowering drugs with
`very different mechanisms of action (in particular,
`saxagliptin and rosiglitazone) had previously been
`found to be associated with an increase in hospitali-
`zations for heart failure (15,19).
`Although compelling, there are several reasons why
`heart failure outcome results should be interpreted
`cautiously. Although hospitalization for heart failure
`was a pre-specified outcome in EMPA-REG OUTCOME,
`it was not the primary outcome and did not have the
`rigor characteristic of heart failure trials. Patients
`could be recruited on the basis of investigator-reported
`heart failure, but there was no formal assessment of
`heart failure status, or cardiac structure or function at
`baseline; for example, no natriuretic peptide mea-
`surement or echocardiography was performed. No
`understanding regarding forms of heart failure (e.g.,
`preserved vs. reduced ejection fraction) was estab-
`lished. Furthermore, it is possible that some of the 76%
`of patients included on the basis of coronary artery
`disease at baseline (including 47% with prior MI) may
`have had unrecognized left ventricular dysfunction.
`In short, the finding of reduced hospitalization for
`heart
`failure is
`impressive, but
`further detail
`
`Novo Nordisk Exhibit 2109
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00004
`
`

`

`2650
`
`Sattar et al.
`The Cardiovascular Importance of New Diabetes Drugs
`
`J A C C V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`M A Y 3 0 , 2 0 1 7 : 2 6 4 6 – 5 6
`
`F I G U R E 2 Potential Pathway Linking Empagliflozin (and Possibly Other SGLT2 Inhibitors) With Lower Risks for Heart Failure
`Hospitalization and Death Due to CV Disease
`
`SGLT2 inhibition
`
`↓ Glucose and sodium
` reabsorption in
` proximal tubule
`↓ Nephron
` hyperfiltration
`
`Slow renal dysfunction
`
`↑ Urinary glucose
`& sodium
`
`Generalized
`decongestion
`
`↓ Cardiac afterload/pre-load
`↓ Systolic & diastolic dysfunction
`↓ Heart failure hospitalization
`↓ Fatal arrhythmias?
`
`By increasing fluid losses via urinary glucose and sodium losses, intravascular volumes and systolic blood pressure are reduced, and there is a
`modest weight loss. These changes in turn lessen cardiac stressors (pre-load and afterload) and may help improve myocardial oxygen supply.
`There may also be a benefit on vascular function. The net result is a likely improvement in cardiac systolic and diastolic function, lessening
`chances of pulmonary congestion, and thus lowering risks of hospitalization for heart failure and fatal arrhythmias. These cardiac function
`benefits could, in turn, feed back to improve renal function. Adapted with permission from Sattar et al. (33). CV ¼ cardiovascular; SGLT2 ¼
`sodium-glucose cotransporter-2.
`
`documenting the patient characteristics and bio-
`markers of heart failure is unavailable. It is possible
`that in some cases empagliflozin prevented the onset
`of clinical heart failure in those with unrecognized
`left ventricular dysfunction, but also that in some
`cases empagliflozin-treated patients already had un-
`recognized clinical heart
`failure. Mechanistic, or
`“bedside to bench,” studies are now trying to clarify
`the mechanistic relationship between empagliflozin
`and heart failure, while large outcome trials investi-
`gating the possible efficacy of SGLT2 inhibitors in
`treating heart
`failure with both preserved and
`reduced ejection fraction are also underway (28–30).
`Other benefits seen in EMPA-REG OUTCOME may
`help to clarify the effect of empagliflozin on CV
`outcomes. For example, empagliflozin also had a
`favorable effect on renal endpoints (31), with reduc-
`tion in incident or worsening nephropathy and
`incident albuminuria. Whether these beneficial renal
`effects are secondary to improved perfusion by car-
`diac or cardiovascular mechanisms or whether they
`are due to primary renal effects is unknown, although
`most consider
`renal benefits (thought
`to reflect
`reversal of maladaptive tubulo-glomerular
`renal
`feedback) to be largely upstream.
`The mechanism of benefit of empagliflozin is not
`fully known, but several are speculated (Figure 2). As
`noted, empagliflozin has numerous possibly benefi-
`cial CV effects including the hemodynamic effects of
`a diuretic agent; beneficial
`renal
`(reduction in
`
`intraglomerular pressure) (32), blood pressure, and
`weight effects; as well as many others, as recently
`reviewed (33,34). Most experts believe the rapid
`reduction in CV death and heart failure hospitaliza-
`tions seen in EMPA-REG OUTCOME is best explained
`by a rapid hemodynamic effect (34,35). Natriuresis, in
`combination with renal glucose losses, is thought to
`lead to a reduction in circulating volume and possibly
`extracellular fluid load, with a consequent lowering
`of cardiac filling and pre-load and afterload pressures.
`Supporting this concept was the rapid and sustained
`increase in hemoglobin and hematocrit demonstrated
`in EMPA-REG OUTCOME (2), as well as preliminary
`evidence for empagliflozin-induced improvements in
`left ventricular mass and diastolic function (36).
`In a more general sense, the data from EMPA-REG
`OUTCOME suggest that many patients with T2D and
`CV disease may have previously unrecognized
`excessive fluid overload, often in association with
`cardiac dysfunction, and that these patients benefit
`rapidly from intravascular decongestion. Some have
`suggested that less left ventricular stretch, arising
`from corrections in intravascular fluid load, might
`also decrease the incidence of atrial and ventricular
`arrhythmias. Another potential mechanism of benefit
`is that patients randomized to empagliflozin were less
`likely to receive other glucose-lowering therapies
`(e.g., insulin and sulfonylureas), drugs that increase
`weight and hypoglycemia risks. Possibly, avoidance
`of these therapies in the treatment arm could have
`
`Novo Nordisk Exhibit 2109
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00005
`
`

`

`J A C C V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`M A Y 3 0 , 2 0 1 7 : 2 6 4 6 – 5 6
`
`Sattar et al.
`The Cardiovascular Importance of New Diabetes Drugs
`
`2651
`
`contributed to the positive outcome. A further pro-
`posed mechanism of benefit of empagliflozin, the
`ketone hypothesis, has been proposed, whereby
`slightly increased ketones with SGLT2 inhibitors
`serve as a better fuel supply for the failing heart (37).
`It is important to understand the potential side
`effects of SGLT2 inhibitors. The most notable adverse
`effect in EMPA-REG OUTCOME was an absolute 4.6%
`increase in genital infections; a greater incidence was
`noted in women. Fortunately, these infections are not
`generally serious, and resolve with a course of anti-
`fungal agents. Once treated, they uncommonly recur.
`From the perspective of a cardiologist, patients
`should be informed of this risk, and shared care with
`primary care physicians (who manage these condi-
`tions on a regular basis) is recommended. It would
`not be prudent to use SGLT2 inhibitors in women or
`men with a history of recurrent genital infections. In
`EMPA-REG OUTCOME,
`there was no increase in
`urinary tract infections, hypoglycemic episodes, or
`diabetic ketoacidosis. Some concern does remain as
`to whether or not SGLT2 inhibitors can increase the
`risk of diabetic ketoacidosis outside of the tightly
`monitored environment of a clinical trial, particularly
`in patients with T2D treated with insulin. It should be
`noted that cases of ketoacidosis have been reported
`in the off-label use of SGLT2 inhibitors in patients
`with type 1 diabetes (38). Patients given these agents
`should be educated about simple warning signs and
`symptoms of potential diabetic ketoacidosis.
`It is worthwhile to emphasize that EMPA-REG
`OUTCOME was not a primary prevention trial.
`Although tempting to speculate, it is impossible to
`conclude that similar benefits would be seen in
`patients without CV disease. This makes the results of
`EMPA-REG OUTCOME all the more important to the
`practicing cardiovascular specialist, given the high
`prevalence of T2D in those with established CV disease
`(39). Several other similar safety trials are being con-
`ducted with SGLT2 inhibitors with slightly differing
`pharmacology. However, these trials also differ in size
`and patient composition. For example, 60% of partic-
`ipants in DECLARE-TIMI 58 (Multicenter Trial to
`Evaluate the Effect of Dapagliflozin on the Incidence of
`Cardiovascular Events-Thrombolysis In Myocardial
`Infarction 58) (NCT01730534) do not have prior CV
`disease, which is important. These trials will report
`results over the next few years.
`
`GLP-1 RECEPTOR AGONISTS. Following on from
`empagliflozin, 2 other drugs, both from the GLP-1
`receptor
`agonist
`family, have been shown to
`improve CV outcomes, albeit in a different pattern
`than EMPA-REG OUTCOME (Figures 1A and B).
`
`Liraglutide is a once-daily injectable GLP-1 receptor
`agonist. It is also associated with weight loss and
`blood pressure lowering.
`In the recent LEADER
`(Liraglutide Effect and Action in Diabetes: Evaluation
`of Cardiovascular Outcome Results)
`trial, 9,340
`patients with a glycosylated hemoglobin (HbA1c)
`of >7.0% (either >50 years of age with established CV
`disease or >60 years of age with 1 or more CV risk
`factors) were randomized to liraglutide or placebo (3).
`The primary endpoint of MACE was reduced by 13%
`(HR: 0.87; 95% CI: 0.78 to 0.97; ARR 1.9%;
`p for superiority ¼ 0.01). The components of the pri-
`mary endpoint were all numerically in favor of lir-
`aglutide, but only CV mortality was statistically
`significantly reduced (HR: 0.78; 95% CI: 0.66 to 0.93;
`p ¼ 0.007; ARR 1.3%); all-cause mortality was also
`reduced (HR: 0.85; 95% CI: 0.74 to 0.97). Subgroup
`analysis did suggest a greater benefit in those with
`established CV disease, rather than those with risk
`factors in the absence of clinically-evident disease.
`Nephropathy events were less common with liraglu-
`tide (1.5 vs. 1.9 per 100 patient-years) but, in contrast
`to the widespread renal benefits with empagliflozin,
`liraglutide-driven renal benefits were driven largely
`by a reduction in new-onset persistent macro-
`albuminuria, with little discernible effects on other
`renal outcomes.
`Subsequently, SUSTAIN (Trial to Evaluate Cardio-
`vascular
`and Other Long-term Outcomes with
`Semaglutide in Subjects with Type 2 Diabetes)-6
`investigated the safety of the once-weekly GLP-1
`receptor agonist, semaglutide. In this phase 3, ran-
`domized, placebo-controlled noninferiority trial,
`3,297 study participants were treated with semaglu-
`tide or placebo. The inclusion criteria were very
`similar to those used in the LEADER trial, as was the
`primary endpoint of 3-point MACE. Treatment with
`semaglutide reduced the primary endpoint by 2.3%
`(HR: 0.74; 95% CI: 0.58 to 0.95; p < 0.001 for non-
`inferiority; p ¼ 0.01 for superiority). The contribution
`of the components of MACE to the reduction in the
`primary endpoint was somewhat different to that
`seen in the LEADER trial, in that CV mortality was not
`affected by semaglutide, but nonfatal stroke was
`improved (ARR: 1.1%; HR: 0.61; 95% CI: 0.38 to 0.99)
`along with a nonsignificant trend toward lower rates
`of incident MI (Figures 1A and 1B). It is important to
`draw an important distinction between the LEADER
`and SUSTAIN-6 trials, as the latter study had a much
`smaller noninferiority design, increasing risk for type
`1 and type 2 error due to underpowering.
`It
`is not yet understood how GLP-1 receptor
`agonists may reduce CV events. Compared with trials
`of SGLT2 inhibitors, the relative benefit of GLP-1
`
`Novo Nordisk Exhibit 2109
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00006
`
`

`

`2652
`
`Sattar et al.
`The Cardiovascular Importance of New Diabetes Drugs
`
`J A C C V O L . 6 9 , N O . 2 1 , 2 0 1 7
`
`M A Y 3 0 , 2 0 1 7 : 2 6 4 6 – 5 6
`
`agonists appeared at a later time following randomi-
`zation,
`in line with atherothrombotic, rather than
`hemodynamic effects. However, both MI and stroke
`were numerically, but not statistically, lower with
`liraglutide than with placebo in the LEADER trial. If
`this was the predominant mechanism of action
`explaining LEADER results, a more convincing
`reduction in MI or stroke might have been antici-
`pated. Other possible mechanisms are blood pressure
`reduction, lessening of arterial stiffness (in keeping
`with lower systolic blood pressure, but higher dia-
`stolic blood pressure, so narrowing of pulse pressure),
`weight loss, and beneficial renal effects. The reduc-
`tion in blood glucose was more pronounced in the
`early years in the LEADER trial than in other recent
`diabetes CVOTs, so one cannot rule out that glucose
`lowering contributed to its beneficial effects; patients
`randomized to liraglutide were less likely to be
`exposed to insulin or
`sulfonylureas, prompting
`speculation that preventing exposure to these
`potentially harmful (or less “net” beneficial) drugs in
`such patients may also be a contributory mechanism.
`As well, severe hypoglycemia was significantly lower
`in patients randomized to liraglutide; emerging
`evidence indicates that hypoglycemia may be more
`harmful in patients with existing CV disease, so less
`hypoglycemia could help explain the CV mortality
`reduction in the LEADER trial. Other direct vascular
`or cardiac effects of GLP-1 receptor agonists have
`been proposed and could have contributed to the CV
`benefits seen (40).
`caveats
`the
`and accepting
`Taken together
`the somewhat
`mentioned in the previous text,
`different pattern of CV benefits in the LEADER versus
`SUSTAIN-6 trials (Figures 1A and 1B) suggests that a
`class effect of GLP-1 receptor agonists on CV outcomes
`cannot be assumed; rather, benefits and potential
`harms may differ between different GLP-1 receptor
`agonists. The results of EXSCEL (Exenatide Study of
`Cardiovascular Event Lowering Trial), testing once-
`weekly exenatide, are eagerly awaited (41). It should
`also be noted that the ELIXA (Evaluation of Lixisena-
`tide in Acute Coronary Syndrome) CVOT did not
`reduce MACE in T2D patients following acute coronary
`syndrome (42), but whether these results were due to
`the short-acting nature of lixisenatide compar