CARMELINA: Linagliptin Reassures for DPP-4 Inhibitors in Diabetes

Liam Davenport

October 08, 2018

BERLIN — Adding the dipeptidyl peptidase-4 (DPP-4) inhibitor linagliptin (Tradjenta, Lilly/Boehringer Ingelheim) to standard of care in patients with type 2 diabetes at high cardiovascular (CV) risk has no impact on CV, heart failure, or renal events, even in those who already have kidney disease, reveal data from the CARMELINA postmarketing outcomes study.

The trial, which was presented here at the European Association for the Study of Diabetes (EASD) 2018 Annual Meeting, involved almost 7000 patients with type 2 diabetes and CV risk factors from 27 countries, three quarters of whom also had chronic kidney disease (CKD). The patients were randomly assigned to linagliptin or placebo on top of standard care and were treated for 2 years.

The primary outcome of three-point major cardiovascular events (MACE), comprising CV death, nonfatal myocardial infarction (MI), and nonfatal stroke, did not significantly differ between linagliptin and placebo.

And following concerns over the potential for DPP-4 inhibitors to increase hospitalizations for heart failure — as seen first of all with saxagliptin (Onglyza, Bristol-Myers Squibb/AstraZeneca) in the SAVOR-TIMI 53 study — linagliptin showed no effect on rates of this event. In addition, rates of a prespecified combined renal endpoint did not differ between the study drug and placebo.

There was nevertheless a significant reduction in microvascular events with linagliptin over placebo, primarily driven by reductions in albuminuria progression. The study did not identify any new safety concerns.

CARMELINA (Cardiovascular and Renal Microvascular Outcome Study With Linagliptin in Patients With Type 2 Diabetes Mellitus) is one in a long line of CV outcome trials (CVOTs) mandated by the US Food and Drug Administration in 2008 to rule out cardiovascular toxicity in the wake of data that linked rosiglitazone (Avandia, GlaxoSmithKline) to an increased risk for MI.

CARMELINA Evens the Field for DPP-4 Inhibitors

Within the DPP-4 inhibitor class, there was dismay when SAVOR-TIMI 53 was reported in 2013. While it showed that saxagliptin was no different from placebo for the composite endpoint of CV death, nonfatal MI, or nonfatal ischemic stroke when added to usual care in patients with T2D, it did increase the risk for heart failure hospitalization, a finding that came out of the blue at the time.

In the EXAMINE trial, reported at the same time, another DPP-4 inhibitor, alogliptin (Nesina, Takeda Pharmaceuticals) was also associated with a trend toward an increased risk for heart failure events. Again, however, the drug did not increase the risk for the combined endpoint of CV events.

These two trials raised a red flag for the DPP-4 inhibitor class in terms of heart failure.

Finally some salvation came for the class came in the form of the TECOS trial of sitagliptin (Januvia, Merck), which demonstrated that the drug did not increase the risk for CV events and was not associated with an increased risk for heart failure.

CARMELINA co-presenter Bernard Zinman, MD, from the University of Toronto and the Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada, said during the session at EASD that the DPP-4 inhibitor studies have shown that there is "clearly no signal for benefit or harm with respect to 3-point MACE."

"However, when we look at hospitalization for heart failure, there's a 27% increase in the SAVOR study, and there is perhaps a signal with EXAMINE," he noted.

"TECOS is right on the line, and now you have CARMELINA to the left of the line of unity: clearly no signal of an increased risk of hospitalization or heart failure."

He also emphasized that "no safety findings for linagliptin were identified" in CARMELINA, which is "reassuring" in this high-risk patient population, although there was a "numerical imbalance" in favor of placebo for acute pancreatitis and pancreatic cancer.

"Total cancer cases were similar across treatment arms, and there's no increased risk of hypoglycemia, including patients most susceptible for hypoglycemia," he added.

"One can conclude that, in this patient population, linagliptin demonstrated cardiovascular safety for atherosclerotic cardiovascular events and a neutral effect on hospitalization for heart failure and kidney outcomes."

CVOTs  Relevant to Only Minority With T2D; Are They a Waste of Money?

Commenting on the findings, Philip D. Home, MD, PhD, professor of diabetes medicine, Newcastle University, United Kingdom, who was not involved in the study, said that CARMELINA was "competently performed" and to the highest standards, but "it's a bit sad perhaps that we have a high discontinuation rate."

He added: "It certainly establishes the safety of linagliptin in a very high-risk population, at least for CV events in people with diabetes in what I call the medium term, because, after all, we look after people for 10, 20 years, and of course its useful sensitivity and subgroup analyses are consistent."

Home also said that the results are "consistent with all the other major studies in more typical populations," although he noted that CARMELINA has a lower power than other studies and the number of safety events recorded was lower.

He noted studies such as these in high-risk populations "do tend to have higher event rates," adding that, in common with the HARMONY Outcomes study of the glucagon-like peptide-1 agonist albiglutide (Tanzeum, GlaxoSmithKline) also presented at EASD, "we got used recently to event rates of over 3%."

"We were told last year that these were applicable to about 16% of our regular common people with diabetes, so there is a concern with the studies already about what the percent of the population we follow in diabetes services they are applicable to."

"I think it's more than 16% myself, but it still means the majority…are not being covered by these studies."

Referring to the inclusion of both patients with CV disease and CKD in CARMELINA, he asked, "Does it make sense to perform one study on two largely distinct populations?"

"The reason I ask this is how do I translate that back to people and back into diabetes care?" Home explained. "We have populations here of 24% with cardiovascular disease and albuminuria only, 41% with prevalent CKD population, and 33% in between."

"But you'll notice you've not been presented the data for these three different populations."

He also pointed out that CARMELINA was different from the other DPP-4 inhibitor studies and the sodium glucose cotransporter 2 inhibitor trials EMPA-REG, involving empagliflozin (Jardiance, Boehringer Ingelheim Pharmaceuticals), and CANVAS, which looked at canagliflozin (Invokana, Janssen Pharmaceuticals), because it had a far higher CV event rate.

This was driven largely by CV death, and Home noted that it is not uncommon for "a whole load people who have uncertain causes of death being allocated to the cardiovascular area."

He also took issue with the conclusion by the CARMELINA researchers that microvascular events significantly differed between linagliptin and placebo. "I do not accept this…. It was driven by albuminuria progression and surely albuminuria progression is not a microvascular outcome in type 2 diabetes."

"Albuminuria and microalbuminuria are markers of vascular inflammation, and we do see them elevated in people without diabetes."

Looking more widely at the recent slew of trials assessing safety data in new T2D drugs, Home said, "We've now had a large number of these outcomes studies, the CVOTs, and none of them have shown safety concerns from MACE, which was the original rationale — as it turns out wrong — when we analyzed the rosiglitazone results."

"We've spent about 3 to 5 billion dollars on studies, to really no particular good effect."

While the results of several further CVOTs are awaited — including for DECLARE-TIMI58, due to be reported at the American Heart Association meeting in Chicago in November; REWIND; CREDENCE; and CAROLINA — he said, "I hope the paradigm will be abandoned, but there is a need for long-term studies, longer than this one, 5 to 10 years, so we actually collect good longitudinal safety data, and find out about things like fractures, amputations, and other things which are recurrent with our drugs."

CARMELINA: "Wanted to Look at This Population," Diabetes, CVD, and CKD

Investigator Julio Rosenstock, MD, Dallas Diabetes and Endocrine Center at Medical City, Texas, explained that the reason CARMELINA had a relatively high proportion of patients with T2D who had CKD was not because it was a regulatory requirement but because the steering committee wanted to look at outcomes in this population.

"The reason to do that is we know that cardiovascular risk is increased in people with type 2 diabetes complicated with chronic kidney disease," he said, adding that "people with type 2 diabetes have a two to three times increased risk cardiovascular risk, and that risk is then doubled [further] in people with chronic kidney disease."

"Despite that, patients with type 2 diabetes and kidney disease have been under-represented in most of the cardiovascular outcome trials."

Rosenstock highlighted that both a low estimated glomerular filtration rate (eGFR) and microalbuminuria are associated with an increased risk for CV events, and CARMELINA therefore included a predefined secondary composite kidney outcome of death from kidney disease, end-stage renal disease, or a sustained eGFR decrease of 40% or more.

He pointed out that linagliptin is unusual in that it does not require dose adjustment for kidney function because it is excreted primarily via the gut/bile, and its high-affinity DPP-4 binding means it has minimal kidney excretion.

Next, Robert Toto, MD, professor of medicine at University of Texas Southwestern Medical Center in Dallas, explained that CARMELINA involved 6991 patients with T2D from 605 sites in 27 countries.

The patients also had either macrovascular disease, such as a history of MI infarction or advanced coronary artery disease, and albuminuria, defined as a urinary albumin-to-creatinine ratio (UACR) of 30 mg/g or less, or they had impaired kidney function with or without albuminuria. Impaired kidney function was defined as an eGFR of 15 to less than 45 mL/min per 1.73 m2 or an eGFR of at least 45 to 75 mL/min per 1.73 m2 and a UACR greater than 200.

The patients were randomly assigned to linagliptin 5 mg/day (n = 3494) or placebo (n = 3485) in addition to local standard of care.

All CV and neurologic events, all aspects of the composite kidney endpoint, and diagnoses of heart failure and pancreatitis were adjudicated independently by the clinical event committee.

Presenting the baseline characteristics, Steve E. Kahn, MD, VA Puget Sound Health Care System, University of Washington, Seattle, said that the mean patient age was 66 years, just over half of patients were men, and around 17% of participants were at least 75 years of age. The mean duration of diabetes was 15 years, and the mean hemoglobin A1c level was approximately 8.0%. Almost all (97%) patients were taking at least one glucose-lowering drug at study entry, with just over half taking metformin and the same proportion taking insulin.

A majority of patients (about 70%) received statins, and up to 60% of patients in both groups received a range of antihypertensive drugs .

Just 19.9% of patients had normoalbuminuria, with 41.5% having microalbuminuria and 38.5% macroalbuminuria. The majority (62.3%) of patients had stage 3 to 5 CKD.

The prevalence of CKD overall was calculated to be 74%; 57% of patients had established CV disease, and 33% had both conditions.

The patients were treated for a median of 1.9 years and were included in the study for a median of 2.2 years. In the linagliptin group, 23.9% of the patients discontinued the study drug prematurely, as did 27.4% of the placebo group.

Hemoglobin A1c levels were reduced by a small but significant degree with linagliptin vs placebo, at a mean overall reduction of 0.36% (P < .0001).

Linagliptin was also associated with a significant reduction in the introduction of new glucose-lowering therapies after baseline, at a hazard ratio vs placebo of 0.76 (P < .0001), a reduction that was similar for both insulin and noninsulin therapies.

The active drug was also associated with a significant reduction in the risk of initiating or increasing the dose of insulin during follow-up vs placebo, at a hazard ratio of 0.72 (P < .0001).

There were no significant differences in body weight or blood pressure during follow-up, and no significant changes over time, within or between the two study groups. Lipid values also remained largely unchanged, with no significant differences.

No Sign of Heart Failure, Reduction in Albuminuria, No Safety Concerns

Next, Darren K. McGuire, MD, a cardiologist from UT Southwestern Medical Center, Dallas, presented the CV, mortality, and heart failure outcomes.

During follow-up, there were 854 three-point MACE events, of which 33.5% were nonfatal MI, 14.3% were nonfatal stroke, and 52.2% were CV death.

There were no significant differences between the linagliptin and placebo groups in the time to the first occurrence of three-point MACE, at a hazard ratio of 1.02 (P = .0002 for noninferiority; P = .7398 for superiority).

This held true regardless of the type of sensitivity analysis used, when the patients were stratified by baseline characteristics, and when the components of MACE were considered separately.

The linagliptin and placebo groups also did not significantly differ in the occurrence of three-point MACE plus unstable angina hospitalization, all-cause mortality, CV mortality, and non-CV mortality.

And importantly in relation to previous CVOTs with DPP-4 inhibitors, there was no significant difference in the time to first occurrence of adjudicator-confirmed hospitalization for heart failure, at a hazard ratio of 0.90 (95% confidence interval, 0.74 -1.08; P = .2635 for superiority).

Turning to the kidney and microvascular outcomes, Vlado Perkovic, MD, PhD, executive director of the George Institute for Global Health, Sydney Australia, said there was a significant reduction in albuminuria progression with linagliptin vs placebo, at a hazard ratio of 0.86 (P = .0034).

Further analysis showed that this result was driven by a significant reduction in progression to macroalbuminuria in patients who had microalbuminuria at baseline, at a hazard ratio of 0.84 (P = .0067).

There was, however, no significant difference in occurrence of the composite kidney endpoint with linagliptin vs placebo, which remained the case when the patients were stratified by baseline characteristics.

Linagliptin was associated with a significant reduction in a composite microvascular outcome, at a hazard ratio of 0.86 (P = .0032).

Perkovic acknowledged that that was driven entirely by a reduction in albuminuria progression, which accounted for most events in the composite endpoint in both treatment groups.

Finally, Mark E. Cooper, MD, PhD, head of the Department of Diabetes, Monash University, Melbourne, Australia, presented the adverse events data from the trial. He reported similar event rates in the linagliptin and placebo groups, at 101.01 per 100 patient-years and 106.88 per 100 patient-years, respectively. That pattern was repeated for individual safety events of particular interest, including hypersensitivity reactions, hepatobiliary events, arthralgia-related events, and renal adverse events.

The adjudicated event rate for acute pancreatitis was higher with linagliptin than with placebo, at 0.12 per 100 patient-years and 0.06 per 100 patient-years, respectively, as it was for pancreatic cancer, at 0.14 per 100 patient-years and 0.05 per 100 patient-years.

Hypoglycemia events did not significantly differ between the linagliptin and placebo groups, both in the overall analysis and in insulin-treated, sulfonylurea-treated, and kidney disease subgroups.

The study was funded by Boehringer Ingelheim and Eli Lilly. Rosenstock has served on scientific advisory boards and received honoraria or consulting fees from AstraZeneca, Boehringer Ingelheim, Daiichi Sankyo, Eli Lilly, Intarcia Therapeutics, Janssen, Lexicon, Merck, Novo Nordisk, and Sanofi. He has also received grants/research support from Asahi, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Eli Lilly, GlaxoSmithKline, Hanmi Pharmaceutical, Intarcia Therapeutics, Janssen, Lexicon, MannKind, Merck, Novartis, Novo Nordisk, Pfizer, Sanofi, and Takeda Pharmaceuticals. Toto is a consultant to Amgen, Boehringer Ingelheim, ZS Pharma, Relypsa, Novo Nordisk, Reate, AstraZeneca, and receives grant support from the National Institutes of Health.

Kahn is a paid consultant on advisory boards for Novo Nordisk. Buse receives research support from, owns stock in, and/or is an advisor for Adocia, American Diabetes Association, AstraZeneca, Dexcom, Elcelyx, Eli Lilly, Fractyl, Intarcia, Lexicon, Metavention, National Institute of Diabetes and Digestive and Kidney Diseases, National Institute of Environmental Health Sciences, NovaTarg, Novo Nordisk, Sanofi, Shenzhen Hightide Biopharmaceutical, VTV Therapeutics, Boehringer Ingelheim, Johnson & Johnson, National Center for Advancing Translational Sciences, National Heart, Lung, and Blood Institute, Patient-Centered Outcomes Research Institute, and Theracos. McGuire has served as clinical trial leadership for AstraZeneca, sanofi-aventis, Ortho-McNeil-Janssen, Boehringer Ingelheim, Merck, Novo Nordisk, Lexicon Pharmaceuticals, Eisai, GlaxoSmithKline, and Esperion Therapeutics. He has served as a consultant for AstraZeneca, sanofi-aventis, Eli Lilly, Boehringer Ingelheim, Merck, Pfizer, and Novo Nordisk.

Perkovic reports personal fees from Retrophin, Janssen, Merck, and Servier; has served on steering committees for trials funded by AbbVie, Boehringer Ingelheim, GlaxoSmithKline, Janssen, and Pfizer; and has participated in scientific presentations or served on advisory boards with AbbVie, Astellas, Astra Zeneca, Bayer, Baxter, Bristol-Myers Squibb, Boehringer Ingelheim, Durect, Eli Lilly, Gilead, GlaxoSmithKline, Novartis, Novo Nordisk, Pfizer, Pharmalink, Relypsa, Sanofi, and Vitae, with fees paid to his institution.

Cooper reports fees for advisory services from Boehringer Ingelheim. Zinman has received grant support from Boehringer Ingelheim, AstraZeneca, and Novo Nordisk and consulting fees from AstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Novo Nordisk, and sanofi aventis. Home declares that he or his institution has received funding for lecturing, advisory, and/or research activities from all manufacturers of DPP-4 inhibitors, including Boehringer Ingelheim, and from manufacturers of competitor products.

European Association for the Study of Diabetes (EASD) 2018 Annual Meeting. Presented October 4, 2018.

For more diabetes and endocrinology news, please follow us on Twitter and on Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.