WASHINGTON – The remarkable and unexpected findings first reported from the EMPA-REG OUTCOME trial in late 2015 – that treatment of type 2 diabetes patients with the SGLT-2 inhibitor empagliflozin led to significantly reduced rates of heart failure hospitalization and all-cause death – received its first major confirmation in an analysis of observational data collected from more than 300,000 patients with type 2 diabetes treated in six countries including the United States.

The new findings also, for the first time, extended the EMPA-REG OUTCOME results (N Engl J Med. 2015 Nov 26;373[22]:2117-28) beyond empagliflozin with evidence that the heart failure and mortality benefit also occurred with other drugs from the sodium glucose cotransporter–2 (SGLT-2) inhibitor class, specifically canagliflozin and dapagliflozin, Mikhail Kosiborod, MD, said at the annual meeting of the American College of Cardiology.

The analysis showed that type 2 diabetes patients who were newly started on treatment with one of these SGLT-2 inhibitors had during follow-up a 39% reduced rate of heart failure hospitalizations, a 51% reduced mortality rate, and a 46% reduced rate of the combined endpoint of heart failure hospitalization or death, compared with patients treated with any other type of oral glucose-lowering drug, reported Dr. Kosiborod. The risk reductions were “remarkably similar to those seen in EMPA-REG OUTCOME,” he noted.

The findings address three “key questions” raised by the EMPA-REG OUTCOME results, Dr. Kosiborod, a cardiologist and professor of medicine at Saint Luke’s Mid America Heart Institute in Kansas City, Mo., said in an interview:

• Is this a class effect? The data “seem to suggest that the benefits seen in the EMPA-REG OUTCOME study are likely a class effect.” The study population of 154,523 patients in the heart failure hospitalization analysis who began treatment with an SGLT-2 inhibitor included 53% who received canagliflozin (Invokana), 42% who received dapagliflozin (Farxiga), and 6% who received empagliflozin (Jardiance) (percentages total 101% because of rounding). Dr. Kosiborod also highlighted that within several of the six countries that contributed data to this analysis – the United States, Denmark, Germany, Norway, Sweden, and the United Kingdom – the percentages of patients on each of these three drugs varied substantially, but despite that the relative reduced risks for heart failure hospitalization and mortality were roughly the same within each country, giving further credence to the notion that a class effect exists.

• Do lower-risk patients benefit? “The benefits of SGLT-2 inhibitor treatment appeared to extend to lower risk patients” than those enrolled in EMPA-REG OUTCOME. In the randomized trial, which enrolled 7,028 patients, more than 99% had established cardiovascular disease. In the new analysis patients had a 13% prevalence of any cardiovascular disease at baseline, and the prevalence of heart failure was 3%.

• Is this relevant to clinical practice? Unlike the highly selected patients entered in the EMPA-REG OUTCOME trial, the patients started on an SGLT-2 inhibitor in the observational study were unselected and came from routine practice situations, “suggesting that the benefits seen in EMPA-REG OUTCOME translate into real-world clinical practice,” Dr. Kosiborod said. “With these data we see for the first time in a large number of patients from multiple countries important evidence suggesting that the SGLT-2 inhibitors may provide in the real world a similar benefit to what was observed in EMPA-REG OUTCOME.”

“The lesson from Dr. Kosiborod’s study is that among patients with type 2 diabetes, treatment with an SGLT-2 inhibitor seems to result in lowered rates of heart failure hospitalizations and mortality, and it’s a safe class of drugs. In the past, we worried about worsening heart failure in patients at risk for developing heart failure” such as patients with type 2 diabetes, said Adrian F. Hernandez, MD, professor of medicine and a cardiologist at Duke University in Durham, N.C. The new data make it seem like using an SGLT-2 inhibitor to treat patients with type 2 diabetes “is an appropriate strategy.” But Dr. Hernandez added that in his opinion metformin remains the top drug for type 2 diabetes, while SGLT-2 inhibitors are now “the next drug class to add,” he said in an interview.

Dr. Kosiborod had a somewhat different take. “If a patient with type 2 diabetes did not want to enter a trial or couldn’t get into a trial and fit the profile of a patient who could benefit, I would absolutely treat that patient with an SGLT-2 inhibitor. I’m using these medications clinically as a cardiologist,” he said. “Treatments that have significant benefits for important outcomes should be prioritized over treatments that may reduce hemoglobin A_1c but do not have similar benefits.”

The CVD-REAL (Comparative Effectiveness of Cardiovascular Outcomes in New Users of SGLT-2 Inhibitors) study used data from adult, previously untreated patients with type 2 diabetes in national registries from the five included European countries. U.S. data were from the Truven Health MarketScan database and from Medicare. This produced a total pool of 160,010 patients who began treatment on an SGLT-2 inhibitor and 1,139,905 patients who began treatment with another oral antidiabetes drug.

Dr. Kosiborod and his associates then performed propensity-score matching to identify 154,523 patients started on an SGLT-2 inhibitor who each closely matched a patient from the other subgroup for baseline demographic and clinical features, producing an analysis dataset of just over 309,000 matched patients. The average age of the included patients was 57 years; 45% were women. During follow-up, 961 patients had a heart failure hospitalization and 1,334 patients died.

Dr. Kosiborod noted that while a potential limitation to his findings is residual confounding not eliminated by the propensity score matching, he was confident about the results because the incidence of other outcomes not expected to be influenced by treatment with SGLT-2 inhibitors were similar in the two study subgroups, suggesting that the linkages between the kind of drug used and differences in heart failure hospitalization rates and in mortality weren’t spurious.

“If there was residual confounding, you’d expect to see similar associations for other endpoints, which we didn’t see,” he said. In addition, the heart failure hospitalization rate differences seen between the SGLT-2 inhibitor recipients and the other patients were consistent in a trio of sensitivity analyses, further buttressing the findings’ plausibility.

CVD-REAL was sponsored by AstraZeneca, the company that markets dapagliflozin (Farxiga). Dr. Kosiborod has been a consultant to and/or received research funding from AstraZeneca, Boehringer Ingelheim, Sanofi-Aventis and Gilead. Dr. Hernandez has received honoraria and/or research support from AstraZeneca, Amgen, Janssen, Merck, Novartis, and Portola. Several of the coauthors on the CVD-REAL study were AstraZeneca employees.

mzoler@frontlinemedcom.com

On Twitter @mitchelzoler

Gliflozins’ heart failure protection in type 2 diabetes confirmed