American Diabetes Association Scientific Sessions

American Diabetes Association Scientific Sessions

Issue: August 2018
June 24, 2018
4 min read

OBSERVE-4D: No amputation risk with canagliflozin in type 2 diabetes

Issue: August 2018
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

John Buse

ORLANDO, Fla. — Adults with type 2 diabetes with and without established cardiovascular disease treated with the SGLT2 inhibitor canagliflozin did not have an increased risk for below-the-knee amputation compared with patients assigned similar SGLT2 inhibitors or other antidiabetes therapies, according to findings from the OBSERVE-4D study.

In a real-world analysis of more than 700,000 patients with type 2 diabetes, researchers also observed a substantial reduction in risk for hospitalization for heart failure — including patients with established CVD — a benefit noted in other SGLT2 inhibitor trials that suggests a possible class effect.

Data that suggested a possible increased risk for below-the-knee amputation with canagliflozin (Invokana, Janssen), first revealed in the CANVAS trial presented the 2017 American Diabetes Association Scientific Sessions, have been the “fly in the ointment in an otherwise very attractive set of results,” John Buse, MD, PhD, director of the Diabetes Center, director of the North Carolina Translational and Clinical Sciences Institute and executive associate dean for clinical research at the University of North Carolina School of Medicine in Chapel Hill, said during a press conference.

As Endocrine Today previously reported, findings from the integrated CANVAS program demonstrated that patients with type 2 diabetes at high risk for CVD assigned canagliflozin had a 33% reduction in risk for hospitalization for HF and were 40% less likely to experience renal decline vs. patients assigned placebo. However, researchers also noted that participants assigned canagliflozin were twice as likely to experience an amputation, primarily at the toe or metatarsal, compared with placebo treatment. In 2017, the FDA required the addition of a Boxed Warning to the canagliflozin label describing the increased amputation risk.

The findings from OBSERVE-4D, a retrospective, real-world, observational study, suggest the amputation risk previously observed with canagliflozin is similar to risk with other SGLT2 inhibitors and other classes of antidiabetes medications. The benefits of the agent outweigh any risks for most patients, Buse said.

“The SGLT2 inhibitor class is exceptionally promising,” Buse said. “Probably 60% to 70% of patients with type 2 diabetes will die of either cardiovascular disease, heart failure or kidney failure, and this class of drugs has been shown to reduce those outcomes substantially. On the other hand, only about 5 to 10% of patients in the United States with type 2 diabetes have availed themselves of the potential benefits of SGLT2 inhibitors.”

Head-to-head comparison

In OBSERVE-4D, Buse and colleagues analyzed patient-level data from four U.S. administrative claims databases: Truven MarketScan Commercial Claims and Encounters, Multi-state Medicaid, Medicare Supplemental Beneficiaries and OptumInsight’s Clinformatics Datamart. Data included 142,000 new users of canagliflozin, 110,000 users of other SGLT2 inhibitors (dapagliflozin [Farxiga, AstraZeneca] and empagliflozin [Jardiance, Boehringer Ingelheim/Lilly]) and 460,000 users of non-SGLT2 antihyperglycemic agents, excluding metformin. The researchers compared risk for both hospitalization for HF and below-the-knee amputation for canagliflozin vs. non-SGLT2 therapies and for canagliflozin vs. other agents in the class, in both the overall population and among a subgroup with established CVD. Researchers used propensity score adjustment to reduce confounding.

“Up until now, no real-world studies have looked at a head-to-head comparison of agents in the class on heart failure or amputation,” Buse said.

Amputation, HF risk

Researchers observed no evidence of increased risk of below-the-knee amputation with canagliflozin vs. dapagliflozin and empagliflozin in both the overall population (HR = 1.14; 95% CI, 0.67-1.93) and among patients with established CVD (HR = 1.08; 95% CI, 0.63-1.82). Similar results were observed when comparing canagliflozin with all non-SGLT2 therapies, with no increased risk in the overall type 2 diabetes population (HR = 0.75; 95% CI, 0.4-1.41) or among those with established CVD (HR = 0.72’ 95% CI, 0.34-1.51), Buse said.

Additionally, new users of canagliflozin experienced a 61% reduction in risk for hospitalization for HF vs. new users of non-SGLT2 therapies in the overall population (HR = 0.39; 95% CI, 0.26-0.6) and a 56% reduction in risk among those with established CVD (HR = 0.44; 95% CI, 0.36-0.54). Those findings reflect similar results observed across the SGLT2 class in other trials and were reflected in comparisons with other SGLT2 agents in OBSERVE-4D: the HR for hospitalization for HF for new users of canagliflozin vs. other SGLT2 therapies was 0.9 (95% CI, 0.71-1.13) for the overall population and 0.7 (95% CI, 0.3-1.63) for those with established CVD.

“The confidence intervals are relatively wide, suggesting these [SGLT2] agents are, more or less, equivalent in their effects on heart failure,” Buse said.

Buse said the hospitalization for HF findings are important in OBSERVE-4D specifically because the data validate internally — as a positive control — the ability to detect a meaningful difference between drugs in a clinical outcome.

“To show a clinical benefit for heart failure suggests that if we can’t show a harm for amputation, it’s not because the data set or the methods were inadequate, it is perhaps there is no harm of amputation,” Buse said.

Risk-benefit profile

Median treatment exposure to canagliflozin was less than 6 months. Buse noted that future observational studies with longer duration are needed. However, OBSERVE-4D data published in Diabetes, Obesity and Metabolism suggests that longer-term exposure does not increase amputation risk.

“In the published paper, there are Kaplan-Meier curves that show exposure out to about 2 years, and there really isn’t any evidence that we have today that there is going to be a different result in a population of patients followed for longer periods of time,” Buse said. “The overall benefit-risk profiles for canagliflozin and the SGLT2 inhibitor class are very positive and physicians and patients should feel comfortable and confident in prescribing this class of drugs to patients.”

Buse cautioned that physicians should consider factors that may increase the risk for amputation when evaluating the optimal therapy for patients, such as a history of prior amputation, peripheral vascular disease, neuropathy and diabetic foot ulcers. – by Regina Schaffer


Buse JB, et al. 4-LB. Presented at: American Diabetes Association 78th Scientific Sessions; June 22-26, 2018; Orlando, Fla.

Disclosures: Buse reports he receives research support from or serves as an advisor for Adocia, AstraZeneca, Boehringer Ingelheim, Dexcom, Elcelyx Therapeutics, Eli Lilly, Intarcia, Johnson & Johnson, Lexicon, Metavention, NovaTarg, Novo Nordisk, Sanofi, Senseonics, Theracos and vTv Therapeutics, and holds stock options in Mellitus Health and PhaseBio Pharmaceuticals. Please see the abstract for all other authors’ relevant financial disclosures.