ORLANDO – but clinicians should still favor other options in patients at risk for amputations, according to investigator John Buse, MD, PhD, chief of the division of endocrinology at the University of North Carolina at Chapel Hill.
Canagliflozin is the only sodium-glucose transporter 2 (SGLT2) inhibitor that carries a black box warning of “lower limb amputations, most frequently of the toe and midfoot” but also the leg. The drug doubled the risk versus placebo in its approval trials, particularly in patients with baseline histories of prior amputations, peripheral vascular disease, neuropathy, or diabetic foot ulcers.
One trial, for instance, reported 7.5 amputations per 1,000 patient years versus 4.2 with placebo, according to labeling.
The new, observational study, which was funded by canagliflozin’s maker Johnson & Johnson and, with the exception of Dr. Buse, conducted by its employees, found no such connection. Investigators reviewed claims data from 142,800 new users of canagliflozin, 110,897 new users of the competing SGLT2 inhibitors empagliflozin (Jardiance) and dapagliflozin (Farxiga), and 460,885 new users of other diabetes drugs except for metformin, Dr. Buse said when he presented the results at the annual scientific sessions of the American Diabetes Association.
The hazard ratio for below-knee amputations with canagliflozin versus non-SGLT2 inhibitors was 0.75 (95% confidence interval, 0.40-1.41; P = 0.30). The ratio versus other SGLT2 inhibitors was 1.14 (95% CI, 0.67-1.93; P = 0.53). Overall, there were 1-5 amputations per 1,000 patient years with the drug.
However, the median follow-up was a few months, far shorter than the median follow-up of over 2 years in the randomized trials. “Therefore, the current study had limited statistical power to detect differences in the 6-12 month time period, the time at which amputation risk began to emerge” in the trials, the study report noted. Also, the investigators didn’t parse out results according to baseline amputation risk. Overall, “none of the analyses were sufficiently powered to rule out the possibility of a modest effect” on amputation rates (Diabetes Obes Metab. 2018 Jun 25. doi: 10.1111/dom.13424).
When moderator Robert H. Eckel, MD, a professor in the division of endocrinology, metabolism, and diabetes at the University of Colorado at Denver, Aurora, asked the 150 or so people who heard the presentation if they use SGLT2 inhibitors in their practices, only a small number raised their hands. Few, if any, raised their hands when he asked if the new results would make them more comfortable prescribing canagliflozin.
“I find [the study] somewhat informative,” Dr. Eckel said in an interview afterwards, “but I think the issue is that the prescribing label still demands that patients be informed of the black box warning. I think we are going to have to wait for the longer term outcomes to determine if [amputation] is a molecule effect or a class effect.”
Dr. Buse later said that “I think for the general population of patients with diabetes, they are at low risk for an amputation,” but “if you are at high risk for having an amputation, we really have to take this risk very seriously. [Canagliflozin] may increase your risk for amputation.
“If I have a patient who has had an amputation and I want to use an SGLT2 inhibitor, I wouldn’t use canagliflozin because of the label. I would use empagliflozin because [amputation] is not in the label, and there was no evidence” of it in trials, he added.
The new study, meanwhile, confirmed the cardiac benefits of SGLT2 inhibitors in type 2 patients. Canagliflozin, for instance, reduced the risk of hospitalization for heart failure by about 60%, compared with non-SGLT2 inhibitors in patients with cardiovascular disease, but it offered no statistically significant heart benefit over other members of its class.
Dr. Buse is an investigator for Johnson and Johnson.