CANVAS: Experts Spar on Canagliflozin Risk/Benefit in Diabetes

Miriam E Tucker

June 16, 2017

SAN DIEGO — The CANVAS program trial results with canagliflozin may have confirmed that cardiovascular benefit is likely to be a class effect of the sodium-glucose cotransporter 2 (SGLT2) inhibitors for type 2 diabetes, but the extent to which the agents may be interchangeable is a subject of heated debate among experts who spoke with Medscape Medical News after the study results were presented.

Moreover, even those who don't believe that the approximate doubling of lower-limb amputation risk seen in the CANVAS program is specific to canagliflozin or even drug-related acknowledge that the finding may put this particular agent at a disadvantage until more information becomes available.

In the combined CANVAS and the CANVAS renal-end-points trial (CANVAS-R), the SGLT2 inhibitor canagliflozin (Invokana, Janssen Pharmaceuticals) reduced the primary composite outcome of cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke by 14% and cut the rate of renal decline by 40% among high-risk type 2 diabetes patients.

With some exceptions, the data were very similar to what was seen with another SGLT2 inhibitor, empagliflozin (Jardiance, Boehringer Ingelheim/Lilly) in the landmark 2015 Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME) study.

"With regard to clinical practice, we are probably dealing here with at least cardiovascular and renal benefits that could apply to the class," Clifford J Bailey, PhD, professor of clinical science at Aston University, Birmingham, United Kingdom, said in his remarks as an independent commentator immediately following presentation of the CANVAS results at the American Diabetes Association (ADA) 2017 Scientific Sessions.

Ele Ferrannini, MD, chief of the metabolism unit at the National Research Council Institute of Clinical Physiology and professor of internal medicine at the University of Pisa, Italy, agrees. "The relative risk reduction in cardiovascular and renal end points in CANVAS suggest — beyond some differences yet to be fully analyzed — a class effect of SGLT2 inhibitors, which therefore clinicians should keep in high consideration for type 2 diabetic patients with advanced vascular and renal disease," he said.

However, there were some important differences between the two trials and their results, most notably an increased amputation risk seen in CANVAS but not EMPA-REG (6.3 vs 3.4 cases per 1000 patient-years; hazard ratio, 1.97).

Moreover, neither the individual end points of all-cause nor cardiovascular death were significantly reduced in CANVAS as they were in EMPA-REG.

Experts interviewed by Medscape Medical News expressed disparate views on the relative importance of these and other differences.

Canagliflozin and Empagliflozin Interchangeable…or Are They?

Dr Bailey pointed out that the reason for the lack of significance in prevention of death may have had to do with the fact that, while nearly all of the EMPA-REG subjects had established cardiovascular disease, only about two-thirds of CANVAS patients did, "which means that the EMPA-REG study is basically looking at people who are further down the line and sicker, whereas CANVAS is looking closer to the type of patients you're going to get in primary care."

Dr Ferrannini noted: "In CANVAS, the inclusion of patients with decidedly lower cardiovascular and renal risk likely diluted the protective effect size of treatment, as would be expected….The data, however, suggest a similar pattern of improved outcomes in both higher- and lower-risk subgroups."

Indeed, said Robert H Eckel, MD, professor of medicine at the University of Colorado Anschutz Medical Campus, Aurora, "I think the CANVAS study was largely confirmatory. I look at it as a similar result as EMPA-REG. The other subcomponents of the primary outcome differ a bit, but I think we're interested in the primary outcome....I would look at the drugs as interchangeable."

But cardiologist Sanjay Kaul, MD, of Cedars Sinai Medical Center, Los Angeles, sees it differently.

While he agrees that canagliflozin's 14% reduction in the primary outcome meets the US Food and Drug Administration's margin for noninferiority (< .001), the P value of .02 that the CANVAS authors reported for superiority "is not robust enough to meet the FDA statutory criterion of 'substantial evidence' that requires a persuasive P value of <0.001 based on a single trial….I would not be surprised if canagliflozin fails to garner a superiority claim for major adverse cardiovascular events [MACE] by the FDA," he told Medscape Medical News.

Moreover, Dr Kaul said, "The favorable renal-outcomes results…are driven by end points that are of questionable clinical relevance —progression of albuminuria or 40% reduction in eGFR — which are not approvable end points."

In fact, Janssen is addressing the renal issue further in a phase 3, double-blind, placebo-controlled study launched in 2014, Canagliflozin and Renal Events in Diabetes with Established Nephropathy Clinical Evaluation (CREDENCE). Enrolling over 4000 patients with type 2 diabetes and nephropathy, that trial is specifically evaluating canagliflozin's ability to slow nephropathy progression. Results are expected in 2019.

Similarly, Boehringer Ingelheim and Eli Lilly have just announced they are to conduct a new, large clinical-outcomes trial investigating empagliflozin for the treatment of 5000 people with chronic kidney disease, both with and without type 2 diabetes.

Amputations: A Deal Breaker?

Despite the cardiovascular and renal effects with canagliflozin, the amputation issue looms large. Some experts also cited the increased fracture risk found with canagliflozin (15.4 vs 11.9 per 1000 patient-years; = .02), but this was only significant in CANVAS and not CANVAS-R.

Neither elevations in amputations nor fractures had been noted with empagliflozin in EMPA-REG, but those data weren't specifically collected in that trial, as had been stipulated in CANVAS.

In CANVAS, the highest absolute risk for amputation occurred among patients with a prior amputation or peripheral vascular disease (PVD), but the relative amputation risk for canagliflozin vs placebo was similar across subgroups.

Asked for his overall take on CANVAS, interventional cardiologist Steven P Marso, MD, chief medical officer for HCA Midwest Health cardiovascular services, Kansas City, Missouri, said: "It's really exciting and reassuring that the point estimate [for canagliflozin] was in the right direction and consistent with empagliflozin, but I think that until we know a lot more about the amputation risk, the use of that drug will be limited.

"You have to think about the net clinical benefit — so it's a MACE benefit minus the amputation, an irreversible complication.…I'd be hesitant to recommend prescribing that medication [canagliflozin] until we know more," he added.

But Dr Bailey noted that because the amputation risk hadn't been seen in any of the other canagliflozin trials or those of other SGLT2 inhibitors so far, "you have to question whether or not there is something in either the population or the way the drug has been introduced that might account for this.

"We can see clearly that patients with a prior amputation or significant PVD are the ones at most risk for another amputation. Providing you're cautious over that particular issue at the moment, I can't see enough evidence to discriminate one agent in the class from another."

Dr Eckel concurs. "Amputations and potentially fracture earlier in the study vs later are issues that need to come up.…But I think the amputation data are really inconsequential.…There's nothing about canagliflozin as a molecule that would make me think it would be more likely to cause peripheral vascular disease or necrotic ulcers or anything else that would lead to amputation."

Nonetheless, he added, "I must admit that if a physician has a choice to prescribe one or the other and both drugs are paid for equally by a third party, you'd tend to go for empagliflozin rather than canagliflozin, just because of this whole issue of early fractures — which weren't significant for the entire [CANVAS program] — and the amputations, but I honestly think the amputations just can't be drug-related."

And Simeon I Taylor, MD, PhD, professor of medicine at the University of Maryland, Baltimore, and a researcher in diabetes drug safety and pharmacogenomics, noted, "There are some real safety issues [with SGLT2 inhibitors], particularly the genitourinary infections, which is more of a tolerability issue, the bone fractures and bone loss — which particularly for older women may be an issue — and of course the amputations, which is probably only an issue for people who have foot ulcers.

"It seems unlikely to me that somebody with a totally healthy lower extremity is going to suddenly have an amputation," Dr Taylor noted. "But if, for example, a patient had a nonhealing foot ulcer I would probably either not prescribe an SGLT2 inhibitor or stop the medication."

However, according to Dr Kaul, "The twofold increase in risk for amputation, including major amputation, is concerning, especially when history of amputation or PAD at baseline does not identify those at higher risk — that is, the interaction is not significant, which implies that the amputation risk is increased similarly in those with or without baseline history of amputation or PAD. This has implication for risk-mitigation strategy."

Overall, Dr Kaul said that together with the increased risk of fracture, "the amputation risk adversely impacts the overall benefit/risk balance of canagliflozin," with the calculation that for every 1000 patient-years of exposure, treatment with canagliflozin prevents 4.6 MACE events at a cost of 2.9 amputations and 3.5 fractures. Based on these data, it is hard to argue in favor of a highly desirable benefit/risk profile of canagliflozin."

Nonetheless, with regard to another concern about the SGLT2 inhibitor class, Dr Kaul also commented, "It is reassuring to note that the incidence of [diabetic ketoacidosis] is very low [in CANVAS] and the risk not materially different from placebo."

Addressing That "Second-Drug" Question…

Dr Taylor went on to add some perspective for doctors to consider when prescribing that "second drug" for type 2 diabetes after metformin.

"For some patients the question will be moot because of their financial situation or their insurance. Their only option may be a generic drug. For others, an SGLT2 inhibitor may be a very attractive choice because of the cardiovascular effect and the renal effect."

He noted that the glucagon peptide-1 (GLP-1) receptor agonists are also good options for some, particularly as there is evidence that liraglutide (Victoza, Novo Nordisk) decreases the risk of major adverse cardiovascular events (as seen in the LEADER trial).

Still, "they're the most expensive and also injectable, so they may not be acceptable options for all patients."

With the dipeptidyl peptidase-4 (DPP-4) inhibitors, on the other hand, "you get modest glucose lowering but no major tolerability or safety issues. My personal belief is that the cardiovascular benefit of SGLT2 inhibitors may be due to blood-pressure lowering, and therefore it may be that a DPP-4 inhibitor in combination with a generic antihypertensive might actually also give you the cardiovascular protection and some degree of glucose lowering, without all the safety issues associated with SGLT2 inhibitors."

Overall, Dr Taylor pointed out that the FDA-mandated cardiovascular end-point trials for type 2 diabetes drugs, such as CANVAS and EMPA-REG, while informative, are not designed to provide direct comparisons among individual agents and that there is considerable intra-individual variability in patient response.

Nonetheless, "by applying common sense and having a discussion with the patient, you can eventually reach a reasonable choice."

Dr Bailey is on advisory panels for Abbott, AstraZeneca, Boehringer Ingelheim, Elcelyx, Lilly, Janssen, Merck, Novo Nordisk, and Sanofi. Dr Kaul is a consultant for Boehringer-Ingelheim, Novo Nordisk and holds equity interest in Johnson & Johnson. Dr Ferrannini is on advisory boards for, is a speaker for, and/or receives research support from Sanofi, AstraZeneca, Boehringer Ingelheim/Lilly, Tanabe-Mitsubishi, Novo Nordisk, and Eli Lilly. Dr Marso has received grants and personal fees from Novo Nordisk, Abbott Vascular, and AstraZeneca.  Dr Taylor was previously employed by Bristol-Myers Squibb (2002–2013) but has sold all of his Bristol-Myers Squibb stock. He currently consults for Ionis Pharmaceuticals and Aegerion Pharmaceuticals and owns stock in Gilead, Celgene, Akorn, and United Healthcare.

For more diabetes and endocrinology news, 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.