CANVAS trial – cardiovascular benefits, amputations and fractures

The CANVAS (Canagliflozin Cardiovascular Assessment Study) outcome trial has been eagerly anticipated since the publication of the EMPA-REG OUTCOME trial (Zinman et al, 2015), which demonstrated dramatic reductions in all-cause and cardiovascular (CV) mortality in empagliflozin-treated patients with type 2 diabetes and high CV risk.

The data were merged from two outcome trials, CANVAS and CANVAS-R (CANVAS-Renal). Although both had similar inclusion criteria, patient populations and interventions, the trial durations were different and there were differences in the significance level of some findings in the different arms.

The pooled data (summarised alongside) showed that the risk of the primary outcome of three-point major adverse cardiovascular events (MACE) – CV death, non-fatal myocardial infarction and non-fatal stroke – was 14% lower in canagliflozin recipients compared with placebo among people with type 2 diabetes who were known to have or who were at high risk of cardiovascular disease (CVD). However, despite a trend towards reduction in all-cause death, CV death, myocardial infarction and stroke, none of these showed statistical significance as individual outcomes. The canagliflozin-treated patients also had a 33% reduction in heart failure admissions and a 27% reduction in progression of albuminuria.

During the average follow-up of 296 weeks in CANVAS and 108 weeks in CANVAS-R, the rate of the primary MACE outcome per 1000 person-years was 26.9 in the canagliflozin group versus 31.5 with placebo.

However, the CANVAS trial also raised significant safety concerns, with a 97% increase in lower-limb amputations and a 26% increase in fracture rates in the canagliflozin group. For every 1000 person-years, those treated with canagliflozin had an increased risk of amputation of the toes, feet or legs, with a rate of 6.3 versus 3.4 in the placebo group. In the pooled data, those treated with canagliflozin had more fractures, with respective rates in the canagliflozin and placebo groups of 15.4 versus 11.9 per 1000 person-years. Interestingly, this increase was statistically significant in the longer CANVAS trial, but not in CANVAS-R.

Since the EMPA-REG study, there has been interest in seeing if the substantial CV benefits would be replicated with other agents in the sodium–glucose cotransporter 2 (SGLT2) inhibitor class. The CANVAS study, along with the recently presented retrospective database analysis, CVD-REAL (also summarised alongside), would suggest that the reductions in heart failure admissions and CV and renal benefits probably are a class effect, extending to canagliflozin and dapagliflozin.

The CANVAS results also expand the group of patients who are likely to benefit from treatment with this class, as the participants in this study included those at high risk of CV events and not just those with established CVD. This would suggest that, as a class, SGLT2 inhibitors do provide cardiorenal protection, at least for high-risk patients.

What are the implications for clinical practice?
I believe we now have enough data from EMPA-REG, CANVAS and, to a lesser extent, CVD-REAL, to have to consider SGLT2 inhibitors as a class with additional benefits beyond glucose-lowering for people with type 2 diabetes and high CV risk. However, the increased risk of amputations and fractures seen in CANVAS certainly impacts the benefit-to-risk calculation for canagliflozin.

For every 1000 person-years of exposure, treatment with canagliflozin prevented 4.6 MACE events at a cost of 2.9 amputations and 3.5 fractures. A history of amputation or peripheral artery disease at baseline did not help to identify those at higher risk of subsequent amputation, making it difficult to recommend specific higher-risk groups who should avoid canagliflozin.

Given that these problems have not been identified with empagliflozin (Kohler et al, 2017), I find it hard to imagine a situation where an informed patient would choose canagliflozin as their preferred SGLT2 inhibitor. Indeed, until we see the CV safety results for dapagliflozin in the DECLARE-TIMI 58 trial (estimated completion in 2019), empagliflozin would perhaps seem to be the safest option for providing this CV benefit to high-risk patients.

To read the article summaries, please download the PDF