Primary and Secondary Cardiovascular and Kidney Prevention With Canagliflozin: Insights From the CANVAS Program and CREDENCE Trial

Background This study evaluated the effects of canagliflozin in patients with type 2 diabetes with and without prevalent cardiovascular disease (secondary and primary prevention). Methods and Results This was a pooled participant‐level analysis of the CANVAS (Canagliflozin Cardiovascular Assessment Study) Program and CREDENCE (Canagliflozin and Renal Events in Diabetes With Established Nephropathy Clinical Evaluation) trial. The CANVAS Program included participants with type 2 diabetes at elevated cardiovascular risk, whereas the CREDENCE trial included participants with type 2 diabetes and albuminuric chronic kidney disease. Hazard ratios (HRs) with interaction terms were obtained from Cox regression models to estimate relative risk reduction with canagliflozin versus placebo across the primary and secondary prevention groups. We analyzed 5616 (38.9%) and 8804 (61.1%) individuals in the primary and secondary prevention subgroups, respectively. Primary versus secondary prevention participants were on average younger (62.2 versus 63.8 years of age) and more often women (42% versus 31%). Canagliflozin reduced the risk of major adverse cardiovascular events (HR, 0.84 [95% CI, 0.76–0.94]) consistently across primary and secondary prevention subgroups (P interaction=0.86). Similarly, no treatment effect heterogeneity was observed with canagliflozin for hospitalization for heart failure, cardiovascular death, end‐stage kidney disease, or all‐cause mortality (all P interaction>0.5). Conclusions Canagliflozin reduced cardiovascular and kidney outcomes with no statistical evidence of heterogeneity for the treatment effect across the primary and secondary prevention subgroups in the CANVAS Program and CREDENCE trial. Although studies on the optimal implementation of canagliflozin within these populations are warranted, these results reinforce canagliflozin's role in cardiorenal prevention and treatment in individuals with type 2 diabetes. Registration URL: https://www.clinicaltrials.gov; Unique identifiers: NCT01032629, NCT01989754, NCT02065791.

kidney failure. 2 Prevention of cardiovascular and kidney events in individuals with T2D remains a public health priority; global population-level prevalence studies have identified that two-thirds of all individuals with T2D do not have overt cardiovascular disease 3 ; therefore, prevention of cardiovascular and kidney disease will have significant health and economic benefits at a population level. 46][7][8][9] In the CANVAS (Canagliflozin Cardiovascular Assessment Study) Program, canagliflozin, an SGLT2i, significantly reduced the risk of major adverse cardiovascular events, defined as the composite of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke (ie, 3-point major adverse cardiovascular event [MACE-3P]), in participants with T2D and elevated cardiovascular risk. 10Furthermore, canagliflozin demonstrated significant reductions in the risk of kidney failure, MACE-3P, and HF events in participants in the CREDENCE (Canagliflozin and Renal Events in Diabetes With Established Nephropathy Clinical Evaluation) trial. 11espite this evidence, use of SGLT2i among eligible people with diabetes remains low. 12,13rior studies have demonstrated substantial cardiovascular and kidney benefits of SGLT2i in people with established cardiovascular disease (secondary prevention). 14Yet, there has been less attention paid to the prevention of cardiovascular and kidney events in individuals with T2D without overt cardiovascular disease (primary prevention).Given the substantial global proportion of individuals with T2D who do not have established cardiovascular disease, it is important to understand the impact of SGLT2i across both primary and secondary prevention populations.The evaluation of outcomes within the primary and secondary prevention subgroups was prespecified in the CANVAS Program. 14,15In the CREDENCE trial, analysis of the primary and secondary outcomes was planned for hierarchical testing, with the analyses for the primary outcome prespecified in both subgroups.In this analysis, we evaluated the efficacy of canagliflozin for primary and secondary prevention subgroups from pooled data across the CANVAS Program and CREDENCE trial.

CLINICAL PERSPECTIVE
What Is New?
• This study evaluated the effects of canagliflozin, an SGLT2i (sodium-glucose cotransporter 2 inhibitor) used to treat type 2 diabetes in patients with and without prevalent cardiovascular disease.• The study did not detect differences in the treatment effects of canagliflozin between individuals with type 2 diabetes and prevalent cardiovascular disease and those without cardiovascular disease.
What Are the Clinical Implications?
• Canagliflozin should be considered as a valuable option for cardiovascular and renal prevention and treatment in individuals with type 2 diabetes, irrespective of their cardiovascular disease status.• These findings emphasize the need for optimizing the use of canagliflozin in both primary and secondary prevention populations, and further research is warranted to determine the best strategies for implementing this medication effectively.

Data Source
In the present analysis, we conducted an individual participant data analysis of the CANVAS Program and CREDENCE trial.The designs of the CANVAS Program (CANVAS and CANVAS-R) and the CREDENCE trial, including their setting, locations, relevant dates, periods of recruitment, and data collection practices, have previously been published in detail. 10,11,16To summarize, the CANVAS Program included 2 randomized, double-blind, placebo-controlled trials of canagliflozin in participants with T2D at elevated cardiovascular risk.
The CREDENCE trial was a randomized, double-blind, placebo-controlled trial enrolling participants with T2D and albuminuric CKD.Both the CANVAS Program and CREDENCE trial included participants with and without cardiovascular disease (primary and secondary prevention populations, respectively).

Participants and Definition of Primary and Secondary Prevention Subgroups
The CANVAS Program enrolled individuals with T2D and an estimated glomerular filtration rate (eGFR) >30 mL/min per 1.73 m 2 .Participants were either ≥30 years of age with a history of symptomatic atherosclerotic cardiovascular events (ASCVD; defined as stroke, myocardial infarction, hospitalization for unstable angina, coronary artery bypass grafting, percutaneous coronary intervention, peripheral revascularization [surgical or percutaneous], and symptomatic with documented hemodynamically significant carotid or peripheral vascular disease or amputation secondary to vascular disease [secondary prevention subgroups]) or ≥50 years of age with no prior cardiovascular events but with ≥2 of the following cardiovascular risk factors: duration of diabetes ≥10 years, systolic blood pressure >140 mm Hg on ≥1 antihypertensive agent, current smoker, microalbuminuria or macroalbuminuria, or high-density lipoprotein cholesterol <1 mmol/L (primary prevention subgroups).
The CREDENCE trial enrolled individuals with both T2D and CKD, which was defined by an eGFR between 30 and 90 mL/min per 1.73 m 2 with a urine albumin/creatinine ratio (UACR) of 300 to 5000 mg/g (33.9-565.6 mg/mmol).The definition of primary and secondary prevention subgroups was the same as in the CANVAS Program.All participants in the CREDENCE trial were ≥30 years of age and were receiving a stable dose of an angiotensin receptor blocker or angiotensin-converting enzyme inhibitor for 4 weeks before randomization.
For both trials, the protocols were approved by the ethics committees at each site, and the study was conducted in accordance with the ethical principles of the Declaration of Helsinki.All participants provided written informed consent.F.G.A. had full access to all of the data in the study and takes responsibility for their integrity and the data analysis.

Details on Randomization, Treatment, and Follow-Up
Across studies, randomization was performed following a 2-week, single-blinded, run-in period.In the CANVAS Program, participants were assigned a 1:1:1 randomization to canagliflozin 300 mg, canagliflozin 100 mg, or matched placebo.In CANVAS-R, participants were randomized 1:1 to canagliflozin 100 mg or placebo, with an option to increase to 300 mg at week 13.CREDENCE trial participants were randomized 1:1 to receive canagliflozin 100 mg or matched placebo.Other therapy for glycemic control and risk management was continued or instituted as per best practice in line with guideline-based care.All participants receiving any of the above dosing regimens were included in this subgroup analysis.All participants, investigators, and care providers were blinded to the study drug throughout the trial periods.

Outcomes of Interest
The primary outcome for the CANVAS Program was MACE-3P.For the CREDENCE trial, the primary outcome was the composite of end-stage kidney failure (dialysis for ≥30 days, kidney transplantation, or eGFR <15 mL/min per 1.73 m 2 ), doubling of serum creatinine (dSCr), or death because of kidney or cardiovascular disease.In the present post hoc analysis, the outcomes of interest were MACE-3P, hospitalization for HF (HHF), cardiovascular death or HHF, all-cause mortality, and kidney outcomes including progression to end-stage kidney disease (ESKD), dSCr, and the composite of time to first ESKD or dSCr.

Statistical Analysis
Post hoc analyses were based upon the integrated data sets from the CANVAS Program and the CREDENCE trial, and subjects were analyzed according to randomization assignment (intention to treat, canagliflozin versus placebo).Analyses are restricted to subjects with nonmissing values of imbalanced covariates (duration of diabetes, sex, baseline eGFR, and baseline UACR).Baseline characteristics were compared Hazard ratios (HRs) and 95% CIs were calculated for participants assigned to canagliflozin versus participants assigned to placebo separately for the primary and secondary prevention groups.Outcomes were analyzed using Cox proportional hazards regression models, with treatment as the exploratory variable and adjusted for duration of diabetes ≤13 versus >13 years (the median in this sample), sex, baseline eGFR <60 versus ≥60 mL/min per 1.73 m 2 , and baseline UACR <300 versus ≥300 mg/g (33.9 mg/mmol).The selection of the adjustment variables and eGFR and UACR thresholds were determined based on clinical significance.Adjustment was required to address confounding in the primary and secondary prevention groups due to inherent differences between the CANVAS Program and CREDENCE trial.The proportional hazards assumption was assessed by including the log-time-by-treatment interaction in the proportional hazards models, inspection of Kaplan-Meier curves, and inspection of Schoenfeld residuals.Homogeneity of treatment effects across the primary and secondary prevention groups was examined by adding the main effect of the prevention group and the treatment-byprevention group interaction to these models.Event rates for key outcomes of interest per 1000 participants through 2.5 years of follow-up were estimated using Poisson regression, with adjustments for the same covariates using the same functional form as described for the proportional hazards models.The adjusted event rate serves to estimate the hypothetical rate if the entire population shared the same probability distribution of the specified covariates in the overall cohort (Table S1).More granular estimates are additionally available in Table S2.This standardization of rates, based on covariate distribution, facilitates comparisons that considers relevant covariates.Absolute risk reductions and 95% CIs between treatment groups were obtained using the delta method after postestimation from the Poisson regression model.Analyses were undertaken using SAS version 9.4.

Risk of MACE-3P and HHF
The risk of MACE-3P and cardiovascular events was lower in the primary prevention subgroup compared with the secondary prevention subgroup as reflected by placebo group event rates (Figures 1 and 2).The adjusted placebo event rate for MACE-3P in the primary and secondary prevention group was 20.1 per 1000 participant-years and 45.7 per 1000 participantyears, respectively.Overall, canagliflozin significantly reduced MACE-3P risk (29.6 versus 34.96 per 1000 participant-years; HR, 0.84 [95% CI, 0.76-0.94])with no effect modification by baseline cardiovascular disease (HR, 0.87 [95% CI, 0.69-1.05]versus HR, 0.84 [95% CI, 0.75-0.94]for primary and secondary prevention subgroups, respectively; P interaction =0.864).Canagliflozin similarly reduced the risk of HHF in both the primary prevention subgroup (4.1 versus 6.6 per 1000 participant-years; HR, 0.62 [95% CI, 0.43-0.88])and secondary prevention subgroup (9.0 versus 13.7 per 1000 participant-years; HR, 0.64 [95% CI, 0.52-0.80];P interaction =0.785) with no significant treatment effect heterogeneity.For HHF, the event rate in the secondary prevention group for canagliflozin closely approximated the event rate observed in the primary prevention subgroup receiving placebo (Figure 1).

Risk of Cardiovascular Death and Mortality Outcomes
The effect of canagliflozin on cardiovascular death (HR, 0.86 [95% CI, 0.74-1.00])was not modified by history of cardiovascular disease (P interaction =0.828; Figures 1  and 2).Similar results were obtained for the composite end point consisting of cardiovascular death or HHF.Overall, canagliflozin significantly reduced .00]for primary and secondary prevention subgroups, respectively; P interaction =0.991; Figure 1B).

DISCUSSION
The present study evaluated the effects of canagliflozin for primary and secondary prevention subgroups with combined participant-level data from the randomized CANVAS Program and CREDENCE trial.We demonstrated that canagliflozin reduced the risks of cardiovascular and kidney events similarly in primary and secondary prevention subgroups of participants with T2D.Canagliflozin reduced the risk of MACE-3P, cardiovascular death, HHF, and ESKD, with no statistical evidence of treatment effect heterogeneity for primary or secondary prevention subgroups.The magnitude of the absolute reductions assessed for the cardiovascular events with canagliflozin were numerically greater in the secondary prevention subgroup compared with the primary prevention subgroup.
A prior prespecified analysis from the CANVAS Program alone demonstrated that canagliflozin reduced MACE-3P, cardiovascular, and kidney outcomes, with no statistical evidence of heterogeneity of the treatment effect across the primary and secondary prevention subgroups. 14Likewise, a similar analysis in primary and secondary prevention subgroups from the CREDENCE trial showed no treatment effect heterogeneity across cardiovascular and renal outcomes. 15A non-participant-level meta-analysis focusing on cardiovascular outcomes across primary and secondary prevention subgroups also aligns with our results but did not evaluate kidney outcomes. 17SGLT2i appears to have greater risk reduction for cardiovascular events, such as HF in individuals with worsened kidney function. 5,18,19UACR and eGFR are established markers for risk of future worsening of kidney function and cardiovascular events, such as HHF.Our analysis extends these prior analyses by using participant-level data from both the CANVAS Program and CREDENCE trial.The CREDENCE trial enrolled participants with T2D and albuminuric CKD, whereas in the CANVAS Program, 69.8% of participants had normoalbuminuria.The present analysis provides a more demographically diverse population with regard to kidney function and  levels of albuminuria to evaluate the impact of canagliflozin in primary and secondary prevention subgroups.The consistency of our results highlights the role of canagliflozin across primary and secondary prevention populations among individuals with a broad distribution of cardiovascular and kidney risk profiles.Among those with T2D, in addition to MACE-3P, HHF has emerged as an important clinical event that has historically been ignored by regulators as an important event in clinical trials.The present analysis reinforces the role of canagliflozin in MACE-3P risk reduction in primary and secondary prevention subgroups.Our results showed consistent treatment effects irrespective of cardiovascular disease history.Although prevalent ASCVD is a major risk factor for the development of HF, even individuals without ASCVD are at increased risk of HF events. 20Because ≈60% to 65% of all individuals with T2D do not have clinical atherosclerotic disease, prevention of cardiovascular death or HHF across primary and secondary prevention populations remains an important population health issue.In addition to the CANVAS Program and CREDENCE trial, the DECLARE-TIMI 58 (Multicenter Trial to Evaluate the Effect of Dapagliflozin on the Incidence of Cardiovascular Events) enrolled participants with T2D with ASCVD (40.6%) or multiple cardiovascular risk factors (59.4%). 21The DECLARE-TIMI 58 demonstrated that dapagliflozin was neutral for the primary efficacy outcome of MACE-3P (HR, 0.93 [95% CI, 0.84-1.03)without statistical heterogeneity between those with and without ASCVD (P interaction =0.25).Dapagliflozin did reduce the risk of the second primary efficacy outcome, the composite of cardiovascular death, or HHF (HR, 0.83 [95% CI, 0.73-0.95]).A subgroup analysis demonstrated reductions in the risk of cardiovascular death or HHF in both primary prevention (HR, 0.83 [95% CI, 0.71-0.98])and secondary prevention (HR, 0.84 [95% CI, 0.67-1.04];P interaction =0.99) populations.The effect of sotagliflozin in the SCORED (Cardiovascular and Renal Events in Patients With Adjusted rates were calculated with Poisson regression, whereby the adjusted rate serves to estimate the hypothetical rate if the entire population shared the same probability distribution of specified covariates.Here, this comprises individuals with specific characteristics: 67.8% exhibit an eGFR >60, 67.9% possess a UACR <300, 50.6% have lived with type 2 diabetes for <13 years, and 64.8% are men.This standardization of rates, based on covariate distribution, facilitates comparisons that consider these factors.*Cox proportional hazards models and Poisson model were adjusted for diabetes duration ≤13 vs >13 years, sex, UACR <300 vs ≥300 mg/g (33.9 mg/mmol), eGFR <60 vs ≥60 mL/min per 1.73 m 2 .Our results align with these analyses demonstrating the usefulness of SGLT2i to reduce the risk of cardiovascular death or HHF across the primary and secondary prevention subgroups.Prior meta-analyses, which have included the CANVAS Program and CREDENCE trial, have demonstrated that SGLT2i reduced the of kidney outcomes, including progression to ESKD (SGLT2i versus placebo; HR, 0.65 [95% CI, 0.53-0.81]). 5Our results expand on this, demonstrating that canagliflozin reduces the risk of kidney outcomes across primary and secondary prevention populations.Globally, of the longterm T2D-related complications, CKD imparts some of the highest burden both in terms of burden to health care systems and risk of subsequent morbidity and mortality. 23Therefore, prevention of CKD represents an important public health initiative.Biomarker studies from the CANVAS Program demonstrated that, at baseline, higher levels of the inflammatory markers TNFR-1 (tumor necrosis factor receptor 1) and TNFR-2 (tumor necrosis factor receptor 2) were associated with increased risk of kidney outcomes among normoalbuminuric participants. 24,25These results were independent of the presence of baseline atherosclerotic disease.Furthermore, canagliflozin, compared with placebo, reduced TNFR-1 and TNFR-2 over time even after adjustment for baseline ASCVD history.In totality, these data reinforce the kidney benefit of canagliflozin among individuals with T2D regardless of the presence of ASCVD.
Our analysis was limited by the fact that these trials were not designed to discern treatment differences with respect to primary and secondary prevention.However, analysis by primary and secondary prevention subgroups represented a prespecified analysis for the CANVAS Program.In contrast, analysis of the primary and secondary outcomes in the CREDENCE trial was planned for hierarchical testing, with subgroup analyses for the primary outcome prespecified in both subgroups.The increased number of events within our pooled subgroups expanded the prior analysis from the CANVAS Program that evaluated the impact of canagliflozin across primary and secondary subgroups.The inclusion criteria of the CANVAS Program led to a population with higher cardiovascular events, and the inclusion criteria of the CREDENCE trial led to a population at higher risk of progression of diabetic kidney disease; therefore, our analysis may not generalize to real-world primary prevention populations.Because our primary prevention subgroup was smaller and had a lower risk for cardiovascular events, there may have been limited statistical power to exclude heterogeneity between the primary and secondary prevention subgroups.The mean duration of follow-up from the CANVAS Program (3.6 years) and CREDENCE trial (2.6 years) may have also limited the ability to determine greater magnitudes of benefit in primary and secondary subgroups.
The CANVAS Program enrolled participants with T2D at high cardiovascular risk and the CREDENCE trial enrolled participants with T2D and albuminuric kidney disease.In this pooled, participant-level analysis of the CANVAS Program and CREDENCE trial, canagliflozin reduced the risk of MACE-3P, cardiovascular, and kidney outcomes, without statistical evidence of treatment effect modification between the primary and secondary prevention subgroups.These findings were maintained for the outcomes of cardiovascular death and all-cause mortality.Future implementation studies are warranted that aim to optimize the use of canagliflozin among eligible individuals as a part of standard of care.These results reinforce the role of canagliflozin in cardiorenal prevention and treatment in individuals with T2D.

Figure 1 .
Figure1.Forest plot for CV and kidney outcomes in primary and secondary prevention subgroups.Adjusted rates were calculated with Poisson regression, whereby the adjusted rate serves to estimate the hypothetical rate if the entire population shared the same probability distribution of specified covariates.Here, this comprises individuals with specific characteristics: 67.8% exhibit an eGFR >60, 67.9% possess a UACR <300, 50.6% have lived with type 2 diabetes for <13 years, and 64.8% are men.This standardization of rates, based on covariate distribution, facilitates comparisons that consider these factors.*Cox proportional hazards models and Poisson model were adjusted for diabetes duration ≤13 vs >13 years, sex, UACR <300 vs ≥300 mg/g (33.9 mg/mmol), eGFR <60 vs ≥60 mL/min per 1.73 m 2 .ARR indicates absolute risk reduction; CV, cardiovascular; dSCr, doubling of serum creatinine; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; HHF, hospitalization for heart failure; HR, hazard ratio; MACE-3P, 3-point major adverse cardiovascular events; MI, myocardial infarction; T2DM, type 2 diabetes mellitus; and UACR, urine albumin/creatinine ratio.

Figure
Figure preparation support was provided by K. Caldwell of Lumanity Communications Inc., and was funded by Janssen Canada Inc.This study was sponsored by Janssen, Inc.The CANVAS Program and CREDENCE trial were sponsored by Janssen Research & Development, LLC, and were conducted collaboratively by the sponsor, an academic-led steering committee, and the Academic Research Organization, George Clinical.Canagliflozin has been developed by Janssen Research & Development, LLC, in collaboration with Mitsubishi Tanabe Pharma Corporation.Disclosures Dr Sharma has received support from the FRSQ-Junior 1 clinician scientist award, the Alberta Innovates Health Solution Clinician Scientist fellowship, the European Society of Cardiology Young Investigator research The data sharing policy of Janssen Pharmaceutical Companies of Johnson & Johnson is available at https:// www.janss en.com/ clini caltr ials/ trans parency.As noted on this site, requests for access to the study data can be submitted through the Yale Open Data Access Project site at http:// yoda.yale.edu Sharma et al Primary/Secondary Prevention With Canagliflozin between participants in the primary and secondary prevention groups using the t test or Wilcoxon rank sum test for continuous variables and the χ 2 test or Fisher exact test, as appropriate, for categorical variables.Categorical variables are presented as participant numbers and percentages.Continuous variables are presented as means (SDs) or medians (interquartile ranges).
Table.Baseline Characteristics of Primary and Secondary Prevention Subgroups From the CANVAS Program and CREDENCE Trial BMI indicates body mass index; CANVAS, Canagliflozin Cardiovascular Assessment Study; CREDENCE, Canagliflozin and Renal Events in Diabetes With Established Nephropathy Clinical Evaluation; DBP, diastolic blood pressure; DPP-4, dipeptidyl peptidase 4; eGFR, estimated glomerular filtration rate; GLP-1, glucagon-like peptide 1; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; IQR, interquartile range; LDL-C, low-density lipoprotein cholesterol; RAAS, renin-angiotensin-aldosterone system; SBP, systolic blood pressure; and UACR, urine albumin/creatinine ratio.*Other includes American Indian or Alaska Native, Native Hawaiian or other Pacific Islander, multiple, other, unknown, and not reported.

Table . Continued
Sharma et alPrimary/Secondary Prevention With Canagliflozin