Impact of Canagliflozin on Kidney and Cardiovascular Outcomes by Type 2 Diabetes Duration: A Pooled Analysis of the CANVAS Program and CREDENCE Trials

Type 2 diabetes remains the most common reason for kidney replacement therapy requiring dialysis. There is persistent residual kidney risk from type 2 diabetes, even after achieving blood pressure control, including the use of renin angiotensin aldosterone system inhibitors (1). To address the benefit of sodium–glucose cotransporter inhibitor (SGLT2i) on cardiovascular (CV) and kidney outcomes, we conducted a post hoc analysis to assess the effects of canagliflozin versus placebo on CV and kidney events in participants with type 2 diabetes and high CV risk and/or CKD, according to baseline duration of diabetes.

This post hoc analysis is an integrated, pooled, patient-level data meta-analysis from the Canagliflozin Cardiovascular Assessment Study (CANVAS) Program and the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Participants With Diabetic Nephropathy (CREDENCE) trials. Both were randomized, double-blinded, placebo controlled, multicenter studies. The CANVAS Program (N = 10,142) consisted of two multicenter, double-blind, placebo-controlled, randomized trials, CANVAS and Canagliflozin Cardiovascular Assessment Study–Renal (CANVAS-R). Eligible participants had type 2 diabetes (HbA_1c of 7.0–10.5% [53–91 mmol/mol]), an estimated glomerular filtration rate (eGFR) >30 mL/min/1.73 m², and were either aged ≥30 years with a history of symptomatic atherosclerotic CV disease or aged ≥50 years with at least two CV risk factors (2,3) The CREDENCE trial consisted of 4,401 participants with type 2 diabetes with HbA_1c of 6.5–12.0% (48–108 mmol/mol), eGFR of 30–89 mL/min/1.73 m², and albuminuria of >300 to ≤5,000 mg/g (>33.9–565.6 mg/mmol) (4). Participants in each study were randomized to receive canagliflozin or placebo (in addition to standard of care therapy). Although the study participants across the two treatment groups had similar characteristics, there were differences between the studies due to inclusion criteria. In CREDENCE, all participants at baseline had severely increased albuminuria. In the CANVAS Program, 18% of participants had abnormal urine albumin-to-creatinine ratio (ACR) at baseline and ∼20% had an eGFR <60 mL/min/1.73 m². With respect to CV risk profile, 66% of participants in the CANVAS Program had established CV disease compared with 50% in CREDENCE. Detailed study methods, statistical analysis plan, and reporting of the CANVAS Program and CREDENCE trials have been previously published (5,6). The CREDENCE and CANVAS Program studies were approved by ethics committees at each site and registered at Clinicaltrials.gov (CREDENCE study NCT02065791, CANVAS Program studies NCT01032629 and NCT01989754).

This post hoc analysis was based on the intent-to-treat analysis set of the CANVAS Program (N = 10,142) and CREDENCE trial (N = 4,401), which included participants who were randomized to canagliflozin or placebo and had values for all selected outcomes (4,7). Categorical variables are represented as counts and percentage of cohort totals, and continuous variables are represented as mean and SD or median and interquartile range. The effects of canagliflozin versus placebo were examined for CV and kidney outcomes in study participants according to baseline duration of diabetes. Diabetes duration was divided into four time periods: <5, ≥5 to ≤10, >10 to ≤15, and >15 years. This analysis pooled patient-level data from the CANVAS Program and the CREDENCE trial to determine the effect of canagliflozin versus placebo on the following outcomes:

• CV events (major adverse CV event [MACE] and the composite of CV death or hospitalization for heart failure)

• CKD progression (doubling of serum creatinine)

• The composite of kidney and CV events (kidney failure requiring therapy or doubling of serum creatinine, MACE, and hospitalization for heart failure).

• Albuminuria progression and regression (change in albuminuria class plus a change in urine ACR by at least 30% from baseline). Albuminuria was defined as normo-, micro-, or macroalbuminuria, now known as normal (stage A1; <30 mg/g [3.39 mg/mmol]), moderate (stage A2; 30–300 mg/g [3.39–33.9 mg/mmol]), and severe (stage A3; >300 mg/g [33.9 mg/mmol]) (8).

Hazard ratios (HR) and 95% CI for each outcome were estimated separately using Cox regression models, by duration of diabetes group. Interaction P values were calculated in a single model for each outcome by including treatment group by diabetes duration interaction. Absolute risk reductions per 1,000 patients over 2.5 years were calculated. A Poisson model was used to estimate the event rates per 100 patient-years of albuminuria progression or regression to compare the impact of canagliflozin on the risk of improved and worsening albuminuria.

A two-sided P < 0.05 for the interaction term was deemed probable to reflect a difference beyond chance. While this is a post hoc combined analysis, the analyses of these outcomes were prespecified for each of the included trials, and the results should be interpreted in this context. Analyses were performed using SAS 9.4 software (SAS Institute, Cary, NC).

The data sharing policy of Janssen Pharmaceutical Companies of Johnson & Johnson is available at https://www.janssen.com/clinical-trials/transparency. As noted on this site, requests for access to the study data can be submitted through the Yale Open Data Access (YODA) Project site at https://yoda.yale.edu.

With increasing duration of diabetes, participants in these studies were older, less likely to have a history of heart failure, but more likely to have a lower eGFR and higher ACR (Table 1). There was no significant interaction between the duration of type 2 diabetes and the effect of canagliflozin on cardiovascular outcomes. The HR for MACE ranged from 1.08 to 0.77 in those with diabetes duration <5 to >15 years, and the absolute risk reduction ranged from 2.5 to −13.6 per 1,000 patients over 2.5 years (number needed to treat [NNT] 73) (Fig. 1). The HR for the composite outcome of CV death or hospitalization for heart failure ranged from 0.75 to 0.80 in those with diabetes duration <5 to >15 years, and the absolute risk reduction ranged from −8.5 to −7.2 per 1,000 patients over 2.5 years. Similarly, the HR for the composite CV and kidney end point ranged from 0.98 to 0.68 in those with diabetes duration <5 to >15 years, and the absolute risk reduction ranged from −0.1 to −23.9 per 1,000 patients over 2.5 years.

There was no significant interaction between the duration of type 2 diabetes and the effect of canagliflozin on the doubling of serum creatinine, on the doubling of serum creatinine or kidney failure requiring therapy, or between the duration of type 2 diabetes and the effect of canagliflozin on albuminuria (Fig. 1). Within the diabetes duration subgroups, for the albuminuria end points of progression and regression, canagliflozin reduced the incidence of progression, and increased the incidence of regression, across each subgroup of diabetes duration, including participants with type 2 diabetes of <5 years. In study participants with type 2 diabetes duration of <5 years, the incidence of albuminuria progression was reduced with canagliflozin compared with placebo (HR 0.76, 95% CI 0.61–0.94; absolute risk reductions for diabetes duration <5 years 37.0 per 1,000 patients over 2.5 years) and for duration >15 years (76.8 per 1,000 patients over 2.5 years; NNT 27 and 13, respectively) (Fig. 1). Similarly, albuminuria regression was more frequently observed with canagliflozin than placebo in participants with diabetes duration <5 years (HR 1.70; 95% CI 1.31–2.21; absolute risk improvement for diabetes duration <5 years 132.9 per 1,000 patients over 2.5 years, NNT 8).

The incidence of any serious adverse event was significantly lower with canagliflozin compared with placebo in study participants with type 2 diabetes of any duration years (P for interaction = 0.74) (Table 2).

In this post hoc analysis of participants with type 2 diabetes and high CV risk or moderate to severe albuminuria, canagliflozin reduced the incidence of CV and kidney events consistently, regardless of diabetes duration. Importantly, canagliflozin reduced the incidence of albuminuria progression and increased albuminuria regression, regardless of diabetes duration, including within the first 5 years from the diagnosis of type 2 diabetes.

This analysis highlights the clinical importance of reducing kidney and CV outcomes even for people with diabetes of short duration. These findings suggest that clinicians should not wait for rising albuminuria or declining kidney function to initiate SGLT2i therapy, but rather consider earlier intervention in people at highest risk of developing CKD or CV events. This can be assessed by risk calculators (9). Indeed, SGLT2 inhibition reduces CV, kidney, and mortality outcomes irrespective of metformin use (10). The European Society of Cardiology now has SGLT2i above metformin in the treatment algorithm for drug-naive patients with type 2 diabetes at high CV risk (11).

This analysis has several limitations. It is a post hoc analysis of three randomized controlled trials. As an exploratory analysis, it was not designed with adequate statistical power to evaluate outcomes in subgroups stratified by diabetes duration. Further, the trials included moderate- to high-risk CV disease and CKD populations; hence, the findings may not be generalizable to populations with low- to moderate-risk type 2 diabetes. However, the large sample size, the inclusion of participants with different baseline CV risk, and different stages of diabetic kidney disease constitutes important strengths of our analysis.

In this analysis, earlier treatment with canagliflozin prevented albuminuria progression and promoted albuminuria regression. Canagliflozin demonstrated CV and kidney benefits, including a reduction in CKD progression regardless of diabetes duration. These findings may assist clinicians treating individuals with type 2 diabetes through shared clinical decision making to manage their disease more effectively. Treatment with canagliflozin confers meaningful and consistent cardiorenal benefits to individuals living with type 2 diabetes regardless of the duration of diabetes.

Acknowledgments. Figure preparation support was provided by Kim Caldwell, PhD, of Lumanity Communications Inc. Canagliflozin was developed by Janssen Research & Development, LLC, in collaboration with Mitsubishi Tanabe Pharma Corporation. The CANVAS Program and CREDENCE trial were supported by Janssen Research & Development, LLC.

Funding and Duality of Interest. This work was funded by Janssen Canada Inc. S.W.T. has received a KMH Clinic unrestricted grant and in-kind support for the Zero to Five study paid to Sunnybrook Research Institute, has received consulting fees from AstraZeneca paid to Sunnybrook Research Institute, has received payment or honoraria for lectures from Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly and Company, Janssen, Novartis, Novo Nordisk, Pfizer, and Sanofi paid to the CHEP Plus education program, and has served as a volunteer for the American Hypertension Specialist Certification Program. T.A.M. is supported by an FRQS Junior 1 Clinician Scholar Award and received speaker honoraria from Daiichi Sankyo, BMS Canada, Janssen, AstraZeneca, and Pfizer, has served on advisory boards for Boehringer Ingelheim, Bayer, GSK, and Servier outside the submitted work, and has also received an unrestricted research grant from AstraZeneca. H.S.B. reports research support paid to his institution by Amgen, AstraZeneca, Boehringer Ingelheim, Ceapro, Eli Lilly and Company, Gilead, Janssen, Kowa Pharmaceuticals Co. Ltd., Madrigal Pharmaceuticals, Merck, Novartis, Novo Nordisk, Pfizer, Sanofi, and Tricida. A.L. received research funding from Canadian Institutes of Health Research and the Kidney Foundation of Canada, served as a scientific advisor to Boehringer Ingelheim, AstraZeneca, National Institute of Diabetes and Digestive and Kidney Diseases, OccuRx, and Chinook Therapeutics, served on a data safety monitoring board or scientific committee for National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Kidney Precision Medicine, CURE consortium, and the University of Washington Kidney Research Institute Scientific Advisory Committee, GSK, AstraZeneca, and Boehringer Ingelheim, and received fees for time as CREDENCE national coordinator from Janssen, directed to her academic team. N.T. has received research support from the National Kidney Foundation, received consulting fees and research support from AstraZeneca, GSK, Boehringer Ingelheim, Bayer, Roche, Otsuka, Janssen, and ProKidney, and has equity in ClinPredict, Klinrisk, ProKidney, Marizyme, and Quanta. A.S. is an employee of New Arch Consulting and received funding from Janssen for this analysis. B.L.N. received fees for travel support, advisory boards, scientific presentations, and steering committee roles from AstraZeneca, Bayer, Boehringer Ingelheim, Cambridge Healthcare Research, and Janssen, with all honoraria paid to his institution. V.P. received fees for an advisory board, steering committee, or scientific presentation from AbbVie, Amgen, Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Chinook Therapeutics, CSL Vifor, Gilead, GlaxoSmithKline, Janssen, MedImmune, Mitsubishi-Tanabe, Mundipharma, Novartis, Novo Nordisk, Otsuka, and Travere Therapeutics, Inc., received speaker fees from Janssen, served on a data and safety monitoring committee for Dimerix, served on the board of directors for George Clinical, Garvan Institute, George Institute, Victor Chang Cardiac Research Institute, Ingham Institute, and Mindgardens Neuroscience Network, and has stock in George Clinical. K.W.M. has received research grants or contracts that support his research projects from the following companies: AHA, Apple, Inc, Bayer, California Institute Regenerative Medicine, Eidos, Ferring, Gilead, Google (Verily), Idorsia, Johnson & Johnson, Luitpold, Novartis, PAC-12, Precordior, Sanifit. K.W.M. provided consulting or other services (including CME) for the following companies (personal income): Amgen, Applied Therapeutics, Bayer, BMS, BridgeBio, CSL Behring, Elsevier, Fibrogen, Fosun Pharma, Johnson & Johnson, Lexicon, Moderna, Myokardia, Novartis, Novo Nordisk, Otsuka, Phasebio, Portola, Quidel, Sanofi, Theravance. K.W.M. has equity in the following companies: Precordior, Regencor. F.G.A. and W.R. are employees of Janssen. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. S.W.T. and T.A.M. drafted the manuscript. S.W.T., T.A.M., H.S.B., A.L., N.T., B.L.N., V.P., K.W.M., W.R., and F.G.A. contributed to the study design and data interpretation. A.S. performed the statistical analysis. All authors critically revised and approved the final version of the manuscript. F.G.A. is the guarantor of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract and poster form at the American Society of Nephrology 55th Annual Meeting & Scientific Exposition, Orlando, FL 2–6 November 2022.