Effects of dapagliflozin on development and progression of ... · 09/06/2019 · Effects of...

www.thelancet.com/diabetes-endocrinology Published online June 9, 2019 http://dx.doi.org/10.1016/S2213-8587(19)30180-9 1

Articles

Effects of dapagliflozin on development and progression of kidney disease in patients with type 2 diabetes: an analysis from the DECLARE–TIMI 58 randomised trialOfri Mosenzon, Stephen D Wiviott, Avivit Cahn, Aliza Rozenberg, Ilan Yanuv, Erica L Goodrich, Sabina A Murphy, Hiddo J L Heerspink, Thomas A Zelniker, Jamie P Dwyer, Deepak L Bhatt, Lawrence A Leiter, Darren K McGuire, John P H Wilding, Eri T Kato, Ingrid A M Gause-Nilsson, Martin Fredriksson, Peter A Johansson, Anna Maria Langkilde, Marc S Sabatine, Itamar Raz

SummaryBackground Sodium-glucose co-transporter-2 (SGLT2) inhibitors have shown beneficial effects on renal outcomes mainly in patients with established atherosclerotic cardiovascular disease. Here we report analyses of renal outcomes with the SGLT2 inhibitor dapagliflozin in the DECLARE–TIMI 58 cardiovascular outcomes trial, which included patients with type 2 diabetes both with and without established atherosclerotic cardiovascular disease and mostly with preserved renal function.

Methods In DECLARE–TIMI 58, patients with type 2 diabetes, HbA1c 6·5–12·0% (47·5–113·1 mmol/mol), with either established atherosclerotic cardiovascular disease or multiple risk factors, and creatinine clearance of at least 60 mL/min were randomly assigned (1:1) to 10 mg dapagliflozin or placebo once daily. A prespecified secondary cardiorenal composite outcome was defined as a sustained decline of at least 40% in estimated glomerular filtration rate [eGFR] to less than 60 mL/min per 1·73m², end-stage renal disease (defined as dialysis for at least 90 days, kidney transplantation, or confirmed sustained eGFR <15mL/min per 1·73 m²), or death from renal or cardiovascular causes; a prespecified renal-specific composite outcome was the same but excluding death from cardiovascular causes. In this renal analysis, we report findings for the components of these composite outcomes, subgroup analysis of these composite outcomes, and changes in eGFR at different timepoints. DECLARE–TIMI 58 is registered with ClinicalTrials.gov, number NCT01730534.

Findings The trial took place between April 25, 2013, and Sept 18, 2018; median follow-up was 4·2 years (IQR 3·9–4·4). Of the 17 160 participants who were randomly assigned, 8162 (47·6%) had an eGFR of at least 90 mL/min per 1·73 m², 7732 (45·1%) had an eGFR of 60 to less than 90 mL/min per 1·73 m², and 1265 (7·4%) had an eGFR of less than 60 mL/min per 1·73 m² at baseline (one participant had missing data for eGFR); 6974 (40·6%) had established atherosclerotic cardiovascular disease and 10 186 (59·4%) had multiple risk factors. As previously reported, the cardiorenal secondary composite outcome was significantly reduced with dapagliflozin versus placebo (hazard ratio [HR] 0·76, 95% CI 0·67–0.87; p<0·0001); excluding death from cardiovascular causes, the HR for the renal-specific outcome was 0·53 (0·43–0·66; p<0·0001). We identified a 46% reduction in sustained decline in eGFR by at least 40% to less than 60 mL/min per 1·73 m² (120 [1·4% vs 221 [2·6%]; HR 0·54 [95% CI 0·43–0·67]; p<0·0001). The risk of end-stage renal disease or renal death was lower in the dapagliflozin group than in the placebo group (11 [0·1%] vs 27 [0·3%]; HR 0·41 [95% CI 0·20–0·82]; p=0·012). Both the cardiorenal and renal-specific composite outcomes were improved with dapagliflozin versus placebo across various prespecified subgroups, including those defined by baseline eGFR (cardiorenal outcome pinteraction=0·97; renal-specific outcome pinteraction=0·87) and the presence or absence of established atherosclerotic cardiovascular disease (cardiorenal outcome pinteraction=0·67; renal-specific outcome pinteraction=0·72). 6 months after randomisation, the mean decrease in eGFR was larger in the dapagliflozin group than in the placebo group. The mean change equalised by 2 years, and at 3 and 4 years the mean decrease in eGFR was less with dapagliflozin than with placebo.

Interpretation Dapagliflozin seemed to prevent and reduce progression of kidney disease compared with placebo in this large and diverse population of patients with type 2 diabetes with and without established atherosclerotic cardiovascular disease, most of whom had preserved renal function.

Funding AstraZeneca.

Copyright © 2019 Elsevier Ltd. All rights reserved.

Lancet Diabetes Endocrinol 2019

Published Online June 9, 2019 http://dx.doi.org/10.1016/S2213-8587(19)30180-9

See Online/Comment http://dx.doi.org/10.1016/S2213-8587(19)30183-4

Diabetes Unit, Hadassah Medical Center, Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel (O Mosenzon MD, A Cahn MD, A Rozenberg MA, I Yanuv MSc, Prof I Raz MD); TIMI Study Group, Division of Cardiovascular Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA (S D Wiviott MD, E L Goodrich MS, S A Murphy MPH, T A Zelniker MD, Prof D L Bhatt MD, Prof M S Sabatine MD); University Medical Center Groningen, University of Groningen, Groningen, Netherlands (Prof H J L Heerspink PhD); Vanderbilt University Medical Center, Nashville, TN, USA (Prof J P Dwyer MD); Li Ka Shing Knowledge Institute, St Michael’s Hospital, University of Toronto, Toronto, ON, Canada (Prof L A Leiter MD); University of Texas Southwestern Medical Center, Dallas, TX, USA (Prof D K McGuire MD); University of Liverpool, Liverpool, UK (Prof J P H Wilding MD); Kyoto University Graduate School of Medicine, Kyoto, Japan (E T Kato PhD); and AstraZeneca, Mölndal, Sweden (I A M Gause-Nilsson MD, M Fredriksson MD, P A Johansson MSc, A M Langkilde MD)

IntroductionFor many years, optimal glucose control plus blood pressure control with angiotensin-converting enzyme

(ACE) inhibitors or angiotensin receptor blockers (ARBs)1,2 has been the basis of treatment of diabetic kidney disease.3–5 Early identification of renal impairment

Articles

2 www.thelancet.com/diabetes-endocrinology Published online June 9, 2019 http://dx.doi.org/10.1016/S2213-8587(19)30180-9

Correspondence to: Dr Ofri Mosenzon, Hadassah Hebrew University Hospital,

POB 12000, Jerusalem 9112001, Israel

[email protected]

and appropriate interventions are more effective than later interventions for the prevention of adverse renal outcomes.3 Even if patients with type 2 diabetes are treated with ACE inhibitors or ARBs, their residual risk of adverse renal and cardiovascular outcomes remains higher than that in age-matched and sex-matched counterparts without diabetes,4,5 and diabetes remains the leading cause of end-stage renal disease in most parts of the world.6,7 Novel treatments are therefore needed to both prevent and slow progression of chronic kidney disease in patients with type 2 diabetes.

Sodium-glucose co-transporter-2 (SGLT2) inhibitors are a class of drugs that have been shown to decrease the rate of decline in renal function in patients with type 2 diabetes and cardiovascular disease. In the EMPA-REG OUTCOME cardiovascular outcomes trial,8 treatment with the SGLT2 inhibitor empagliflozin reduced the risk of the main composite renal outcome (defined as progression to macroalbuminuria; a doubling of the serum creatinine level accompanied by an eGFR of ≤45 mL/min per 1·73 m²; initiation of renal-replacement therapy; or death from renal disease) by 39% compared with placebo in a high-risk population of patients with previous atherosclerotic cardiovascular disease and a high prevalence of chronic kidney disease (baseline median estimated glomerular filtration rate [eGFR] 74·1 mL/min per 1·73 m², with 25·9% [1819/7020] with an eGFR below 60 mL/min per 1·73 m² and 39·6% [2782/7020] with microalbuminuria or

macroalbuminuria).9 In the CANVAS trial Program,10–12 the SGLT2 inhibitor canagliflozin improved renal outcomes compared with placebo in a population of patients predominantly with atherosclerotic cardiovascular disease (65·6% [6656/10 142]) and with renal function similar to that of the EMPA-REG OUTCOME population (mean eGFR 76·5 mL/min per 1·73 m² and median urinary albumin-to-creatinine ratio [UACR] 12·3 mg/g). In the CREDENCE trial,13 patients with type 2 diabetes and albuminuric chronic kidney disease (eGFR 30 to <90 mL/min per 1·73 m² and UACR >300 to 5000 mg/g) were randomly assigned to canagliflozin or placebo. Canagliflozin was associated with a 30% reduction in the primary composite outcome of end-stage renal disease (ie, dialysis, transplantation, or a sustained eGFR <15 mL/min per 1·73 m²), doubling of serum creatinine concentrations, or death from renal or cardiovascular causes.13

In the placebo-controlled Dapagliflozin Effect on Cardiovascular Events (DECLARE)–TIMI 58 trial,14 the SGLT2 inhibitor dapagliflozin reduced the frequency of one of the dual primary composite outcomes (cardiovascular death or hospital admission for heart failure) but did not significantly reduce the frequency of the other (major adverse cardiovascular events). As previously reported,14 a cardiorenal secondary composite outcome (≥40% decrease in eGFR to <60 mL/min per 1·73 m², new end-stage renal disease, or death from renal

Research in context

Evidence before this studyWe searched PubMed for all English-language publications from Jan 1, 2000, to May 28, 2019, using the search terms “SGLT2”, “CKD”, “kidney disease”, “diabetic nephropathy”, and “eGFR”. Previous trials have shown that the sodium-glucose co-transporter-2 (SGLT2) inhibitors empagliflozin and canagliflozin slowed the progression of nephropathy in patients with type 2 diabetes. However, most of the patients in these trials had established atherosclerotic cardiovascular disease or mild to moderate chronic kidney disease, or both. In the primary report of the DECLARE–TIMI 58 trial, a cardiorenal secondary composite outcome (≥40% decrease in estimated glomerular filtration rate [eGFR] to <60 mL/min per 1·73 m², new end-stage renal disease, or death from renal or cardiovascular causes) was reduced by 24%, and a renal-specific composite outcome excluding cardiovascular death was reduced by 47%.

Added value of this studyIn this renal analysis of the DECLARE–TIMI 58 trial, we identified a 46% reduction in sustained decline in eGFR by at least 40% to less than 60 mL/min per 1·73m² (hazard ratio 0·54 [95% CI 0·43–0·67]; p<0·0001). Despite small numbers of events, we also showed a reduction in the combined risk of end-stage renal disease or renal death (0·41 [95% CI 0·20–0·82]; p=0·012).

Furthermore, we have shown the consistent beneficial effects of dapagliflozin on composite renal outcomes compared with placebo in predefined subgroups of patients, including those defined by eGFR status and the presence or absence of atherosclerotic cardiovascular disease at baseline. Thereby, our results show the effect of an SGLT2 inhibitor on both early prevention and reduction in progression of chronic kidney disease in patients with type 2 diabetes. DECLARE–TIMI 58 is the first study to show the effects of an SGLT2 inhibitor on clinically important renal outcomes and on changes in eGFR in a large cohort of patients with type 2 diabetes with and without previous atherosclerotic cardiovascular disease, most of whom had normal or only mildly reduced renal function.

Implications of all the available evidenceOn the basis of available evidence, SGLT2 inhibitors seem to reduce the risk of both progression and development of nephropathy in patients with type 2 diabetes, irrespective of the presence of atherosclerotic cardiovascular disease or baseline renal function. The effect of SGLT2 inhibitors on nephropathy is being examined in dedicated studies of renal outcomes, both in patients with and without type 2 diabetes. However, these trials focus on populations with nephropathy at baseline, and therefore should be considered as complementary to our findings.

Articles


or cardiovascular causes) was reduced by 24% and the same composite outcome excluding cardiovascular death (renal-specific outcome) was reduced by 47%. Notably, the study population of DECLARE–TIMI 58 included a mixture of patients with established atherosclerotic cardiovascular disease and those multiple risk factors for cardiovascular disease; additionally, most of the participants had preserved renal function (mean eGFR 85·2 mL/min per 1·73 m²). Here we report results of detailed analyses of renal outcomes of the DECLARE–TIMI 58 trial, including components of the cardiorenal and renal-specific composite outcomes, subgroup analysis of these composite outcomes, and change in eGFR at different timepoints, in order to investigate the renal effects of dapagliflozin in this large and diverse study population.

MethodsStudy design and participantsThe DECLARE–TIMI 58 trial design, baseline characteristics of participants, and main results have been previously reported.14–16 Briefly, patients with type 2 diabetes and either established atherosclerotic cardio-vascular disease (age ≥40 years and either ischaemic heart disease, cerebrovascular disease, or peripheral arterial disease), or multiple risk factors for atherosclerotic cardiovascular disease (age ≥55 years for men or ≥60 years for women plus at least one of dyslipidaemia, hyper-tension, or current tobacco use) were eligible to be enrolled. Participants were also required to have HbA1c between 6·5% and 12.0% (47·5–113·1 mmol/mol) and creatinine clearance (estimated by the Cockcroft-Gault equation17) of 60 mL/min or higher. A full list of inclusion and exclusion criteria is provided in the appendix.

The trial protocol was approved by the institutional review board at each participating site and all participants provided written informed consent.

Randomisation and maskingParticipants entered a 4–8 week, single-blind, placebo run-in period, which was designed to ensure completion of all necessary blood and urine tests and to identify immediate non-compliers. Patients who remained eligible after the run-in period were randomly assigned (1:1) via web-based response system to dapagliflozin 10 mg once daily or matching placebo. Dapagliflozin and placebo tablets were identical in appearance. Participants and all study staff were masked to treatment allocations until study completion. Randomisation was stratified by cardiovascular risk category (established disease vs multiple risk factors) and baseline haematuria status (present vs absent; because of the regulatory requirement to investigate the risk for bladder cancer14).

ProceduresFollowing the baseline (randomisation) visit, patients had an in-person follow-up visit every 6 months for

assessment of adherence to the trial regimen and clinical and safety events and for laboratory testing. Patients were contacted by telephone every 3 months between in-person visits. In addition to assigned treatments, all participants also received standard-of-care therapy for type 2 diabetes, cardiovascular diseases, and risk factors. The trial was designed to continue until at least 1390 patients had a major adverse cardiovascular event.

Laboratory tests used in the analysis reported here including measurement of both serum creatinine and urinary albumin and creatinine, were done at the central laboratories (LabCorp Clinical Trials [Covance], various locations) at screening, baseline, 6 months, 12 months, and yearly thereafter. Serum creatinine and urine albumin and creatinine concentrations were also measured at the end of the trial, or at the last on-treatment visit, in patients who prematurely discontinued study drug. eGFR was calculated with the Chronic Kidney Disease Epidemiology Collaboration equation.

According to the trial protocol, an unscheduled serum creatinine test was done if central or local laboratory analyses identified doubling of serum creatinine concen-trations from baseline, serum creatinine concentrations higher than 6·0 mg/dL (530 µmol/L), a decrease from baseline in eGFR by 30% or more to an eGFR of less than 60 mL/min per 1·73 m², or an eGFR of less than 15 mL/min per 1·73 m². In these instances, a new central laboratory measurement was obtained as soon as possible (ideally within 4 days), and another central laboratory measurement was obtained after at least 4 weeks. The baseline value for each laboratory test for safety was the last assessment on or before the date of randomisation. The change from this baseline was calculated for these parameters, and time to onset of the outcomes was calculated according to the first of the two subsequent laboratory assessments.

OutcomesAs per the final protocol,14,15 the trial had dual primary efficacy endpoints: major adverse cardiovascular events (a composite of cardiovascular death, myocardial infarction, or ischaemic stroke) and a composite of cardio-vascular death or hospital admission for heart failure. The trial had two secondary efficacy outcomes: a composite cardiorenal outcome and all-cause death. Because the trial met only one of its dual primary outcomes for superiority (cardiovascular death or hospital admission for heart failure), all other analyses of additional outcomes should be considered hypothesis generating only.

In the present renal analysis, we report data for a cardiorenal secondary composite outcome, a renal-specific composite outcome (both previously reported for the overall study population14), and the individual components of these composite outcomes. We also report findings from predefined subgroup analyses of these composite outcomes and comparisons of eGFR changes at different timepoints in the treatment and

See Online for appendix

Articles


eGFR* ≥90 (n=8162) eGFR* 60 to <90 (n=7732) eGFR* <60 (n=1265) p value†

Demographic characteristics

Sex ·· ·· ·· 0·18

Men 5057 (62·0%) 4866 (62·9%) 814 (64·3%) ··

Women 3105 (38·0%) 2866 (37·1%) 451 (35·7%) ··

Age, years 61·2 (6·1) 66·2 (6·5) 67·3 (6·6) <0·0001

Age ≥75 years 95 (1·2%) 818 (10·6%) 183 (14·5%) <0·0001

BMI (kg/m²) 31·6 (6·1) 32·1 (5·9) 34·5 (6·0) <0·0001

Race ·· ·· ·· <0·0001

White 6251 (76·6%) 6313 (81·6%) 1088 (86·0%) ··

Non-white 1911 (23·4%) 1419 (18·4%) 177 (14·0%) ··

Medical history

Duration of type 2 diabetes 10·9 (7·2) 12·5 (8·0) 14·5 (8·9) <0·0001

Established atherosclerotic cardiovascular disease 3193 (39·1%) 3138 (40·6%) 643 (50·8%) <0·0001

History of congestive heart failure 688 (8·4%) 809 (10·5%) 227 (17·9%) <0·0001

History of dyslipidaemia‡ 6370 (78·0%) 6327 (81·8%) 1098 (86·8%) <0·0001

History of hypertension§ 7133 (87·4%) 7088 (91·7%) 1205 (95·3%) <0·0001

Cardiovascular drugs used

Antiplatelet drugs 4813 (59·0%) 4790 (62·0%) 884 (69·9%) <0·0001

ACE inhibitors or ARBs 6434 (78·8%) 6418 (83·0%) 1097 (86·7%) <0·0001

β blockers 3978 (48·7%) 4235 (54·8%) 816 (64·5%) <0·0001

Statins or ezetimibe 5934 (72·7%) 5903 (76·3%) 1031 (81·5%) <0·0001

Diuretics 2752 (33·7%) 3442 (44·5%) 773 (61·1%) <0·0001

Mineralocorticoid receptor antagonists 262 (3·2%) 386 (5·0%) 114 (9·0%) <0·0001

Glucose-lowering drugs used

Metformin 6961 (85·3%) 6263 (81·0%) 843 (66·6%) <0·0001

Insulin 3018 (37·0%) 3284 (42·5%) 711 (56·2%) <0·0001

Sulfonylureas 3671 (45·0%) 3205 (41·5%) 445 (35·2%) <0·0001

Dipeptidyl peptidase-4 inhibitors 1366 (16·7%) 1331 (17·2%) 191 (15·1) 0·17

Glucagon-like peptide-1 receptor agonists 347 (4·3%) 331 (4·3%) 72 (5·7%) 0·058

Laboratory and clinical measurements

HbA1c (%) 8·5 (1·2) 8·1 (1·1) 8·2 (1·2) <0·0001

HbA1c (mmol/mol) 68·9 (13·6) 65·3 (12·5) 66·5 (12·9) <0·0001

eGFR* (mL/min per 1·73 m²) 98·3 (6·5) 77·0 (8·5) 51·4 (7·2) <0·0001

UACR group (mg/g)

N¶ 8026 7582 1234 ··

<30 5691 (70·9%) 5267 (69·5%) 686 (55·6%) <0·0001

30–300 1887 (23·5%) 1761 (23·2%) 381 (30·9%) ··

>300 448 (5·6%) 554 (7·3%) 167 (13·5%) ··

Blood pressure (mm Hg)

Systolic 134·9 (15·0) 135·3 (15·6) 133·5 (16·6) 0·0012

Diastolic 78·9 (8·8) 77·5 (9·2) 75·3 (9·4) <0·0001

Lipids (mg/dL)

LDL cholesterol 90·3 (35·9) 85·4 (34·5) 83·5 (36·4) <0·0001

HDL cholesterol 47·4 (13·1) 47·4 (13·0) 44·2 (12·0) <0·0001

Triglycerides 179·4 (141·8) 173·9 (121·7) 197·4 (155·3) <0·0001

Data are n (%) or mean (SD), unless otherwise specified. eGFR=estimated glomerular filtration rate. ACE=angiotensin-converting enzyme. ARBs=angiotensin receptor blockers. UACR=urine albumin-to-creatinine ratio. *eGFR (measured in mL/min per 1·73m²) was calculated with the Chronic Kidney Disease Epidemiology Collaboration formula. †p values were calculated for comparison of all 3 categories; p values for continuous variables were calculated with the Kruskal-Wallis test and for categorical variables with the χ² test. ‡History of dyslipidaemia was defined as LDL cholesterol concentration >130 mg/dL (3·36 mmol/L) within the past 12 months or on lipid-lowering therapy prescribed by a physician for hypercholesterolemia (documented LDL cholesterol >130 mg/dL [3·36 mmol/L]) for more than 12 months. §History of hypertension was defined as blood pressure >140/90 mm Hg at enrolment visit (with both an elevated systolic blood pressure [>140 mm Hg] and an elevated diastolic blood pressure [>90 mm Hg] at both measurements) or antihypertensive therapy prescribed by a physician for blood pressure lowering at any time. ¶UACR was not measured at baseline for all patients, so N values are smaller for UACR group than for the overall population.

Table: Baseline characteristics by eGFR categories at baseline

Articles


placebo groups (in the overall population and by baseline eGFR subgroup). The cardiorenal secondary composite outcome was defined as time to first event of a composite of sustained confirmed decrease in eGFR by at least 40% (as confirmed by two tests at the central laboratory at least 4 weeks apart)18 to less than 60 mL/min per 1·73 m², end-stage renal disease (defined as either dialysis for 90 days or more, kidney transplantation, or sustained—ie, two measurements at the central laboratory at least 4 weeks apart—eGFR of <15 mL/min per 1·73 m²), or cardiovascular or renal death. The renal-specific outcome, which was the same as the cardiorenal composite but excluding cardiovascular death, was a prespecified exploratory outcome. The individual components of the composite renal outcomes were prespecified exploratory endpoints. Here we also report findings from sensitivity analyses in which a one-time decrease in eGFR of 40% or more to an eGFR of less than 60 mL/min per 1·73 m² was used to define the eGFR decline component of the cardiorenal and renal specific composite outcomes, without the requirement of confirmation by repeat testing.

The analyses reported here were predefined in the trial statistical analysis plan14 or specified in a separate academic analysis plan created prior to database lock (appendix).

Statistical analysisIn the present analysis, we report findings for the cardiorenal and renal-specific composite outcomes and their components, subgroup analyses of these composite outcomes, and comparison of eGFR change by treatment group at different timepoints.

We divided participants into subgroups according to their baseline eGFR, based on the Chronic Kidney Disease Epidemiology Collaboration equation, in order to test if the renal effect of dapagliflozin is affected by patients baseline renal function. Participants were classified into one of three groups: eGFR equal to or greater than 90 mL/min per 1·73 m², eGFR 60 to less than 90 mL/min per 1·73 m², and eGFR less than 60 mL/min per 1·73 m². Baseline characteristics are reported as absolute numbers and percentages for categorical variables and as mean and SDs for continuous variables. We used the χ² test to compared categorical variables, and the Kruskal-Wallis test to compared continuous variables, between eGFR subgroups.

Analyses were done according to the intention-to-treat principle, using data for all randomly assigned participants. Adjudicated outcome data were used to define cardiovascular death and renal death events.

We used the Kaplan-Meier method to generate cumulative incidence curves for the cardiorenal and renal-specific composite outcomes and for sustained decrease in eGFR by at least 40% to less than 60 mL/min per 1·73 m². Hazard ratios (HRs) and 95% CIs were calculated with the Cox proportional-hazard model for the cardiorenal and renal-specific composite outcomes

(previously reported only for the entire population14), and their individual components, both in the entire population and in prespecified subgroups according to patients’ baseline demographics, medical histories, background medications, and baseline measure ments. For these subgroup analyses, an interaction term for the subgroup with randomisation group was included in the Cox model. Treatment effect models included the stratification factors of baseline atherosclerotic cardiovascular disease (ie, established disease vs multiple risk factors) and haematuria (ie, present vs absent) at baseline.

Change in eGFR was calculated with a mixed model for repeated measures, which produced least-squares mean estimates and 95% CIs for each treatment group, the entire trial population, and by eGFR categories at baseline.

No adjustments for multiplicity were made in any analyses. We used SAS (version 9.4) and Stata (version 14.2) for all analyses.

DECLARE–TIMI 58 is registered with ClinicalTrials.gov, number NCT01730534.

Role of the funding sourceThe DECLARE–TIMI 58 trial was a collaboration between the funder and two academic research organizations (TIMI Study Group and Hadassah Medical Organization). The funder was involved in the study design, data collection, data analysis, interpretation, and writing of this report. IAMG-N, MF, PAJ, AML are employed by the study funder. Data analyses were done by the TIMI Study Group, which has access to the complete study database, allowing independent analyses of the results; any discrepancies were resolved by discussion. The DECLARE–TIMI 58 publication committee made the decision to submit for publication.

ResultsThe trial took place between April 25, 2013, and Sept 18, 2018; median follow-up was 4·2 years (IQR 3·9–4·4). Of the 17 160 participants who were randomly assigned, 6974 (40·6%) had established atherosclerotic cardiovascular disease and 10 186 (59·4%) had multiple risk factors for atherosclerotic cardiovascular disease.

Of the 17 159 participants with available baseline eGFR data (one participant had missing data for eGFR), 8162 (47·6%) had an eGFR of at least 90 mL/min per 1·73 m² and 7732 (45·1%) had an eGFR of 60 to less than 90 mL/min per 1·73 m², reflecting the enrolment criteria. Because of the difference between creatinine clearance (estimated by the Cockcroft-Gault equation17) and eGFR calculations, and because inclusion criteria were applied at the screening visit but baseline eGFR was calculated at the randomisation visit, there were 1265 (7·4%) participants who entered the trial with an eGFR of less than 60 mL/min per 1·73 m² (table). The mean eGFR was 85·2 mL/min per 1·73 m² (SD 15·9) in the overall population, 98·3 mL/min

Articles


per 1·73 m² (6·5) in the subgroup of patients with an eGFR of at least 90 mL/min per 1·73 m², 77·0 mL/min per 1·73 m² (8·5) in the subgroup with an eGFR of 60 to less than 90 mL/min per 1·73 m², and 51·4 mL/min per 1·73 m² (7·2) in the subgroup with an eGFR less than 60 mL/min per 1·73 m² (table). At baseline, 11 644 (69·1%) of the 16 843 patients with available data for UACR had normoalbuminuria (ie, <30 mg/g), 4030 (23·9%) had microalbuminuria (ie, ≥30 to ≤300 mg/g), and 1169 (6·9%) had macroalbuminuria (ie, >300 mg/g).

Participants with lower eGFR at baseline were older, had been diagnosed with type 2 diabetes for longer, and had higher BMIs than those with higher eGFRs at baseline (table). Patients in the lower eGFR categories had a higher baseline prevalence of atherosclerotic cardiovascular disease and were more likely to have a history of heart failure and cardiovascular risk factors (such as hypertension and hyperlipidaemia) than those with a baseline eGFR of 90 mL/min per 1·73 m² or higher (table). The use of various cardiovascular drugs was consistently more common in the lower eGFR categories than in those with a baseline eGFR of 90 mL/min per 1·73 m² or higher (table). ACE inhibitor or ARB use ranged from 78·8% in those with a baseline eGFR of 90 mL/min per 1·73 m² or higher to 86·7% in those with a baseline eGFR below 60 mL/min per 1·73 m². Metformin and sulfonylurea were less commonly used in the lower eGFR subgroups than in those with a baseline eGFR of 90 mL/min per 1·73 m² or higher; insulin use was more common in the lower eGFR subgroups (table). At baseline, patients in the lower eGFR subgroups had lower HbA1c and LDL cholesterol but higher albuminuria than those with

eGFR of 90 mL/min per 1·73 m² or higher (table). There were no meaningful differences between treatment groups in terms of baseline characteristics in the different eGFR subgroups (appendix pp 1–2).

As previously reported,14 the frequency of the secondary cardiorenal composite outcome was significantly lower in the dapagliflozin group than in the placebo group in the overall study population (HR 0·76 [95% CI 0·67–0·87]; p<0·0001; figures 1, 2A). A sensitivity analysis in which a one-time decrease in eGFR of 40% or more to an eGFR of less than 60 mL/min per 1·73 m² was used to define meeting the eGFR decline component of the cardiorenal endpoint, without requirement of confirmation by repeat testing, also showed a significant reduction in the cardiorenal composite outcome with dapagliflozin compared with placebo (HR 0·77 [95% CI 0·70–0·85]; p<0·0001). As previously reported,14 the renal-specific composite outcome was 47% less common in the dapagliflozin group than in the placebo group (HR 0·53 [95% CI 0·43–0·66]; p<0·0001; figures 1, 2B). The sensitivity analysis in which a one-time decrease in eGFR of 40% or more to an eGFR of less than 60 mL/min per 1·73 m² was used, without requirement of confirmation by repeat testing, also showed a significant reduction in the renal-specific outcome with dapagliflozin compared with placebo (HR 0·69 [95% CI 0·62–0·78]; p<0·0001).

In assessing the components of the composite renal outcomes, the risk of a sustained confirmed decrease in eGFR of more than 40% to an eGFR of less than 60 mL/min per 1·73 m² was significantly lower in the dapagliflozin group than in the placebo group (HR 0·54 [95% CI 0·43–0·67]; p<0·0001; figures 1, 2C). End stage

Figure 1: Composite cardiorenal and renal-specific outcomes and their individual components for dapagliflozin versus placebo in the overall trial populationThe composite cardiorenal outcome consisted of a sustained decrease in eGFR by at least 40% to less than 60 mL/min per 1·73 m², end-stage renal disease, or cardiovascular or renal death. The composite renal-specific outcome consisted of a sustained decrease in eGFR by at least 40% to less than 60 mL/min per 1·73 m², end-stage renal disease, or renal death. eGFR was calculated with the Chronic Kidney Disease Epidemiology Collaboration equation. Data for the composite cardiorenal and renal-specific outcomes and cardiovascular death were previously reported in reference 14. eGFR=estimated glomerular filtration rate.

Hazard ratio(95% CI)

p value

5·3%

2·6%

2·5%

0·2%

0·1%

2·7%

0·3%

0·76 (0·67–0·87)

0·53 (0·43–0·66)

0·54 (0·43–0·67)

0·31 (0·13–0·79)

0·60 (0·22–1·65)

0·98 (0·82–1·17)

0·41 (0·20–0·82)

<0·0001

<0·0001

<0·0001

0·013

0·32

0·83

0·012

Kaplan-Meier event rate (4 years)

1·00·50·1 1·7

Favours placeboFavours dapagliflozin

Dapagliflozin group

n/N (%)

Composite cardiorenal outcome

Composite renal-specific outcome

Sustained eGFR decrease ≥40% to eGFR<60mL/ min per 1·73m²

End-stage renal disease

Renal death

Cardiovascular death

End-stage renal disease or renal death

Kaplan-Meier event rate (4 years)

370/8582 (4·3%) 127/8582 (1·5%)

120/8582 (1·4%)

6/8582 (0·1%)

6/8582 (0·1%)

245/8582 (2·9%)

11/8582 (0·1%)

4·2%

1·5%

1·4%

0·1%

0·1%

2·7%

0·1%

480/8578 (5·6%) 238/8578 (2·8%)

221/8578 (2·6%)

19/8578 (0·2%)

10/8578 (0·1%)

249/8578 (2·9%)

27/8578 (0·3%)

Placebo group

n/N (%)

Articles


renal disease or renal death occurred in 11 of 8582 patients in the dapagliflozin group and 27 of 8578 in the placebo group (HR 0·41 [95% CI 0·20–0·82]; p=0·012; figure 1).

We noted no significant interactions between most prespecified subgroups defined by demographics, medical history, background medication, and baseline laboratory measurements and the favourable effect of dapagliflozin compared with placebo on both the cardiorenal composite outcome and the renal-specific composite outcome (figure 3; appendix pp 3–4). There was no difference between the stratum with established atherosclerotic cardiovascular disease and those with multiple risk factors for atherosclerotic cardiovascular disease, both for the renal-specific composite outcome (pinteraction=0·72; figure 3) and for the cardiorenal composite outcome (pinteraction=0·67; appendix p 3). An absence of interaction was also seen for eGFR group at baseline (renal-specific outcome pinteraction=0·87 [figure 3]; cardiorenal outcome pinteraction=0·97 [appendix p 4]). The exceptions were an interaction between treatment groups and UACR subgroups for the cardiorenal composite outcome (pinteraction=0·020; appendix p 4), but not the renal-specific outcome (pinteraction=0·30; figure 3), and an interaction between treatment groups and diuretics use for the renal-specific outcome (pinteraction=0·0021; figure 3)—although benefit was apparent both with and without diuretics, the positive effect of dapagliflozin seemed to be greater in patients without diuretic use at baseline.

Figure 4 shows the mean change in eGFR over time in the two treatments groups for the overall population and by eGFR subgroups. 6 months after randomisation, the mean decrease in eGFR was larger in the dapagliflozin group than in the placebo group. The mean change equalised by 2 years, and at 3 and 4 years the mean decrease in eGFR was less with dapagliflozin than with placebo. A similar pattern was seen in subgroups defined by baseline eGFR.

DiscussionIn the DECLARE–TIMI 58 trial, patients randomly assigned to dapagliflozin had significantly reduced frequencies of composite cardiorenal and renal-specific

Figure 2: Kaplan-Meier curves for the composite cardiorenal outcome (A), the composite renal-specific outcome (B), and sustained decrease in eGFR by at

least 40% to less than 60 mL/min per 1·73 m² (C)The composite cardiorenal outcome consisted of a sustained decrease in eGFR by

at least 40% to less than 60 mL/min per 1·73 m², end-stage renal disease, or cardiovascular or renal death. The composite renal-specific outcome consisted of a

sustained decrease in eGFR by at least 40% to less than 60 mL/min per 1·73 m², end-stage renal disease, or renal death. eGFR was calculated with the Chronic

Kidney Disease Epidemiology Collaboration equation. The Kaplan-Meier curve for the composite cardiorenal outcome (A) was previously reported in reference 14.

HR=hazard ratio. eGFR=estimated glomerular filtration rate.

Number at riskPlacebo group

Dapagliflozingroup

85788582

85088533

84228436

83268347

82008248

80568136

79328009

74097534

53895472

Cum

ulat

ive

prob

abili

ty o

f eve

nt

0 0·50

0·01

1·0 1·5 2·0 2·5 3·0 3·5 4·0

0·02

0·03

0·04

0·05

0·06

APlacebo groupDapagliflozin groupHR 0·76 (95% CI 0·67–0·87; p<0·0001)


Dapagliflozingroup

85788582

85048523

84158422

83218338

81938242

80568127

79258004

74037522

53825464

Cum

ulat

ive

prob

abili

ty o

f eve

nt

0 0·50

0·01

1·0 1·5 2·0 2·5 3·0 3·5 4·0

0·02

0·03

0·04

0·05

0·06

B

HR 0·53 (95% CI 0·43–0·66; p<0·0001)


Dapagliflozingroup

85788582

85038523

84168421

83238338

Time since randomisation (years)

81978242

80538128

79298004

74067522

53835464

Cum

ulat

ive

prob

abili

ty o

f eve

nt

0 0·50

0·01

1·0 1·5 2·0 2·5 3·0 3·5 4·0

0·02

0·03

0·04

0·05

0·06

C

HR 0·54 (95% CI 0·43–0·67; p<0·0001)

Articles


outcomes compared with those in the placebo group.14 Components of the composite outcomes were also significant reduced with dapagliflozin, including a sustained decrease in eGFR by at least 40% to less than 60 mL/min per 1·73 m². End-stage renal disease, although a rare event in the trial, as would be expected in a population with near-normal baseline eGFR, was also significantly reduced with dapagliflozin compared with

placebo. These benefits occurred in a large and diverse population of patients with type 2 diabetes, irrespective of the presence of established atherosclerotic cardiovascular disease. Only 7·4% of the study population had moderate renal impairment (ie, an eGFR <60 mL/min per 1·73 m²), suggesting that the effects of dapagliflozin are not limited to patients with renal impairment. Collectively, these data suggest that treatment with SGLT2 inhibitors could be

Figure 3: Subgroup analyses of dapagliflozin versus placebo for the composite renal-specific outcomeThe composite renal-specific outcome consisted of a sustained decrease in eGFR by at least 40% to less than 60 mL/min per 1·73 m², end-stage renal disease, or renal death. eGFR was calculated with the Chronic Kidney Disease Epidemiology Collaboration equation. ACE=angiotensin-converting enzyme. ARB=angiotensin receptor blocker. eGFR=estimated glomerular filtration rate. UACR=urine albumin-to-creatinine ratio.

Hazard ratio(95% CI)

p value pinteractionDapagliflozin group

n/N (%)

Placebo group

n/N (%)Kaplan-Meierevent rate(4 years)

Kaplan-Meierevent rate(4 years)

<0·0001

<0·0001

<0·0001

<0·0001

<0·0001

0·0035

0·39

0·0051

0·0061

0·0014

0·0015

0·0001

<0·0001

0·022

<0·0001

<0·0001

0·046

<0·0001

0·37

0·023

<0·0001

0·028

<0·0001

0·12

0·0002

0·0003

<0·0001

0·059

0·0002

0·0082

<0·0001

0·64

0·97

0·69

0·44

0·72

0·78

0·41

0·16

0·0021

0·26

0·87

0·30

Male

Female

Age (years)

≥65

<65

Race

White

Non-white

Diabetes duration (years)

≤5

>5–10

>10–15

>15–20

>20

Medical history

Established cardiovascular disease

Multiple cardiovascular risk factors

History of heart failure

No history of heart failure

History of hypertension

No history of hypertension

Background medication

ACE inhibitor or ARB use at baseline

No ACE inhibitor or ARB use at baseline

Diuretic use at baseline

No diuretic use at baseline

Laboratory measurements

Hba1c <7%

Hba1c 7 to <8%

Hba1c 8 to <9%

Hba1c ≥9%

eGFR ≥90 mL/min per 1·73m2

eGFR 60 to <90 mL/min per 1·73m2

eGFR <60 mL/min per 1·73m2

UACR <30 mg/g

UACR 30–300 mg/g

UACR >300 mg/g

79/5411 (1·5%)

48/3171 (1·5%)

62/3951 (1·6%)

65/4631 (1·4%)

103/6843 (1·5%)

24/1739 (1·4%)

25/1887 (1·3%)

32/2375 (1·3%)

31/2015 (1·5%)

20/1246 (1·6%)

19/1059 (1·8%)

65/3474 (1·9%)

62/5108 (1·2%)

27/852 (3·2%)

100/7730 (1·3%)

122/7769 (1·6%)

5/813 (0·6%)

106/6977 (1·5%)

21/1605 (1·3%)

81/3488 (2·3%)

46/5094 (0·9%)

9/773 (1·2%)

38/3317 (1·1%)

41/2193 (1·9%)

39/2297 (1·7%)

41/4137 (1·0%)

65/3838 (1·7%)

21/606 (3·5%)

50/5819 (0·9%)

39/2017 (1·9%)

31/594 (5·2%)

1·5%

1·4%

1·6%

1·4%

1·5%

1·5%

1·3%

1·3%

1·6%

1·6%

1·6%

1·9%

1·2%

3·3%

1·3%

1·6%

0·5%

1·5%

1·3%

2·3%

0·9%

1·0%

1·2%

1·9%

1·7%

1·0%

1·6%

3·8%

0·9%

2·0%

4·8%

139/5327 (2·6%)

99/3251 (3·0%)

117/3956 (3·0%)

121/4622 (2·6%)

188/6810 (2·8%)

50/1768 (2·8%)

32/1949 (1·6%)

59/2356 (2·5%)

55/1937 (2·8%)

44/1187 (3·7%)

48/1147 (4·2%)

118/3500 (3·4%)

120/5078 (2·4%)

48/872 (5·5%)

190/7706 (2·5%)

222/7658 (2·9%)

16/920 (1·7%)

211/6973 (3·0%)

27/1605 (1·7%)

111/3479 (3·2%)

127/5099 (2·5%)

21/774 (2·7%)

83/3309 (2·5%)

59/2327 (2·5%)

75/2164 (3·5%)

79/4025 (2·0%)

121/3894 (3·1%)

38/659 (5·8%)

95/5825 (1·6%)

66/2013 (3·3%)

75/575 (13·0%)

2·5%

2·8%

2·8%

2·5%

2·5%

3·2%

1·5%

2·2%

2·8%

3·7%

4·0%

3·2%

2·3%

5·4%

2·3%

2·7%

1·8%

2·9%

1·4%

2·9%

2·4%

2·2%

2·2%

2·6%

3·6%

2·0%

2·8%

5·8%

1·5%

3·3%

12·8%

0·55 (0·42–0·73)

0·50 (0·35–0·70)

0·53 (0·39–0·72)

0·53 (0·39–0·71)

0·54 (0·42–0·69)

0·48 (0·30–0·79)

0·79 (0·47–1·34)

0·54 (0·35–0·83)

0·54 (0·35–0·84)

0·42 (0·25–0·72)

0·42 (0·25–0·72)

0·55 (0·41–0·75)

0·51 (0·37–0·69)

0·58 (0·36–0·92)

0·52 (0·41–0·66)

0·54 (0·43–0·67)

0·36 (0·13–0·98)

0·50 (0·39–0·63)

0·77 (0·44–1·37)

0·72 (0·54–0·95)

0·36 (0·26–0·50)

0·42 (0·19–0·91)

0·45 (0·31–0·66)

0·73 (0·49–1·09)

0·48 (0·32–0·70)

0·50 (0·34–0·73)

0·54 (0·40–0·73)

0·60 (0·35–1·02)

0·52 (0·37–0·74)

0·59 (0·39–0·87)

0·38 (0·25–0·58)

1·00·50·1 1·5

Favours placeboFavours dapagliflozin

Sex

Articles


useful for the early prevention of chronic kidney disease in patients with type 2 diabetes.

There are several important differences between the renal outcome data from DECLARE–TIMI 58 and those from the EMPA-REG OUTCOME (empagliflozin)8,9 and CANVAS (canagliflozin) trials.10–12 The DECLARE–TIMI 58 population had a lower prevalence of established atherosclerotic cardiovascular disease (40·6% vs 100% in EMPA-REG OUTCOME and 65·6% in CANVAS) and better renal function (mean eGFR 85·2 mL/min per 1·73 m² vs 74·1 mL/min per 1·73 m² in EMPA-REG OUTCOME and 76·5 mL/min per 1·73 m² in CANVAS) at baseline. Median follow-up was 4·2 years in DECLARE–TIMI 58, compared with 3·1 years in EMPA-REG

OUTCOME and 2·4 years in CANVAS. The cardiorenal and renal-specific composite renal outcomes analysed in DECLARE–TIMI 58 included only outcomes based on sustained change in eGFR and clinical endpoints and not on softer endpoints based on changes in albuminuria, unlike the main renal outcomes in EMPA-REG OUTCOME8 and the CANVAS Program.10 The inclusion of sustained eGFR changes only (ie, with two consecutive tests >4 weeks apart) adds to the robustness of our analysis.

CREDENCE13 was the first large-scale outcome trial of an SGLT2 inhibitor (canagliflozin) with a renal primary outcome to be completed. CREDENCE and other trials that are underway (NCT03036150, NCT03315143) will

Figure 4: Change in mean eGFR in the overall population (A) and in patients with baseline eGFRs ≥90 mL/min per 1·73 m² (B), of 60 to <90 mL/min per 1·73 m² (C), and <60 mL/min per 1·73 m² (D)eGFR means were adjusted for the stratification variables in a mixed-effects model. Error bars represent 95% CIs. eGFR was calculated with the Chronic Kidney Disease Epidemiology Collaboration equation. eGFR=estimated glomerular filtration rate.

Dapagliflozin groupPlacebo group

p<0·0001 p<0·0001

68007098

57706050

3·0 4·0

Sample sizePlacebo group

Dapagliflozin group

0 0·5

85788581

82238273

78847978

73167513

–12

–8

Adj

uste

d m

ean

chan

ge in

eGF

R (m

L/m

in p

er 1

·73m

²)

1·0 2·0

–4

0

4A

0 0·5

40254137

38604004

37293869

34773666

32393483

27482982

1·0 2·0 3·0 4·0

B


0 0·5

659606

617576

576544

523493

472454

391382

1·0 2·0 3·0 4·0

D

p<0·0001

p<0·0001

p<0·0001 p<0·0001

p<0·0001

p<0·0001

p<0·0001 p<0·0001p<0·0001

p<0·0001

p=0·0005 p=0·042p=0·35

p=0·44

p=0·053

p=0·077

p=0·79

p=0·24

Sample sizePlacebo group

Dapagliflozin group

0 0·5

38943838

37463693

35793565


33163354

30893161

26312686

–12

–8

Adj

uste

d m

ean

chan

ge in

eGF

R (m

L/m

in p

er 1

·73m

²)

1·0 2·0 3·0 4·0

–4

0

4C

Articles


help to define the role of SGLT2 inhibitors in the pharmacological management of diabetic kidney disease and possibly other nephropathies, due to the inclusion of participants without diabetes (NCT03594110, NCT03190694). However, the participants in these dedicated renal outcomes trials all have prevalent nephropathy. In CREDENCE, albuminuric chronic kidney disease was an inclusion criterion. By contrast, the population of DECLARE–TIMI 58 mainly did not have substantially reduced eGFR at baseline, and 69·1% of the study population had normoalbuminuria. Therefore, the population was at much lower risk of adverse renal outcomes. Thus, our findings for the effects of dapagliflozin on renal outcomes could have important implications for the early prevention of diabetic kidney disease. Notably, the renal benefits of dapagliflozin were seen in almost all subgroups analysed, which further emphasises the consistency of this effect.

The mechanisms underlying the beneficial effects of SGLT2 inhibitors on renal outcomes are unknown, but could include haemodynamic, metabolic, and possibly other mechanisms.23–27 The beneficial effect of dapagliflozin on the renal-specific composite outcome seems to be relatively greater in the subgroup of patients not treated with any diuretics (pinteraction=0·0021)—although this might be a chance finding and should be interpreted with caution, it could be related to the patient population prescribed diuretics, or might be associated with the potential renal protective mechanisms of SGLT2 inhibitors. ACE inhibitors and ARBs were extensively used in DECLARE–TIMI 58, but did not attenuate the beneficial renal effects of dapagliflozin (pinteraction=0·16). The positive renal effect of dapagliflozin when combined with ACE inhibitors or ARBs further supports the haemodynamic theory of the renal effect of SGLT2 inhibitors—ie, reduction of intraglomerular pressure driven both by vasoconstriction of the afferent arteriole with SGLT2 inhibitors and vasodilation of the efferent arteriole with ACE inhibitors or ARBs.23–27

The mean decrease in eGFR was larger in the dapagliflozin group than in the placebo group at 6 months, but the mean change had equalised by 2 years—by 3 and 4 years, the mean decrease in eGFR was less with dapagliflozin than with placebo. These finding can be explained by the difference between the acute versus chronic effects of SGLT2 inhibitors on renal function, associated with their haemodynamic effect.23–27 The DECLARE–TIMI 58 population had well preserved renal function and therefore the rate of deterioration in both treatment groups was slow, meaning that it took longer to show the benefit of dapagliflozin on eGFR compared with previous cardiovascular outcome trials of SGLT2 inhibitors,8,10 The low frequency of testing of renal function throughout the trial limits our ability to accurately estimate the exact timing of changes in eGFR. The between difference in eGFR between treatment groups was small, and probably does not fully account

for the larger between-group differences in terms of the cardiorenal and renal-specific composite outcomes—specifically the lower frequency of eGFR decreases of at least 40% to sustained eGFR less than 60 mL/min per 1·73 m² in the dapagliflozin group compared with the placebo group that seemed to be apparent at 2 years. This discrepancy, however, is not unique to DECLARE–TIMI 58. In CREDENCE,13 the between-group difference in the rate of decline in eGFR was only 1·52 mL/min per 1·73 m² per year (95% CI 1·11–1·93), yet the between-group difference in the cardiorenal composite outcome was already significant at the 1-year follow-up. When assessing differences in eGFR between treatment groups, group size should be considered: in the smallest eGFR subgroup in the present analysis (eGFR <60 mL/min per 1·73 m²), the difference between the dapagliflozin and placebo groups was not statistically significant, although the trend was similar to what was seen in the other, larger eGFR subgroups (figure 4).

On the basis of post-marketing reports, the US Food and Drug Administration published a warning28 about the risk of acute kidney injury in patients taking SGLT2 inhibitors. However, in DECLARE–TIMI 58,14 there was a 31% decrease in the risk of acute kidney injury in the dapagliflozin group compared with the placebo group. No increase in acute kidney injury was associated with the use of SGLT2 inhibitors in the CANVAS Program10 or in observational retrospective cohorts,29,30 and a decrease in acute kidney injury was identified in EMPA-REG OUTCOME8 and in CREDENCE.13

A limitation of our analysis is that the cardiorenal composite outcome was a secondary outcome in the DECLARE–TIMI 58 trial, and because one of the dual primary outcomes was not superior to placebo, all other outcomes should be considered hypothesis generating only. However, the extent, consistency, and robustness of the renal findings in all cardiovascular outcome trials of SGLT2 inhibitors and in retrospective cohorts29,30 strongly supports the conclusion that SGLT2 inhibitors are potent renoprotective drugs. Another limitation is that creatinine concentrations and spot urine-based UACR were measured only at screening, baseline, 6 months, 12 months, and yearly thereafter, except when specific criteria required an early repeated test; therefore, subtle changes in renal function might have been missed. Finally, the choice to use only confirmed sustained outcomes and not any one-off change in eGFR (like in the previous cardiovascular outcome trials of SGLT2 inhibitors8,10) might have led to a reduction in the numbers of events, but adds to the robustness of our findings and helps to differentiate between transient acute kidney injuries and permanent changes in renal function.

In conclusion, in the DECLARE–TIMI 58 trial, dapagliflozin treatment led to substantial reduction in the risk of clinically significant renal deterioration in a large and diverse population of patients with type 2 diabetes. These results emphasise the value of SGLT2

Articles


inhibitors as an important component of both prevention and treatment of chronic kidney disease among patients with type 2 diabetes.ContributorsOM, SDW, JPD, ETK, IAMG-N, MF, AML, MSS, and IR contributed to the study design. OM, SDW, IAMG-N, AML, MSS, and IR did the literature search. OM, SDW, AR, IY, ELG, SAM, IAMG-N, AML, MSS, and IR designed the figures. OM, SDW, ETK, IAMG-N, MF, AML, MSS, and IR contributed to data collection and OM, SDW, AC, AR, IY, ELG, SAM, DLB, LAL, DKM, JPHW, ETK, IAMG-N, MF, PAJ, AML, MSS, and IR contributed to data analysis. OM, SDW, AC, AR, IY, HJLH, TAZ, DLB, LAL, DKM, JPHW, ETK, IAMG-N, MF, AML, MSS, and IR contributed to data interpretation. OM, SDW, AC, AR, IY, HJLH, TAZ, JPD, DLB, LAL, DKM, JPHW, ETK, IAMG-N, MF, AML, MSS, and IR contributed to the writing of the report.

Declaration of interestsOM reports grants and personal fees from AstraZeneca, Bristol-Myers Squibb, and Novo Nordisk and personal fees from Eli Lilly, Sanofi, Merck Sharp & Dohme, Boehringer Ingelheim, Johnson & Johnson, and Novartis. SDW reports grants from AstraZeneca, Bristol-Myers Squibb, Sanofi Aventis, and Amgen; grants and personal fees from Arena, Daiichi Sankyo, Eisai, Eli Lilly, and Janssen; grants and consulting fees from Merck (additionally his spouse is employed by Merck); and personal fees from Aegerion, Allergan, AngelMed, Boehringer Ingelheim, Boston Clinical Research Institute, Icon Clinical, Lexicon, St Jude Medical, Xoma, Servier, AstraZeneca, and Bristol-Myers Squibb. AC reports grants and personal fees from AstraZeneca and personal fees from Novo Nordisk, Eli Lilly, Sanofi, Boehringer Ingelheim, Merck Sharp & Dohme, and GlucoMe. ELG and SAM report research grant support through Brigham and Women’s Hospital from Abbott Laboratories, Amgen, AstraZeneca, Critical Diagnostics, Daiichi-Sankyo, Eisai, Genzyme, Gilead, GlaxoSmithKline, Intarcia, Janssen Research and Development, The Medicines Company, MedImmune, Merck, Novartis, Poxel, Pfizer, Roche Diagnostics, and Takeda. HJLH reports consulting for AbbVie, Astellas, AstraZeneca, Boehringer Ingelheim, Fresenius, Gilead, Janssen, Merck, Mitsubishi Tanabe, and Mundi Pharma, with all honoraria paid to his employer. TAZ reports grants to his institution from AstraZeneca and grants from Bristol-Myers Squibb. JPD reports research support from AstraZeneca and Sanofi. DLB has served on advisory boards for Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, PhaseBio, and Regado Biosciences; has served on the boards of directors for Boston VA Research Institute, the Society of Cardiovascular Patient Care, and TobeSoft; has chaired the American Heart Association Quality Oversight Committee, the NCDR-ACTION registry steering committee, and the VA CART research and publications committee; has served on data monitoring committees for the Baim Institute for Clinical Research (for the PORTICO trial, funded by St Jude Medical), Cleveland Clinic (including for the ExCEED trial, funded by Edwards), Duke Clinical Research Institute, Mayo Clinic, Mount Sinai School of Medicine (for the ENVISAGE trial, funded by Daiichi Sankyo), and the Population Health Research Institute; has received honoraria from the American College of Cardiology, the Baim Institute for Clinical Research (RE-DUAL PCI clinical trial steering committee funded by Boehringer Ingelheim), Belvoir Publications, Duke Clinical Research Institute (for serving on clinical trial steering committees), HMP Global, the Journal of the American College of Cardiology, Medtelligence/ReachMD (for serving on continuing medical education [CME] steering committees), the Population Health Research Institute (for serving as the US national co-leader of COMPASS, funded by Bayer, and on the operations committee, publications committee, and steering committee), Slack Publications, the Society of Cardiovascular Patient Care (for serving as secretary and treasurer), and WebMD (for serving on CME steering committees); has served as the deputy editor for Clinical Cardiology; has received research funding from Abbott, Amarin, Amgen, AstraZeneca (including for serving on the DECLARE–TIMI 58 trial executive committee), Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Chiesi, Eisai, Ethicon, Forest Laboratories, Idorsia, Ironwood, Ischemix, Lilly, Medtronic, PhaseBio, Pfizer, Regeneron, Roche, Sanofi Aventis, Synaptic, and The Medicines Company; has received book royalties from Elsevier; has been a site co-investigator for Biotronik, Boston

Scientific, St Jude Medical, and Svelte; is a trustee for the American College of Cardiology; and has done unfunded research for FlowCo, Fractyl, Merck, Novo Nordisk, PLx Pharma, and Takeda. LAL reports grants and personal fees from AstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Merck, Novo Nordisk, and Sanofi; personal fees from Servier; and grants from GlaxoSmithKline. DKM reports personal fees from AstraZeneca, Boehringer Ingelheim, Janssen Research and Development, Sanofi US, Merck Sharp & Dohme, Lilly USA, Novo Nordisk, GlaxoSmithKline, Lexicon, Eisai, Esperion, Metavant, Pfizer, and Applied Therapeutics. JPHW reports grants, consultancy fees (paid to his institution), and personal fees for lectures and trial steering committee participation from AstraZeneca; grants, consultancy fees (paid to his institution), and personal fees for lectures from Novo Nordisk; consultancy fees (paid to his institution) and personal fees for lectures from Boehringer Ingelheim, Janssen, Napp, Mundipharma, Lilly, Takeda, and Sanofi; and consultancy fees (paid to his institution) from Wilmington Healthcare. ETK reports personal fees from Daiichi Sankyo, AstraZeneca, Bristol-Myers Squibb, and Tanabe-Mitsubishi Pharma and grants and personal fees from Ono. MSS reports research grant support through Brigham and Women’s Hospital from Abbott Laboratories, Amgen, AstraZeneca, Bayer, Daiichi-Sankyo, Eisai, Gilead, GlaxoSmithKline, Intarcia, Janssen Research and Development, The Medicines Company, MedImmune, Merck, Novartis, Poxel, Pfizer, Quark, Roche Diagnostics, and Takeda and consulting fees from Alnylam, Amgen, AstraZeneca, Bristol-Myers Squibb, CVS Caremark, Dyrnamix, Esperion, IFM Therapeutics, Intarcia, Ionis, Janssen Research and Development, The Medicines Company, MedImmune, Merck, MyoKardia, and Novartis. IR reports personal fees from AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Concenter BioPharma and Silkim, Eli Lilly, Merck Sharp & Dohme, Novo Nordisk, Orgenesis, Pfizer, Sanofi, SmartZyme Innovation, Panaxia, FuturRx, Insuline Medical, Medial EarlySign, CameraEyes, Exscopia, Dermal Biomics, Johnson & Johnson, Novartis, Teva, GlucoMe, and DarioHealth. IAMG-N, MF, PAJ and AML are employees of AstraZeneca. AR and IY declare no competing interests.

Data sharingIndividual participant data will not be made available. However, we encourage parties interested in collaboration to contact the corresponding author directly for further discussions.

AcknowledgmentsThe DECLARE–TIMI 58 trial was initially funded by AstraZeneca and Bristol-Myers Squibb; by the time of publication AstraZeneca was the sole funder. TAZ was supported by a research grant from Deutsche Forschungsgemeinschaft (ZE 1109/1–1).

References1 Roscioni SS, Heerspink HJL, de Zeeuw D. The effect of RAAS

blockade on the progression of diabetic nephropathy. Nat Rev Nephrol 2014; 10: 77–87.

2 Molitch ME, Adler AI, Flyvbjerg A, et al. Diabetic kidney disease: a clinical update from kidney disease: improving global outcomes. Kidney Int 2015; 87: 20–30.

3 Schievink B, Kröpelin T, Mulder S, et al. Early renin-angiotensin system intervention is more beneficial than late intervention in delaying end-stage renal disease in patients with type 2 diabetes. Diabetes Obes Metab 2016; 18: 64–71.

4 de Boer IH, Rue TC, Hall YN, Heagerty PJ, Weiss NS, Himmelfarb J. Temporal trends in the prevalence of diabetic kidney disease in the United States. JAMA 2011; 305: 2532–39.

5 Afkarian M, Sachs MC, Kestenbaum B, et al. Kidney disease and increased mortality risk in type 2 diabetes. J Am Soc Nephrol 2013; 24: 302–08.

6 Liyanage T, Ninomiya T, Jha V, et al. Worldwide access to treatment for end-stage kidney disease: a systematic review. Lancet 2015; 385: 1975–82.

7 United States Renal Data System. 2016 USRDS annual data report: epidemiology of kidney disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2016.

8 Wanner C, Inzucchi SE, Lachin JM, et al. Empagliflozin and progression of kidney disease in type 2 diabetes. N Engl J Med 2016; 375: 323–34.

Articles


9 Cherney DZI, Zinman B, Inzucchi SE, et al. Effects of empagliflozin on the urinary albumin-to-creatinine ratio in patients with type 2 diabetes and established cardiovascular disease: an exploratory analysis from the EMPA-REG OUTCOME randomised, placebo-controlled trial. Lancet Diabetes Endocrinol 2017; 5: 610–21.

10 Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017; 377: 644–57.

11 Perkovic V, de Zeeuw D, Mahaffey KW, et al. Canagliflozin and renal outcomes in type 2 diabetes: results from the CANVAS Program randomised clinical trials. Lancet Diabetes Endocrinol 2018; 6: 691–704.

12 Neuen BL, Ohkuma T, Neal B, et al. Cardiovascular and renal outcomes with canagliflozin according to baseline kidney function. Circulation 2018; 138: 1537–50.

13 Perkovic V, Jardine MJ, Neal B, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med 2019; published online April 14. DOI:10.1056/NEJMoa1811744.

14 Wiviott SD, Raz I, Bonaca MP, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med 2019; 380: 347–57.

15 Wiviott SD, Raz I, Bonaca MP, et al. The design and rationale for the Dapagliflozin Effect on Cardiovascular Events (DECLARE)–TIMI 58 Trial. Am Heart J 2018; 200: 83–89.

16 Raz I, Mosenzon O, Bonaca MP, et al. DECLARE-TIMI 58: participants’ baseline characteristics. Diabetes Obes Metab 2018; 20: 1102–10.

17 Schwandt A, Denkinger M, Fasching P, et al. Comparison of MDRD, CKD-EPI, and Cockcroft-Gault equation in relation to measured glomerular filtration rate among a large cohort with diabetes. J Diabetes Complications 2017; 31: 1376–83.

18 Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150: 604–12.

19 Ninomiya T, Perkovic V, de Galan BE, et al. Albuminuria and kidney function independently predict cardiovascular and renal outcomes in diabetes. J Am Soc Nephrol 2009; 20: 1813–21.

20 Gerstein HC, Mann JF, Yi Q, et al. Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals. JAMA 2001; 286: 421–26.

21 Hillege HL, Fidler V, Diercks GFH, et al. Urinary albumin excretion predicts cardiovascular and noncardiovascular mortality in general population. Circulation 2002; 106: 1777–82.

22 Scirica BM, Mosenzon O, Bhatt DL, et al. Cardiovascular outcomes according to urinary albumin and kidney disease in patients with type 2 diabetes at high cardiovascular risk. JAMA Cardiol 2018; 3: 155–63

23 Heerspink HJL, Kosiborod M, Inzucchi SE, Cherney DZI. Renoprotective effects of sodium-glucose cotransporter-2 inhibitors. Kidney Int 2018; 94: 26–39.

24 de Albuquerque Rocha N, Neeland IJ, McCullough PA, Toto RD, McGuire DK. Effects of sodium glucose co-transporter 2 inhibitors on the kidney. Diab Vasc Dis Res 2018; 15: 375–86.

25 DeFronzo RA, Norton L, Abdul-Ghani M. Renal, metabolic and cardiovascular considerations of SGLT2 inhibition. Nat Rev Nephrol 2017; 13: 11–26.

26 Heerspink HJL, Perkins BA, Fitchett DH, Husain M, Cherney DZI. Sodium glucose cotransporter 2 inhibitors in the treatment of diabetes mellitus: cardiovascular and kidney effects, potential mechanisms, and clinical applications. Circulation 2016; 134: 752–72.

27 Fioretto P, Zambon A, Rossato M, Busetto L, Vettor R. SGLT2 inhibitors and the diabetic kidney. Diabetes Care 2016; 39 (suppl 2): S165–71.

28 US Food and Drug Administration. FDA Drug Safety Communication: FDA strengthens kidney warnings for diabetes medicines canagliflozin (Invokana, Invokamet) and dapagliflozin (Farxiga, Xigduo XR). https://www.fda.gov/drugs/drugsafety/ucm505860.htm (accessed April 1, 2019).

29 Fadini GP, Solini A, Manca ML, et al. Effectiveness of dapagliflozin versus comparators on renal endpoints in the real world: a multicentre retrospective study. Diabetes Obes Metab 2019; 21: 252–60.

30 Cahn A, Melzer-Cohen C, Pollack R, Chodick G, Shalev V. Acute renal outcomes with sodium-glucose co-transporter-2 inhibitors: real-world data analysis. Diabetes Obes Metab 2019; 21: 340–48.

Effects of dapagliflozin on development and progression of ... · 09/06/2019 · Effects of...

Documents

Transcript of Effects of dapagliflozin on development and progression of ... · 09/06/2019 · Effects of...