CADTH Health Technology Review

Dapagliflozin for Chronic Kidney Disease

Rapid Review with Expert Input

Authors: Sara D. Khangura, Melissa Severn

Reviewers

External Reviewers

This document was externally reviewed by content experts and the following individuals granted permission to be cited.

Swapnil Hiremath, MD, MPH

Associate Professor

Faculty of Medicine, and School of Epidemiology and Public Health, University of Ottawa

Key Messages

• Recent, large, high-quality trials have demonstrated some benefits of dapagliflozin for the treatment of chronic kidney disease (most often in patients with type 2 diabetes) as compared to placebo.

• Data describing the clinical effectiveness of dapagliflozin have identified both relative benefits and no differences compared to placebo in various measures of renal and cardiovascular health and function, as well as health care utilization, mortality, and adverse events.

• A large proportion of the available data has been generated from the same randomized controlled trial that has recently been reported in multiple publications describing various patient subgroups and outcomes.

• No evidence was identified describing the cost-effectiveness of dapagliflozin for the treatment of patients with chronic kidney disease.

Context and Policy Issues

Chronic kidney disease (CKD) is a common condition, with estimates in the literature that 10% to 12% of the world's population lives with CKD.¹ This finding that is consistent with Canadian data indicating that approximately 10% of adults in Canada are living with the condition.² CKD contributes to reduced quality of life,³ and often progresses to kidney failure and death⁴; it is currently 1 of the most rapidly rising causes of death worldwide, with estimates suggesting that CKD could become the fifth most common global cause of death by the year 2040.^5,6

There are multiple risk factors for developing CKD, including older age, type 2 diabetes (T2D), obesity, ethnic origin and/or family history.⁷ CKD is 1 of several possible comorbid conditions (often co-occurring with cardiovascular [CV] disease), in as many as half of all patients with T2D.⁸ And while early intervention has been identified as an important mitigating factor for deleterious outcomes of CKD — including end-stage kidney disease (ESKD) and death — many individuals do not experience symptoms early in the course CKD.² This often leads to cases of CKD going undiagnosed, with data from 1 US study showing that as many as 22% of people with later stages 3 to 5 CKD may go undetected in the primary care system.⁷ Often, patients are not diagnosed with CKD until they experience CV symptoms, which is associated with later stages of the disease.⁹

The mainstay of current medical treatment for CKD has relied heavily on angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARB) for several decades.^7,10,11 Nonetheless, CKD persistently progresses to ESKD across time, despite currently available interventions, which emphasizes the need for more effective therapies.^10,12 Indeed, it has been highlighted in the literature that no new medical interventions have come available to mitigate the progression of CKD since the 1990s.¹³

Sodium-glucose co-transporter protein 2 inhibitors (SGLT2is) were developed for patients with T2D and aimed at reducing blood glucose and A1C levels in these patients; however, their beneficial effects on renal and CV outcomes in these patients have since led to the investigation of this class of drugs on cardiorenal outcomes in patients with CV and kidney diseases.¹¹ Notably, the protective effects of SGLT2is on kidney function have since been hypothesized to be independent of their effects on reducing glucose,^10,14 and so represent an important potential advancement in available treatments for CKD. In the literature, enthusiasm around the benefits of SGLT2is on renal and CV health is readily apparent, having been hailed as “a milestone discovery in medicine”(p. 6)⁹; having “revolutionized the treatment of cardiovascular and diabetic kidney disease,”(p. 335)¹ and; ushering in “a new era”¹⁵(p. 144)⁵(p. 1090) for patients with these diseases.

Of the available SGLT2is, several have been studied in the context of CKD, including dapagliflozin.¹⁶ The benefits of dapagliflozin in treating the dual epidemic of T2D with CV comorbidities have been recognized, and more recently, dapagliflozin has been approved by the US FDA for treating adults with CKD at high risk for disease progression.^13,17 Approval was also recently granted in the European Union for the use of dapagliflozin in patients with CKD (regardless of their diabetes status).¹⁸

Given the deleterious impacts of CKD among Canadians and recent advancements in the evidence describing the effects of dapagliflozin, there is a need to consult the available research literature to help inform health care and policy decision-making. Thus, the aim of this review is to identify, assemble and summarize available evidence describing the clinical effectiveness and cost-effectiveness of dapagliflozin for the treatment of CKD.

Research Questions

1. What is the clinical effectiveness of dapagliflozin for adults with chronic kidney disease?

2. What is the cost-effectiveness of dapagliflozin for adults with chronic kidney disease?

Methods

Literature Search Methods

A limited literature search was conducted by an information specialist on key resources including MEDLINE, Embase, the Cochrane Database of Systematic Reviews, the International Health Technology Assessment (HTA) Database, the websites of Canadian and major international health technology agencies, as well as a focused internet search. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s Medical Subject Headings (MeSH), and keywords. The main search concepts were dapagliflozin and CKD. No filters were applied to limit the retrieval by study type. Conference abstracts were removed from the search results. The search was also limited to English language documents published between January 1, 2016 and December 11, 2021.

Selection Criteria and Methods

One reviewer screened citations and selected studies. In the first level of screening, titles and abstracts were reviewed and potentially relevant articles were retrieved and assessed for inclusion. The final selection of full-text articles was based on the inclusion criteria presented in Table 1. Of note, studies were considered eligible regardless of any variation in the definitions used for CKD. All papers describing patients with CKD of any stage were considered eligible and included.

Exclusion Criteria

Articles were excluded if they did not meet the selection criteria outlined in Table 1, they were duplicate publications, or were published before 2021. Systematic reviews in which all relevant studies were captured in other more recent or more comprehensive systematic reviews were excluded. Primary studies retrieved by the search were excluded if they were captured in 1 or more included systematic reviews. Guidelines with unclear methodology were also excluded.

Critical Appraisal of Individual Studies

The included publications were critically appraised by 1 reviewer using the following tools as a guide: A MeaSurement Tool to Assess systematic Reviews 2 (AMSTAR 2)¹⁹ for systematic reviews and the Downs and Black checklist²⁰ for randomized and non-randomized studies. Summary scores were not calculated for the included studies; rather, the strengths and limitations of each included publication were described narratively.

Summary of Evidence

Quantity of Research Available

A total of 229 citations were identified in the literature search. Following screening of titles and abstracts, 165 citations were excluded and 64 potentially relevant reports from the electronic search were retrieved for full-text review. There were no relevant publications retrieved from the grey literature search for full-text review. Of the potentially relevant articles, 45 publications were excluded for various reasons, and 19 publications met the eligibility criteria for the review and were included in this report. These comprised 2 systematic reviews (SRs), 6 randomized controlled trials (RCTs) — 1 of which was reported across 11 eligible papers included in this review — and 1 non-randomized study. Appendix 1 presents the PRISMA²¹ flow chart of the study selection.

Additional references of potential interest are provided in Appendix 5.

Summary of Study Characteristics

Study Design

The SRs identified in this review were published in 2021²² and 2019.²³ Both SRs had broader eligibility criteria than those informing this review i.e., 1 SR sought studies investigating sodium-glucose co-transporter-2 inhibitors (SLGT2is) among patients with type 2 diabetes (T2D) — some of whom also had chronic kidney disease (CKD)²² — and the other SR sought studies investigating GLP-1 receptor agonists and SGLT2i in patients with both T2D and CKD.²³ Accordingly, both SRs included 1 RCT each that was eligible for inclusion and therefore summarized in this review.^22,23

The 1 RCT included in the 2021 SR²² was the Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease (DAPA-CKD) trial, which was also identified and reported in the 11 papers that met eligibility criteria for this review.^24-34 Whereas eligible primary studies that are included within eligible SRs are generally not otherwise included in the review, because the 11 papers contained a much larger amount of data and information than was reported in the SR, all of the papers describing DAPA-CKD were retained for inclusion in this report (including the SR describing the DAPA-CKD trial). The RCT included in the 2019 SR was also described in another eligible publication that conducted a longer-term, post-hoc analysis of the trial data (i.e., a greater amount of data and information) and so, was also included in this review.³⁵

The DAPA-CKD study was a randomized, placebo-controlled, double-blind trial with 11 publications identified and included in this review.^24-34 Dapagliflozin was also investigated by some of the same investigators in another randomized, placebo-controlled, double-blind trial — the Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure (DAPA-HF) study — which produced 1 paper examining the subgroup of patients in that study with CKD and so, was eligible for inclusion in this review.³⁶ Other RCTs included the dapagliflozin on proteinuria in non-diabetic patients with chronic kidney disease (DIAMOND) study, which was a placebo-controlled, double-blind crossover RCT³⁷; the dapagliflozin alone and in combination with saxagliptin and effect of dapagliflozin and saxagliptin on glycemic control in patients with type 2 diabetes and chronic kidney disease (DELIGHT) RCT, which also used a placebo-controlled and double-blind design³⁸; and the Dapagliflozin on Blood Glucose Level and Renal Safety in Patients With type 2 Diabetes (DERIVE) RCT, which similarly used a placebo-controlled, double-blind design.³⁹ Lastly, 1 non-randomized study (NRS) investigating dapagliflozin compared with empagliflozin was identified in this review, and used a longitudinal, retrospective cohort design.⁴⁰

Country of Origin

The SRs identified in this review were conducted in China²² and the US.²³ The long-term follow-up study³⁵ of the same RCT that was included in the 2019 SR²³ was conducted in 111 sites across 13 countries: US, Argentina, Canada, India, Mexico, Peru, Italy, Australia, France, Spain, Denmark, Puerto Rico, and Singapore.⁴¹

The DAPA-CKD study was conducted in 386 sites across 21 countries, which were not reported individually in the papers included in this review, but were reported in the published protocol: Argentina, Brazil, Canada, China, Denmark, Germany, Hungary, India, Japan, Mexico, Peru, Philippines, Poland, Russia, South Korea, Spain, Sweden, Ukraine, UK, US, and Vietnam.⁴² DAPA-HF was conducted in 410 sites across 20 countries³⁶ which were also not reported individually in the paper included in this review, but were available from the published protocol: Argentina, Brazil, Bulgaria, Canada, China, Czech Republic, Denmark, Germany, Hungary, India, Japan, Netherlands, Poland, Russia, Slovakia, Sweden, Taiwan, UK, US, and Vietnam.⁴³ The DIAMOND study was conducted in 6 sites across 3 countries: Canada, Malaysia and the Netherlands.³⁷ The DELIGHT RCT was conducted in 116 sites across 9 countries: Australia, Canada, Japan, South Korea, Mexico, South Africa, Spain, Taiwan, and the US.³⁸ Similarly, the DERIVE RCT was conducted in multiple sites across multiple countries: 88 sites in Bulgaria, Canada, Czech Republic, Italy, Poland, Spain, Sweden, and the US.³⁹ And finally, the NRS included in this review was conducted in Taiwan.⁴⁰

Patient Population

The 2021 SR sought data describing patients with T2D, some of whom also had CKD (i.e., patients in the 1 included RCT from the SR, which was the DAPA-CKD study) and so, were eligible for inclusion in this review; 4,304 study participants.²² The 2019 SR sought studies describing patients with T2D and CKD,²³ which was consistent with the patient eligibility criterion for this review, and the single RCT that was eligible from the SR examined 252 patients. Accordingly, the report of long-term findings from the 1 RCT included in the 2019 SR also described patients with T2D and CKD, but examined the subgroup of 166 patients with stage 3 CKD.³⁵

The DAPA-CKD study described adults with CKD — some of whom also had T2D and some of whom did not.^24-34 Several subgroup analyses were conducted and published, describing study patients by estimated glomerular filtration rate (eGFR) status (i.e., less than or greater than, or equal to 45 mL/min per 1.73 m²²⁶); cardiovascular (CV) disease status (i.e., with and without CV disease) and heart failure (HF) status (i.e., with and without HF)^28,29; glycemic status (i.e., normoglycemic, pre-diabetic and diagnosed T2D),³⁰ T2D status (i.e., with or without T2D)³²; stage 4 CKD (i.e., 624 patients)²⁴; focal segmental glomerulosclerosis (FSGS) (i.e., 104 patients)³¹; and; immunoglobulin A (IgA) nephropathy (i.e., 270 patients).³³

The DAPA-HF trial was necessarily focused on patients with HF, but the included paper by Jhund and colleagues reported on the subgroup of 1,926 patients with CKD (defined as an eGFR of < 60 mL/min per 1.73 m²).³⁶ The DIAMOND study recruited and assessed 53 non-diabetic patients with CKD.³⁷ The DELIGHT RCT included 293 patients with moderate to severe CKD and T2D receiving stable doses of an ACE inhibitor or and ARB.³⁸ The DERIVE trial included 321 patients with stage 3 CKD.³⁹ The NRS included in this review reported findings on 7,624 adult patients with both CKD and T2D.⁴⁰

Interventions and Comparators

All of the studies in the review examined dapagliflozin at a dosage of 10 mg daily^22-40 and 2 also included data describing dapagliflozin at a dosage of 5 mg daily.^23,35 In addition to investigating the safety and efficacy of dapagliflozin (10 mg) versus placebo, 1 RCT also evaluated a third group of patients who received combination therapy including dapagliflozin (10 mg/day) and saxagliptin³⁸; however, data from this arm of the trial were not included or summarized in this review, which excluded combination therapy.

The comparator groups described in all but 1 of the studies⁴⁰ included in this review received a placebo.^22-39 The 1 NRS included in this review compared dapagliflozin to empagliflozin (i.e., another SGLT2i) at dosages of 10 mg or 25 mg.⁴⁰

Outcomes

All of the studies included in this review described outcomes of relevance to the efficacy and/or safety of dapagliflozin in CKD, that is, CV and/or renal complications; health care utilization; mortality and/or; adverse events (including serious adverse events).^22-40

Outcomes describing the efficacy of dapagliflozin included those of renal function; for example, changes in eGFR (measured using mL/min/1.73 m²)^{23,24,26,28-31,34,38-40}; outcomes describing end-stage kidney disease (ESKD)^{23,24,28-30,34,35,37-40}; as well as those describing composite and other measures of renal function.^{24,25,28-34,36,40} Similarly, multiple studies reported on outcomes of relevance to CV function; that is, risk of atrial fibrillation (AF) and stroke.^{23,24,26,28-31,34,38-40}

Additional outcomes of relevance to this review included health care utilization (measured by hospitalizations)^28,29,36; and mortality (i.e., numbers of patients who died).^{24,27-29,32,34-36,38} The comparative safety of dapagliflozin was also reported in many of the papers by describing adverse events (AEs)^{24,28,29,31-39} and serious adverse events (SAEs).^{24,25,28,29,32-39}

Details regarding the characteristics of included publications are provided in Appendix 2.

Summary of Critical Appraisal

Systematic Reviews

The 2021 SR demonstrated both strengths and limitations. The methods described were generally sound, including clear eligibility criteria and a comprehensive search — the latter of which is a critical component of a well-conducted SR, as it assures the reader that efforts have been made to identify a maximum amount of eligible information. In addition, authors described duplicate screening and data abstraction, which are important features of SRs. Duplicate screening reduces the potential for bias and error, and helps to ensure that all eligible studies are identified and included in the review. Similarly, duplicate data abstraction is important for ensuring that data have been accurately and comprehensively identified. As well, the 2021 SR reported appropriate statistical methods for meta-analyses, and an assessment of both risk of bias and of publication bias.²² However, the report was limited in its description of a rationale for the selection of study designs. A description of excluded studies (including the rationale behind their exclusion) was missing, and there was an absence of any mention of a protocol or reference to an a priori design of review methods and criteria.²² These features are important in systematic reviews to ensure transparency of the methods and reproducibility of the findings.⁴⁴

The 2019 SR had many limitations and few strengths.²³ The PICO (population, intervention, comparator and outcomes) criteria were made clear, which is important for framing and establishing the aim and research question(s) of a review. In addition, information about the included studies was sufficiently detailed. However, there was no reference to a protocol, the literature search was limited by lack of a description of a grey literature search, no rationale was provided for the selection of study designs, no description of duplicate screening of citations or data abstraction, and no mention of an assessment of risk of bias of publication bias.²³ For instance, a clear and comprehensive assessment of the risk of bias for included studies is a key component of all SRs,¹⁹ because understanding the quality of a study is an important part of weighing the value and contribution of the evidence it provides in answer to a research question (i.e., a higher quality of evidence provides more confidence in the findings reported, while lower levels of evidence are cause for caution in the interpretation of findings). The limitations of the 2019 SR included in this review introduce uncertainty as to the extent of the review’s utility, as the methods undertaken were not clearly rigorous, drawing into question whether the review may be biased in the studies it summarized and the findings it produced.

Randomized Controlled Trials

Most of the RCTs demonstrated many strengths and few limitations, with generally clear descriptions of robust methods and clear reporting of findings.^24-39 The DAPA-CKD RCT, in particular, demonstrated few limitations, with generally clear reporting and little evidence of threats to internal validity.^24-34 Nonetheless, while there was insufficient information reported in any of the reports to adequately assess the extent to which external validity was robust, because the trial employed a multinational, multi-centre design with thousands of patients across the world, there is a reasonable level of confidence in the external validity of the findings. Further, while a power calculation describing sufficient study power in the full set of study patients was provided,^25,26,34 several of the papers describing subgroup analyses from the trial either did not address power specific to their subgroup analyses,^24,27-30,32 or acknowledged that the analyses had insufficient power to demonstrate a clinically important difference between treatment groups.^31,33 Finally, with 11 papers published in 2 years (as identified by this review), 1 criticism of the DAPA-CKD trial reporting may be that the authors engaged in “salami slicing,” that is, the publication of 1 trial across multiple papers.⁴⁵ This practice has been described as 1 that can be used as self-serving on the part of study co-authors (e.g., increasing the number of journal publications to support career advancement)⁴⁶ and may be problematic if it distorts the findings of the study (e.g., introducing the opportunity for “cherry picking” of data and potentially compromising the power of the analyses to detect a clinically important effect), and/or; diverging from the statistical plan by generating post-hoc analyses that may not have been pre-specified.⁴⁵ Importantly, salami slicing may legitimately be used to manage the reporting of large studies and/or datasets that cannot reasonably be described in 1 paper.⁴⁶ In the case of the DAPA-CKD trial, the publication of multiple papers describing subgroup analyses is unlikely to have introduced an important risk of bias, because the sufficiently powered findings from the overall trial data demonstrate a significant benefit of the intervention.³⁴ Nonetheless, the number of papers generated from the DAPA-CKD trial (as identified here) is arguably high, with some of the analyses explicitly described as being post hoc, and representing a potential source of bias in the reporting of results for this RCT.^28,30,31

Four of the 5 remaining RCT reports included in this review were generally reported clearly, including sufficient detail and demonstrating several strengths and few limitations.^36-39 The papers were clear in their description of the study aims, patients, interventions and outcomes, and demonstrated features of internal and external validity i.e., randomized, double-blind, placebo-controlled designs conducted in multiple sites in multiple countries.^36-39 Although 3 of the reports described a power calculation demonstrating the extent to which the sample size was sufficient to detect a clinically important difference between treatment groups,^37-39 the report describing a sub-analysis of findings from the DAPA-HF trial did not.³⁶ Nonetheless, the main report of findings for the DAPA-HF trial did describe sufficient power to detect a statistically significant difference between the treatment groups in the primary outcome.^47,36 One of the RCTs used a crossover design (i.e., 6 weeks of dapagliflozin followed by a 6-week washout period and then 6 weeks of placebo, or vice versa),³⁷ which can introduce the risk of aliasing (i.e., that effects from the first intervention may carry over to the time period during which the next intervention is being assessed, even with a washout period), and contribute to the potential for type II error.⁴⁸ The fifth report describing a secondary analysis of trial data demonstrated some important limitations that render it of limited value. There was insufficient detail provided describing the methods used in the RCT from which the data were taken, with a reference to the main trial paper, leaving the reader without access to the information.³⁵ In addition, no information on the representativeness of the patients assessed was provided, preventing the reader from considering this as a threat to external validity.³⁵ While the authors acknowledged that their analyses were post hoc, there were insufficient details provided to assess the potential for bias and confounding; for example, simple outcome data (i.e., numerators and denominators) were not reported; statistical methods were not described in detail, and; no description of the power of the analysis to detect a clinically important difference between treatment groups was provided.³⁵

Finally, it was noted that all of the RCTs included in this review were funded by the same for-profit, private industry pharmaceutical manufacturer,^24-39 which manufactures dapagliflozin under the brand name Farxiga. This may or may not introduce risk of bias; for instance, 1 paper in this review reporting subgroup analyses from the DAPA-HF study described oversight by an academic team not employed by the pharmaceutical manufacturer.³⁶ External (and presumably objective) oversight of an RCT generally represents a strength of the study. Nonetheless, conflict of interest statements for several of these academic co-authors acknowledged the receipt of funds from the same for-profit, private industry pharmaceutical manufacturer in the form of speaking and/or consultation fees, as well as grant monies.³⁶

While it is beyond the scope of this Rapid Review to investigate the extent to which the sole-reported source of funding for all RCTs eligible for and included in this review^24-39 may have introduced a risk of bias to the findings summarized herein, it remains an important consideration when assessing the possible impact that a conflict of interest may have on the potential for risk of bias as it concerns these included studies.⁴⁹

Non-Randomized Study

The NRS demonstrated both strengths and limitations. There was a clear report of the aim, study objectives, patient characteristics, interventions, potential confounders, and estimates of random variability.⁴⁰ All eligible patients from a large, regional database were included in the analyses, which contributes to the confidence that can be placed in the external validity of the findings.⁴⁰ The findings were generated from planned analyses, with data from patients being observed across the same time period and adjustments made to account for potentially confounding factors⁴⁰ — all of which contribute to the confidence that can be placed in the study’s internal validity. Nonetheless, some limitations were apparent as well; most importantly, the study was necessarily not randomized by virtue of its retrospective, observational design, which introduces a threat to the internal validity of the findings.⁴⁰ In addition, some details were either not reported or not clearly reported, including simple outcome data and actual P values for some outcomes, as well as adverse events that were missing from the repot of findings.⁴⁰ And while the study reported a large sample size, there was no discussion about the power of the study to detect a clinically important difference between the treatment groups.⁴⁰

Additional details regarding the strengths and limitations of included publications are provided in Appendix 3.

Summary of Findings

Clinical effectiveness of dapagliflozin for adults with chronic kidney disease

Renal Health/Function

Changes in eGFR

Eleven of the 19 papers included in this review reported on changes in eGFR (in mL/min/1.73 m²)^{23,24,26,28-31,34,38-40}; 7 of which reported findings from the DAPA-CKD trial.^{24,26,28-31,34} The primary report of findings describing data for all of the 4,304 study patients by treatment group found that statistically significantly fewer patients experienced a decline of at least 50% in eGFR among the dapagliflozin (10 mg) group (i.e., 112/2,152; 5.2%) as compared to those receiving placebo (i.e., 201/2,152; 9.3%), producing a comparative hazard ratio (HR) of 0.53 (95% CI, 0.42 to 0.67) that favoured dapagliflozin (10 mg).³⁴ This statistically significant improvement favouring patients receiving dapagliflozin (10 mg) as compared to placebo was also observed in several subgroup analyses of DAPA-CKD patients.^26,28,29,31 On the other hand, no significant difference in the number of patients experiencing a decline of at least 50% eGFR was found between dapagliflozin (10 mg) and placebo in several other subgroup analyses.^26,31

Other RCTs examining changes in eGFR between dapagliflozin (10 mg) and placebo in patients with CKD and T2D produced similar findings favouring dapagliflozin (10 mg). One RCT including data describing 293 patients with moderate to severe CKD and T2D found a statistically significant difference in change of mean eGFR from baseline to 24 weeks favouring dapagliflozin (10 mg) as compared to placebo; that is −2.35 mL/min/1.73 m² (95% CI, −4.16 to −0.53, P = 0.011).³⁸ Similarly, another trial comparing dapagliflozin (10 mg) with placebo in 321 patients with stage 3 CKD reported a statistically significant difference in change of mean eGFR from baseline to 24 weeks favouring dapagliflozin (10 mg) over placebo; that is −2.49 mL/min/1.73 m² (95% CI, −1.59 to −0.02).³⁹ One RCT reported in an SR described changes in mean eGFR across 24 weeks of follow-up for each treatment group only (i.e., no comparative statistics reported), finding a difference of −4.80 mL/min/1.73 m² in patients receiving dapagliflozin (10 mg), −2.38 mL/min/1.73 m² in patients receiving dapagliflozin (5 mg) and −0.25 mL/min/1.73 m² in the placebo group, with authors narratively reporting no statistically significant difference between the groups. Similarly, the 1 NRS included in this review reported no statistically significant difference in mean eGFR between patients receiving dapagliflozin (10 mg) as compared to empagliflozin (10 mg); that is P = 0.145 or empagliflozin (25 mg) i.e., P = 0.217.⁴⁰

ESKD

There were 5 papers that reported on the occurrence of ESKD, all of which used data from the DAPA-CKD trial.^24,28-30,34 The main report describing all 4,304 DAPA-CKD patients found fewer patients with ESKD at the end of follow-up in those receiving dapagliflozin (10 mg); that is 109/2,152 (5.1%) as compared to those receiving placebo; that is 161/2,152 (7.5%).³⁴ This difference between treatment groups was statistically significant i.e., HR 0.64 (95% CI, 0.50 to 0.82).³⁴ In an analysis of sub-components of the ESKD outcome, statistically significant benefits were found in the dapagliflozin (10 mg) group with regard to the number of patients experiencing an eGFR of less than 15 mL/min/1.73 m² (i.e., HR 0.67 [95% CI, 0.51 to 0.88]) and long-term dialysis (i.e., HR 0.66 [95% CI, 0.48 to 0.90]).³⁴ The 4 remaining papers reporting on the occurrence of ESKD described subgroup analyses from DAPA-CKD with variable findings reported; some of which were concordant with the statistically significant benefit of dapagliflozin (10 mg) found in the main trial, and some of which were not.^24,28-30 (Table 6)

Composite and Other Measures of Kidney Function

Ten publications described composite and other measures of kidney function^{24,28-34,36,40}; 8 of these reported data from the DAPA-CKD trial^24,28-34 and 2 reported data from other RCTs.^36,40

The primary and secondary outcomes from the DAPA-CKD trial were both composed of several component outcomes. The primary outcome was a composite of the number of patients experiencing a first occurrence of decline in eGFR of at least 50%, ESKD, or death from CV or renal causes; the secondary outcome was similar to the primary outcome, but did not include death from CV causes (i.e., the number of patients experiencing a first occurrence of decline in eGFR of at least 50%, ESKD, or death from renal causes). Both outcomes were included in the main report for the trial describing all 4,304 study patients by treatment group only, with investigators finding statistically significantly fewer patients in the dapagliflozin (10 mg) as compared to the placebo group experiencing the primary outcome; that is HR 0.61 (95% CI, 0.51 to 0.72, P < 0.001).³⁴ Similarly, the secondary composite outcome demonstrated a benefit of dapagliflozin (10 mg) versus placebo with statistically significantly fewer patients experiencing a first occurrence of any of the outcome components; that is HR 0.56 (95% CI, 0.45 to 0.68, P < 0.001).³⁴ The benefit of dapagliflozin (10 mg) was also reported in several additional papers from the DAPA-CKD trial examining subgroups of patients .Statistically significantly fewer patients receiving dapagliflozin (10 mg) experienced the primary or secondary composite outcomes as compared to those receiving placebo, regardless of the presence or absence of comorbid CV disease²⁸; HF²⁹; or T2D.³² Several additional subgroup analyses also indicated a significant benefit of dapagliflozin (10 mg) as compared to placebo in both the primary and secondary composite outcomes (i.e., among DAPA-CKD patients with Stages 2 or 3 CKD²⁴; IgA nephropathy³³; diabetic nephropathy or glomerulonephritis.³²) Similarly, another paper included analyses from the primary composite outcome only, and likewise found statistically significantly fewer patients among those receiving dapagliflozin (10 mg) versus placebo with pre-diabetes (i.e., HR 0.37 [95% CI, 0.21 to 0.66]) or T2D (i.e., HR 0.64 [95% CI, 0.52 to 0.79]).³⁰ Nonetheless, several subgroup analyses found no statistically significant difference between the treatment groups in either the primary or secondary composite outcomes, including DAPA-CKD patients with stage 4 CKD²⁴; normoglycemia³⁰; FSGS³¹; ischemia or hypertension.³²

The DAPA-HF trial also reported on the same composite outcome as was reported in the DAPA-CKD trial as a secondary outcome; that is patients experiencing a first occurrence of decline in eGFR of ≥ 50%, ESKD, or death from renal causes.³⁶ However, the group of patients that were eligible for inclusion in this review had HF and reduced ejection fraction (as well as CKD with or without T2D), with authors reporting a non-significant difference between dapagliflozin (10 mg) and placebo i.e., HR 0.95 (95% CI, 0.50 to 1.82).³⁶

Other measures of renal health and/or function reported from the DAPA-CKD trial included a detailed analysis of abrupt decline in kidney function, defined as a doubling of serum creatinine between study visits (median interval of 100 days).²⁵ Investigators found statistically significantly fewer patients receiving dapagliflozin (10 mg) with an abrupt decline in kidney function as compared to patients receiving placebo i.e., HR 0.68 (95% CI, 0.49 to 0.94; P = 0.02).²⁵ Authors also conducted subgroup analyses, finding a statistically significant benefit of dapagliflozin (10 mg) by number of events observed (per 100 patient-years) in subgroups of patients who were older than 65 years (but not in the subset 65 years of age and younger); female (but not in male patients); diagnosed with T2D or not; found to have an eGFR of lower than 45 mL/min/1.73 m² (but not in patients with an eGFR of at least 45 mL/min/1.73 m²), as well as those not diagnosed with HF (but not patients diagnosed with HF).²⁵

The NRS included in this review described the difference in mean serum creatinine (mg/dL) from baseline to at least 28 days among patients with CKD and T2D, reporting a significant benefit of empagliflozin (10 mg) as compared to dapagliflozin (10 mg); that is P = 0.010, but no statistically significant difference between empagliflozin (25 mg) and dapagliflozin (10 mg); that is P = 0.163.⁴⁰

Cardiovascular Health/Function

AF

One RCT described in 1 SR reported on the risk of AF in 4,304 patients with CKD (with or without T2D), reporting an OR of 0.47 (95% CI, 0.2 to 1.09) between dapagliflozin (10 mg) and placebo, indicating no statistically significant difference between the groups.

Composite and Other Measures of Cardiovascular Health/Function

The main report from the DAPA-CKD trial describing all 4,304 study patients reported findings from 1 secondary composite outcome including CV components; that is hospitalization for HF or death from CV causes.³⁴ Study authors reported a significant difference between groups favouring dapagliflozin (10 mg) compared to matching placebo; that is hazard ratio (HR) 0.71 (95% CI, 0.55 to 0.92, P = 0.009).³⁴ Related subgroup analyses of this same outcome from the DAPA-CKD RCT were reported in several additional papers, producing variable results across patient characteristics; that is in patients with stage 2 or 3 CKD²⁴; CV disease²⁹; without HF,²⁹ or with T2D,³² a significant benefit of dapagliflozin (10 mg) was found as compared to placebo; whereas no statistically significant difference between dapagliflozin (10 mg) and placebo was observed in this outcome among patients with stage 4 CKD²⁴; without CV disease²⁹; without T2D, or³²; with HF.²⁹ Notably, the DAPA-HF trial reported on a similar composite outcome (i.e., worsening HF/hospitalization for HF or CV death) in 1,926 patients with HF and CKD (defined as an eGFR of < 60 mL/min per 1.73 m2), finding a significant benefit of dapagliflozin (10 mg) relative to placebo; that is HR 0.72 (95% CI, 0.59 to 0.86).³⁶ This apparent discrepancy in findings concerning the effectiveness of dapagliflozin (10 mg) among patients with CKD and HF may be due to the smaller number of patients included in the subgroup analysis of the DAPA-CKD trial (i.e., N = 468) as compared the number of patients in DAPA-HF, widening the CI in the former study of this subgroup of patients and rendering the finding not statistically significant. Thus, more confidence can be placed in the finding from the DAPA-HF finding due to its assessment of a larger group of patients.

The paper examining data from DAPA-CKD trial patients by CV disease status reported on several additional composite measures of CV outcomes, including a pre-specified exploratory investigation of MI, stroke or death from CV causes, and a post-hoc exploratory analysis of MI, stroke, hospitalization for HF or death from CV causes, as well as MI, stroke, hospitalization for heart failure, ESKD or death from any cause.²⁹ Whereas no significant differences between dapagliflozin (10 mg) and matching placebo were found in the first 2 of these 3 composite measures for either patients with or without CV disease, a significant benefit in favour of dapagliflozin (10 mg) was reported in the latter outcome in both patients with (HR: 0.72 [95% CI, 0.58 to 0.89]) and without CV disease (HR: 0.68 [95% CI, 0.54 to 0.85]).²⁹

One RCT reported on change in hematocrit ratio from baseline to 24 weeks in 293 patients with moderate to severe CKD and T2D, finding a statistically significant benefit in favour of dapagliflozin (10 mg) compared to placebo i.e., difference in mean percentage 0.03 (95% CI, 0.02 to 0.04), P < 0.0001.³⁸

Stroke

One RCT described in 1 SR reported on the risk of stroke in 4,304 patients with CKD (with or without T2D), reporting an odds ratio (OR) of 0.86 (95% CI, 0.51 to 1.47) between dapagliflozin (10 mg) and placebo, indicating no statistically significant difference between the groups.

Health Care Utilization

Hospitalization

Three reports from 2 RCTs reported on outcomes including hospitalization for HF in patients with CKD and CV disease (with or without T2D).^28,29,36 The DAPA-CKD trial found statistically significantly fewer patients receiving dapagliflozin (10 mg) were hospitalized, or experienced an urgent visit for HF as compared to those receiving placebo; that is HR 0.66 (95% CI, 0.52 to 0.83).³⁶ Similarly, subgroup analyses from the DAPA-CKD trial found that statistically significantly fewer patients receiving dapagliflozin (10 mg) experienced a first hospitalization for HF compared with those receiving placebo, whether or not they had CV disease or HF.^28,29 Finally, the DAPA-HF trial likewise found a statistically significant benefit favouring dapagliflozin (10 mg) versus placebo in a composite outcome measuring total hospitalizations for HF or death from CV causes; that is HR 0.79 (95% CI, 0.64 to 0.97).³⁶

Mortality

Nine reports of RCT data described mortality among patients with CKD, with or without T2D or HF^{24,27-29,32,34-36,38}; 6 of which reported findings from the DAPA-CKD trial,^{24,27-29,32,34} and 2 reported data from other RCTs.^35,36,38

All-Cause Mortality

Both the main report from the DAPA-CKD RCT and a sub-analysis focusing on mortality as the sole outcome of interest found statistically significantly fewer of the 4,304 patients in the study died from any cause in the dapagliflozin (10 mg) group as compared to those receiving placebo; that is HR 0.69 (95% CI, 0.53 to 0.88), P = 0.004).^27,34 The paper describing a detailed sub-analysis of mortality also reported findings on all-cause mortality by various subgroups, observing statistically significantly fewer patients in the dapagliflozin (10 mg) group who died from any cause as compared to placebo among those who were older than 65 years of age; male; either had T2D or not; had an eGFR of less than 45 mL/min/1.73 m²; had a urine albumin-creatinine ratio (UACR) of greater than 1,000 mg/g or at least 1,000 mg/g; had a systolic blood pressure (SBP) of greater than 130 mm Hg, or; had a serious infection.²⁷ Conversely, there was no statistically significant difference in all-cause mortality found between the treatment arms in subgroups of patients who were 65 years of age or younger; female; had an eGFR of at least 45 mL/min/1.73 m²; had a SBP of 130 mm Hg or less, or; had a serious malignancy.²⁷ Chertow and colleagues also reported on death from any cause in DAPA-CKD patients by stage of disease, finding no statistically significant benefit of dapagliflozin (10 mg) as compared to placebo in patients with stage 4 CKD i.e., HR 0.68 (95% CI, 0.39 to 1.21) but a statistically significant benefit of dapagliflozin (10 mg) in patients with stages 2 and 3 CKD i.e., HR 0.69 (95% CI, 0.52 to 0.92).²⁴ Two additional subgroup analyses by CV disease and HF status reported by McMurray and colleagues also found statistically significantly fewer patients receiving dapagliflozin (10 mg) died from any cause as compared to placebo, in both patients with and without CV disease or HF.^28,29 Likewise, another subgroup analysis of DAPA-CKD patients by T2D status confirmed the finding summarized above from Heerspink and colleagues²⁷ observing a statistically significant benefit of dapagliflozin (10 mg) as compared to placebo in all-cause mortality for patients with or without T2D.³² Similarly, additional subgroup analyses of patients identified statistically significantly fewer patients with diabetic nephropathy or glomerulonephritis experienced all-cause mortality with dapagliflozin (10 mg) as compared to placebo; however, there was no difference between treatment groups found in patients with ischemia or hypertension, or those with another or unknown cause of CKD.³²

All-cause mortality was also reported in the DAPA-HF trial, with the authors reporting no significant difference between the treatment groups; that is a HR 0.85 (95% CI, 0.68 to 1.07).³⁶ And in 2 RCTs, deaths were reported as an adverse event; with 1 death in the dapagliflozin (10 mg) group and no deaths in the placebo group in 1 RCT, and³⁸; 3 in the dapagliflozin (10 mg) group, 1 in the dapagliflozin (5 mg) and 4 in the placebo group,³⁵ with no characterization of the difference between groups was described in either of these latter 2 trials.^35,38

Death From Renal Causes

Both the main report from the DAPA-CKD RCT and a sub-analysis focusing on mortality as the sole outcome of interest reported no statistically significant difference in death from renal causes (i.e., kidney failure) among patients receiving dapagliflozin (10 mg) as compared to those receiving placebo, that is 2 of 2,152 in the dapagliflozin (10 mg) group and 6 of 2,152 in the placebo group (HR 0.35 [95% CI, 0.07 to 1.73]).^27,34 These data were also reported in subgroup analyses of patients by CV disease and HF status, but the numbers per group were too small to compare statistically and so, no characterization of the differences between groups was reported.^27,34

Death From CV Causes

The main report of findings from the DAPA-CKD trial found no statistically significant difference between dapagliflozin (10 mg) and placebo in the full set of 4,304 study patients i.e., HR 0.81 (95% CI, 0.58 to 1.12).³⁴ This lack of difference between dapagliflozin (10 mg) and placebo in CV death was also found in a detailed analysis of mortality (i.e., all CV deaths)²⁷; subgroup analyses of patients with or without CV disease or HF,^28,29 and; in DAPA-HF patients i.e., 0.88 (95% CI, 0.69 to 1.13).³⁶ Similarly, a detailed investigation of CV deaths in DAPA-CKD patients found no difference between treatment groups in patients experiencing sudden cardiac death, acute MI or stroke; but did find statistically significantly fewer patients who died from HF in the dapagliflozin (10 mg) as compared to placebo; that is HR 0.27 (95% CI, 0.08 to 0.98).²⁷

Death From Non-CV Causes

The detailed report of mortality in DAPA-CKD patients also reported all deaths caused by reasons other than CV and found significantly fewer non-CV deaths among patients receiving dapagliflozin (10 mg) as compared to placebo i.e., HR 0.54 (95% CI, 0.36 to 0.82).²⁷ A sub-analysis of components of this outcome identified malignancy as a likely driver of this statistically significant difference; that is HR 0.42 (95% CI, 0.19 to 0.97), while no statistically significant difference was found between treatment groups in deaths from infection or kidney failure.²⁷ Similarly, no significant difference was found between dapagliflozin (10 mg) and placebo in deaths with no determined cause; that is HR 0.80 (95% CI, 0.47 to 1.38), P = 0.426.²⁷

Safety

Adverse Events

Thirteen reports described adverse events (AEs) in patients with CKD^24,28-39; 8 of which described data from the DAPA-CKD trial^24,28-34 and 5 described data from other RCTs.^35-39 The main report of findings from the full set of study patients in the DAPA-CKD trial found no statistically significant difference between the numbers of patients in either treatment group experiencing renal events, bone fractures, amputations, diabetic ketoacidosis, or discontinuing study medication; however, there were more patients in the dapagliflozin (10 mg) group (5.9%) who experienced symptoms of volume depletion as compared to those in the placebo group (4.2%), P = 0.01.³⁴ A subgroup analysis of the DAPA-CKD patients by stage of CKD produced similar findings, with no statistically significant difference identified between treatment groups in either the stage 2 and 3 or stage 4 CKD patients for most of the AEs that were assessed in the main report of findings (i.e., bone fractures, amputations, diabetic ketoacidosis, or discontinuation of study medication).²⁴ However, volume depletion was experienced by statistically significantly more patients in the dapagliflozin (10 mg) group as compared to placebo in the stage 2 and 3 patients (with no statistically significant difference in stage 4 patients between treatment groups).²⁴ Likewise, renal AEs were experienced by statistically significantly more stage 2 and 3 patients receiving dapagliflozin (10 mg) as compared to placebo (with no statistically significant difference in stage 4 patients between treatment groups).²⁴ Similarly, another subgroup analysis of the DAPA-CKD patients by T2D status also produced similar findings, with no statistically significant difference identified between treatment groups in either the patients with or without T2D for most of the AEs that were assessed in the main report of findings (i.e., fractures, amputations, diabetic ketoacidosis, kidney-related AEs or discontinuation of study medication).³² However, volume depletion was experienced by statistically significantly more patients in the dapagliflozin (10 mg) group as compared to placebo in the patients without T2D (with no statistically significant difference in T2D patients between treatment groups).³² On the other hand, the total number of any AE were experienced by statistically significantly more T2D patients receiving placebo as compared to dapagliflozin (10 mg) (with no statistically significant difference in patients without T2D between treatment groups).³² The remaining papers describing subgroup analyses from the DAPA-CKD trial did not characterize the difference between treatment groups, reporting only raw numbers of patients per group and making interpretation of the treatment comparison less clear; data from these papers are detailed in Appendix 4.^28,29,31,33

The DAPA-HF trial assessed the same AEs as were assessed in the DAPA-CKD trial, with no statistically significant difference found between treatment groups for any of the AEs.³⁶ The remaining studies did not characterize the difference between treatment groups, reporting only raw numbers of patients per group and making interpretation of the treatment comparison less clear; data from these papers are detailed in Appendix 4.^35,37-39

Serious Adverse Events

Twelve papers described SAEs in patients with CKD^{24,25,28,29,32-39}; 7 of which reported data from the DAPA-CKD trial^{24,25,28,29,32-34} and 5 of which reported data from other RCTs.^35-39

The main report of findings from the full set of study patients in the DAPA-CKD trial found statistically significantly more patients in the placebo group experiencing any SAE (i.e., P = 0.002) or a major episode of hypoglycemia (i.e., P = 0.04) as compared to those in the dapagliflozin (10 mg) group.³⁴ Similarly, the subgroup analysis of the DAPA-CKD patients by stage of CKD indicated a statistically significant difference between treatment groups in the stage 2 or 3 patients, with fewer patients experiencing any SAE or an episode of major hypoglycemia; however, no statistically significant difference was observed in the stage 4 CKD patients between treatment groups.²⁴ Similarly, another subgroup analysis of the DAPA-CKD patients by T2D status also produced similar findings, with statistically significantly more patients with T2D receiving placebo experiencing an episode of major hypoglycemia.³² One paper focused on abrupt decline in kidney function and acute kidney injury (AKI) in DAPA-CKD patients, reporting no significant difference between treatment groups in AKI-related SAEs; that is HR 0.77 (95% CI, 0.54 to 1.10), P = 0.15.²⁵ The remaining papers describing subgroup analyses from the DAPA-CKD trial did not characterize the difference between treatment groups, reporting only raw numbers of patients per group and making interpretation of the treatment comparison less clear; data from these papers are detailed in Appendix 4.^28,29,33

The DAPA-HF found statistically significantly more SAEs in patients receiving placebo as compared to those receiving dapagliflozin (10 mg); P = 0.003.³⁶ The remaining studies did not characterize the difference between treatment groups, reporting only raw numbers of patients per group and making interpretation of the treatment comparison less clear; data from these papers are detailed in Appendix 4.^35,37-39

Cost-Effectiveness Of Dapagliflozin for Adults With Chronic Kidney Disease

Because there were no studies identified assessing the cost-effectiveness of dapagliflozin in adult patients with CKD, no summary can be provided.

Appendix 4 presents the main study findings by outcome.

Limitations

This review identified a large number of publications describing the clinical effectiveness of dapagliflozin, but is limited by the lack of available evidence describing the cost-effectiveness of dapagliflozin, as no eligible economic evaluations were identified.

Many of the papers in this review came from the same trial, which indicates that the evidence base describing dapagliflozin for CKD could be smaller than it may appear. This seems to be consistent with the SRs that were included in this review;^22,23 that is, there were few studies included that addressed the use of dapagliflozin in patients with CKD, and there was some overlap between the studies included in the eligible SRs with the primary studies included in this review (though, the data reported in the SRs were very limited compared to those described in the primary study reports).

While this report was not focused on a particular definition or stage of CKD, the studies included focused on patients with stages 3 and 4 CKD, providing informative analyses of the effects of dapagliflozin across a spectrum of disease severity. Nonetheless, research describing a broader range — and more specific subgroups — of patients, dosages of dapagliflozin and/or alternative comparisons, and a wider variety of outcomes may provide additional, useful, and important insights into the clinical effectiveness of dapagliflozin, as well as considerations for implementing its use into clinical practice. For instance, the limitations of the characteristics of the DAPA-CKD study population, which currently represents the largest and most current available dataset describing dapagliflozin in CKD patients, have been highlighted in the literature (i.e., only patients with proteinuria were included⁴; without type 1 diabetes, and; without other forms of CKD; for example those with polycystic kidney disease, lupus nephritis and anti-neutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis).⁹ Data describing these kinds of patients were also not described in the other studies included in this review and represent opportunities for continued research into the clinical effectiveness of dapagliflozin. And while the DAPA-CKD trial did include patients with non-diabetic CKD, there have been recent calls in the literature for additional research investigating the effects of SGLT2is in this population, as well.¹⁴

In addition, while almost all of the studies summarized in the review provided data describing the comparison between dapagliflozin and placebo, there was only 1 study that described dapagliflozin in comparison with another SGLT2i (i.e., empagliflozin),⁴⁰ which retrospectively relied on real-world data and necessarily used a non-randomized design, and is therefore methodologically less robust than the RCTs that were summarized in this review. Nonetheless, it is worth highlighting that this sole study of dapagliflozin compared with another SGLT2i in this review found few statistically significant differences between patients receiving either SGLT2i; for example, no statistically significant differences in changes to eGFR.⁴⁰ These findings have the potential to be hypothesis generating and emphasize the importance of RCTs comparing various SGLT2is in patients with CKD (and possibly other conditions). Moreover, no information was identified by this review comparing dapagliflozin to ACE inhibitors, ARBs, GLP-1 agonists, or other treatments for managing CKD which represents another potential area for future research.

With regard to outcomes, the DAPA-CKD (from which a large proportion of the data were taken to inform this review) and DAPA-HF trials relied on the use of composites as primary and secondary end points, which have been highlighted in the literature as having the potential to introduce uncertainty or inflated treatment effects.^50-52 In addition, the DAPA-CKD RCT was ended early by an independent committee due to the demonstrated efficacy of dapagliflozin, and this could ostensibly have affected the study’s power to establish a clinically relevant difference between treatment groups in the CKD patient population.¹¹ Finally, there was a broad range of data describing renal and CV function, as well as mortality and safety identified in this review; however, there were no data identified describing quality of life or health-related quality of life, which would provide insight into the patient experience of being treated with dapagliflozin.

Lastly, while most of the RCTs included in this review were found to demonstrate more strengths than limitations, all were funded by the same private industry, for-profit pharmaceutical manufacturer. This represents a possible conflict of interest that could have introduced bias or other threats to the validity of the findings.

Conclusions and Implications for Decision- or Policy-Making

Nineteen reports describing dapagliflozin in patients with CKD were identified and found to be eligible for inclusion in this review: 2 SRs,^22,23 16 reports of RCT data,^24-39 and 1 NRS.⁴⁰ Most studies reported findings describing the clinical effectiveness of dapagliflozin compared with placebo, and no studies describing cost-effectiveness were identified.

The strengths of the studies identified in this review include the identification of data from multiple publications describing trials that used a randomized, double-blind, and placebo-controlled design.^24-39 Two of these studies (1 of which was reported across multiple publications) investigated large groups of patients across multiple sites and centres, which increases confidence in the external validity of the findings.^{24-30,32,34,36} All of the RCTs included in this review recruited and observed outcomes in patients from Canadian study sites,^24-39 which increases confidence in the relevance of the findings from these studies as they pertain to the Canadian context.

Benefits of dapagliflozin (10 mg daily) were found among patients with CKD in measures of kidney function (e.g., composite measures of renal function, including beneficial changes in the components of these composite measures e.g., improved eGFR and reduced ESKD).^{23,24,26,28-31,34,38,39} Measures of CV health and/or function produced variable results, with some benefits of dapagliflozin demonstrated in some studies and subgroups for some outcomes,^{24,28,29,32,34,36,38,39} but the findings of benefit were not consistent across patient subgroups and outcomes.^{22,24,28,29,32,34,36,38,39} Importantly, findings for many of the subgroups and outcomes were not (or likely were not) sufficiently powered to detect a statistically significant difference between treatment groups; these findings are concordant or discordant with sufficiently powered findings, and to an extent they are hypothesis generating, but their capacity to detect an actual difference is limited, and may contribute to variability across the findings summarized in this review. Salient findings from the main report of the DAPA-CKD trial that included all study patients and focused on the primary and secondary outcomes as compared between treatment groups did report statistically significant improvements in the dapagliflozin group in both renal and CV composite outcomes (as well as several of the outcomes’ sub-components).³⁴ This and other findings from the DAPA-CKD RCT and other trials of SGLT2is have been highlighted in much of the editorial literature as demonstrating sufficient cardiorenal protection so as to justify incorporating this class of drugs into standard care for patients with CKD (with or without T2D).^9,10,15 It is worth noting, however, that much of the editorial literature expressing urgency around implementing the use of dapagliflozin and other SGLT2is into clinical care as soon as possible also lists conflicts of interests for the editorials’ authors that include the private industry pharmaceutical manufacturer, which has also funded all of the trials in this area of research, constituting a potential source of bias.^5,9,10,15

With regard to health care utilization, most of the findings summarized in this review indicated a statistically significant protective effect of dapagliflozin as compared to placebo (though, there were only 3 papers that reported on this outcome, which was limited to hospitalizations and did not consider other measures of health care usage).^28,29,36 Nonetheless, reductions in health care utilization, in general, and hospitalization, in particular, are important considerations for patients with CKD, who require significant health care resources to manage their condition.³ Mortality was significantly reduced in patients receiving dapagliflozin in several of the studies and subgroups summarized in this review; notably, the DAPA-CKD RCT's analysis of all-cause death in the large sample of patients with CKD (and with or without T2D) found statistically significantly fewer patients receiving dapagliflozin who died from any cause as compared to those receiving placebo. However, not all of the studies in this review found a significant benefit in various measures of mortality across various subgroups, so, it may be that some groups could benefit more than others.

Finally, there were multiple analyses of the comparative safety of dapagliflozin with placebo,^24,28-39 with a preponderance of data suggesting no significant difference between dapagliflozin and placebo, further indicating the favourability of dapagliflozin. Whereas some data indicated a risk of increased volume depletion or renal AEs in some patients receiving dapagliflozin as compared to placebo, there were also data in the main report of findings from the DAPA-CKD RCT suggesting a statistically significant protective effect of dapagliflozin from the SAEs described therein.³⁴ This overall finding of the relative safety of dapagliflozin is corroborated in the published literature; for example, that SGLT2is do not appear to increase the risk of hypoglycemia.⁵³

As it concerns the benefits of dapagliflozin, subgroup analyses of the DAPA-CKD data indicated that the these may favour some patients more than others; for example, patients with CKD and T2D were demonstrated to experience statistically significant benefits with dapagliflozin as compared to placebo across most outcomes observed in the trial.³² The focus on treatment with SGLT2i among some subgroups of patients has been the subject of commentary and recommendations; that is in patients with T2D and a high risk of HF or progression of CKD and treatment with SGLT2is has been emphasized as an important intervention.⁵³ Although much of the published literature has described the effectiveness and safety of SGLT2is in patients with T2D and CKD, there remains less clinical data available describing the effects of SGLT2is, in general (and dapagliflozin, in particular), in non-diabetic CKD^14,54,55; type 1 diabetes, pediatric populations (e.g., adolescents with diabetic kidney disease), kidney transplant patients¹⁰; older adults¹¹; racial minority and disadvantaged communities (who bear a greater burden of CKD)¹³; as well as CKD patients with various levels of renal function (e.g., moderate or severe).⁸ While additional research on the renoprotective effects of the SGLT2i empagliflozin is under way (i.e., the EMPA-KIDNEY RCT investigating a comparison to placebo in patients with or without T2D),^6,13 there remains a need for additional research on the effects of dapagliflozin in various subsets of patients.

Given the preponderance of data and evidence found by this review and discussed in the literature that appears to support the benefit of dapagliflozin in patients with CKD, there remains a need for additional information describing particular patient subgroups, comparisons with other interventions, and additional outcomes (in particular, those which are patient-oriented). This, alongside the unknown cost-effectiveness of dapagliflozin for CKD warrants careful deliberation for decision- and policy-makers when considering the implementation of dapagliflozin into standard care for patients with CKD in Canada.

References

1.Patel AB, Mistry K, Verma A. DAPA-CKD: Significant Victory for CKD with or without Diabetes. Trends Endocrinol Metab. 2021;32(6):335-337. PubMed

2.Smekal MD, Donald M, Beanlands H, et al. Development and Preliminary Psychometric Testing of an Adult Chronic Kidney Disease Self-Management (CKD-SM) Questionnaire. Can J Kidney Health Dis. 2021;8:20543581211063981-20543581211063981. PubMed

3.Davidson JA. SGLT2 inhibitors in patients with type 2 diabetes and renal disease: overview of current evidence. Postgrad Med. 2019;131(4):251-260. PubMed

4.Patrick SA, Dabal TD, Jackson CD. Dapagliflozin Improves the Clinical Outcomes of Patients with Chronic Kidney Disease and Albuminuria. J Gen Intern Med. 2021;36(9):2915-2917. PubMed

5.Sarafidis P, Ortiz A, Ferro CJ, et al. Sodium--glucose co-transporter-2 inhibitors for patients with diabetic and nondiabetic chronic kidney disease: a new era has already begun. J Hypertens. 2021;39(6):1090-1097. PubMed

6.Fernandez-Fernandez B, Sarafidis P, Kanbay M, et al. SGLT2 inhibitors for non-diabetic kidney disease: drugs to treat CKD that also improve glycaemia. Clin Kidney J. 2020;13(5):728-733. PubMed

7.Mende CW. Chronic Kidney Disease and SGLT2 Inhibitors: A Review of the Evolving Treatment Landscape. Adv Ther. 2021;30:30. PubMed

8.Giorgino F, Vora J, Fenici P, Solini A. Renoprotection with SGLT2 inhibitors in type 2 diabetes over a spectrum of cardiovascular and renal risk. Cardiovascular Diabetology. 2020;19(1) (no pagination).

9.Anders HJ, Peired AJ, Romagnani P. SGLT2 inhibition requires reconsideration of fundamental paradigms in chronic kidney disease, 'diabetic nephropathy', IgA nephropathy and podocytopathies with FSGS lesions. Nephrology Dialysis Transplantation. 2020;13:13.

10.Almaimani M, Sridhar VS, Cherney DZI. Sodium-glucose cotransporter 2 inhibition in non-diabetic kidney disease. Curr Opin Nephrol Hypertens. 2021;30(5):474-481. PubMed

11.Mateo KF, Hirai T, Isaac D, et al. Dapagliflozin Reduces Adverse Renal and Cardiovascular Events in Patients with Chronic Kidney Disease. Journal of Clinical Outcomes Management. 2020;27(6):248-251.

12.Dekkers CCJ, Wheeler DC, Sjostrom CD, Stefansson BV, Cain V, Heerspink HJL. Effects of the sodium-glucose co-transporter 2 inhibitor dapagliflozin in patients with type 2 diabetes and Stages 3b-4 chronic kidney disease. Nephrology Dialysis Transplantation. 2018;33(11):2005-2011. PubMed

13.Jafar TH. FDA approval of dapagliflozin for chronic kidney disease: a remarkable achievement? Lancet. 2021;398(10297):283-284. PubMed

14.Onyali CB, Anim-Koranteng C, Shah HE, et al. Role of Selective Sodium-Glucose Co-Transporter-2 Inhibitors in Managing Cardio-Renal Complications in Type 2 Diabetes Mellitus: Beyond Glycemic Control. Cureus. 2021;13(8):e17452. PubMed

15.Halimi JM. SGLT2 inhibitors: A new era for our patients. Nephrologie et Therapeutique. 2021;17(3):143-148. PubMed

16.Kanduri SR, Kovvuru K, Hansrivijit P, et al. SGLT2 Inhibitors and Kidney Outcomes in Patients with Chronic Kidney Disease. J. 2020;9(9):2723.

17.Balkanski S. Dapagliflozin - structure, synthesis, and new indications. Pharmacia. 2021;68(3):591-596.

18.Broderick JM. Dapagliflozin approved in Europe for CKD, regardless of diabetes status. 2021: https://www.urologytimes.com/view/dapagliflozin-approved-in-europe-for-ckd-regardless-of-diabetes-status. Accessed 15 December 2021.

19.Shea BJ, Reeves BC, Wells G, et al. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ. 2017;358:j4008. PubMed

20.Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377-384. PubMed

21.Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700. PubMed

22.Zheng RJ, Wang Y, Tang JN, Duan JY, Yuan MY, Zhang JY. Association of SGLT2 inhibitors with risk of atrial fibrillation and stroke in patients with and without type 2 diabetes: a systemic review and meta-analysis of randomized controlled trials. J Cardiovasc Pharmacol. 2021;22:22. PubMed

23.Kelly MS, Lewis J, Huntsberry AM, Dea L, Portillo I. Efficacy and renal outcomes of SGLT2 inhibitors in patients with type 2 diabetes and chronic kidney disease. Postgrad Med. 2019;131(1):31-42. PubMed

24.Chertow GM, Vart P, Jongs N, et al. Effects of Dapagliflozin in Stage 4 Chronic Kidney Disease. Journal of the American Society of Nephrology. 2021;32(9):2352-2361. PubMed

25.Heerspink HJL, Cherney D, Postmus D, et al. A pre-specified analysis of the Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease (DAPA-CKD) randomized controlled trial on the incidence of abrupt declines in kidney function. Kidney International. 2021;22:22. PubMed

26.Heerspink HJL, Jongs N, Chertow GM, et al. Effect of dapagliflozin on the rate of decline in kidney function in patients with chronic kidney disease with and without type 2 diabetes: a prespecified analysis from the DAPA-CKD trial. Lancet Diabetes Endocrinol. 2021;9(11):743-754. PubMed

27.Heerspink HJL, Sjostrom CD, Jongs N, et al. Effects of dapagliflozin on mortality in patients with chronic kidney disease: a pre-specified analysis from the DAPA-CKD randomized controlled trial. European Heart Journal. 2021;42(13):1216-1227. PubMed

28.McMurray JJV, Wheeler DC, Stefansson BV, et al. Effect of Dapagliflozin on Clinical Outcomes in Patients With Chronic Kidney Disease, With and Without Cardiovascular Disease. Circulation. 2021;143(5):438-448. PubMed

29.McMurray JJV, Wheeler DC, Stefansson BV, et al. Effects of Dapagliflozin in Patients With Kidney Disease, With and Without Heart Failure. JACC Heart Fail. 2021;9(11):807-820. PubMed

30.Persson F, Rossing P, Vart P, et al. Efficacy and Safety of Dapagliflozin by Baseline Glycemic Status: A Prespecified Analysis From the DAPA-CKD Trial. Diabetes Care. 2021;44(8):1894-1897. PubMed

31.Wheeler DC, Jongs N, Stefansson BV, et al. Safety and efficacy of dapagliflozin in patients with focal segmental glomerulosclerosis: A prespecified analysis of the DAPA-CKD trial. Nephrology Dialysis Transplantation. 2021;25:25. PubMed

32.Wheeler DC, Stefansson BV, Jongs N, et al. Effects of dapagliflozin on major adverse kidney and cardiovascular events in patients with diabetic and non-diabetic chronic kidney disease: a prespecified analysis from the DAPA-CKD trial. Lancet Diabetes Endocrinol. 2021;9(1):22-31. PubMed

33.Wheeler DC, Toto RD, Stefansson BV, et al. A pre-specified analysis of the DAPA-CKD trial demonstrates the effects of dapagliflozin on major adverse kidney events in patients with IgA nephropathy. Kidney International. 2021;100(1):215-224. PubMed

34.Heerspink HJL, Stefansson BV, Correa-Rotter R, et al. Dapagliflozin in Patients with Chronic Kidney Disease. N Engl J Med. 2020;383(15):1436-1446. PubMed

35.Fioretto P, Stefansson BV, Johnsson E, Cain VA, Sjostrom CD. Dapagliflozin reduces albuminuria over 2 years in patients with type 2 diabetes mellitus and renal impairment. Diabetologia. 2016;59(9):2036-2039. PubMed

36.Jhund PS, Solomon SD, Docherty KF, et al. Efficacy of Dapagliflozin on Renal Function and Outcomes in Patients With Heart Failure With Reduced Ejection Fraction: Results of DAPA-HF. Circulation. 2021;143(4):298-309. PubMed

37.Cherney DZI, Dekkers CCJ, Barbour SJ, et al. Effects of the SGLT2 inhibitor dapagliflozin on proteinuria in non-diabetic patients with chronic kidney disease (DIAMOND): a randomised, double-blind, crossover trial. Lancet Diabetes Endocrinol. 2020;8(7):582-593. PubMed

38.Pollock C, Stefansson B, Reyner D, et al. Albuminuria-lowering effect of dapagliflozin alone and in combination with saxagliptin and effect of dapagliflozin and saxagliptin on glycaemic control in patients with type 2 diabetes and chronic kidney disease (DELIGHT): a randomised, double-blind, placebo-controlled trial. Lancet Diabetes Endocrinol. 2019;7(6):429-441. PubMed

39.Fioretto P, Del Prato S, Buse JB, et al. Efficacy and safety of dapagliflozin in patients with type 2 diabetes and moderate renal impairment (chronic kidney disease stage 3A): The DERIVE Study. Diabetes Obes Metab. 2018;20(11):2532-2540. PubMed

40.Lin YH, Huang YY, Hsieh SH, Sun JH, Chen ST, Lin CH. Renal and Glucose-Lowering Effects of Empagliflozin and Dapagliflozin in Different Chronic Kidney Disease Stages. Front Endocrinol (Lausanne). 2019;10:820. PubMed

41.Kohan DE, Fioretto P, Tang W, List JF. Long-term study of patients with type 2 diabetes and moderate renal impairment shows that dapagliflozin reduces weight and blood pressure but does not improve glycemic control. Kidney International. 2014;85(4):962-971. PubMed

42.Heerspink HJL, Stefansson BV, Chertow GM, et al. Rationale and protocol of the Dapagliflozin And Prevention of Adverse outcomes in Chronic Kidney Disease (DAPA-CKD) randomized controlled trial. Nephrology Dialysis Transplantation. 2020;35(2):274-282. PubMed

43.McMurray JJV, DeMets DL, Inzucchi SE, et al. A trial to evaluate the effect of the sodium-glucose co-transporter 2 inhibitor dapagliflozin on morbidity and mortality in patients with heart failure and reduced left ventricular ejection fraction (DAPA-HF). European Journal of Heart Failure. 2019;21(5):665-675. PubMed

44.Page MJ, Moher D, Fidler FM, et al. The REPRISE project: protocol for an evaluation of REProducibility and Replicability In Syntheses of Evidence. Systematic Reviews. 2021;10(1):112. PubMed

45.Werner MU. Salami-slicing and duplicate publication: gatekeepers challenges. Scandinavian Journal of Pain. 2021;21(2):209-211. PubMed

46.Menon V, Muraleedharan A. Salami Slicing of Data Sets: What the Young Researcher Needs to Know. Indian J Psychol Med. 2016;38(6):577-578. PubMed

47.McMurray JJV, Solomon SD, Inzucchi SE, et al. Dapagliflozin in Patients with Heart Failure and Reduced Ejection Fraction. N Engl J Med. 2019;381(21):1995-2008. PubMed

48.Heesen P. Crossover trials: what are they and what are their advantages and limitations? 2020: https://s4be.cochrane.org/blog/2020/09/07/crossover-trials-what-are-they-and-what-are-their-advantages-and-limitations/. Accessed 22 Jan 2022.

49.Romain PL. Conflicts of interest in research: looking out for number one means keeping the primary interest front and center. Curr Rev Musculoskelet Med. 2015;8(2):122-127. PubMed

50.Shaikh A, Ochani RK, Khan MS, et al. Contribution of individual components to composite end points in contemporary cardiovascular randomized controlled trials. Am Heart J. 2020;230:71-81. PubMed

51.Ferreira-González I, Busse JW, Heels-Ansdell D, et al. Problems with use of composite end points in cardiovascular trials: systematic review of randomised controlled trials. BMJ. 2007;334(7597):786. PubMed

52.Freemantle N, Calvert M, Wood J, Eastaugh J, Griffin C. Composite outcomes in randomized trials: greater precision but with greater uncertainty? JAMA. 2003;289(19):2554-2559. PubMed

53.Zoungas S, de Boer IH. Sglt2 inhibitors in diabetic kidney disease. Clinical Journal of the American Society of Nephrology. 2021;16(4):631-633. PubMed

54.Papazafiropoulou AK, Melidonis A, Antonopoulos S. Effects of glucagon-like peptide-1 receptor agonists and sodium-glucose co-transporter 2 inhibitors on cardiorenal and metabolic outcomes in people without diabetes. Current Pharmaceutical Design. 2021;27(8):1035-1042. PubMed

55.Sternlicht HK, Bakris GL. Reductions in albuminuria with SGLT2 inhibitors: a marker for improved renal outcomes in patients without diabetes? Lancet Diabetes Endocrinol. 2020;8(7):553-555. PubMed

Appendix 1: Selection of Included Studies

Figure 1: Selection of Included Studies

Appendix 2: Characteristics of Included Publications

Note that this appendix was not copy-edited.

Table 2: Characteristics of Included Systematic Review

A1C = glycated hemoglobin; CKD = chronic kidney disease; DAPA = dapagliflozin; eGFR = estimated glomerular filtration rate; FPG = fasting plasma glucose; g = gram(s); GLP-1 receptor agonists = Glucagon-like peptide receptor agonists; m² = metre(s) squared; MA = meta-analysis; mg = milligram; min = minute(s); mL = millilitre; N/n = number; OR = odds ratio; pt/pts = patient(s); RCT = randomized controlled trial; RR = relative risk; SGLT2i = sodium/glucose co-transporter-2 inhibitor; SD = standard deviation; SR = systematic review; T2D = type 2 diabetes; UACR = urine albumin-creatinine ratio; wk = week(s); yr = year(s).

Table 3: Characteristics of Included Primary Clinical Studies

AE = adverse event(s); ACE inhibitor(s) = angiotensin-converting enzyme inhibitor(s); ARB = angiotensin receptor blockers; BMI = body mass index; BP = blood pressure; CKD = chronic kidney disease; CV = cardiovascular; DAPA = dapagliflozin; dx = diagnosis; eGFR = estimated glomerular filtration rate; Empa10 = empagliflozin 10 mg; Empa25 = empagliflozin 25 mg; ESKD = end-stage kidney disease; FPG = fasting plasma glucose; FSGS = focal segmental glomerulosclerosis; g = gram(s); h = hour(s); A1C = glycated hemoglobin; HF = heart failure; IgA = immunoglobulin A; IQR = interquartile range; kg = kilogram; m2 = metre(s) squared; mg = milligram; mGFR = measured glomerular filtration rate; mg/mmol = milligrams per millimole; MI = myocardial infarction; min = minute(s); mL = millilitre; mmol/mol = millimoles per mol; mo = month(s); N/n = number; pt/pts = patient(s); RCT = randomized controlled trial; SAE = serious adverse event(s); SBP = systolic blood pressure; SD = standard deviation; SGLT2i = sodium/glucose co-transporter-2 inhibitor; T2D = type 2 diabetes; UACR = urine albumin-creatinine ratio; wk = week(s); yr = year(s).

Appendix 3: Critical Appraisal of Included Publications

Note that this appendix was not copy-edited.

Table 4: Strengths and Limitations of Systematic Reviews Using the AMSTAR Checklist¹⁹

AMSTAR 2 = A MeaSurement Tool to Assess systematic Reviews 2.

Table 5: Strengths and Limitations of Clinical Studies Using the Downs and Black Checklist²⁰

Appendix 4: Main Study Findings

Note that this appendix was not copy-edited.

Table 6: Summary of Findings by Outcome — Renal Outcomes

AR = absolute risk; CI = confidence interval; CKD = chronic kidney disease; CV = cardiovascular; d = day(s); dL = deciletre; Empa10 = empagliflozin 10 mg; Empa25 = empagliflozin 25 mg; eGFR = estimated glomerular filtration rate; ESKD = end-stage kidney disease; FSGS = focal segmental glomerulosclerosis; g = gram(s); h = hour(s); A1C = glycated hemoglobin; HF = heart failure; HR = hazard ratio; IgA = immunoglobulin A; m2 = metres squared; mg = milligram; mGFR = measured glomerular filtration rate; min = minute(s); mL = millilitre; N/n = number; NC = not calculable; NR = not reported; NRS = non-randomized study; NS = not significant; OR = odds ratio; P = probability; P-value = P value; pt/pts = patient(s); RCT = randomized controlled trial; SGLT2i = sodium/glucose co-transporter-2 inhibitor; SE = standard error; T2D = type 2 diabetes; UACR = urine albumin-creatinine ratio; vs. = versus; wk = week(s); yr = year(s).

Table 7: Summary of Findings — Cardiovascular Outcomes