[107] EMPA-REG OUTCOME Trial: What does it mean?

[107] EMPA-REG OUTCOME Trial: What does it mean?

Empagliflozin (Jardiance) inhibits the sodium-glucose co-transporter 2 (SGLT2), which is the predominant mechanism responsible for reabsorption of glucose from the glomerular filtrate back into the circulation. By inhibiting SGLT2, empagliflozin increases urinary glucose excretion.1 Empagliflozin was approved by Health Canada in 2015 to improve glycemic control in adult patients with type 2 diabetes (T2DM).2 In 2016, Health Canada and the U.S. Food and Drug Administration (FDA) approved an additional claim that empagliflozin reduces the incidence of cardiovascular death in patients with T2DM who have established cardiovascular disease. This was based on the results of a single trial, EMPA-REG OUTCOME, which was conducted as a requirement by the FDA to rule out a 30% (relative) increase in cardiovascular events by empagliflozin.3,4

The EMPA-REG OUTCOME Trial tested adding empagliflozin to ‘standard of care’

In this trial, 7028 T2DM patients with established cardiovascular disease were randomized to 1 of 3 arms: empagliflozin 10 mg or empagliflozin 25 mg or placebo and were followed for a median of 3.1 years.3 Participants were eligible if their glycated haemoglobin (HbA1c) was 7.0% to 10.0% (mean 8.1%). Other parameters: mean age 63, 72% white, 72% male, mean BMI 31 kg/m2, 95% taking antihypertensives, 81% taking lipid-lowering drugs, and 83% taking antiplatelet medications. Most had a diagnosis of T2DM for >5 years, more than half had a diagnosis for >10 years. At baseline 98% of participants were receiving other glucose-lowering medications; 68% were receiving 2 or more. Metformin, insulin and sulfonylureas were prescribed to 74%, 48%, and 43% of participants, respectively. The net health effect of the various glucose-lowering drugs and drug combinations prescribed to participants at baseline and throughout the trial is not known.5

The results using our usual outcome hierarchy are shown in Table 1. The 2 empagliflozin arms have been combined; the outcomes were similar for the 2 doses.

Table 1: The EMPA-REG OUTCOME Trial Outcomes3,6

Outcome Placebo
N=2333
Empagliflozin
N=4687
RR [95%CI] ARR
Death from any cause (ITT) 8.3% 5.7% 0.69 [0.57, 0.82] 2.6%
Death from any cause (On treatment) 4.7% 3.4% 0.72 [0.57, 0.92] 1.3%
Total serious adverse events (On treatment) 42.3% 38.2% 0.90 [0.85, 0.96] 4.1%
Total CV serious adverse events (ITT) 12.1% 10.5% 0.86 [0.75, 0.99] 1.6%

If this trial had been conducted in a way that empagliflozin was the only difference between the groups, these results would suggest that empagliflozin causes a net health benefit in people with T2DM and a history of cardiovascular disease.  Unfortunately, empagliflozin was not the only difference between the groups. During the trial, HbA1c was not blinded and investigators could escalate medications in an effort to achieve a glucose target ≤6.5-7.0% in accordance with aggressive ‘standards of care’. As a result, other glucose-lowering medications were added more frequently and at higher doses in the placebo group3,7 (see Table 2).

Table 2: The EMPA-REG OUTCOME Trial Medication changes after randomization3

Participants with additional: Placebo Empagliflozin
Glucose-lowering medications added in concordance with an escalated ‘standard of care’ 31.5% 19.5%
Insulin 11.5% 5.8%
Dipeptidyl peptidase 4-inhibitor 8.3% 5.6%
Sulfonylurea 7.0% 3.8%
Thiazolidinedione 2.9% 1.2%

Interpretation

There are thus at least 3 possible interpretations of the EMPA-REG OUTCOME Trial:

  1. Empagliflozin decreases mortality and serious adverse events when added to ‘standard of care’.
  2. The more aggressive use of other glucose-lowering medications in the placebo group increases mortality and serious adverse events.
  3. A combination of 1 and 2.

After a detailed examination of the published trial3, the U.S. FDA review7, the European Medicines Agency (EMA) review8, and the German Institute for Quality and Efficiency in Health Care (IQWiG) review9, we are not confident in explanation #1. In support of explanation #2 there are 2 RCTs that have shown that more intensive therapy in T2DM increases total mortality and cardiovascular mortality.10,11 Whether these findings are attributable to the pursuit of an intensive HbA1c target or to the use of any specific drug or drug-combination remains uncertain.

We are not alone in not accepting the results of this trial. In June 2016, a 23-member FDA advisory committee voted on whether, based on this single trial, the company could claim that empagliflozin reduces cardiovascular mortality. The vote was 12 for and 11 against.12,13

Did empagliflozin cause harm?

Harms were increased in those taking empagliflozin in the EMPA-REG OUTCOME Trial ( Table 3). The increase in genital infections in both men and women means that during a 3 year time period, 1 in 29 men and 1 in 14 women will have their quality of life adversely affected by a genital infection.

Table 3: The EMPA-REG OUTCOME Trial Harms3

Outcome Placebo
N=2333
Empagliflozin
N=4687
RR [95%CI] ARI
Urosepsis 0.1% 0.4% 2.83 [0.83, 9.66] 0.3%
Genital infection (women) 2.6% 10.0% 3.84 [2.34, 6.30] 7.4%
Genital infection (men) 1.5% 5.0% 3.47 [2.27, 5.30] 3.5%

Other reasons for scepticism

The FDA review of the EMPA-REG OUTCOME Trial rejected the manufacturer’s claim that empagliflozin reduces the risk of heart failure and nephropathy.7 In addition, analyses of outcomes by geography identified unexplained regional differences: the magnitude of effect of empagliflozin on mortality was less in North America and Europe compared to Latin America and Asia.6 Another independent drug bulletin has critically appraised this trial and identified these and many other concerns.14 It should also be clear that this trial tells us nothing about the use of empagliflozin in other T2DM clinical settings.

Conclusions

  • The EMPA-REG OUTCOME Trial tested the addition of empagliflozin to a ‘standard of care’ for T2DM whose impact on clinically important outcomes is currently unknown.
  • It is uncertain whether the reduction in mortality and serious adverse events in the EMPA-REG OUTCOME Trial is attributable to empagliflozin or to less use of other glucose-lowering therapies.
  • The results of this trial are not applicable to people with T2DM in other clinical settings.
  • Until there is a body of evidence informed by large, independently conducted comparative effectiveness trials of different therapeutic strategies, we will not know the optimal treatment of T2DM at various stages of the diagnosis.
The draft of this Therapeutics Letter was submitted for review to 75 experts and primary care physicians in order to correct any inaccuracies and to ensure that the information is concise and relevant to clinicians.
The Therapeutics Initiative is funded by the BC Ministry of Health through a grant to the University of BC. The Therapeutics Initiative provides evidence-based advice about drug therapy, and is not responsible for formulating or adjudicating provincial drug policies.
ISSN 2369-8691 (Online) <||> ISSN 2369-8683 (Print)

References

  1. Health Canada. Product Monograph. JARDIANCE empagliflozin tablets. Drug Product Database. August 25, 2017; [Internet] https://health-products.canada.ca/dpd-bdpp/index-eng.jsp
  2. Health Canada. Summary Basis of Decision. JARDIANCE. Drug and Health Product Register. August 27, 2015; [Internet] https://hpr-rps.hres.ca/reg-content/summary-basis-decision.php
  3. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med 2015; 373(22):2117-28. DOI: 10.1056/NEJMoa1504720
  4. U.S. Food & Drug Administration. Guidance for Industry. Diabetes Mellitus – Evaluating Cardiovascular Risk in New Antidiabetic Risk in New Antidiabetic Therapies to Treat Type 2 Diabetes. December 2008; [Internet] http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm071627.pdf
  5. Maruther NM, Tseng E, Hutfless S, et al. Diabetes medications as monotherapy or metformin-based combination therapy for type 2 diabetes. A systematic review and meta-analysis. Ann Intern Med 2016; 164(11):740-51. DOI: 10.7326/M15-2650
  6. Boehringer Ingelheim. EMPA-REG OUTCOME Trial. U.S. Food & Drug Administration Endocrinologic and Metabolic Drugs Advisory Committee Briefing Document. NDA 204629 / NDA 206111. May 19, 2016; [Internet] https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/UCM508423.pdf
  7. U.S. Food & Drug Administration. 2016 Meeting materials, Endocrinologic and Metabolic Drugs Advisory Committee. June 28, 2016 ; [Internet] https://www.fda.gov/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/ucm491062.htm
  8. European Medicines Agency. Committee for Medicinal Products for Human Use. Jardiance (empagliflozin) Assessment report. EMA/11728/2017. Procedure No. EMEA/H/C/002677/II/0014. December 15, 2016; [Internet] http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Assessment_Report_-_Variation/human/002677/WC500224837.pdf
  9. Institute for Quality and Efficiency in Health Care. Empagliflozin – Benefit assessment according to 35a Social Code Book V [Translation of Sections 2.1 to 2.7 and Appendix A of the dossier assessment Empagliflozin – Nutzenbewertung gemäß 35a SGB V]. IQWiG Reports – Commission No. A16-12. Version 1.0, May 30, 2016; [Internet] https://www.iqwig.de/download/A16-12_Empagliflozin_Extract-of-dossier-assessment.pdf
  10. Action to Control Cardiovascular Risk in Diabetes (ACCORD) Study group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008; 358(24):2545-59. DOI: 10.1056/NEJMoa0802743
  11. Duckworth W, Abraira C, Moritz T, et al. Glucose control and vascular complications in Veterans with type 2 diabetes. N Engl J Med 2009; 360(2):129-39. DOI: 10.1056/NEJMoa0808431
  12. U.S. Food & Drug Administration. Center for Drug Evaluation and Research. Summary Minutes of the Endocrinologic and Metabolic Drugs Advisory Committee Meeting. June 28, 2016; [Internet] https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/UCM518911.pdf
  13. U.S. Food & Drug Administration. Center for Drug Evaluation and Research. Transcript for the Meeting of the Endocrinologic and Metabolic Drugs Advisory Committee (EMDAC). June 28, 2016; [Internet] https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/UCM522096.pdf
  14. Saiz, LC. The EMPA-REG OUTCOME trial (empagliflozin). A critical appraisal. The power of truth, the truth of power. Drug and Therapeutics Bulletin of Navarre. 2016; Vol 24, No 3. http://www.navarra.es/NR/rdonlyres/53670132-6111-4C40-A17C-ACC734822AA5/369046/Bit_v24n3_e.pdf

 

7 Comments
  • Randy Minion
    Posted at 11:55h, 20 November Reply

    did you look at the later European study that used mostly Farxiga and Invokana, as they seemed to have similar efficacy?

    • Alan Cassels @ TI
      Posted at 10:11h, 25 June Reply

      We will be addressing those products in future newsletters. Thanks for the comment.
      cheers,
      Alan Cassels
      Communications Director
      Therapeutics Initiative

  • Chris Richardson
    Posted at 10:24h, 08 December Reply

    First let me say that I agree that we need to be cautious in our interpretation of positive results from a single trial of a new drug. So we need to see the results replicated. (Fortunately trials of other drugs in this class also seem to carry similar signals).
    But. Your criticism of the empagliflozin not being the only difference between the two groups seems to be a strength to me. It is much more pragmatic, and therefore applicable to practice, to allow manipulation of the other medications. Whether the reduction in mortality is due to adding empagliflozin or less use of other anti-hyperglycemics isn’t a particularly important distinction to the patient who ‘doesn’t die’ as a result of this intervention.

  • Dr. Matthew Bernard
    Posted at 19:12h, 10 January Reply

    1) Neither empagliflozin group alone achieved statistical significance for the primary outcome. In fact, the Kaplan-Meier curves for the 2 doses are not even represented in the main article, only the outcome for the groups combined.

    2) The study isn’t really blinded. By prohibiting uptitration of other drugs, the A1C difference between the two groups was remarkable after 3 months. I don’t think it would have been difficult to surmise which individual patients were on active drug vs placebo – just look at the A1Cs. Empagliflozin does reduce A1C.

    3) There is no clear a priori hypothesis as to how this drug would keep people alive longer. It was statistical happenstance in a trial designed to ensure they weren’t harming patients.

    4) What happens to the survival curves of the placebo groups in the last 6 months of their analysis. There is a near exponential increase in events in that group. What is the rationale?

  • Greg Streppel
    Posted at 16:31h, 23 January Reply

    Thank-you for this excellent critical appraisal.

  • Joan King
    Posted at 09:49h, 31 January Reply

    Diabetes Canada appreciates the opportunity to comment on the Therapeutics Initiative’s assessment of the EMPA-REG OUTCOME trial. We are concerned with several points made in Therapeutics Letter 107 and with the overall tone and message of the review. It gives the impression of a bias against medication use in people with diabetes. This is a position we patently disagree with. We believe that people living with diabetes deserve every available opportunity to achieve their full health potential. This includes having access both to medications and to nonpharmacologic interventions that could improve their outcomes and reduce their risk of disease-related complications. It is important for Therapeutics Initiative reviewers to declare their perspectives against medications and specific populations up-front when writing editorials about clinical trials and medical therapy. There are significant policy implications to a publicly funded institution taking a position that appears biased against people living with a progressive, chronic condition with no known cure.

    1. In interpreting the EMPA-REG OUTCOME trial results, the authors posit that “the more aggressive use of other glucose-lowering medications in the placebo group increases mortality and serious adverse events”. As the authors likely know, the Food and Drug Administration (FDA), through its 2008 guidance document , mandated that all the cardiovascular outcome trials (CVOTs) with antihyperglycemic agents be carried out with a design of glycemic equipoise (both the drug and placebo arms should continue to be treated during the trial to standard of care and HbA1c targets according to local guidelines in every country). All CVOTs, since this regulation was applied, have had an increased utilization of glucose lowering agents in the placebo and comparator arms of the trials and resulted in minimal HbA1c difference between the two arms (ranging from 0.2%-0.3%). This is relevant because the HbA1c effects are minimized between the groups and unlikely to be the reason for differences in serious adverse events.

    2. Empagliflozin was noted to cause harm to study participants, in the form of genital infections for 1 in 29 men and 1 in 14 women over a three year period, and that this adversely affected their quality of life. Indeed, increased risk of genitourinary infections is a known side effect of empagliflozin. It is an effect that many people living with diabetes would be willing to chance and/or to bear for an opportunity at longer life. Decreased risk of cardiovascular mortality – i.e. survival – is the most important outcome for the majority of people with diabetes. However, we strongly support patients being informed of the potential benefits and harms of therapy and being part of the decision-making for their own treatment.

    3. The FDA’s rejection of the claim that empagliflozin reduces the risk of nephropathy was also noted in the Therapeutics Letter as another “reason for scepticism”. We agree that this evidence should direct regulatory approval and clinical recommendations. In a 2016 update to Diabetes Canada’s 2013 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada, empagliflozin was recommended as a treatment for suboptimal glycemic control in people with type 2 diabetes and clinical cardiovascular disease. Though the potential microvascular benefits of the drug are recognized, the guidelines suggest the addition of empagliflozin specifically for its cardioprotective effects, as follows:
    “In adults with type 2 diabetes with clinical cardiovascular disease in whom glycemic targets are not met, an antihyperglycemic agent with demonstrated cardiovascular outcome benefit should be added to reduce the risk of major cardiovascular events (Grade 1, Level 1A for empagliflozin).”

    The lack of demonstrated clinical benefit to reduce the risk of nephropathy does not negate the demonstrated cardiovascular outcome benefits.

    4. The conclusion states that “the EMPA-REG OUCTOME Trial tested the addition of empagliflozin to a ‘standard of care’ for T2DM whose impact on clinically important outcomes is currently unknown”. Recommendations for current standards of care are based on trials that have measured HbA1c. Glycemic control measured by HbA1c is a surrogate outcome used in clinical trials that has been a standard metric for decades. In the absence of outcome data, HbA1c is useful to guide decision making. Therapeutics Initiative reviewers have previously questioned the validity of HbA1c as a surrogate outcome. Now that health outcome data are available, these data are also being discarded. The reviewers seem to be systematically biased against people with diabetes having access to medications that can improve their health outcomes.

    Diabetes Canada asserts that education, behavioural interventions and support are essential for optimal diabetes management. Medications can be added to therapy as an important adjunct to the care regimen for many Canadians. Therapy must always be individualized, and should include evidence-based options for people living with this disease. Diabetes Canada highlights that the evidence for the role of empagliflozin in the population with type 2 diabetes and clinical cardiovascular disease is robust and we continue to recommend this medication as a choice for those clinicians and patients who wish to access it. We would welcome a discussion with you about these very important issues and the evidence that supports this position.

    Sincerely,

    Dr. Jan Hux
    President
    Diabetes Canada

    Conflict of interest declaration:
    Diabetes Canada has declared financial relationships with the following entities in the bio-medical arena that could be perceived to influence, or that give the appearance of potentially influencing this comment: Abbott Diabetes Care, Ascencia Diabetes Care, AstraZeneca Canada Inc., Bayer Pharmaceuticals, Boehringer Ingleheim (Canada) Ltd., Eli Lilly Canada Inc., Innovative Medicines Canada, Ipsen, Janssen Inc., McNeil Consumer Healthcare, Merck Canada Inc., Novartis Pharmaceuticals Canada Inc., Novo Nordisk Canada Inc., Paladin Labs Inc., Roche Diabetes Care, Sanofi Canada, Valeant Canada LP.
    Diabetes Canada receives operating money from general fundraising, donations, foundations, governments, and private corporations including, but not exclusively, medical education, pharmaceutical, medical device and medical supplies companies. Funding sources include: Abbott Nutrition, Auto Control Medical Inc., BD Medical Diabetes Care, Dexcom Canada, ForaCare Technology Canada Inc., Insulet Canada Corporation, LifeScan Canada Ltd., mdBriefCase Group Inc., Medtronic, Medtronic of Canada Ltd., Montmed, Tykess Pharmaceuticals, VitalAire Canada Inc., the Government of Newfoundland and Labrador, the Government of Saskatchewan – Drug Plan and Extended Benefits, the Government of Saskatchewan – Health, the British Columbia Provincial Health Services Authority, the Northern Ontario Heritage Fund Corporation, the Province of New Brunswick, the Nova Scotia Health Authority, the Ontario Government Seniors Services, the Healthy Kids Community Challenge – Timiskaming, the Timiskaming Public Health Unit, Manitoba Health, the Government of Canada – HR Development, the New Brunswick Regional Health Authority, the Government of New Brunswick Department of Health, Seniors Community Grant – Timmins, Service Canada, the First Nations and Inuit Health Branch, the Government of Canada, the Government of New Brunswick Department of Health, the Northern Ontario Heritage Fund Corporation, the File Hills Qu’appelle Tribal Council, and the Saskatchewan Liquor and Gaming Authority.

    • Therapeutics Initiative
      Posted at 17:53h, 13 March Reply

      Thank you for your feedback on our Therapeutics Letter #107 about the EMPA-REG OUTCOME trial. We appreciate all comments on our Therapeutics Letters.
      We have responded to your four key points as follows:

      1. The lack of certainty of the overall impact of a single trial suggests the need for a cautious approach. Therapeutics Letter #107 interpreting the EMPA-REG OUTCOME trial results was corroborated by reviews performed by the US FDA, the European Medicines Agency (EMA) and the German Institute for Quality and Efficiency in Health Care (IQWiG). Those reviews also cast doubt on the hypothesis that empagliflozin decreases mortality and serious adverse events when added to ‘standard of care’. The placebo patients were more aggressively prescribed glucose-lowering medications and at least 2 RCTs have shown that more intensive therapy in T2DM increases total mortality and cardiovascular mortality. Whether the increased risk is attributable to differences in HbA1c or to the drugs used has not been reconciled.

      2. In terms of the harms of empagliflozin, we agree with you and also strongly support patients being “informed of the potential benefits and harms of therapy and being part of the decision-making for their own treatment,” and that is why we quantified the absolute harms in Table 3 of our Therapeutics Letter #107.

      3. Drugs all have potential harms as well as potential benefits. You make the point that “the lack of demonstrated clinical benefit to reduce the risk of nephropathy does not negate the demonstrated cardiovascular outcome benefits” yet we would add that all outcomes should be part of the decision making regarding empagliflozin.

      4. You note that “glycemic control measured by HbA1c is a surrogate outcome used in clinical trials that has been a standard metric for decades.” We emphasize that there is global heightened concern about the mis-use of surrogate markers, and thus caution in interpreting a drug’s influence on these markers is warranted. Our group is always working to find and analyze the best available evidence so that we can identify effective and safe medications.

      We believe that in the interests of the health of people with diabetes you would support our main conclusion: “Until there is a body of evidence informed by large, independently conducted comparative effectiveness trials of different therapeutic strategies, we will not know the optimal treatment of T2DM at various stages of the diagnosis.

      JM Wright
      Editor-in-Chief
      Therapeutics Letter

      Conflict of interest declaration:
      No conflicts of interest.

Post A Comment