Saxagliptin Assessment of Vascular Outcomes Recorded in Patients With Diabetes Mellitus–Thrombolysis in Myocardial Infarction 53 - SAVOR-TIMI 53

Contribution To Literature:

The SAVOR-TIMI 53 trial showed that saxagliptin is not associated with an excess of CV events as compared with placebo in patients with DM2 and either established CV disease or risk factors.


Saxagliptin is a novel selective dipeptidyl peptidase 4 (DPP-4) inhibitor that lowers glucose. The current trial sought to study the cardiovascular (CV) safety of saxagliptin in patients with type 2 diabetes mellitus (DM2).

Study Design

  • Placebo Controlled
  • Randomized
  • Parallel
  • Stratified

Patient Populations:

  • DM2
  • Age ≥40 years, and
  • Documented HbA1c ≥6.5% in the previous 6 months, and
  • High risk for a CV event with established CV disease, or multiple risk factors
    - Must be ≥55 years old (male) or ≥60 years old (female)
    - Dyslipidemia, hypertension, or current smoker

    Number of enrollees: 16,492
    Duration of follow-up: 2 years
    Mean patient age: 65.1 years
    Percentage female: 33%


  • Concomitant use of other DPP-4 inhibitors or GLP-1 agonists (within 6 months)
  • End-stage renal disease or on dialysis
  • Renal transplantation
  • Serum creatinine >6.0 mg/dl

Primary Endpoints:

  • Composite of CV death/nonfatal MI/nonfatal stroke

Secondary Endpoints:

  • CV death/MI/stroke/unstable angina/coronary revascularization/heart failure

Drug/Procedures Used:

Patients with DM2 and established CV disease or multiple risk factors were randomized in a double-blind fashion to either saxagliptin 5 mg daily (2.5 mg if glomerular filtration rate [GFR] <50 ml/min) or placebo. Stratification was performed based on established CV disease versus risk factors only and renal function.

Concomitant Medications:

Aspirin (75%), statin (78%), angiotensin-converting enzyme inhibitors (55%), angiotensin-receptor blockers (28%), beta-blockers (62%), insulin (42%), and metformin (70%)

Principal Findings:

A total of 16,492 patients were randomized, 8,280 to saxagliptin and 8,212 to placebo. Baseline characteristics were fairly similar between the two arms. The median duration of diabetes was 10.3 years, with a mean glycated hemoglobin (HbA1c) of 8%. Established atherosclerotic disease was noted in 78.5% of patients, including prior MI in 38%. The mean GFR was 72.6 ml/min (13.6% with GFR 30-50).

HbA1c levels were lowered modestly at 1 year in the saxagliptin arm compared with placebo (7.6% vs. 7.9%, p < 0.001) and at 2 years (7.5% vs. 7.8%, p < 0.001). There was correspondingly a 23% decrease in the intensification of anti-hyperglycemic medications with saxagliptin compared to control (p < 0.001), and a 30% decrease in the initiation of insulin therapy for more than 3 months with saxagliptin compared to control (p < 0.001).

The primary endpoint of CV death/MI/stroke was similar between the saxagliptin and placebo arms at 2 years (7.3% vs. 7.2%; HR = 1.00, 95% CI 0.89-1.12; p < 0.001 for noninferiority; p = 0.99 for superiority). The secondary endpoint of CV death/MI/stroke/unstable angina/coronary revascularization/heart failure was also similar (12.8% vs. 12.4%, p = 0.66). Individual endpoints including CV death (2.9% vs. 3.2%, p = 0.72), MI (3.4% vs. 3.2%, p = 0.52), and all-cause mortality (4.2% vs. 4.9%, p = 0.15) were all similar. In patients with a high N-terminal pro-brain natriuretic peptide level at baseline, hospitalization for congestive heart failure was higher (10.9% vs. 8.9%, p = 0.024).

Hypoglycemia was more common with saxagliptin (15.3% vs. 13.4%, p < 0.001), mostly due to minor hypoglycemia (self-limiting symptoms or glucose <54 mg/dl) (14.2% vs. 12.5%, p = 0.002). However, major hypoglycemia requiring active intervention was also higher (2.1% vs. 1.7%). The incidences of acute or chronic pancreatitis (0.3% vs. 0.3%) and pancreatic cancer (5 vs. 12 cases) were similar. Nonfatal angioedema was more common with saxagliptin (8 vs. 1 case, p = 0.04); no fatal cases were reported.

Utility of urinary albumin:creatinine ratio (UACR): There was a linear increase with each progressive UACR category (<10 mg/g, 10-30 mg/g, 30-300 mg/g, and >300 mg/g) in the incidence of the primary composite endpoint (cardiovascular death, MI, or ischemic stroke) (3.9%, 6.9%, 9.2%, and 14.3%); CV death (1.4%, 2.6%, 4.1%, and 6.9%); and hospitalization for heart failure (1.5%, 2.5%, 4.0%, and 8.3%) (adjusted p < 0.001 for trend). After adjusting for baseline levels of N-terminal pro–B-type natriuretic peptide, high-sensitivity troponin T, and high-sensitivity C-reactive protein, levels of UACR remained significantly associated with CV outcomes; however, the relationship was attenuated.


The results of the SAVOR-TIMI 53 trial indicate that saxagliptin, a DPP-4 inhibitor, is not associated with an excess of CV events as compared with placebo in patients with DM2 and either established CV disease or risk factors; the upper margin of 95% CI of 1.12 for the primary endpoint was lower than the threshold set by the Food and Drug Administration (FDA) (1.3) for postmarketing trials. Modest reductions in HbA1c were noted, with an increase in hypoglycemia events. No increase in pancreatitis or pancreatic cancer was noted.

Following the much publicized CV safety concerns with rosiglitazone, the FDA mandated that all new diabetes drugs conduct studies demonstrating CV safety. The upper limit of the 95% CI for the HR had to be <1.8 for premarketing studies and <1.3 for postmarketing studies. This trial thus establishes the CV safety profile of saxagliptin for use in patients with DM2. As has been true for almost all diabetes medications, lowering HbA1c was not associated with improved CV outcomes (macrovascular outcomes) in the short-term. Most diabetic drugs have been associated with improvement in microvascular outcomes; however, those were not studied in the current trial. For now, clinicians need to continue to focus on other modifiable risk factors for CV risk reduction among individuals with DM2, such as hypertension and dyslipidemia, in addition to glycemic control.

Baseline UACR was independently associated with total mortality as well as CV events, such as CV death, MI, ischemic stroke, and hospitalization for heart failure, thus expanding prior observations and supporting the hypothesis that UACR provides complementary insight into the association between diabetic kidney disease and CV risk among patients with DM2.


Scirica BM, Mosenzon O, Bhatt DL, et al. Cardiovascular Outcomes According to Urinary Albumin and Kidney Disease in Patients With Type 2 Diabetes at High Cardiovascular Risk: Observations From the SAVOR-TIMI 53 Trial. JAMA Cardiol 2017;Dec 6:[Epub ahead of print].

Scirica BM, Bhatt DL, Braunwald E, et al., on behalf of the SAVOR-TIMI 53 Steering Committee and Investigators. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes mellitus. N Engl J Med 2013;369:1317-26.

Presented by Dr. Deepak Bhatt at the European Society of Cardiology Congress, Amsterdam, Holland, September 2, 2013.

Clinical Topics: Diabetes and Cardiometabolic Disease, Dyslipidemia, Heart Failure and Cardiomyopathies, Prevention, Acute Heart Failure, Heart Failure and Cardiac Biomarkers, Hypertension

Keywords: Diabetes Mellitus, Type 2, Risk Factors, Hypoglycemia, Glucose, Hemoglobin A, Glycosylated, Pancreatitis, Chronic, Dyslipidemias, United States Food and Drug Administration, Thiazolidinediones, Hypertension, Natriuretic Peptide, Brain, Myocardial Infarction, Insulin, Stroke, Dipeptidyl Peptidase 4, Risk Reduction Behavior, Pancreatic Neoplasms, Pharmaceutical Preparations, Heart Failure, Hypoglycemic Agents, Glomerular Filtration Rate, Kidney Diseases, Creatinine

< Back to Listings