1. Clinical Outcomes with Newer Antihyperglycemic Agents FDA-Mandated CV Safety Trials 1

2. TECOS (TRIAL EVALUATING CARDIOVASCULAR OUTCOMES WITH SITAGLIPTIN) Clinical Outcomes with Antihyperglycemic Agents 2

3. Clinical Outcomes with Sitagliptin Study Design N=14,671 patients with T2D and CVD Randomization –Sitagliptin: n=7332 (6972 completed) –Placebo: n=7339 (6905 completed) Noninferiority study: 1.3 marginal upper boundary of 2-sided 95% CI –Primary composite outcome: cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina –Secondary composite outcome: cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. Key Results Median follow-up: 3.0 years Least squares mean difference in A1C: -0.29% (95% CI -0.32 to -0.27) for sitagliptin vs placebo Noninferior to placebo for cardiovascular outcomes –Primary HR: 0.98 (0.88-1.09); P<0.001 –Secondary HR: 0.99 (0.89-1.11); P<0.001 No difference between sitagliptin and placebo in incidence of infections, cancer, renal failure, hypoglycemia, or noncardiovascular death 3 TECOS CI, confidence interval; HR, hazard ratio; TECOS, Trial Evaluating Cardiovascular Outcomes with Sitagliptin. Green JB, et al. N Engl J Med. 2015;373:232-242.

4. Primary and Secondary Outcomes with Sitagliptin 4 TECOS Per Protocol Analysis (n=14,523) *Cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina. † Secondary composite: cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. NF, noninferiority; TECOS, Trial Evaluating Cardiovascular Outcomes with Sitagliptin. Green JB, et al. N Engl J Med. 2015;373:232-242. Hazard ratio (95% CI)P value Primary composite endpoint*0.98 (0.88-1.09)<0.001 (NF) Secondary composite endpoint † 0.99 (0.89-1.11)<0.001 (NF) Acute pancreatitis1.80 (0.86-3.76)0.12 Any cancer (except nonmelanoma skin cancer) 0.93 (0.89-1.44)0.38 Pancreatic cancer0.91 (0.37-2.25)0.85 Severe hypoglycemia1.13 (0.89-1.44)0.31 Favors sitagliptin

5. Individual Secondary Outcomes with Sitagliptin 5 TECOS Intent to Treat Analysis (n=14,671) CV, cardiovascular; MI, myocardial infarction; NF, noninferiority; TECOS, Trial Evaluating Cardiovascular Outcomes with Sitagliptin. Green JB, et al. N Engl J Med. 2015;373:232-242. Favors sitagliptin Hazard ratio (95% CI)P value CV death1.03 (0.89-1.19)0.71 Hospitalization for unstable angina0.90 (0.70-1.16)0.42 Fatal or nonfatal MI0.95 (0.81-1.11)0.49 Fatal or nonfatal stroke0.97 (0.79-1.19)0.76 Death from any cause1.01 (0.90-1.14)0.88 Hospitalization for heart failure1.09 (0.83-1.20)0.98 Hospitalization for heart failure or CV death1.02 (0.90-1.15)0.74

6. Clinical Outcomes with Sitagliptin 6 TECOS (n=14,671) TECOS, Trial Evaluating Cardiovascular Outcomes with Sitagliptin. Green JB, et al. N Engl J Med. 2015;373:232-242.

7. EXAMINE (EXAMINATION OF CARDIOVASCULAR OUTCOMES WITH ALOGLIPTIN VERSUS STANDARD OF CARE) Clinical Outcomes with Antihyperglycemic Agents 7

8. Clinical Outcomes with Alogliptin Study Design N=5380 patients with T2D and ACS Randomization –Alogliptin: n=2701 –Placebo: n=2679 Noninferiority study: prespecified HR margin = 1.3 for primary endpoint –Primary composite endpoint: CV death, nonfatal MI, or nonfatal stroke –Secondary: CV death, nonfatal MI, nonfatal stroke, urgent revascularization for unstable angina Key Results Median follow-up: 18 months Least squares mean difference in A1C: -0.36% (95% CI -0.43 to -0.28; P<0.001) for alogliptin vs placebo CV outcomes –Primary HR: 0.96 (≤1.16); P=0.32 –Secondary HR: 0.95 (≤1.14*); P=0.26 No difference between alogliptin and placebo in incidence of acute and chronic pancreatitis, cancer, renal impairment, angioedema, or severe hypoglycemia 8 EXAMINE *Upper boundary of 1-sided repeated CI, alpha level 0.01. CI, confidence interval; CV, cardiovascular; EXAMINE, Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care; HR, hazard ratio; MI, myocardial infarction. White W, et al. N Engl J Med. 2013;369:1327-1335.

9. Alogliptin CV Outcomes and Mortality 9 EXAMINE EXAMINE, Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care. White W, et al. N Engl J Med. 2013;369:1327-1335. CV Death, Nonfatal MI, or Nonfatal StrokeCV Death All-Cause Death

10. Hazard ratio (95% CI)P value Primary composite0.96 (≤1.16)*0.32 Primary endpoint components CV death0.79 (0.6-1.04)0.10 Nonfatal MI1.08 (0.88-1.33)0.47 Nonfatal stroke0.91 (0.55-1.50)0.71 Primary secondary endpoint † 0.95 (≤1.14)*0.26 Death from any cause0.85 (0.66-1.10)0.21 Clinical Outcomes with Alogliptin 10 EXAMINE Safety Endpoints (n=5380) *Upper boundary of 1-sided repeated CI, alpha level 0.01. † CV death, nonfatal MI, nonfatal stroke, urgent revascularization for unstable angina. CI, confidence interval; CV, cardiovascular; EXAMINE, Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care; MI, myocardial infarction. White W, et al. N Engl J Med. 2013;369:1327-1335. Favors alogliptin

11. SAVOR-TIMI (SAXAGLIPTIN ASSESSMENT OF VASCULAR OUTCOMES RECORDED IN PATIENTS WITH DIABETES MELLITUS–THROMBOLYSIS IN MYOCARDIAL INFARCTION) Clinical Outcomes with Antihyperglycemic Agents 11

12. Clinical Outcomes with Saxagliptin Study Design N=16,492 patients with T2D and CVD or CVD risk Randomization –Saxagliptin: n=8280 –Placebo: n=8212 Superiority study with provision to test for noninferiority –Primary composite endpoint: CV death, nonfatal MI, or nonfatal ischemic stroke –Secondary: CV death, nonfatal MI, nonfatal ischemic stroke, hospitalization for HF, coronary revascularization, or unstable angina Key Results Median follow-up: 2.1 years Endpoint A1C –Saxagliptin: 7.7% ± 1.4% (P<0.001 vs placebo) –Placebo: 7.9% ± 1.5% CV outcomes –Primary HR: 1.00 (0.89-1.27); P=0.99 –Secondary HR: 1.02 (0.94-1.11); P=0.66 Higher incidence of HF hospitalization in saxagliptin group No difference between groups in incidence of acute or chronic pancreatitis; fewer cases of pancreatic cancer in saxagliptin group; more cases of nonfatal angioedema in saxagliptin group (8 vs 1) 12 SAVOR-TIMI CI, confidence interval; CV, cardiovascular; HF, heart failure; HR, hazard ratio; MI, myocardial infarction; SAVOR-TIMI, Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus–Thrombolysis in Myocardial Infarction. Scirica BM, et al. N Engl J Med. 2013;369,1317-1326.

13. Clinical Outcomes with Saxagliptin 13 SAVOR-TIMI Prespecified Composite Endpoints and Mortality (n=16,492) Hazard ratio (95% CI)P value Primary composite endpoint*1.00 (0.89-1.27)0.99 Secondary composite endpoint † 1.02 (0.94-1.11)0.66 Death from any cause1.11 (0.96-1.27)0.15 CV death1.03 (0.87-1.22)0.52 *CV death, nonfatal MI, or nonfatal ischemic stroke; † CV death, nonfatal MI, nonfatal ischemic stroke, hospitalization for HF, coronary revascularization, or unstable angina. CI, confidence interval; CV, cardiovascular; HF, heart failure; MI, myocardial infarction; SAVOR-TIMI, Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus–Thrombolysis in Myocardial Infarction. Scirica BM, et al. N Engl J Med. 2013;369,1317-1326. Favors saxagliptin

14. Individual Secondary Outcomes with Saxagliptin 14 *Doubling of creatinine, initiation of dialysis, renal transplantation, or creatinine >6.0 mg/dL CI, confidence interval; CV, cardiovascular; SAVOR-TIMI, Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus–Thrombolysis in Myocardial Infarction. Scirica BM, et al. N Engl J Med. 2013;369,1317-1326. SAVOR-TIMI Prespecified Individual Endpoints (n=16,492) Favors saxagliptin Hazard ratio (95% CI)P value Myocardial infarction0.95 (0.80-1.12)0.52 Ischemic stroke1.11 (0.80-1.12)0.38 Hospitalization for unstable angina1.19 (0.89-1.60)0.24 Hospitalization for heart failure1.27 (1.07-1.51)0.007 Hospitalization for coronary revascularization0.91 (0.80-1.04)0.18 Renal endpoint*1.08 (0.88-1.32)0.46 Hospitalization for hypoglycemia1.22 (0.82-1.83)0.33

15. Hazard ratio (95% CI) P value eGFR ≤60 mL/min1.36 (1.07-1.71)0.03 eGFR >60 mL/min1.16 (0.89-1.51)0.27 No prior heart failure1.30 (1.03-1.65)0.03 Prior heart failure1.23 (0.94-1.59)0.13 No risk factors*1.15 (0.81-1.63)0.03 1 risk factor1.35 (1.06-1.72)0.02 2 risk factors1.22 (0.86-1.73)0.27 Q4 NT-proBNP (333-46,627 pg/mL)1.31 (1.04-1.66)0.02 Baseline Characteristics and Risk of HF Hospitalization With Saxagliptin 15 *eGFR ≤60 mL/min or history of previous HF. HF, heart failure; NT-proBNP, N-terminal pro B-type natriuretic peptide; Q, quartile; SAVOR-TIMI, Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus–Thrombolysis in Myocardial Infarction. Scirica BM, et al. Circulation. 2014;130:1579-1588. SAVOR-TIMI Post-hoc Analysis (n=16,492) Favors saxagliptin

16. eGFR (mL/min) HF historyNo. HF risk factors † NT-proBNP quartiles (pg/mL) No. excess HHF events in patients treated with saxagliptin vs placebo per 1000 pt-y (5-64)(65-141) (142- 333) (334- 46,647) n =1163748551438721051041851888663076 3073 15 160490107 Risk of HF Hospitalization with Saxagliptin vs Placebo 16 Absolute risk difference* *Saxagliptin vs placebo. † eGFR ≤60 mL/min or history of previous HF. HF, heart failure; HHF, hospitalizations for heart failure. Scirica BM, et al. Circulation. 2014;130:1579-1588. SAVOR-TIMI Post-hoc Analysis (n=16,492)

17. EMPA-REG OUTCOME (EMPAGLIFLOZIN CARDIOVASCULAR OUTCOME EVENT TRIAL IN TYPE 2 DIABETES MELLITUS PATIENTS) Clinical Outcomes with Antihyperglycemic Agents 17

18. Clinical Outcomes with Empagliflozin Study Design N=7020 patients with T2D and CVD Randomization –Empagliflozin: n=4687 –Placebo: n=2333 Noninferiority study: prespecified HR margin = 1.3 for primary endpoint –Primary endpoint: composite of CV death, nonfatal MI (excluding silent MI), or nonfatal stroke –Secondary endpoint: composite of CV death, nonfatal MI (excluding silent MI), nonfatal stroke, and hospitalization for unstable angina Key Results Median follow-up: 3.1 years Week 206 A1C, difference from placebo –Empagliflozin 10 mg: -0.24% (955 CI, -0.40% to -0.08%) –Empagliflozin 25 mg: -0.36% (95% CI, -0.51% to -0.20%) CV outcomes (pooled analysis) –Primary: HR 0.86 (95% CI 0.74 to 0.99); P=0.04 for superiority –Secondary HR: 0.89 (95% CI 0.78 to 1.01); P<0.001 for noninferiority and P=0.08 for superiority Significantly lower rates of all-cause death, CV death, and HF hospitalization with empagliflozin Increased rates of genital infections in empagliflozin-treated patients 18 EMPA-REG OUTCOME CI, confidence interval; CV, cardiovascular; HF, heart failure; HR, hazard ratio; MI, myocardial infarction. Zinman B, et al. N Engl J Med. 2015;373:2117-2128.

19. Clinical Outcomes with Empagliflozin 19 EMPA-REG OUTCOME Pooled Analysis (N=7020) *CV death, nonfatal MI (excluding silent MI), or nonfatal stroke; † CV death, nonfatal MI (excluding silent MI), nonfatal stroke, and hospitalization for unstable angina. CI, confidence interval; CV, cardiovascular; HF, heart failure; HR, hazard ratio; MI, myocardial infarction. Zinman B, et al. N Engl J Med. 2015;373:2117-2128. Hazard ratio (95% CI)P value Primary composite endpoint*0.86 (0.74-0.99)0.04 Secondary composite endpoint † 0.89 (0.78-1.01)0.08 Death from any cause0.68 (0.57-0.82)<0.001 CV death0.62 (0.49-0.77)<0.001 Fatal or nonfatal MI0.87 (0.70-1.09)0.23 Hospitalization for HF0.65 (0.50-0.85)0.002 Hospitalization for HF or CV death0.66 (0.55-0.79)<0.001 Favors empagliflozin

20. Clinical Outcomes with Empagliflozin 20 EMPA-REG OUTCOME Pooled Analysis (N=7020) *CV death, nonfatal MI (excluding silent MI), or nonfatal stroke; † CV death, nonfatal MI (excluding silent MI), nonfatal stroke, and hospitalization for unstable angina. CI, confidence interval; CV, cardiovascular; HF, heart failure; HR, hazard ratio; MI, myocardial infarction. Zinman B, et al. N Engl J Med. 2015;373:2117-2128.