2. 2011 US Diabetes Statistics Diabetes affects 25.8 million people in the US (8.3% of the U.S. population) – 18.8 million diagnosed, 7.0 million undiagnosed 10.9 million (26.9%) of those aged ≥65 have diabetes 215,000 people <20 79 million US adults >20 years estimated to have had prediabetes in 2010 7 th leading cause of death in the U.S. www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf

3. Percentage of U.S. Adults With Diagnosed Diabetes 1994 2000 No Data 9.0% 2009 www.cdc.gov/diabetes/statistics

4. Cost of Diabetes Total (direct and indirect) estimated diabetes costs in the US in 2007 = $174 billion – Medical expenses for people with diabetes are more than two times higher than for people without diabetes A 50 year old with diabetes dies, on average, 6 years earlier than someone without diabetes Emerging Risk Factors Collaboration. NEJM. 2011; www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf

5. Type 2 Diabetes Pathophysiology 1 Bergenstal RM, et al. Endocrinology. 2001; 2 DeFronzo RA. Diabetes. 1988; 3 Poitout V, et al. Endocrinology. 2002. Gluco- lipotoxicity  Production of glucose in the liver 1,2 Acquired/genetic factors (obesity) 1,2  FFA 1-3 Type 2 DM 1 Inherited/acquired factors  Glucose uptake 1,2 Insulin deficiency, inappropriate glucagon secretion 1,3 Insulin resistance 1 Hyperglycemia 1-3 FFA=free fatty acid Decreased Incretin Effect

6. Current Therapeutic Targets Dopamine Analogs Pramlintide BRAIN PANCREAS Insulin GLP-1 Agonists DPP-4 Inhibitors Sulfonylureas Pramlintide (α cells only) Meglitinides LIVER Metformin Thiazolidinediones (TZD) GI TRACT GLP-1 Agonists Alpha Glucosidase Inhibitors MUSCLE/FAT Metformin Thiazolidinediones (TZD) ?? KIDNEY ??

7. Updated ADA/EASD Consensus Algorithm Nathan DM, et al. Diabetes Care. 2009. At Diagnosis: Lifestyle + Metformin Lifestyle + Metformin + Sulfonylurea Lifestyle + Metformin + Basal Insulin Lifestyle + Metformin + Intensive Insulin Lifestyle + Metformin + Pioglitazone No hypoglycemia, edema/CHF, bone loss Lifestyle + Metformin + GLP-1 agonist No hypoglycemia, weight loss, nausea/vomiting Lifestyle + Metformin + Basal Insulin Lifestyle + Metformin + Pioglitazone + Sulfonylurea STEP 1STEP 2STEP 3 Tier 2: Less well-validated therapies Tier 1: Well-validated therapies

8. AACE/ACE DIABETES ALGORITHM FOR GLYCEMIC CONTROL American Association of Clinical Endocrinologists. AACE/ACE Diabetes Algorithm for Glycemic Control. Available at https://www.aace.com/publications.

9. Risks of Current Therapies: Weight Gain Met-for min DPP-4 Inhib- itor GLP-1 Agonist SUGlinideTZDAGIInsulin Pram- lintide Weight Gain Met- formin DPP-4 Inhib- itor GLP-1 Agonist SUGlinideTZDAGIInsulin Pram- lintide Hypoglycemia Neutral Benefits Causes

10. The Kidneys Play an Important Role in the Handling of Glucose Wright EM, et al. J Intern Med. 2007. Total glucose stored in body ~450 g Glucose utilization ~250 g/day Brain ~125 g/day Rest of body ~125 g/day Glucose in Western diet ~180 g/day Renal glucose production (gluconeogenesis + ~70 g/day glycogenolysis) Renal glucose filtration and reabsorption ~180 g/day Urinary glucose0 g

11. Sodium-Glucose Cotransporters SGLT1SGLT2 SiteMostly intestine with some kidneyAlmost exclusively kidney Sugar specificityGlucose or galactoseGlucose Affinity for glucose High Km = 0.4 Mm Low Km = 2 Mm Capacity for glucose transport LowHigh Role Dietary glucose absorption Renal glucose reabsorption Lee YJ, et al. Kidney Int Suppl. 2007.

12. Altered Renal Glucose Control in Diabetes Renal gluconeogenesis is increased in patients with Type 2 DM Renal contribution to hyperglycemia 3-fold increase relative to patients without diabetes Glucose reabsorption Increased SGLT2 expression and activity in renal epithelial cells from patients with diabetes vs. normoglycemic individuals Marsenic O. Am J Kidney Dis. 2009; Bakris GL, et al. Kidney Int. 2009; Rahmoune H, et al. Diabetes. 2005.

13. Rationale for SGLT2 Inhibitors The SGLT2 is a glucose transporter responsible for 90% of glucose reabsorption Selective SGLT2 inhibitors could reduce blood glucose levels due to increased renal excretion of glucose Mutations in the SGLT2 transporter linked to hereditary renal glycosuria, a relatively benign condition in humans Selective SGLT2 inhibition would cause urine loss of the calories from glucose (200-300 kcal/day), also potentially leading to weight loss Brooks AM, Thacker SM. Ann Pharmacother. 2009; Nair S, et al. J Clin Endocrinol Metab. 2010.

14. Effects of SGLT2 Inhibitors Inhibition of renal tubular Na + -glucose cotransporter Reversal of hyperglycemia Reduction of “glucotoxicity” Insulin sensitivity in muscle and liver Gluconeogenesis Improved beta cell function Brooks AM, Thacker SM. Ann Pharmacother. 2009; Nair S, et al. J Clin Endocrinol Metab. 2010.

15. SGLT2 Inhibitors in Phase 3 Development Dapagliflozin Canagliflozin Empagliflozin Ipragliflozin Tofogliflozin

16. Empagliflozin: Change in A1C N = 408 Baseline A1C = 7.9% *P<.001 vs. placebo †500 mg BID for four weeks, then 1000 mg BID or the maximum tolerated dose Placebo Change in A1C (%) Ferrannini E, et al. Abstract 877. EASD 2010. Randomized, double-blind, 12 week trial comparing empagliflozin and open-label metformin †

17. Empagliflozin: Change in FPG N = 408 *P<.001 vs. placebo †500 mg BID for four weeks, then 1000 mg BID or the maximum tolerated dose Ferrannini E, et al. Abstract 877. EASD 2010. * * * * Placebo 5 mg10 mg25 mgMetformin Change in FPG (mg/dl) Randomized, double-blind, 12 week trial comparing empagliflozin and open-label metformin †

18. Empagliflozin: Change in Weight * N = 408 Baseline BMI = 29 kg/m 2 *P<.001 vs. placebo †500 mg BID for four weeks, then 1000 mg BID or the maximum tolerated dose Ferrannini E, et al. Abstract 877. EASD 2010. * * * Placebo5 mg10 mg25 mgMetformin Change in weight (%) Randomized, double-blind, 12 week trial comparing empagliflozin and open-label metformin †

19. Empagliflozin: Safety Considerations Side effects: – Polyuria (3.3% vs. 0% in placebo), thirst (3.3% vs. 0% in placebo), and nasopharyngitis (2% vs. 1.2% in placebo) were the most frequently reported side effects – UTI 1.2% vs. 1.2% in placebo and 1.3% in metformin Ferrannini E, et al. Abstract 877. EASD 2010. Randomized, double-blind, 12 week trial comparing empagliflozin and open-label metformin †

20. Canagliflozin: Change in A1C *P<.001 vs. placebo calculated using LS means Rosenstock J, et al. Abstract 77-OR. ADA 2010. N = 451 Change in A1C (%) * * * * Mean Baseline A1C (%) 7.718.017.817.577.707.717.62 * * Canagliflozin add-on to metformin, double-blind, placebo-controlled, dose-ranging study (Phase 2)

21. Canagliflozin: Change in FPG Change in FPG (mg/dl) Rosenstock J, et al. Abstract 77-OR. ADA 2010. N = 451 Baseline FGP 162 mg/dl) *P<.001 vs. placebo calculated using LS means * * * * * * Canagliflozin add-on to metformin, double-blind, placebo-controlled, dose-ranging study (Phase 2)

22. Canagliflozin: Change in Weight N = 451 Baseline weight 87 kg *P<.01 vs. placebo calculated using LS means Rosenstock J, et al. Abstract 77-OR. ADA 2010. Change in weight (%) Mean Baseline Weight (kg) 85.587.587.7 87.886.387 * * * * * Canagliflozin add-on to metformin, double-blind, placebo-controlled, dose-ranging study (Phase 2)

23. Canagliflozin: Safety Considerations Side effects: – Mild-to-moderate, transient – Non dose-dependent increase in symptomatic genital infections: 3-8% in canagliflozin vs. 2% in placebo and 2% in sitagliptin – UTI: 3-9% in canagliflozin (no dose-dependency) vs. 6% in placebo and 2% in sitagliptin – Hypoglycemia: 0-6% in canagliflozin (no dose dependency) vs. 2% in placebo and 5% in sitagliptin Rosenstock J, et al. Abstract 77-OR. ADA 2010. Canagliflozin add-on to metformin, double-blind, placebo-controlled, dose-ranging study (Phase 2)

24. Effects of Canagliflozin on Vulvovaginal Candida Colonization (VCC) – Relative to PBO/SITA, CANA treatment increased conversion to positive vaginal Candida culture and VVC events – Incidence of VVC in female subjects was 2.9% and 3.7% with PBO and SITA, respectively, and 10.4% with CANA – None of the VVC events were serious or led to discontinuation; most were treated with topical or oral antifungals, and resolved without study drug interruption Nyirjesy P, et al. Abstract 0032-LB. ADA 2011.

25. Bacteria or UTI Incidence with Canagliflozin –At Week 12, CANA decreased A1C (0.45%-0.73% relative to PBO), lowered weight (1.3%-2.3% relative to PBO), and increased urinary glucose excretion (UGE) (35.4-61.6 mg/mg creatinine) –Conversion from negative baseline urine bacterial culture to positive culture did not differ in pooled CANA group vs pooled PBO/SITA group (4.8% vs. 3.7%, respectively) –UTI AEs (both symptomatic and positive post-baseline urine culture reported as a UTI) occurred in 16 (5.0%) in pooled CANA group and 5 (3.8%) in pooled PBO/SITA group –All UTI AEs considered mild or moderate, and none led to discontinuation Nicolle L, et al. Abstract 0043-LB. ADA 2011.

26. Dapagliflozin Phase 3 Studies: Change in A1C Change in A1C (%) Placebo Dapa 2.5mg Dapa 5mg Dapa 10mg Wilding JPH, et al. Abstract 871. EASD 2010; Strojek K, et al. Abstract 870. EASD 2010; Ferrannini E, et al. Diabetes Care. 2010; Bailey CJ, et al. Lancet. 2010. Baseline- Monotherapy (N=591): 8.6% Met add-on (N=546): 8% SU add-on (N=597): 8.1% Insulin add-on (N=808): 8.5%

27. Change in FPG (mg/dl) Placebo Dapa 2.5mg Dapa 5mg Dapa 10mg Baseline- Monotherapy (N=591): 179 mg/dl Met add-on (N=546): 163.9 mg/dl Insulin add-on (N=808): 178 mg/dl Wilding JPH, et al. Abstract 871. EASD 2010; Ferrannini E, et al. Diabetes Care. 2010; Bailey CJ, et al. Lancet. 2010. Dapagliflozin Phase 3 Studies: Change in FPG

28. Change in weight (kg) Placebo Dapa 2.5mg Dapa 5mg Dapa 10mg Baseline- Monotherapy (N=591): 89.7 kg Met add-on (N=546): 85.9 kg SU add-on (N=597): 81.1 kg Insulin add-on (N=808): 94 kg Wilding JPH, et al. Abstract 871. EASD 2010; Strojek K, et al. Abstract 870. EASD 2010; Ferrannini E, et al. Diabetes Care. 2010; Bailey CJ, et al. Lancet. 2010. Dapagliflozin Phase 3 Studies: Change in Weight

29. Dapagliflozin, Metformin XR, or Both as Initial Therapy Change in A1CChange in FPG Change in Body Weight Henry R, et al. Abstract 307-OR. ADA 2011.

30. Dapagliflozin: Effect on BP and Lipids HDL (% change) – Placebo: +0.4 – Dapa 2.5mg: +1.8 – Dapa 5mg: +3.3 – Dapa 10mg: +4.4 Triglycerides (% change) – Placebo: +2.1 – Dapa 2.5mg: -2.4 – Dapa 5mg: -6.2 – Dapa 10mg: -6.2 Systolic Blood Pressure (mmHg) – Placebo: -0.2 – Dapa 2.5mg: -2.1 – Dapa 5mg: -4.3 – Dapa 10mg: -5.1 Diastolic Blood Pressure (mmHg) – Placebo: -0.1 – Dapa 2.5mg: -1.8 – Dapa 5mg: -2.5 – Dapa 10mg: -1.8 Bailey CJ, et al. Lancet. 2010.

31. Dapagliflozin: Safety Considerations Based on all trials (Monotherapy, metformin add-on, sulfonylurea add-on, and insulin add-on) Side effects: – UTI*: 3.9-13.2% in dapagliflozin vs. 4-6.2% in placebo 1-3 – Genital infections*: 3.9-12.9% in dapagliflozin vs. 0.7-5% in placebo 1-4 – Hypoglycemia: 0-7.9% in dapagliflozin vs. 2.7-4.8% in placebo 2-4 * Most occurred during the first 24 weeks, were generally mild, and responded to routine management 1 Wilding JPH, et al. Abstract 871. EASD 2010; 2 Strojek K, et al. Abstract 870. EASD 2010; 3 Ferrannini E, et al. Diabetes Care. 2010; 4 Bailey CJ, et al. Lancet. 2010.

32. Dapagliflozin: Sulfonylurea Comparator Study Results Average A1C: 7.72% Change in A1C: -0.52% for dapagliflozin vs. -0.52% for glipizide Weight change: -3.2 kg for dapagliflozin vs. +1.4 kg for glipizide Hypoglycemic episodes: 3.5% for dapagliflozin vs. 40.8% with glipizide Significant reductions in diastolic and systolic blood pressure and improvement in HDL with dapagliflozin vs. glipizide (P<.0001) Side effects: UTI: 10.8% with dapagliflozin vs. 6.4% with glipizide (actively solicited) Genital infection: 12.3% with dapagliflozin vs. 2.7% with glipizide (actively solicited) Renal impairment: 5.9% with dapagliflozin vs. 3.4% with glipizide Nauck MA, et al. Diabetes Care. 2011. Randomized, double-blind, parallel-group, multicenter trial comparing dapagliflozin to glipizide as add-on to metformin

33. Dapagliflozin + Insulin for 48 Weeks Wilding J, et al. Abstract 0021-LB. ADA 2011. A1C reductions from baseline for PLA and DAPA 2.5, 5, and 10mg at 24 weeks were maintained at 48 weeks – (−0.43%, −0.74%, −0.94%, −0.93%, respectively) 24 week body weight reductions with DAPA were maintained at 48 weeks – −1.5kg with DAPA 10mg vs. +0.9kg with PLA AEs, serious AEs, and discontinuations were balanced across all groups Actively solicited s/sx suggestive of UTI and genital infections (GI) were more with DAPA vs PLA – UTI 7.9%-10.8% vs. 5.1%; GI 6.4%-10.7% vs. 2.5% – most events were reported during the first 24 weeks, were of mild/moderate intensity and responded to standard treatment AE = Adverse events s/sx = signs and symptoms

34. Dapagliflozin + Insulin for 48 Weeks: Insulin Dose Wilding J, et al. Abstract 0021-LB. ADA 2011. 15 10 5 0 -5 04812162024283236404448 Timepoint (weeks) PLA + INS DAPA 2.5 mg + INS DAPA 5 mg + INS DAPA 10 mg + INS Insulin dose (IU/day) Adjusted mean change from baseline ± SE

35. Long-Term Efficacy of Dapagliflozin + Metformin Week 102 Results Adj. Mean ∆ From Baseline PBO+MET DAPA 2.5mg + MET DAPA 5mg + MET DAPA 10mg + MET A1C, %0.02-0.48-0.58-0.78 FPG, mg/dL-10.4-19.3-26.5-24.5 Weight, kg-0.7-2.2-3.4-2.8 % with ≥1 hypoglycemic event 5.83.65.15.2 Bailey CJ, et al. Abstract 0988-P. ADA 2011.

36. Potential SGLT2 Safety Considerations??? Evidence Demonstrates Urinary tract/genital infections Questions Hepatic toxicity? Breast and bladder cancer?? Intravascular volume depletion due to osmotic diuresis?? Nephrotoxicity (AGEs)?? Drug-drug interactions?? Evidence Does Not Demonstrate Electrolyte imbalance (Na +, K +, Ca ++, PO 4 ) Increased risk for hypoglycemia Nocturia

37. Dapagliflozin PDUFA Date The FDA issued a Complete Response Letter to the makers of dapagliflozin on January 19, 2012 requesting additional information.

38. SGLT2 Inhibitors: A New Era in Diabetes Treatment?? In treatment-naive patients with newly-diagnosed Type 2 DM, SGLT2 inhibitors resulted in: Clinically meaningful decreases in A1C and fasting plasma glucose with no increased risk of hypoglycemia Also improved glycemic control in combination with a variety of other antihyperglycemic agents Metformin, sulfonylureas, insulin Side effects generally appear to be mild and transient, while avoiding increased risk of hypoglycemia or weight gain