Improving Outcomes with SGLT2 Cotransporter Inhibitors in Challenging T2DM Patients Part 2

Efficacy of SGLT2 Inhibition in T2DM and Application to Clinical Practice: Approved Agents Ele Ferrannini, MD Department of Clinical & Experimental Medicine University of Pisa Pisa, Italy

Dapagliflozin: A1C Ferrannini E et al. Diabetes Care. 2010;33: Change from baseline in A1C (%) Week 24 value (95% CI) Placebo (n=75)-0.23 (0.43 to -0.02) Dapagliflozin 2.5 mg (n=65)-0.58 (-0.80 to -0.36) Dapagliflozin 5 mg (n=64)-0.77 (-0.99 to -0.55) Dapagliflozin (n=70)-0.89 (-1.1 to -0.67)

Dapagliflozin: FPG Ferrannini E et al. Diabetes Care. 2010;33: Change from baseline in FPG (mg/dL) Week 24 value (95% CI) Placebo (n=75)-4.1 (-11.8 to 3.5) Dapagliflozin 2.5 mg (n=65)-15.2 (-23.5 to -7.0) Dapagliflozin 5 mg (n=64)-24.1 (-32.5 to -15.6) Dapagliflozin (n=70)-28.8 (-36.8 to -20.9)

Introduction to SGLT2 Inhibitors Mode of action Efficacy Durability Limitations

Durability of Dapagliflozin vs Glipizide as Add-on to Metformin: 4-year Data * Data are adjusted mean change from baseline ± 95% CI derived from a longitudinal repeated measures model DAPA, dapagliflozin; GLIP, glipizide; MET, metformin, SBP, systolic blood pressure Sample size (excluding data after rescue). * DAPA + MET GLIP + MET Study week Sample size (excluding data after rescue). * Study week Rescue therapy available Change to A1C (%)* DAPA + MET GLIP + MET Rescue therapy NOT available Dapagliflozin + MetforminGlipizide+ Metformin Week 208 values 0.20% (95% CI: 0.05, 0.36) DBL -0.30% (95% CI: -0.51, -0.00) -0.10% (95% CI: -0.25, 0.06) Sample size (excluding data after rescue). * Study week Change in weight (kg)* Abstract results exclude data after rescue 0.73 kg (95% CI: 0.06, 1.40) DBL kg (95% CI: ) kg (95 CI: -4.30, -3.01) Change in SBP (mm Hg)* mmHg (95% CI: -1.66, 1.61) DBL –3.67 mm Hg (95% CI: -5.92, -1.41) mm Hg (95% CI: -5.24, -2.14) Langkilde A, et al. Presented at the 49th EASD Annual Meeting; Barcelona, Spain; Sept, Abstract 934. DAPA + MET GLIP + MET

Introduction to SGLT2 Inhibitors Mode of action Efficacy Durability Limitations

GlycosuriaDependson GFR and Glycemia Dependency of postdose glycosuria on eGFR and FPG concentration – SGLT2 inhibitor induced urinary glucose excretion was a direct function of eGFR and fasting glucose levels Ferrannini E, et al. Diabetes Care. 2013;36:

Summary: Application of SGLT2 Inhibitors in Clinical Practice Rapid onset of action Dose-dependent glycosuria and natriuresis No hypoglycemia in monotherapy Potential use in type 1 diabetes Consistent reduction in body weight and blood pressure Durability of action Limited efficacy in severe renal impairment

Efficacy of SGLT2 Inhibition in T2DM: Investigational Agents Mark W. Stolar, MD Associate Professor Clinical Medicine, General Internal Medicine and Geriatrics Feinberg School of Medicine Northwestern University Chicago, IL

SGLT2 Inhibitors Potential Advantages Use associated with weight loss Hypoglycemia low risk as monotherapy Increased urinary Na excretion associated with slight drop in SBP No untoward sequelae in patients with inherited defects of SGLT2 Total metabolic effects? Potential Disadvantages Modest reduction in A1C Polyuria complaints? Any compensatory increase in hunger? Urinary tract infections or genitourinary fungal infections? Exacerbate diabetic nephropathy?

RationaleforMultipleSGLT2 Investigational Agents Greater potency at the SGLT2 receptor Higher SGLT2 specificity may lead to fewer side effects Partial SGLT1 agonism might improve glycemic efficacy Horse race effect

SGLT2 Inhibitors Rosenwasser RF, et al. Diab Met Syn Obes. 2013;6:453–467. SGLT2 Inhibitor Company Dapagliflozin Empagliflozin Ipragliflozin Luseogliflozin Tofogliflozin Remogliflozin Janssen Bristol-Myers Squibb, AstraZeneca Boehringer Ingelheim, Eli Lilly Astellas, Kotobuki Taisho Chugai GlaxoSmithKline Approved by FDA Phase of Development Phase 3 Phase 2b CanagliflozinApproved by FDA

ComparisonofPotencyand Selectivity of SGLT2 Inhibitors Meng W, et al. J Med Chem. 2008;51:1145–1149. Liang Y, et al. Presented at American Diabetes Association Annual Meeting; June 2009; New Orleans, LA. Abstract 521-P. *Investigational; not approved by the US FDA or available for use in patients Grempler R et al. Presented at American Diabetes Association Annual Meeting; June 2009; New Orleans, LA. Abstract 521-P. Fujimori Y, et al. J Pharmacol Exp Ther. 2008;327:268–276. Selectivity SGLT1/2IC 50 Potency IC 50 vs phlorizin O-glycosides Phlorizin91 T-1095A*~4~3 Sergliflozin A*~300~8 Remogliflozin*~360~3 C-glycosides Dapagliflozin~1200~32 Canagliflozin~220~9 Empagliflozin*>2500~27

Empagliflozin: Change in A1C *P ˂.001 vs. placebo †500 mg BID for 4 weeks, then 1000 mg BID or the maximum tolerated dose Reprinted with permission from Ferrannini E, et al. Randomized, double-blind, 12-week trial comparing empagliflozin and open-label metformin ∆A1C (%)∆A1C (%) * Presented at the 46th EASD Annual Meeting; Sept 2010; Stockholm, Sweden. Abstract * mg 10 mg 25 mg MET 500 mg BID† * *

BRIGHTEN Trial: Ipragliflozin Kashiwagi A, et al. Presented at the 47th EASD Annual Meeting; Sept 2011; Lisbon, Portugal. Abstract 149. Phase 3, double-blind, placebo-controlled study:16 weeks (n=129) Ipragliflozin treatment group 50mg QD (n=62) – Ipragliflozin achieved a mean decrease of 1.23% in A1C compared with placebo (P<0.001) – FPG was reduced by 45.8mg/dL compared with placebo (P<0.001) – Total body weight was reduced by 1.47 kg compared with placebo (P<0.001) – Similar adverse events occurredin about half the patients in both cohorts – 1 case of hypoglycemia and 2 cases of genital infections in ipragliflozin group

MetabolicResponse to Empagliflozin in Patients with T2DM N=66; dose 25 mg/day Study I: Single dose Study II: Following 4 weeks of treatment Study I: Glycosuria (7.8 [4.4] g over 3 hours of fasting, median), which led to an increase in endogenous glucose production (EGP, 13.8 [5.2] to 17.6 [4.8] μmol.kgFFM-1.min-1, P<0.0001) matching the glycosuria Changes were maintained in Study II Empagliflozin lowers fasting and postprandial glycemia – Increasing total glucose removal despite a compensatory increase in endogenous glucose production – Improving ß-cell function – Shifting substrate utilization from glucose to lipid Ferrannini E, et al. Presented at the 49th EASD Annual Meeting; Sept 2013; Barcelona, Spain. Abstract 932.

Empagliflozin Improves Glycemic Parameters and Cardiovascular Risk Factors in Patients with T2DM Hach T, et al. Presented at the 49th EASD Annual Meeting; Sept 2013; Barcelona, Spain. Abstract 943. Pooled data from 4 phase 3 trials (N=2,477) 10 mg or 25 mg given for 24 weeks as monotherapy, add-on to metformin, add-on to metformin + SU, or add-on to pioglitazone ± metformin Adjusted means based on ANCOVA with last observation carried forward imputation; values on rescue medication were excluded from analysis of A1C, FPG, weight and BP. † Full analysis set (all randomized and treated patients who had a baseline A1C value). ‡ Inclusion criteria: A1C ≥7.0% to ≤10.0%. ***P<0.001 vs placebo. PlaceboEmpagliflozin 10 mg Empagliflozin 25 mg Baseline A1C % (SE) †‡ 8.02 (0.03)7.98 (0.03)7.96 (0.03) Change from baseline in A1C at week 24 (SE) ‡ (0.03)-0.70 (0.03)***-0.76 (0.03)*** Baseline FPG, mg/dL (SE) † (1.3)152.6 (1.2) Change from baseline in FPG at week 24, mg/dL (SE) † 7.4 (1.0)-20.5 (1.0)***-23.2 (1.0)*** Baseline body weight, kg (SE) † (0.66)78.77 (0.65)79.10 (0.66) Change from baseline in body weight at week 24, kg (SE) † (0.09)-2.05 (0.09)***-2.25 (0.09)*** Baseline SBP, mm Hg (SE) † (0.5)129.6 (0.5)129.0 (0.5) Change from baseline in SBP at week 24, mm Hg (SE) † -0.5 (0.4)03.9 (0.4)***-4.3 (0.4)*** Baseline DBP, mm Hg (SE) † 78.0 (0.3)78.7 (0.3)78.3 (0.3) Change from baseline in DBP at week 24, mm Hg (SE) † -0.6 (0.2)-1.8 (0.2)***-2.0 (0.2)***

Empagliflozin as Add-on to Basal Insulin for 78 Weeks Johansen OE, et al. Presented at the 49th EASD Annual Meeting; Sept 2013; Barcelona, Spain. Abstract 931. A1C and insulin dose: Adjusted means based on ANCOVA in full analysis set of “completers” (patients who completed minimum treatment duration of 119 days [for week 18 data] and 532 days [for week 78 data] with last observation carried forward imputation. FPG and body weight: Adjusted means based on ANCOVA in full analysis set (all treated patients with baseline A1C) with last observation carried forward imputation. **P<0.01 vs placebo; *** P<0.001 vs placebo. Randomized, double-blind to receive empagliflozin 10 mg (n=169) or 25 mg qd (n=155) or placebo (n=170) for 78 weeks PlaceboEmpagliflozin 10 mg Empagliflozin 25 mg Baseline ±SE A1C week 18 analysis set (%)8.10± ± ±0.08 Change from baseline ±SE in A1C at week 18 (%)-0.01± ±0.07***-0.71±0.07*** Baseline ±SE A1C week 78 analysis set (%)8.09± ± ±0.08 Change from baseline ±SE in A1C at week 78 (%)-0.02± ±0.08***-0.64±0.09*** Baseline ±SE insulin dose (IU)47.84± ± ±2.79 Change from baseline ±SE in insulin dose at week 78 (IU)5.45± ±1.48**-0.47±1.59** Baseline ±SE in FPG (mg/dL)142±4138±4146±4 Change from baseline ±SE in FPG at week 78 (mg/dL)3±3-10±3**-15±3*** Baseline ±SE body weight (kg)90.5± ± ±1.7 Change from baseline ±SE in body weight at week 78 (kg)0.7± ±0.5***-2.0±0.5***

Empagliflozin ImprovesBlood Pressure in Patients with T2DM Tikkanen I, et al. Presented at the 49th EASD Annual Meeting; Sept 2013; Barcelona, Spain. Abstract 942. Adjusted mean based on ANCOVA in full analysis set with last observation carried forward (LOCF) imputation. For blood pressure parameters, values following a change in antihypertensive therapy were set to missing and imputed via LOCF. Values after start of antidiabetic rescue medication were set to missing and imputed via LOCf for all parameters. For each dose group, statistical testing of primary and key secondary endpoints was hierarchical at alpha = Further endpoints were called statistically significant if P-values were smaller than nominal alpha = *P<0.05 vs. placebo; **P<0.01 vs. placebo; ***P<0.001 vs. placebo. PlaceboEmpagliflozin 10 mg Empagliflozin 25 mg Baseline A1C % (SE)7.90 (0.04)7.87 (0.05)7.92 (0.04) Change from baseline in A1C % (SE)0.03 (0.04)-0.59 (0.04)***-0.62 (0.04)*** Baseline 24-hour SBP (ABPM), mm Hg (SE) (0.72) (0.78) (0.73) Change from baseline in 24-hour SBP (ABPM) mm Hg (SE)0.48 (0.49)-2.95 (0.48)***-3.68 (0.48)*** Baseline mean seated office SBP, mm Hg (SE) (0.75) (0.73) *** (0.76) *** Change from baseline in mean seated office SBP, mm Hg (SE)-0.67 (0.70)-4.60 (0.69)-5.47 (0.69) Baseline 24-hour DBP, mm Hg (SE)75.16 (4.5)75.13 (0.50)74.64 (0.45) Change from baseline in 24-h DBP, mm Hg (SE)0.32 (0.29)-1.04 (0.28)*** Baseline mean seated office DBP, mm Hg (SE)83.67 (0.43)84.13 (0.44)83.82 (0.41) Change from baseline in mean seated office DBP, mm Hg (SE)-1.13 (0.39)-3.06 (0.39)***-3.02 (0.39)***

Change in Body Weight in Week Monotherapy Studies of SGLT2 Inhibitors Rosenstock J, et al. Diabetologia. 2010;53(Suppl 1):S349. Seman L, et al. Presented at the 47th EASD Annual Meeting; Sept 2011; Lisbon, Portugal. Abstract Baseline body weight, kg69.37 Kashiwagi A, et al. Presented at the 47th EASD Annual Meeting; Sept 2011; Lisbon, Portugal. Abstract (kg) from -0.5 baseline mg qd100 mg qd 200 mg qd 300 mg qdPlacebo5 mg qd Empagliflozin 12 wk study (N=408) 10 mg qd25 mg qd Placebo Ipragliflozin 16 wk study in Japanese patients (N=129) 50 mgPlacebo Cangliflozin 12 wk study in Japanese patients (N=383) Mean change in body weight Empagliflozin & Ipragliflozin not FDA approved Statistical significance not reported

Change in Body Weight in 12-Week Add-on to Metformin Studies of SGLT2 Inhibitors Baseline body weight, kg87 Rosenstock J, et al. Diabetologia. 2010;53(Suppl 1):S349. Seman L, et al. Presented at the 47th EASD Annual Meeting; Sept 2011; Lisbon, Portugal. Abstract mg qd mg qd mg qd mg qd mg bid SitagliptinSitagliptin 1 mg qd 5 mg qd 10 mg qd 25 mg qd 50 mg qd PlaceboPlacebo S i t a gl i p t i n PlaceboPlacebo Canagliflozin (placebo adjusted values) 12 wk study (N=451) Empagliflozin 12 wk study (N=495) Mean change in body weight (kg) from baseline Empagliflozin not FDA approved Statistical significance not reported