CURRENT ISSUES IN DIABETES MANAGEMENT

CURRENT ISSUES IN DIABETES MANAGEMENT Robert B. Baron MD MS Professor and Associate Dean UCSF School of Medicine Declaration of full disclosure: No conflict of interest Diabetes Mellitus: U.S. Impact DIABETES 16.7 Million IFG (8.3%) 12.3 Million (6.3%) ~1 Million Type 1 ~16 Million Type 2 TOTAL: 29 Million (14.4%) 2/3 Diagnosed 1/3 Undiagnosed (4.9 Million) Estimates of Diagnosed Diabetes in Adults, 2005 CDC, 2005

Case 1 Ms. EH is a 68 y.o. woman with type 2 diabetes, hypertension, and coronary heart disease (s/p MI in 2002). Meds: Metformin, glipizide, aspirin, lisinopril, metoprolol, and simvastatin Exam: BP 135/85, HR 72, BMI 29 kg/m 2 Normal heart, lungs, extremities Abraira, DM, Obes, Met, 2008 Case 1 Her glycemic goal should be: 1. HbA1c <6.0% 2. HbA1c <6.5% 3. HbA1c <7.0% 4. HbA1c <7.5% Abraira, DM, Obes, Met, 2008 2009 Practice Guideline Revisions Testing should begin at age 10, and repeated every 3 years (if overweight plus two risks) Patients with impaired glucose tolerance or impaired fasting glucose should be referred for weight loss (5-10%) and exercise (150 minutes per week) General Hb AIC goal: < 7% ADA Diabetes Care, 2009 Diabetes Care, Jan 2008

2009 Practice Guideline Revisions Selected patients: lower goals (well below 7%) Short duration of DM, long life expectancy, and no CVD Selected patients: higher goals (above 7%) Hypoglycemia, limited life expectancy, advanced vascular complications, co-morbid conditions, and those for whom the goal is difficulty to obtain ADA Diabetes Care, 2009 Diabetes Care, Jan 2008 2009 Practice Guideline Revisions Bariatric Surgery should be considered for those with BMI over 35 Pneumonia vaccine to all over age 2 Hypertension: Goal <130/80: Use ACE or ARB; if needed, thiazide or loop diuretic (depending on GFR above or below 30) Lipids: <100 (<70 optional). If goals not reached with high dose statins, reduction of 30-40% is acceptable goal ADA Diabetes Care, 2009 Diabetes Care, Jan 2008 2009 Practice Guideline Revisions ASA: over age 40 and for those with other CHD risk factors Foot care: annual exam (inspection, pulses, monofilament, vibration, pinprick, reflexes) Critically ill surgical patients: Blood sugar goal close to 110 and generally <140 Critically ill non-surgical patients: Goal <140 ADA Diabetes Care, 2009 Diabetes Care, Jan 2008

Glycemic Control Update 3 new trials ADVANCE ACCORD VA Diabetes Trial Older studies (UKPDS, VA, UGDP) Ongoing studies ADVANCE TRIAL RCT in DM 2; 11,140 patients; 20 countries; 5 yr Intensive vs. standard BS control Intensive HbA1C goal 6.5% or less Intensive: 6.5% Standard: 7.3% Outcome: composite macrovascular and/or microvascular events ADVANCE, NEJM 2008 ADVANCE TRIAL Standard Intensive p Combined Events 20.0% 18.1% 0.01 Microvasc events 10.9% 9.4% 0.01 Nephropathy 5.2% 4.1% 0.006 No differences in: Macrovascular events CV death Death from all causes ADVANCE, NEJM, 2008

ACCORD Trial NIH RCT in DM 2, 10,251 patients, known CVD or risk factors, mean A1c 8.1% Intensive vs. standard BP (120 v. 140) Lipid control (fibrates v. statins + fibrates Normalization v. standard BS control (A1c 6 v. 7-7.9) Outcomes: CV events. Also microvascular events, quality of life, others ACCORD, NEJM, 2008 ACCORD trial Intensive n=5,128 Standard n=5,123 HR (95% CI) A1c achieved: 6.5% 7.5% - 1 outcome: 352 371 0.90 (0.78-1.04) Total mortality 5.0% 3.1% 1.22 (1.01-1.46) CVD mortality 2.6% 1.8% 1.35 (1.04-1.76) Hypoglycemia 10.5% 3.5% - Wt. gain>10 kg 27.8% 14.1% - NEJM, 2008 ACCORD Trial Standard Intensive Deaths 203 257 11/1000/y 14/1000/y Number Needed to Harm: 333 February 2008 (after 3.5 years): NIH stops this arm of study

Differences between ACCORD and ADVANCE Mean age Duration DM Target A1c Statin use in trial TZD use in trial ASA use in trial Death (any cause) Weight gain in trial (kg) Major hypoglycemia ACCORD 62 10 <6.0 88 vs. 88 92 vs. 58 76 vs. 76 5.0 vs. 3.1 3.5 vs. 0.4 3.1 vs. 10 ADVANCE 66 8 6.5 46 vs. 48 17 vs. 11 57 vs. 55 8.9 vs. 9.6 0.0 vs. -1.0 0.7 vs. 0.4 Dluhy, NEJM, 2008 VA Diabetes Trial 20 VA centers; n=1,791; started in 2000 Intervention: intensive glycemic Rx (goal A1c<6.0%) vs. improved treatment (goal A1c 8-9.0%) A1c separation goal of 1.5% Protocol-based intervention (metformin, rosiglitazone, glimepiride, then insulin) BP and lipid treatment goals equal Outcomes: CV events (CV deaths, MI, CVA, CHF, cardiac revascularization, ischemic amputation) Abraira, Obes Met, 2008 Abraira, DM, Obes, Met, 2008 VADT: HbA1c achieved 8.4% 6.9% Abraira, 2008

VADT: Results No significant reduction in CVD with intensive glycemic control (HR 0.87, 0.73-1.04, p=0.12) Subgroups: Advanced subclinical disease: Coronary calcium 100: no benefit Coronary calcium <100: benefit from intensive control Duration of diabetes: shortest time period had the most benefit 12-15 years had neutral effect >16 years had increased risk of CV events ADA meeting, 2008 Study UKPDS-33 Older Trials N Intervention Outcome 3,867 Diet, Chlorpropramide, Glibenclamide, Insulin UKPDS-34 753 Metformin vs. diet VA-CSDM 153 Insulin Std. vs. intensive UGDP 823 Insulin vs. Tolbutamide vs. Placebo 16% RRR MI (NS) (p=0.054) 39% RRR MI 30% RRR agg. 1.6x increased risk (NS) 2x increased risk with Tol. Type 1 Diabetes and CVD NEJM 2005: DCCT F/U HbA1c (%) Cum. Incid. NFMI, CVA, CVD JMNEJM, 1993; Jama, 2002; NEJM, 200

Type 1: reduced complications Lebovitz, 2008 Glucose Lowering to Prevent CVD Ongoing Trials in People with Dysglycemia Yrs from Diagnosis -10-5 0 5 10 15 ACCORD NAVIGATOR ACE VADT ADVANCE ORIGIN Eye, Kidney, Nerve Disease CVD HEART 2D RECORD BARI 2D High IFG &/or IGT Type 2 Diabetes (T2DM) Dysglycemia - - - - - - - - - - - - - - - - - - - - - - - Gerstein, 2008 Critically Ill patients? Meta-analysis of 29 RCTs (n=8,432 patients) Mortality Rates Tight Usual RR (95% CI) Overall 21.6% 23.3% 0.93 (0.85-1.03) Very tight, 110 mg/dl 23.0% 25.2% 0.90 (0.77-10.4) Moderate, <150 mg/dl 17.3% 18.0% 0.99 (0.83-1.18) Medical ICU 26.9% 29.7% 0.92 (0.82-1.04) Surgical ICU 8.8% 10.8% 0.88 (0.63-1.22) Med-Surg ICU 26.1% 27.0% 0.95 (0.80-1.13) Wiener, Jama, 2008

Intensive Glucose Control NICE-SUGAR International RCT; 6104 patients Intervention: Intensive glycemic Rx (81-108 m/dl) vs. conventional control (less than 180 m/dl) Results: Death: 27.5% in intensive vs 24.9% in conventional 14% increase; NNH = 38 Severe hypoglycemia: 6.8% intensive vs. 0.5% conventional Effect seen in both surgical and medical patients NEJM, 2009 Abraira, DM, Obes, Met, 2008 Glycemic Control Summary No consistent evidence that tight glycemic control reduces risk of CVD Possible subgroups with benefit: shorter diabetes duration less CAC (no prior CVD) Strong evidence to support decrease in microvascular disease outcomes with more intensive glucose control More hypoglycemia and weight gain with most intensive regimens No consistent evidence that tight glucose control improves mortality in hospitalized patients. Case 1 Her glycemic goal should be: 1. HbA1c <6.0% 2. HbA1c <6.5% 3. HbA1c <7.0% 4. HbA1c <7.5%

In my practice, I have initiated: 1. Exenatide (Byetta ) 2. Sitagliptin (Januvia ) 3. Both exenatide and sitagliptin 4. Pramlintide (Symlin ) 5. All three of the above 6. None of the above Case 2: 48 yo woman with DM, BMI 33, on diet and exercise. HbA1C is 9.2. Your next best step is: 1. Begin metformin 2. Begin a sulfonylurea 3. Begin a thiazolidinedione 4. Begin insulin 5. Begin exenatide (Byetta ) or sitagliptin (Januvia

48 yo woman with DM, BMI 33, on diet and exercise and metformin. HbA1C is now 8.0. Your next best step is: 1. Continue current therapy 2. Begin a sulfonylurea 3. Begin a thiazolidinedione 4. Begin insulin 5. Begin exenatide (Byetta ) or sitagliptin (Januvia ) Category Sulfonylurea Biguanides α Glucosidase inhibitors Meglitinides DPP-4 inhibitors Incretin mimetics Antihyperglycemic synthetic analogs Diabetes Medications Thiazolidenediones Action Stimulates insulin secretion hepatic glucose production Delay conversion of CHO to glucose in digestion Insulin sensitizers (muscle/fat) Stimulates insulin production Blocks enzyme that deactivates GLP-1 Stimulates insulin; slows digestion (GLP-1 agonists) Analog of amylin hormone Safety Issues with All Classes Sulfonylureas: hypoglycemia, weight gain, increased CVD mortality? Metformin: GI symptoms, lactic acidosis (with impaired renal function) TZD: CHF, edema, weight gain, fractures, increased CVD mortality? DPP-4 inhibitors: HA, URI, nasopharyngitis Incretin mimetics: N/V, acute pancreatitis? Insulin: weight gain, severe hypoglycemia

Generic Oral Hypoglycemic Slide Change from Drug A to B, C, or D Add Drug A to B, or B to A HgA 1c Add Drug C Add Drug D Time Attaining Glycemic Goals Using Monotherapy in Obese Patients With Type 2 Diabetes Proportion of Patients With HbA 1c <7% (%) 60 50 40 30 20 10 0 Diet Alone Sulfonylurea Metformin Insulin 3 Years 6 Years 9 Years Turner RC et al. JAMA. 1999;281:2005-2012. Weight Changes Associated with Anti- Hyperglycemic Therapies for Type 2 Diabetes 12 10 8 6 4 2 0-2 -4-6 10.84 7.32 3.94-5.29 Change in Weight Sulfonylurea Metformin Insulin TZD

ARE THIAZOLIDINEDIONES SUPERIOR? Thiazolidinedione use slows progression to combination therapy Thiazolidinediones reduce microalbuminuria and hyperfiltration Reduced intimal proliferation ARE THIAZOLIDINEDIONES INFERIOR? Meta-analysis of 42 trials of rosiglitazone: Odds CI MI 1.43 (1.03-1.98) Death 1.64 (0.98-2.74) Nissen, NEJM 2007 What about Pioglitazone? PRO-active trial: 5,238 DM with macrovascular disease Pioglitazone vs. placebo Primary endpoint: HR 0.90, 95% CI 0.80-1.02, p=0.095 Composite endpoint: HR 0.84, 0.72-0.98, p=0.027 CHICAGO trial: effect of Pioglitazone vs. glimepiride on CIMT Less change in CIMT in pio group (-0.013 mm difference between groups), p<0.02 PERISCOPE trial: 543 patients with DM and CHD Pio vs. glimepiride for 18 months on IVUS 0.16% in Pio vs. 0.73% in glim. (p=0.002)

0.25 PROactive Primary Endpoint: No Statistically Significant Difference vs Placebo in CV outcome Kaplan-Meier Event Rate 0.20 0.15 0.10 0.05 Pioglitazone Placebo HR 0.90 P 0.095 CI 0.80-1.02 0.0 N at risk: 5238 5018 4786 4619 4433 4268 693 (228) 0 6 12 18 24 30 36 Time From Randomization (mo) Adapted from Dormandy JA, et al. Lancet 2005;366:1279 89; proactive-results.com Composite primary endpoint: all cause mortality, non-fatal MI (including silent MI), stroke, leg amputation, ACS, cardiac intervention, leg revascularization Oral Agent Failure Why does this occur? Changing HbA1c goals Compliance, side effects Wrong diagnosis (LADA--latent autoimmune diabetes in adults 10%) Stress, diabetogenic medications Natural progression of the disease Relative Contributions of Fasting and Postprandial Plasma Glucose to Total Glycemic Excursions as a Function of A1C Contribution (%) 80 60 40 20 Postprandial hyperglycemia Fasting hyperglycemia 0 1 (<7.3) 2 (7.3 8.4) 3 (8.5 9.2) A1C (%) Quintiles 4 (9.3 10.2) 5 (>10.2) Monnier L et al. Diabetes Care. 2003;26:881-885.

Glucose (mg/dl) Relative Function (%) Natural History of Type 2 Diabetes 250 200 150 100 50 0 *IFG = impaired fasting glucose Obesity IFG * Diabetes Uncontrolled hyperglycemia 350 300 250 200 150 100 50 Post-meal Glucose Beta-cell failure -10-5 0 5 10 15 20 25 30 Years of Diabetes Fasting Glucose Insulin Resistance Insulin Level` Natural History of Type 2 Diabetes Thiazolidinedione - Biguanide Glucose (mg/dl) Relative Function (%) 350 300 250 200 150 100 50 250 200 150 100 50 0 Lifestyle SU Post-meal Glucose Beta-cell failure Insulin Fasting Glucose Insulin Resistance Insulin Level -10-5 0 5 10 15 20 25 30 Years of Diabetes Insulin Plus Oral Agents Introduction of insulin Bedtime Intermediate/Long-acting insulins NPH, glargine 10 units Self-monitoring of blood glucose (hypoglycemia education) Insulin plus other oral agent combinations (maintain effect on insulin sensitivity)

When to go to > 1 shot per day HgA1c >7 Glucose in AM at goal but glucose before dinner >140 Options Add premeal lispro/aspart Add bid premixed insulin 70/30, 75/25 Questions Continue metformin? Sulfonylurea,? Thiazolidinedione Function of Insulin in Regimens Meal coverage (carbohydrates) Basal insulin Correction of high blood sugar More Options Incretin mimetics Exenatide (Byetta ) 4/05 Sitagliptin (Januvia ) 6/06 Amylinomimetics (amylin analog) Pramlintide (Symlin ) 3/05

INCRETINS Gut factors that promote insulin secretion in response to nutrients Major incretins: GLP-1, CCK, GIP Oral Glucose Promotes More Insulin Release than IV Glucose - Indicating a Role for Incretins Incretin Drugs GLP Agonists Exenatide Liraglutide CJC-1131 AVE-0010 Albugon Glp-1-transferin Exenatide Lar DPP IV Inhibitors Vildagliptin Sitagliptin Saxagliptin PSN=931 Takeda-Syrrx

A1C (%) Effect (change from baseline) Placebo BID 5 mcg exenatide BID 10 mcg exenatide BID MET 0.1-0.4-0.8 SFU 0.1-0.5-0.9 MET+SFU 0.2-0.6-0.8 Changes in A1C from baseline vs placebo statistically significant Weight (change from baseline) & Hypoglycemia Placebo BID 5 mcg exenatide BID 10 mcg exenatide BID Weight (kg) -1.4-3.1-4.2 Hypoglycemia (%) MET SFU MET + SFU 5.3 3.3 1.26 4.5 14.4 19.2 5.3 35.7 27.8 Open-label extension study to 90 weeks: persistence in weight loss and A1C

Side Effects GI Nausea (44% vs 18% with placebo); incidence lessens over time; 3% dropout rate due to nausea Vomiting (13% vs 4%) Diarrhea (13% vs 6%) Headache (9% vs 6%) Hypoglycemia (see previous slide) Exenatide Linked to Pancreatitis FDA Alert (10/07): 30 postmarketing reports of acute pancreatitis in patients taking Exenatide An association is suspected Improvements in HbA 1C With Initial Co-administration of Sitagliptin and Metformin Mean Baseline HbA 1C = 8.8% N=1091 Placebo -0.5 Sita 100 mg QD HbA 1C (%)* -1.0-0.8-1.0-1.5-2.0-1.3-1.6 Met 500 mg BID Met 1000 mg BID Sita 50 mg BID + Met 500 mg BID Sita 50 mg BID + Met 1000 mg BID -2.5-2.1 * Placebo-subtracted LS mean change form baseline at Week 24. Sita=sitagliptin; Met=metformin. Aschner P, et al. Oral presentation at the EASD 42 nd Annual Meeting; 14-17 September 2006; Copenhagen.

Sitagliptin adverse reactions Number of patients (%) Monotherapy Nasopharyngitis + pioglitazone Upper resp. infection Sitagliptin n = 443 23 (5.2) n = 175 11 (6.3) Placebo n = 363 12 (3.3) n = 178 6 (3.4) Small increase in WBC neutrophil count higher by 200 on Sitagliptin No nausea or vomiting No weight loss Sitagliptin linked to hypersensitivity reactions October 2007 MERCK announces post marketing reports of anaphylaxis, angioedema, and exfoliative skin conditions (including Stevens-Johnson) MERCK concludes not possible to establish causality Natural History of Type 2 Diabetes Incretins Thiazolidinedione - Biguanide Lifestyle SU Insulin Glucose (mg/dl) Relative Function (%) 350 300 250 200 150 100 50 250 200 150 100 50 0 Post-meal Glucose Beta-cell failure Fasting Glucose Insulin Resistance Insulin Level -10-5 0 5 10 15 20 25 30 Years of Diabetes

Drug Cost Comparison Drug and Dose Cost/month Sulfonylurea Generic $4-14 Brand $34 Rapaglinide 2 mg tid $111 Acarbose 100 mg tid $75 Metformin 1000 bid Generic $ 4-60 Brand $104 Rosiglitazone 8 mg qd $175 Pioglitazone 45 mg/d $180 Sitagliptin $146 Exenatide 5mcg $185 10mcg $209 Glargine, 45 U/d $118 24 hour fitness Center $40 YMCA $60 Diet, lifestyle + metformin Step 1 Basal insulin Sulfonylurea Thiazolidinedione Step 2 Intensive Insulin Rx Thiazolidinedione or basal insulin Sulfonylurea or basal insulin Step 3 Add basal insulin Or intensify insulin Rx Step 4 Consensus algorithm ADA. Diabetes Care (2006) Aug. 29:1963 Diet, lifestyle + metformin Step 1 Basal insulin Sulfonylurea Pioglitazone Exenatide Sitagliptin Step 2 Intensive Insulin Rx Pioglitazone or Exenatide or basal insulin Sulfonylurea Sulfonylurea or or Exenatide Pioglitazone or Sitagliptin or basal insulin or basal insulin Step 3 Sulfonylurea or Pioglitazone or basal insulin Add basal insulin or intensify insulin Rx (or add third oral agent or Exenatide) Step 4

48 yo woman with DM, BMI 33, on diet and exercise. HbA1C is 9.2. Your next best step is: 1. Begin metformin 2. Begin a sulfonylurea 3. Begin a thiazolidinedione 4. Begin insulin 5. Begin exenatide (Byetta ) or sitagliptin (Januvia ) 48 yo woman with DM, BMI 33, on diet and exercise and metformin. HbA1C is now 8.0. Your next best step is: 1. Continue current therapy 2. Begin a sulfonylurea (my choice) 3. Begin a thiazolidinedione 4. Begin insulin 5. Begin exenatide (Byetta ) or sitagliptin (Januvia ) Conclusions Tight glycemic control not effective in lowering CHD or CVD outcomes Many newer diabetes agents available, all with some side effects few with hard outcome data Glucose control may be more important early in diabetes Good BP and lipid control is important throughout the diabetes life course