American Diabetes Association - 71st Scientific Sessions

June 24 - 28, 2011; San Diego, CA Report – CV Disease and Complications

Executive Highlights

Complications and comorbidities were an increasingly big topic this year at ADA, with exciting new therapies on the horizon and ongoing questions into the relationship between insulin and cancer. Dr. David Boyer (Retina Vitreous Associates, Los Angeles, CA) took a deep dive into two-year phase 3 data on Roche’s (Genentech’s) Lucentis (ranibizumab) in diabetic macular edema. Compared to sham injections, Lucentis led to a significantly higher percentage of improvements to 20/40 vision (a critical real-world cutoff for driving and reading) and significantly lower rates of progression to proliferative diabetic retinopathy. We think these positive data will be important going into FDA submission (planned for the end of the year) and eventual reimbursement discussions (as a reminder, the similar but less expensive Roche’s (Genentech’s) Avastin [bevacizumab], is currently used off-label to treat DME but is widely believed to be less effective in this indication than Lucentis). In nephropathy, we saw extremely compelling phase 2 data on Reata’s bardoxolone for chronic kidney disease. Although yearlong findings were mentioned only briefly during the company’s corporate symposium, they were published in detail in the New England Journal of Medicine the first day of ADA (published simultaneously as the data was presented at the European Renal Association/ European Dialysis and Transplant Association meeting in Prague – great momentum for the company as it continues enrolling for its phase 3 trial. As a reminder, Reata does not yet have a commercialization partner in the US. Meanwhile, the insulin-cancer link received less attention compared to last year’s ADA, and we were not surprised by a preclinical study indicating that the increased IGF-1R-binding seen in Sanofi’s Lantus (insulin glargine) may not translate to in vivo cancer risk.

Table of Contents 

Cardiovascular Disease and Other Complications/Comorbidities

Symposium: Future Treatment of Vascular-Neural Degenerative Disease

Morten Charles, MD (Aarhus University, Aarhus, Denmark)

After uniquely prefacing his talk with a story of taking a walk with his patients, Dr. Charles reviewed how diabetic peripheral neuropathy (DPN) can be identified, who is at risk for DPN, and whether treatment can help prevent DPN. He described the key symptoms and signs of DPN as well as methods for testing for DPN. In terms of who is at risk for DPN, while the Eurodiab study showed a relation between CVD history and clinically diagnosed neuropathy, the Nerve Conduction Sub-Study did not implicate CVD history but suggested a relationship between other microvascular complications and DPN. Dr. Charles noted that neuropathy should thus be considered upon any indication of microvessel disease. Though the effects of intensive intervention on DPN are not clear for those with type 2 diabetes, Dr. Charles nevertheless suggested starting treatment for type 2 as early as possible to prevent complications.

  • Diabetic peripheral neuropathy (DPN) can be identified through key signs and symptoms. Identifying symptoms include numbness, a dead feeling in the extremities, tingling, burning or shooting pain, or contact pain. Signs include sensory loss, loss of proprioception, or muscular wasting. It was suggested at the most recent Neurodiab conference that possible DPN is characterized by symptoms or signs, probable DPN by symptoms and signs, and confirmed DPN by symptoms or signs and neuroconduction abnormalities. Methods for testing for DPN include clinical assessment (i.e. measuring symptoms and signs), qualitative sensory testing, nerve conduction studies, nerve biopsies, skin biopsies, neuropad tests, LDI flare tests, or corneal confocal microscopy.

  • In terms of who is at risk for diabetes, the Eurodiab study showed a strong association between CVD history and neuropathy. However, in the Nerve Conduction Sub-Study, DPN was associated with other microvascular complications, but not with history of CVD. The Eurodiab Prospective Complications Study enrolled 3,250 participants with type 1 diabetes and followed up with them for 7.3 years. It suggested a relation between DPN and CVD history, hypertension, smoking, diabetes duration, BMI, triglyceride levels, and cholesterol, but not between DPN and retinopathy or albumin excretion. In contrast, the Nerve Conduction Sub-Study, which tested nerve conduction velocity (a measure of nerve function) in conjunction with 456 Eurodiab follow up examinations, found that age, retinopathy, and micro and macroalbuminuria were negatively associated with nerveconduction velocity. However, lipids, blood pressure, and other macrovascular indicators weren’t. Dr. Charles noted this may suggest that there is a common pathophysiological pathway for microvascular complications that isn’t related to macrovascular disease so neuropathy should be considered with any indication of microvessel disease.

  • Though the effect of intensive treatment on DPN for type 2 diabetes has not yet been solidified, Dr. Charles nevertheless suggested starting treatment early. In the Steno Type 2 study, there was a significant effect of an intensive multifactorial intervention on nephropathy, retinopathy, and autonomic neuropathy at 7.8 years of follow up as compared to conventional treatment. In unpublished Danish data from the ADDITION study, the intensively treated group was less likely to develop neuropathy than those conventionally treated, but the effect was not statistically significant. Additionally, in contrast to DCCT’s demonstration that treatment level is important for preventing complications in type 1, UKPDS, ADVANCE, and ACCORD haven’t given us a clear picture for type 2 diabetes. Nonetheless, because diabetic complications can often precede type 2 diagnosis, Dr. Charles noted that to prevent peripheral neuropathy, treatment should be started as early as possible.

Symposium: Joint ADA/The Lancet Symposium

GLYCEMIC CONTROL AND THE INCIDENCE OF HEART FAILURE IN 20,985 PATIENTS WITH TYPE 1 DIABETES: AN OBSERVATIONAL STUDY

Marcus Lind, MD, PhD (Sahlgrenska University Hospital, Gothenburg, Sweden)

Dr. Lind conducted a large observational study (n=20,985) exploring the relationship between glycemic control and the rate of hospitalization for heart failure in patients of type 1 diabetes. His results suggest that there is an increased hazard ratio of 3.98 (95% CI 2.23-7.14) for hospitalization for heart failure in patients of type 1 diabetes with A1c greater than 10.5% compared to those with lower A1c. While this study cannot confirm a causal relationship between glycemic control and risk of heart failure, Dr. Lind believes it supports the use of ECG screening for patients with type 1 diabetes when there is a history of poor glycemic control or long diabetes duration. From our view, this would be expected since 10.5% A1c indicates quite poor control - we are curious the extent to which this remains significant at lower A1cs.

  • Dr. Lind’s study explored the relationship between glycemic control and the rate of hospitalization for heart failure in patients with type 1 diabetes. According to Dr. Lind, no major studies have explored glycemic control and heart failure in type 1 diabetes on the scale seen in type 2 patients. Studies suggest that good glycemic control can reduce the frequency of major vascular events in patients with type 2 diabetes (Turnbull, Diabetologia, 2009), and Dr. Lind hypothesizes that glycemic control may confer cardiovascular benefits to these patients in the long term.

  • Using data from the Swedish National Diabetes Registry (NDR) and national hospital discharge statistics, Dr. Lind tracked hospitalization for heart failure and death in 20,985 adults with type 1 diabetes for a mean period of 9 years. The mean age for those patients was 48.6, mean BMI was 25 kg/m2, and mean A1c was 8.2%. Of the patients tracked, 653 (3.0%) were hospitalized for heart failure, a higher number than the general population rate of 0.5% for those aged between 45 and 54. Adjusted for age, sex, diabetes duration, smoking, cardiovascular risk factors, and other co-morbidities, the hazard ratio for type 1 patient with diabetes with A1c above 10.5% was 3.98 (95% CI 2.23-7.14) compared to those with A1c below 10.5%. This suggests that type 1 patients with A1c elevated above 10.5% were four timesmore likely to be hospitalized for heart failure compared to patients with lower A1c. The hazard ratio per 1.0% increase in A1c was estimated to be 1.3.

  • While this study showed a strong association between poor glycemic control and heart failure, due to its observational nature it is not possible to establish a causal relationship between good glycemic control and the risk of heart failure in patients with high A1c. Nevertheless, Dr. Lind believes that heart failure should be considered a major complication in type 1 diabetes, and that good glycemic control may prevent hospitalization for heart failure. He noted that ECG screening in patients with type 1 diabetes should be considered when there is a history of poor glycemic control, long duration of disease, or other modifiable risk factors of heart failure.

Questions and Answers

Q: You had some data on HDL. I presume you also have non-HDL data. What was the relationship between LDL and hospitalization for heart failure?

A: Dr. Lind: HDL showed a strong relationship with heart failure, but LDL did not have such a significant effect.

Q: Could the higher A1c observed in patients with higher risk of heart failure be an indirect effect of heart problems?

A: Dr. Lind: It is possible. It is hard to know how much glycemic control improved this.

Q: The patients in the study have been diagnosed for an average of 20 years. What percentage of them has other measures of heart dysfunction?

A: Dr. Lind: I’m not sure of the percentage.

Q: In the ACCORD trial, intensive glycemic control did not have any effect on heart failure in patients with type 2 diabetes. However, in the meta-analysis of those trials, A1c was correlated with the risk of heart failure by the exact same amount as what you presented. Any comments?

A: Dr. Lind: Intensive glycemic control can have an effect on heart failure. The question is more to what extent. It is critical to know that. These trials are limited by their time perspective of about five years, and we need longer trials to get a better sense of previous glycemic control on the legacy of heart complications.

EFFECT OF EARLY INTENSIVE MULTIFACTORIAL THERAPY ON 5-YEAR CARDIOVASCULAR OUTCOMES IN INDIVIDUALS WITH TYPE 2 DIABETES DETECTED BY SCREENING (ADDITION-EUROPE) - A CLUSTER-RANDOMIZED TRIAL

Torsten Lauritzen, MD (University of Aarhus, Aarhus City, Denmark)

Dr. Lauritzen studied the feasibility of using stepwise screening programs to identify and treat individuals with undiagnosed diabetes. About 4.0% of those he tested had diabetes, and both routine and intensive care reduced A1c, BMI, blood pressure, and cholesterol levels in these patients. The Michigan Model identified a significant reduction in cardiovascular mortality in patients undergoing both routine and intensive care, leading Dr. Lauritzen to suggest that a population-wide stepwise screening program is feasible. We were glad to see these results as we believe there needs to be far more focus on identifying those with type 2 diabetes earlier and treating them aggressively and optimally at each stage of their disease in an effort to avoid costly, long-term complications and to improve quality of life.

  • Dr. Lauritzen’s study aimed to evaluate the feasibility of stepwise screening programs that identify and treat individuals with undiagnosed diabetes in an effort to lower their risk of cardiovascular disease. Intensive multi-factorial treatment can halve the risk of cardiovascular disease in people with longstanding diabetes. However, the effectiveness of multi-factorial treatment in people with screen-detected diabetes was unknown. To study the feasibility of stepwise screening programs for diabetes, Dr. Laurtizen’s group targeted a population aged 40-69 years, and through a postal questionnaire identified 74,310 high-risk individuals for blood tests. Diabetes was detected in 3,057 of those patients by the WHO criteria. The mean age of those patients was 60 years, and about 30% were current smokers.

  • Screened diabetes patients were randomly assigned intensive and routine care, and drug treatments were recommended for those with high levels of A1c, blood pressure, or cholesterol. Of those randomized, 1,379 diabetes patients were randomly assigned to intensive care, and 1,678 were assigned to routine care. Intensive treatment followed the ADDITION-Europe standards, and funding was given to clinics to enable them to deliver the higher level of care. Routine treatment followed current clinical practices including diet, physical activity, and smoking cessation. Family physicians were given directions to initiate drug treatment if A1c rose above 6.5%, blood pressure rose above 120/80 mmHg, and cholesterol rose above 135 mg/dl. Physicians were instructed to intensify drug treatment if blood pressure rose above 135/85 mg/dl and cholesterol levels rose above 173 mg/dl. The patients were followed for an average of 5.3 years.

  • Both intensive and routine care was found to reduce systolic blood pressure, cholesterol, A1c, and BMI. Systolic blood pressure was reduced by 10 mmHg from baseline to follow up in patients treated with routine care, and 14 mmHg in patients following intensive treatment. Mean total cholesterol was reduced by about 1 mmHg in both groups. Mean A1c dropped from 7.0% (+1.5) to 6.7% (+0.95) in those following routine care, and from 7.0% (+1.6) to 6.6% (+0.95) in those following intensive care. Mean BMI dropped from 31.6 to 31.0 kg/m2 in the routine group and from 31.6 to 31.1 kg/m2 in the intensive group. A non-significant 17% higher reduction in cardiovascular hazard ratio was observed in intensive treatment compared to routine treatment.

  • This results of this study led Dr. Lauritzen to suggest that a population based, step- wise screening treatment program is feasible in primary care. In the complete endpoint simulation through the Michigan Model done by Dr. Herman, routine or intensive care significantly reduced cardiovascular mortality compared to no-screening and no-treatment. Cardiovascular risk factors improved in both treatment groups. As such, Dr. Lauritzen believes that step-wise screening and treatment of diabetes is a feasible option.

Questions and Answers

Q: About 27-28% of the participants are smoking, but you didn’t present any data on smoking cessation?

A: Dr. Lauritzen: About 20% of them stopped smoking during the study. I had to choose what data to present due to time limits.

Q: I understand that the intensive treatment protocols are only recommendations and are not necessarily followed?

A: Dr. Lauritzen: It’s true that the intensive treatment protocol include only recommendations and it’s up to doctors and patients to decide whether to follow them or not.

Q: Only about 3,000 of the 75,000 you tested had diabetes, this seems unusual?

A: Dr. Lauritzen: The yield in this study was low - I must admit. Only 4% of those we tested had diabetes. However, for each person with screen-detected diabetes we found two with prediabetes and five with a high risk of cardiovascular disorders. If you take all these into account it may be economical to do the screening.

Symposium: Heterogeneity Among Asian Populations in Diabetes and Metabolic Risk

CARDIOVASCULAR CONDITIONS AND RISK FACTORS

Baiju Shah, MD, PhD (University of Toronto Department of Medicine, Toronto, Canada)

Dr. Shah discussed the cardiovascular conditions and risk factors that are common in Asian populations in North America and Asia. Outlining several differences between Asians and North Americans in cardiovascular health, Dr. Shah demonstrated that cardiovascular disease (CVD) risk factors vary significantly both across Asian populations and between Asian and North American groups. Perhaps the most intriguing part of his talk involved a comparison of CVD risk between the Asian diaspora and native Asians, showing that migrant Indian populations are heavily affected by environmental factors that increase their CVD risk despite genetic similarity to the native Indian population. He concluded by stating that while CVD risk factors vary among Asian populations, the impact of these risk factors on CVD prevalence within each population is quite similar.

  • The National Health Interview Survey in the United States showed that among Asian Americans, there is heterogeneity in CVD risk factors. More importantly, there isn’t a consistent pattern that can be observed among these risk factors. For example, despite having the highest risk for CVD, Indian-Americans have the highest rates of physical inactivity, but the lowest rates of smoking and hypertension, and the second lowest rates of obesity among the Asian American groups measured.

  • In a 2008 study conducted by Dr. Shah, the cardiovascular risk factors for South Asian and Chinese individuals living in Canada were compared and were correlated to clear differences in obesity and diabetes. It was found that smoking rates are fairly similar and low for both South Asian and Chinese individuals, yet there were clear differences in diabetes and obesity rates. South Asians, specifically Indians, had far higher rates of diabetes and slightly higher rates of obesity. This translated into an almost threefold increase in the rates of heart disease and stroke in South Asians over the Canadian Chinese population.

  • The INTERHEART study in Canada looked into the prevalence of the various CV risk factors in patients with acute myocardial infarction, and added further fuel to the observation that cardiovascular risk factors vary significantly between Asian populations. INTERHEART demonstrated that in Asian populations, hypertension, inadequate exercise, psychosocial stress, cholesterol, non-regular alcohol consumption, smoking, and diabetes accounted for about 90% of the total risk of CVD independent of age, sex, and origin country. Cardiovascular risk factors were also analyzed on the basis of odds ratio, and these risks were consistent across different regions of origin for the patient. Dr. Shah believes this strongly shows that while CV risk factors vary among Asian populations, the impact of risk factors upon CVD prevalence is fairly similar across populations.

  • Dr. Shah concluded by describing a study he conducted on CV risk variation between the Asian diaspora and native Asians, specifically looking at hypertension and myocardial infarction in Gujarat, India, versus Gujarati migrant populations in the UK. He found that the migrant populations had two fold higher risk of hypertension, and slightly higher risk of myocardial infarction. Similar studies done with migrant populations in the UK demonstrated dramatic differences in CV risk factors between countries of origin.

Questions and Answers

Q: What do you think about no longer using BMI to measure obesity in Asian populations? My own data suggests that it would be inaccurate, and I believe that several prominent researchers think the same. What do you think about this?

A: Dr. Shah: I think we’ve actually established that in most populations BMI is no longer as strong of a predictor of central obesity as other measures like waist circumference. We have also found that the incidence rate of diabetes for a certain BMI is not consistent across populations. You would find that incidence rate of diabetes in white populations for a BMI of 30 would be the same incidence for an Indian population at a BMI of 23. I agree that needs to change.

Q: I think that Asian populations need to have a different standard for BMI, or change the measurement standard entirely. I think also that we need to look more closely at risk factors for chronic diabetes in ethnic populations in terms of when these immigrants came to the US, because I think that newer immigrants may be encountering different environmental factors than immigrants from decades ago.

A: Dr. Shah: Absolutely, I think there are a number of Asian populations that support your theory. Different parts of India have people coming at different times, so I think the idea of when people come to the West and where exactly they come from is a lot more complicated because it’s a matter of different people from different parts of the country coming at different times.

Symposium: Atherosclerosis in Type 1 Diabetes - The Same Disease as in Type 2 Diabetes?

THE EPIDEMIOLOGY OF CARDIOVASCULAR DISEASE IN TYPE 1 DIABETES-IS IT THE SAME AS TYPE 2 DIABETES?

John C. Rutledge, MD (University of California Davis, Davis, CA)

Dr. Rutledge compared the epidemiology and characteristics of cardiovascular disease in type 1 and type 2 diabetes. He found that coronary artery disease, systolic heart failure, and macrovascular disease are more common in type 2 diabetes; however, retinopathy, erectile dysfunction, and microvascular disease are more common in type 1 diabetes. His research was unclear in the comparisons of stroke, diabetic nephropathy, diabetic foot syndrome, and dementia.

  • Dr. Rutledge compared the epidemiology and characteristics of cardiovascular disease in type 1 and 2 diabetes. He prefaced his talk by saying that this is an understudied area, and that the existing research is sometimes contradictory and difficult to directly compare. He concentrated on the less extensively studied comparisons, since the other topics are being covered elsewhere.

  • Dr. Rutledge’s research group performed a meta-analysis of the literature on type 1 and 2 diabetes and cardiovascular disease. He reviewed the anatomy and physiology ofcapillaries and arteries, noting that the pathologies of these structures likely have differing etiologies. Thus, he treated micro- and macrovascular outcomes as separate processes in his analysis.

  • Current evidence relating diabetes to dementia suggests that there are different characteristics and etiologies in type 1 and type 2 diabetes. Type 1 diabetes is associated with increased cognitive decline, but it is unclear if this predisposes to dementia or if cognitive decline is associated with hypoglycemia. Type 2 diabetes is strongly associated with minimal cognitive impairment, dementia, vascular dementia, and Alzheimer’s disease.

  • He briefly discussed type 3 diabetes in which the brain reduces insulin production and increases insulin resistance. Interestingly, this is associated with an increased risk of Alzheimer’s disease.

  • He concluded that diabetes is an independent predictor of systolic and diastolic heart failure. Since systolic heart failure can be caused by coronary artery disease, it is more commonly found in patients with type 2 diabetes than type 1 diabetes. Despite the fact that heart failure is associated with increased insulin resistance, the use of insulin is associated with an increased mortality in diabetes patients with systolic heart failure.

Questions and Answers

Q: Are BNP levels different between diastolic and systolic heart failure?

A: Dr. Rutledge: There is no difference.

Q: In terms of risk factors for atherosclerosis and diabetes, are lipids a greater risk factor for type 1 or 2?

A: Dr. Rutledge: If it’s for coronary artery disease, I’d say that lipids are more important in type 2 rather than in type 1. Unfortunately, the role of lipids in microvasculature is not well understood and so its role in type 1 is not yet clear. One thing I didn’t mention is that the microvasculature does not have the same structure or organization in different tissues. It’s totally different in muscle compared to the brain.

THE GENETICS AND PATHOPHYSIOLOGY OF CARDIOVASCULAR DISEASE IN PATIENTS WITH TYPE 1 DIABETES

Trevor J. Orchard, MD (University of Pittsburgh, Pittsburgh, PA)

Dr. Orchard reviewed the pathophysiology of cardiovascular disease in type 1 diabetes. HDL is protective for CAD in men, but this relationship does not seem to hold up in women. Nephropathy and duration of diabetes are the strongest predictors of CAD. He hypothesized that hyperglycemia in diabetes promotes more extensive and more stable atherosclerotic plaques. This leads to a weaker than anticipated association with coronary events, which generally involve the rupture of atherosclerotic plaques. He also talked about a genetic marker that may be relevant to type 1 diabetes, haptoglobin (Hp). Hp has two alleles that can help predict risk of cardiovascular disease. The “2” allele confers a higher risk of cardiovascular disease. Unfortunately, 44% of patients with type 1 diabetes are homozygous for the “2” allele of Hp. Interestingly, Vitamin E has been shown to reduce CV risk in patients who are homozygous for the 2 allele. Therefore, Dr. Orchard suggested that this supplement may be helpful as an adjunct therapy in these patients.

  • The Pittsburgh Epidemiology of Diabetes Complications (EDC) study was a study of 658 patients diagnosed with type 1 diabetes. This is an extremely long-term study, startingin 1950, with tracking continuing even today. He noted that the incidence of renal disease is decreasing in patients more recently diagnosed with diabetes (past 1965) but that there did not seem to be an equivalent benefit to coronary artery disease.

  • There is an inverse relationship between the incidence of CAD and HDL in men, but there did not seem to be the same relationship in women. Nephropathy and duration of diabetes are the strongest predictors of CAD, but HDL is a significant predictor as well.

  • Although insulin resistance is not the primary etiology of type 1 diabetes, it is a strong predictor of coronary artery disease. Patients with a history of type 2 diabetes and insulin resistance are more likely to suffer coronary artery disease. Additionally, insulin resistance in patients with type 1 diabetes are more likely to experience nephropathy.

  • He hypothesized that hyperglycemia in diabetes promotes more extensive and more stable atherosclerotic plaques. This leads to a weaker than anticipated association with coronary events, which generally involve the rupture of atherosclerotic plaques. So despite the increased accumulation of plaque, the pathogenesis does not manifest itself as coronary artery disease but rather as peripheral artery disease and other vascular diseases.

  • The haptoglobin gene (Hp) has two alleles that can help predict the risk of cardiovascular disease. The “2” allele confers a higher risk of cardiovascular disease due to impaired reverse cholesterol transport, in which HDL transports cholesterol from the periphery to the liver. Unfortunately, 44% of patients with type 1 diabetes are homozygous for the “2” allele of Hp. The HOPE and ICARE trials demonstrated that the increased risk of the “2" allele can be partially ameliorated with vitamin E supplements, so Dr. Orchard suggested that vitamin E might be an important adjunct treatment to minimize risk of cardiovascular disease in patients with type 1 diabetes.

ASSOCIATION OF INCREASED RETINAL THICKNESS AND SYSTEMIC COMPLICATIONS IN PATIENTS WITH 50 OR MORE YEARS OF TYPE 1 DIABETES

Jennifer K. Sun, MD (Brigham and Women’s Hospital, Boston, MA)

The study authors investigated the use of optical coherence tomography (OCT) to assess risk for systemic complications of type 1 diabetes. In 557 patients with a history of "50 years of type 1 diabetes, OCT-derived CST was collected along with the extent of diabetic retinopathy, retinal edema risk factors, nephropathy, neuropathy, and cardiovascular disease. The presence of retinal thickening was associated with the presence of worse retinopathy, nephropathy, cardiovascular disease, and neuropathy. Adjusting for age, diabetes duration, and retinopathy severity, only nephropathy was still significantly associated with retinal thickening.

  • The study authors investigated the use of optical coherence tomomgraphy (OCT) to assess risk for systemic complications of type 1 diabetes. OCT is a non-invasive imaging technique that can detect diabetic macular edema and central subfield thickness (CST). The authors theorized that, since the macula is composed of neural and vascular tissues (which are particularly susceptible to microvascular disease), that the macula may reflect how diabetes has reflected the rest of the body.

  • In 557 patients with a history of ≥50 years of type 1 diabetes, OCT-derived CST was collected along with the extent of diabetic retinopathy, retinal edema risk factors, nephropathy, neuropathy, and cardiovascular disease. Average age was 67, duration of diabetes was 55 yeas, and current A1c was 7.3%.

  • Retinal thickening (CST ≥250 um) occurred in 20% of eyes. Notably, 56% of these cases showed no evidence of macular degeneration on photographs and so would not be picked up with routine care. Retinal thickening was associated with older age, longer diabetes duration, older age at diabetes onset, and male gender.

  • The presence of retinal thickening was associated with the presence of worse retinopathy, nephropathy, cardiovascular disease, and neuropathy. Adjusting for age, diabetes duration, and retinopathy severity, only nephropathy was still significantly associated with retinal thickening.

  • The authors concluded that OCT retinal imaging could be used to estimate the severity of diabetic retinopathy, nephropathy, and cardiovascular disease.

ATHEROSCLEROSIS IN TYPE 1 DIABETES–IS THE PLAQUE THE SAME?

Nikolaus Marx, MD (University Hospital Aachen, Aachen, Germany)

Dr. Marx reviewed the limited evidence on atherosclerosis in type 1 diabetes, concluding that these patients are at higher risk than people without diabetes for cardiovascular and circulatory issues. He observed, “atherogenesis all starts with endothelial dysfunction”; discouragingly, children with type 1 diabetes tend to have impaired endothelial function relative to controls. The burden of unstable plaques (the most common cause of myocardial infarction) has been suggested to be similarly high in people with both type 1 and type 2 diabetes. Also, many biomarkers of cardiomyopathic risk (including MCP-1, sICAM-1, sVCAM-1, and sE-Selectin) appear elevated in both types of diabetes relative to matched controls.

Questions and Answers

Comment: It seems we have a paucity of data on atherosclerotic disease in type 1 diabetes.

A: Dr. Marx: The previous presenter noted that more study is needed in this area. I think that could be the summary for this entire session.

RISK FACTOR MODIFICATION IN TYPE 1 DIABETES – IS THERE EVIDENCE OF BENEFIT?

Sarah D. de Ferranti, MD, MPH (Children’s Hospital Boston, Boston, MA)

Children and adolescents with type 1 diabetes typically have elevated risk factors for atherosclerosis, and Dr. de Ferranti believes that these risk factors should be treated early despite the difficulty of assessing long-term effects with any pediatric intervention. She reviewed data and made recommendations on topics including blood pressure (the use of ACE inhibitors was associated with better CV outcomes in the EDC study, so she recommended ACEi therapy to normalize blood pressure in patients with hypertension) and dyslipidemia (non-HDL cholesterol is elevated in patients with high A1c and, she thinks, should be treated; ApoB appears higher in all type 1 patients and may be more widely used as a risk marker in the future). Although glycemic control correlates with other risk factors, Dr. de Ferranti believes that trying to treat atherogenic risk factors through glucose-lowering therapies alone is “like whistling in the dark” given many patients’ difficulty in reaching glycemic targets.

Questions and Answers

Q: Convincing patients to take more drugs for prevention is difficult when they are on so many injections and there is no evidence of benefit.

A: Dr. de Ferranti: The evidence is scant. It depends on if you are thinking of a 10-year or 30-year range. I am thinking about the 30-year range; I’m trying not to deliver high-risk kids to adult care providers. I argue that if you are looking at a 30-year timeframe, the risk is high.

Q: Dr. Hertzel Gerstein (McMaster University, Hamilton, Canada): One important risk factor is retinopathy. Have people started to incorporate eye photographs in risk stratification?

A: Dr. de Ferranti: I think interest is growing in looking at complications in this way. That is a great idea; retinopathy progression gives a great idea of glycemic control and exposure.

Symposium: Ocular Complications

INCREASED RISK OF DIABETIC MACULAR EDEMA (DME) AMONG TYPE 2 DIABETES PATIENTS TREATED WITH PPAR-G AGONISTS: RESULTS OF A LARGE COHORT STUDY

Iskandar Idris, MD (Sherwood Forest Hospitals NHS Trust, Nottingham, UK)

The study authors sought to determine whether there is a relationship between PPAR-gamma agonist use and macular degeneration. A total of 103,386 patients without diabetic macular degeneration (DME) were divided based on whether they were treated with TZDs. The primary outcome was the development of DME at one year, with a 10-year follow-up. The incidence of DME was 1.3% in TZD- users and 0.2% in TZD non-users. This is equivalent to an odds ratio of 5.7 (95% CI: 4.1-7.9). This association held even after compensation for multiple confounders. At 10 years the effect was still significant with a hazard ratio of 5.2 (95% CI: 4.3-6.3). The authors concluded that TZDs are associated with a 3-6 fold increased risk of DME even after adjustment for confounding variables.

  • The study authors sought to determine whether there is a relationship between PPAR-gamma agonist use and macular degeneration. The research group has previously demonstrated that increased permeability of blood vessels could explain the pulmonary edema seen with the use of thiazolidinediones (TZDs).

  • Patient data for this study was drawn from the UK primary care database, The Health Improvement Network (THIN). A total of 103,386 patients without diabetic macular degeneration (DME) were stratified based on whether they were treated with TZDs.

DIABETIC RETINOPATHY SCREENING: IMPLICATIONS WITH EXENATIDE TREATMENT

Lakshminarayanan Varadhan, MRCP (University Hospitals of North Staffordshire NHS Trust, Stoke-on-Trent, UK)

In this study, the authors assessed the impact of exenatide on the incidence of diabetic retinopathy. The authors retrospectively analyzed data from 165 patients who had been on exenatide 10mcg twice daily for more than six months. The mean A1c of the patients at baseline was 9.6%. Stratifying progression to diabetic retinopathy by A1c change, patients with a greater drop in A1c actually showed higher rates of progression. This is likely because these patients had higher A1cs at baseline, although this was difficult to discern directly from the data. Of the 133 patients (80.6%) whose A1c was lowered after beginning exenatide, rates of progression to diabetic retinopathy were 30.1% for those with an A1c reduction of 0- 2%, 43.6% for an A1c reduction of 2-4%, and 45.5% for an A1c reduction >4%. The authors concluded that significant reductions in A1c were associated with progression of diabetic retinopathy. We feel that this conclusion is slightly misleading, since they did not stratify these results based on baseline A1c level, which is predictive of a patient’s response to exenatide as well as the progression to diabetic retinopathy. We were also surprised patients were not on insulin with such high A1cs although we note that the A1c range in this study is broad.

  • The authors retrospectively analyzed data from 165 patients who had been on exenatide 10mcg (Byetta) for more than six months. The mean A1c of the patients at baseline was 9.6%. The lowest average A1c was 8.1%, which was achieved at 176 days after beginning treatment.

  • Of the 165 total patients, 49 (29.7%) showed progression of diabetic retinopathy (16 new cases, 33 showing progress of a pre-existing conditions). In contrast, 32 (19.4%) of the patients showed an improvement in diabetic retinopathy. We would assume over time, patients better controlled (and who adhere better to therapy that is easier, such as exenatide) would have fewer complications though this is difficult to study in a controlled setting and over a short period of time.

Symposium: Pharmacoepidemiology and Diabetes - Defining Unexpected Risks and Benefits

DIABETES MEDICATIONS AND CANCER

Samy Suissa, PhD (McGill University, Montreal, Canada)

Cautioning that “improperly performed observational studies need a health warning,” Dr. Suissa gave an intriguing presentation that challenged a number of the studies surrounding various diabetes medications and cancer risk. On the protective effects of metformin, he argued results were likely “too good to be true,” suggesting the majority were plagued by the “immortal time bias” - referring to the additional survival time factored in when patients switching to metformin are lumped in as overall metformin patients in calculations. He made note of further studies examining time-dependent exposure that demonstrated no preventative effect. In contrast, with insulin glargine, while he acknowledged a number of limitations in the original Diabetologia studies, he found merit in a recent study of the UK’s General Practitioner’s Database in which first-time users with glargine trended toward increased breast cancer risk after five years exposure (HR=1.8; nonsignificant) and prevalent users on glargine showed significantly increased risk (HR=2.7). He suggested this was a much more plausible time course for the analog to affect tumor progression, though he recognized further research would be necessary.

  • Despite the “tsunami” of studies supporting the association, Dr. Suissa contested the notion that metformin use reduces cancer risk. Beginning with an observational study published in 2005 that suggested a 23% reduction in cancer risk with metformin use, Dr. Suissa reviewed a number of studies in support of the protective effects of metformin. However, using the Saskatchewan study (observed a 20% in reduction in cancer risk with metformin vs. sulfonylurea) as an example, he suggested the majority of these studies are plagued by an “immortal time bias.” This bias refers to the labeling of patients who started on one treatment but were switched to metformin as overall metformin users in calculations - given patients had to have been alive for a certain period to be able to be switched to metformin, this adds additional survival time (the “immortal period”) to their tabulated lifespan, biasing results toward benefit. While the authors of the Saskatchewan study eventually took this into account in a future paper, their results showed no change - leading Dr. Suissa to suggest that the corrective calculationswere not performed properly. Further studies (particularly one performed by Kaiser) examining time-dependent exposure with metformin have demonstrated no preventative effect. Given the interest in this area, we assume this will continue to be an area of hot debate - this also puts certain pressure on the randomized trials investigating metformin and cancer underway.

  • Dr. Suissa also put forth evidence supporting an increased risk of breast cancer with long-term insulin glargine use. Though he acknowledged a number of limitations in the original Diabetologia studies (type 1 and type 2 patients were mixed, findings were not cancer-specific, follow-up was short giving implausible effects), he found merit in a recent studyin the UK’s General Practitioner’s Database. In the observational study, 15,227 women with type 2 diabetes treated with insulin from 2002-2006 were matched by age, length of treatment, and diagnoses; first-time insulin users were made distinct from prevalent insulin users (five years of use before cohort entry). Results suggested that in the initial five years of treatment, insulin glargine showed no increased risk of breast cancer compared with other insulin treatments. However, after five years of use, first-time users with glargine trended toward increased risk (HR=1.8; n.s.) and prevalent users on glargine showed significantly increased risk (HR=2.7). Dr. Suissa thus concluded glargine use in the long-term - a more plausible environment for a drug to affect tumor progression - may be associated with increased risk of breast cancer, though certainly further research will be necessary.

  • Dr. Suissa was unclear what to conclude on the association of pioglitazone and bladder cancer risk. In the closing minutes of his presentation, he briefly reviewed the recently published French study (see the June 11th, 2011 Closer Look), suggesting some increased risk with an effect size similar to that seen in the Kaiser trial - however, without further details on the protocol, he conceded he could make no conclusion on the validity of the study.

Questions and Answers

Q: Do you think in your analysis showing association for the longer term that the approach takes care of the within person confounding by allocation?

A: I don’t know. I’m a slow thinker. I’d have to think about this in the comfort of my office. I’m uncomfortable giving an answer otherwise.

Symposium: Intensive Treatment of Diabetes – Focus on Prevention of Hypoglycemia

HOW TO REDUCE THE HYPOGLYCEMIA RISK IN INTENSIFIED TREATMENT OF TYPE 2 DIABETES

Gabriele Perriello, MD, PhD (University of Perugia, Perugia, Italy)

After highlighting a number of risk factors for hypoglycemia in type 2 diabetes (e.g., intensified treatment, the use of secretagogues, age, and duration of diabetes), Dr. Perriello emphasized that sulfonylureas should be avoided in the treatment of type 2 diabetes when possible; instead, GLP-1 agonists and DPP-4 inhibitors would be more appropriate options to limit hypoglycemia. He noted that the addition of DPP-4 inhibitors to insulin reduces the incidence of insulin-related hypoglycemia events, and that combining basal insulin with a GLP-1 agonist could result in reductions in A1c, body weight, and postprandial hyperglycemia without increasing hypoglycemia.

  • Dr. Perriello highlighted that the use of sulfonylureas causes more hypoglycemia than other oral antidiabetics. He even went so far as to say that sulfonylureas are the maincause of hypoglycemia in type 2 diabetes. In a meta-analysis of head-to-head comparisons of several oral antidiabetics with respect to their hypoglycemic potential, sulfonylureas consistently demonstrated higher hypoglycemia risk (Bolen et al., Ann Intern Med 2007). In observational studies, treatment with sulfonylureas was associated with mild hypoglycemia in 39% of patients, and with severe hypoglycemia in 7% of patients (UK Hypoglycemia Study Group, Diabetologia 2007).

  • Subsequently, he listed a number of factors that increase the risk of hypoglycemia with sulfonylureas: impaired drug clearance (e.g., renal impairment, hepatic failure, hypothyroidism), impaired counterregulatory capacity (e.g., Addison’s disease, growth hormone deficiency, hypopituitarism), increased peripheral glucose uptake (e.g., exercise), decreased endogenous glucose production (e.g., liver failure, alcohol), and impaired glucose absorption (malabsorption, anorexia). In addition, Dr. Perriello noted that concurrent medications could also worsen hypoglycemia, including drugs that: 1) decrease renal excretion of sulfonylureas (e.g., aspirin, alluopurinol); 2) displace sulfonylureas from albumin (e.g., aspirin, warfarin, sulfonamides, trimethoprim, fibrates); 3) decrease metabolism of sulfonylureas (e.g., warfarin, monoamine oxidase inhibitors); and 4) increase secretagogue activity (e.g., quinolones, non- steroidal anti-inflammatory drugs).

  • Dr. Perriello commented that GLP-1 agonists and DPP-4 inhibitors confer minimal risk of hypoglycemia. In a study by Nauck et al., sitagliptin reduced the incidence of hypoglycemia by 27% compared to sulfonylurea, when both were used as add-on therapies to metformin (Diabetes Obesity Medicine 2007). Similarly, a significantly lower proportion of patients on saxagliptin plus metformin versus sulfonylurea plus metformin (3% versus 36.3%) experienced hypoglycemia in a recent trial (Goke et al., Int J Clin Pract 2010).

  • The addition of DPP-4 inhibitors to basal insulin could reduce hypoglycemia, while the addition of a GLP-1 agonist to basal insulin could reduce A1c, postprandial hyperglycemia, and body weight without increasing hypoglycemia. In a study in which vildagliptin was used as an add-on to basal insulin, it caused less hypoglycemia than the basal insulin itself, while providing a similar degree of glycemic control (Fonseca et al., Diabetologia 2007). In a more recent study, exenatide plus insulin glargine was found to reduce A1c, postprandial hyperglycemia, and body weight beyond insulin glargine therapy alone, without increasing hypoglycemia (Buse et al., Ann Intern Med 2011).

Questions and Answers

Q: Do GLP-1 agonists and DPP-4 inhibitors only work when someone is newly diagnosed with type 2 diabetes, or are they similarly effective for people with longstanding diabetes who may not have the proper glucagon response or beta cell reserve?

A: In patients with longstanding diabetes, DPP-4 inhibitors and GLP-1 agonists can still be used. They will require less exogenous insulin to have the same effect. Along with the reduction in insulin dose, there could be a reduction in hypoglycemia.

Q: I don’t think you mentioned that one of the risk factors is hypoglycemia unawareness. Are there any strategies we can use (this also applies to type 1) to improve awareness of hypoglycemia?

A: The treatment is to reduce hypoglycemia events. If a patient has many events, you should increase the glucose level to reduce hypoglycemia, until the person is able to experience the symptoms of hypoglycemia again.

TREATMENT PRIORITY OF THE ELDERLY – PREVENTION OF HYPOGLYCEMIA

Bernard Charbonnel, MD (University of Nantes, Nantes, France)

Dr. Charbonnel discussed treatment of the elderly, suggesting that it may be preferable to target A1c levels at or above 8.0% in frail, elderly patients for whom moderate therapy has not been effective. Elderly type 2 diabetes patients are at high risk of hypoglycemia, due to a weakened perception of hypoglycemia related symptoms, a diminished level of hormonal control, and a higher likelihood of dementia. Furthermore, the Collaborations on Trials of Lowering Glucose (CONTROL) group showed that intensive therapy to reduce A1c to less than 7.5% had no cardiovascular benefits in the first five years of treatment.

  • Elderly type 2 diabetes patients are at high risk of hypoglycemia. According to Dr. Charbonnel, this is because of a weakened perception of hypoglycemia related symptoms, a diminished level of hormonal control due to type 2 diabetes, and a higher likelihood of dementia. While hormonal responses to hypoglycemia were similar across age groups, the symptoms of hypoglycemia began at a plasma glucose level of 3.6mmol/l in younger patients and 3.0mmol/l in the elderly patients. Furthermore, hormonal release occurred at higher blood glucose levels in diabetes patients compared to non-diabetes patients. Finally, the Fremantle diabetes study found that dementia at baseline was a strong independent predictor of severe hypoglycemia, but in patients with normal cognition, severe hypoglycemia was not associated with any cognitive decline. To Dr. Charbonnel, this suggested that dementia is linked with a diminished ability for patients with dementia to control hypoglycemia, although the direction of causation is not yet completely settled.

  • Current estimates suggest that the risk of tight glycemic control may exceed its benefits in many frail, elderly patients. The frequency of hypoglycemia is high and appears to be under-reported and under-estimated in elderly patients. Meta-analysis from the four main studies explored by the Collaborations on Trials of Lowering Glucose (CONTROL) group showed that intensive therapy to reduce A1c to less than 7.5% had no cardiovascular benefits over five years of treatment. Dr. Charbonnel commented that blood glucose control may only be beneficial with respect to cardiovascular risk over a long period time through “blood glucose memory.” As a result, Dr. Charbonnel suggested that maximizing the quality of life for many frail elderly patients may involve keeping A1c at or above 8% rather than pursuing an aggressive treatment.

Questions and Answers

Q: Who is the frail elderly patient?

A: Dr. Charbonnel: It must be individualized. It may be difficult to recognize in day to day life, but it should be easy for the clinician. It’s hard to get it down to a strict set of criteria, however. I think it’s not hard to identify patients for whom A1c may be lowered below 7% without too aggressive therapy even without a strict criteria.

Q: The risk of microangio complications is important in elderly diabetes patients, and by not controlling hyperglycemia we might be moving a diabetic patient to a cardiovascular patient, not to mention neglecting kidney and neural disease.

A: Dr Charbonnel: I agree. But we’re speaking of frail elderly patients. Glucose control needs a long time to show something, and the life expectancy of the patients must be taken into account. In diabetes patients of 70 years or older, long-term cardiovascular health may still be an important point of consideration, but not the main one.

Oral Presentations: CV Disease and Other Complications/Comorbidities

RANIBIZUMAB (ANTI-VEGF) FOR VISION LOSS DUE TO DIABETIC MACULAR EDEMA – RESULTS OF TWO PHASE III RANDOMIZED TRIALS

David Boyer, MD (Retina Vitreous Associates, Los Angeles, CA)

Dr. Boyer presented analyses of phase 3 trials of Lucentis (Roche [Genentech]’s ranibizumab) for diabetic macular edema (DME), bolstering previously presented efficacy results with strong data across a range of secondary efficacy endpoints. Compared to those given sham injections, statistically significantly more patients receiving Lucentis achieved 20/40 vision or better at 24 months – a finding Dr. Boyer highlighted, since 20/40 represents the legal threshold for driving a car in most states as well as an approximate cutoff for ability to read a newspaper. Other notable benefits of Lucentis therapy included lower rates of progression to proliferative diabetic retinopathy, as well as improvements in best-corrected visual acuity (BCVA) that became apparent seven days after the start of treatment and were maintained for 24 months. Ocular and systemic safety events were similar to those in past clinical studies of Lucentis, which has been FDA-approved for neovascular age-related macular degeneration (wet AMD) since 2006. As a reminder, the company plans to file a supplementary biologics license application (BLA) by the end of 2011.

  • In the double-blinded, multi-center RIDE and RISE trials (n=382; 377), adult patients with diabetes (type 1 or type 2) and DME were randomized to receive monthly intravitreal injections of 0.3 mg Lucentis (n=125; 125) or 0.5 mg Lucentis (n=127; 125), or monthly sham injections (n=130; 127). Exclusion criteria included a history of vitreoretinal surgery in the study eye, panretinal laser photocoagulation or macular laser photocoagulation in the study eye within three months of screening; previous use of intraocular or periocular corticosteroids in the study eye within three months of screening, previous treatment with Lucentis or other anti-angiogenic drugs in either eye within three months prior to day zero of the study; and a history of myocardial infarction or cerebrovascular accident within three months prior to day zero. Macular laser rescue treatment was made available to all patients starting at month three, based on pre-specified criteria, and panretinal photocoagulation was available for all patients when clinically indicated.

  • Topline two-year results were released earlier this year for RISE and RIDE (see February 15, 2011 Closer Look and the April 8, 2011 Closer Looks, respectively). The trials will continue blinded until month 36 (data expected 1Q12 for both trials); patients in the sham group had the option of crossing over to 5.0 mg Lucentis at month 24 (to preserve the blinding, all patients were asked if they wanted to switch to 5.0 mg Lucentis). Open-label extensions will follow.

  • Patients had a mean age of 61.7-63.5 years, with mean diabetes duration of 14.5-16.6 years and mean A1c of 7.6-7.7%. They were 65-80% male and 77.6-82.7% Caucasian. Baseline vision was 54.7-57.5 ETDRS chart letters, corresponding to vision of roughly 20/80.

  • Dr. Boyer highlighted that statistically significantly more patients in each Lucentis group achieved 20/40 vision or better at 24 months. He explained that 20/40 vision is the legal threshold for being able to drive, and it also roughly represents the cutoff for being able to read a book or newspaper.

  • Lucentis treatment was similarly effective whether or not patients’ baseline A1c was above 8.0%. Patients with better baseline glucose control appeared to achieve better BCVAgains in RISE, but this relationship was not as apparent in RIDE. In both trials, Lucentis showed a significant benefit compared to control regardless of baseline A1c.

  • As previously noted, significantly more Lucentis patients improved their best- corrected visual acuity (BCVA) by 15 letters or more after 24 months (the studies’ primary efficacy endpoint). Lucentis also led to statistically better improvements in mean BCVA, which were apparent one week after starting treatment and sustained for two years, as well as statistically significant improvements in central foveal thickness (less than a 135 decrease in each sham group vs. decrease of 250 or more in each Lucentis group). Also, over 35% of patients in each Lucentis group improved by two or more steps in the Retinopathy Severity Scale, as opposed to 7% or less for the sham groups.

 

RIDE

RISE

Monthly dose

Sham

0.3 mg

0.5 mg

Sham

0.3 mg

0.5 mg

n

130

125

127

127

125

125

Percent of patients with ≥ 15 letter gain at month 24

18.1

44.8

39.2

12.3

33.6

45.7

Percent of patients with ≥ 15 letter gain at month 24, baseline A1c ≤ 8.0 (RISE n=243; RIDE n=246)

17.5

46.9

39.0

13.1

36.7

49.4

Percent of patients with ≥ 15 letter gain at month 24, baseline A1c > 8.0

(RISE n=120; RIDE n=122)

18.6

43.6

42.1

12.2

29.3

40.0

Percent of patients with ≥

20/40 vision at month 24

37.8

60.0

63.2

34.6

54.4

62.2

Percent of patients who progressed to proliferative diabetic retinopathy

11.5

3.2

3.9

15.0

1.6

5.6

Mean change in BCVA (letters)

2.3

10.9

12

2.6

12.5

11.9

  • Fewer patients in the Lucentis groups required macular laser or panretinal photocoagulation, and fewer Lucentis patients experienced a vision loss of 15 letters or more at month 24.
  • Functionality gains were seen in other measures besides BCVA. Patients’ scores on the composite scale of the Visual Function Questionnaire 25 (VFQ-25), a 25-question assessment of perceived visual function, increased by a mean of roughly 3.0 with Lucentis than with sham therapy (by way of reference, a 5-point change is considered equivalent to a 15-letter change). Contrast sensitivity, a measure of people’s ability to distinguish shades of gray from a white background (Pelli-Robson chart), improved by more than 2.0 points in each Lucentis group while declining by 0.2 in both sham groups.
 

RIDE

RISE

Monthly dose

Sham

0.3 mg

0.5 mg

Sham

0.3 mg

0.5 mg

n

130

125

127

127

125

125

Percentage receiving macular laser

70

36

19.7

74

39.2

35.2

Percentage receiving panretinal photocoagulation

12.3

1.6

1.6

11.0

0

0.8

Percentage of subjects with ≥15-letter loss from baseline at month 24

8.5

1.6

3.9

10.2

2.4

2.4

Mean change from baseline in composite VFQ-25 score

4.0

7.3

6.9

4.4

7.0

7.5

  • Lucentis’ safety profile was consistent with Lucentis phase 3 trials for age-related macular degeneration, and the therapy was associated with fewer adverse events associated with diabetic retinopathy, including retinal neovascularization and vitreous hemorrhage, iris neovascularization. Minor ocular adverse events more common in the Lucentis arms included increased intraocular pressure; this included measurements of intraocular pressure taken 60 minutes post-injection.
 

RIDE

RISE

Monthly dose

Sham

0.3 mg

0.5 mg

Sham

0.3 mg

0.5 mg

n

127

125

124

123

125

126

Cataract Conditions

33.1

27.2

31.5

30.1

28.8

24.6

Endophthalmitis

0

0.8

1.6

0

0.8

0

Glaucoma (excl. congenital)

3.1

1.6

0.8

1.6

4.0

3.2

Intraocular inflammation

3.1

3.2

0.8

3.3

4.0

2.4

Intraocular pressure increased

11.0

15.2

18.5

2.4

20.0

14.3

Iris neovascularization

1.6

0

0.8

0

0.8

0

Retinal detachment

1.6

0

0.8

0.8

0.8

0

Retinal neovascularization

5.5

0.8

0.8

13.8

0.8

4.0

Retinal tear

0

0

0

0

0

1.6

Visual acuity reduced

3.1

1.6

1.6

8.9

8.0

6.3

Vitreous hemorrhage

15.0

0.8

2.4

13.0

3.2

3.2

  • Systemic adverse events were generally balanced between groups. Rates of stroke, death, vascular death, and serious adverse events of hypertension were slightly higher among Lucentis users than those receiving sham injections. Arteriothrombotic events as a whole were more common among the sham injection groups.
 

RIDE

RISE

Monthly dose

Sham

0.3 mg

0.5 mg

Sham

0.3 mg

0.5 mg

n

127

125

124

123

125

126

SAEs potentially related to systemic VEGF inhibition

9.4

9.6

5.6

10.6

5.6

11.9

Arteriothrombotic events

8.7

7.2

4.8

7.3

3.2

7.9

Myocardial infarction

4.7

5.6

2.4

2.4

1.6

3.2

Angina

1.6

0

0

0.8

0

0.8

Cerebrovascular accident

1.6

1.6

2.4

1.6

0.8

4.0

Transient ischemic attack

1.6

0.8

0

2.4

0

0.8

Hypertension

0

1.6

1.6

0.8

0.8

3.2

Non-ocular hemorrhage

1.6

0.8

0

2.4

0.8

1.6

Death

1.6

3.2

4.8

0.8

2.4

4.0

Vascular death

1.6

3.2

2.4

0.8

0.8

2.4

ABILITY OF INDICES OF CARDIOVASCULAR DISEASE (CVD) RISK AND COMORBIDITY TO PREDICT CVD OUTCOMES WITH INTENSIVE GLUCOSE CONTROL IN THE VADT

Nalurporn Chokrungvaranon, MD (Honolulu, HI)

Dr. Chokrungvaranon discussed the results from a post-hoc analysis examining the ability of indices of cardiovascular risk and comorbidity to predict cardiovascular outcomes with intensive glucose control in the VADT. The post-hoc analysis found that individuals in the upper tertile of each of the four indices (Framingham risk score, UKPDS risk equation, Charlson comorbidity index, prognostic index for four- year mortality in older adults [developed by Sei J. Lee, et al.]) showed no benefit with intensive glucose control, while those in the low and middle tertiles experienced modest benefit. Interestingly, those in the middle tertile showed significant benefit more so than the low tertile; Dr. Chokrungvaranon hypothesized that this could be due to power issues with the analysis. Based on these findings, Dr. Chokrungvaranon suggested that cardiovascular risk scores and comorbidity indices may be useful tools to identify patients who should be considered for less aggressive treatment, and that A1c goals should be individualized.

  • The post-hoc analysis aimed to determine whether low, moderate, and high cardiovascular risk scores and/or comorbidity indices modulate the effects of intensive glucose control on cardiovascular events in the VADT trial. Cardiovascular risk scores were calculated using the Framingham risk score and the UKPDS risk equation. Comorbidity indices were obtained using the Charlson comorbidity index, and a scale developed by Sei J. Lee et al. for the prognostic index for four-year mortality in older adults.

  • Individuals in the upper tertile of each of the four indices showed no benefit from intensive glucose control; in contrast, those in the low and middle tertiles demonstrated modest benefit. Using the Framingham risk scores to stratify the VADT population, those in the low, middle, and upper tertiles had hazard ratios of 0.85, 0.66 (p=0.01), and 1.06, respectively. The same pattern was observed when stratifying with the UKPDS risk score; patients in the low, middle, and upper tertiles had respective hazard ratios of 0.88, 0.77 (p<0.05), and 1.02. Using the Charlson comorbidity index, patients in the low, middle, and upper tertiles had hazard ratios of 0.99, 0.57 (p<0.01), and 0.94. Finally, using the prognostic index developed by Lee et al., patients in the low, middle, and upper tertiles had hazard ratios of 0.71 (p=0.04), 0.99, and 0.92, respectively.

  • Dr. Chokrungvaranon noted a number of caveats with the study: it was a post-hoc analysis of the VADT, they were not able to obtain all the information needed to calculate cardiac risk scores and comorbidity indices for all patients, and the VADT patient population is highly specific, consisting primarily of elderly male veterans.

Questions and Answers

Q: Did you think about running the analysis using continuous variables, instead of stratifying by risk tertiles?

A: Yes, we are thinking about performing an analysis using risk as a continuous variable.

Q: Thank you for your interesting analysis. The issue I think is that we get too hung up on relative risk reduction and p values. We need to think more about absolute risk and the number needed to treat. If you showed that there was a 10% benefit for people with a high event rate, then that’s a lot of events prevented. But if something is statistically significant in a low-risk group, with a low event rate, the number needed to treat may still be quite large. For high-risk patients, you would need to treat approximately 130 patients for five years to prevent one event, whereas in the low-risk group, you would need to treat about 300.

A: I agree, yes.

Q: I always get concerned when I see an analysis that picks out the middle subgroup in the absence of benefit on either side, particularly when it’s post-hoc analysis. Do you have a hypothesis on why this is the case?

A: We think it may have to do with power. The event rates were much lower in the low tertile compared to the middle tertile; for low-risk patients, we may need more follow-up time to see the difference between the intensive and the standard treatment groups. Or, it might be something unique to the VADT.

PIOGLITAZONE (PIO) REDUCES PROGRESSION OF CAROTID ATHEROSCLEROSIS INDEPENDENTLY OF IMPROVEMENT IN METABOLIC RISK FACTORS

Aramesh Saremi, MD (NIDDK, Phoenix, AZ)

After exploring the relationships between various risk factors and the progression of carotid atherosclerosis in the ACT NOW trial, Dr. Saremi concluded that pioglitazone reduces the progression of carotid atherosclerosis independently of improvements in traditional, metabolic, and inflammatory risk factors, as well as the use of concomitant medications. In the study, metabolic risk factors including A1c, fasting plasma glucose, two-hour glucose, insulin levels, insulin sensitivity indices, HDL, triglycerides, and triglyceride/HDL ratios improved with pioglitazone use, despite increasing BMI; adiponectin levels nearly doubled, while CRP and PAI-1 levels were significantly reduced with pioglitazone treatment. In conclusion, Dr. Saremi suggested that pioglitazone could have direct anti-atherosclerotic effects, which have been previously documented in animal studies.

Questions and Answers

Q: How do you think your results compare to CHICAGO results? That study showed that the reduction in the progression of carotid atherosclerosis was driven by changes in HDL.

A: In our studies, HDL increased too. Our population had IGT, while CHICAGO’s had type 2 diabetes. In my presentation, I showed that the effects weren’t explained by improvements in HDL. I didn’t say there was no effect, but rather, that there was a residual effect of pioglitazone that is protective besides HDL; our findings are consistent with CHICAGO data.

RISK OF CARDIOVASCULAR DISEASE ASSOCIATED WITH SULFONYLUREA OR METFORMIN USE IN OLDER PATIENTS WITH TYPE 2 DIABETES

Alex Fu, PhD (Cleveland Clinic, Cleveland, OH)

Previous studies have suggested that sulfonylurea therapy may increase the risk of cardiovascular disease. Dr. Fu showed the results of a large cohort study of older US patients, in which sulfonylurea therapy was associated with a 33% higher risk of cardiovascular events within two years as compared to metformin.

  • Using a retrospective cohort from the GE electronic medical record database, Dr. Fu and colleagues examined the potential association between sulfonylurea monotherapy and cardiovascular disease. A total of 8502 older patients (>65 years of age) treated with sulfonylurea or metformin retrospectively studied for incidence of cardiovascular disease. The mean age at baseline was 75 years, and differences in baseline characteristics between the groups were corrected for using propensity score matching. Patients at baseline had no history of ischemic heart disease, myocardial infarction, stroke, transient ischemic attack, and peripheral arterial disease within one year of study baseline.

  • Controlling for baseline characteristics, patients treated with sulfonylurea monotherapy had 33% higher risk of cardiovascular disease as compared to patients treated with metformin monotherapy. The rate of cardiovascular disease within two years of initiating therapy was 12.4% in the sulfonylurea arm vs. 10.4% in the metformin arm of the study. Sulfonylurea use was also associated with a shorter time to developing the first cardiovascular event. Most of the differences in cardiovascular disease outcomes between the two groups were driven by a higher incidence of ischemic heart disease in the sulfonylurea arm.

Questions and Answers

Q: Did you compare different SUs? We found looking at the DOD database that there were differences.

A: No we didn’t look at individual SU agents.

LIFESTYLE CHANGE AND METFORMIN IMPROVE BIOMARKERS OF INFLAMMATION, ENDOTHELIAL DYSFUNCTION AND COAGULATION IN THE DIABETES PREVENTION PROGRAM (DPP)

Marinella Temprosa, MS (George Washington University, Washington, DC)

Ms. Temprosa presented biomarker data from the three treatment arms (lifestyle, metformin, placebo) of the DPP study. She showed that lifestyle, and to a lesser degree metformin, can produce favorable effects on multiple markers of inflammation, endothelial function and coagulation, but that these effects may be influenced by age, sex and race/ethnicity.

  • The Diabetes Prevention Program (DPP) randomized 3,234 patients with prediabetes into three treatment groups: lifestyle intervention, metformin 850 mg twice a day, or placebo. Approximately half of the participants who enrolled in the study were from minority groups. The present study examined changes in biomarkers of inflammation from serum samples collected at baseline and after one year from 3,194 DPP participants.

  • Of the three DPP groups, only the lifestyle intervention group had significant favorable changes across a broad range of inflammation, endothelial dysfunction and coagulation biomarkers. Lifestyle intervention produced significant favorable changes in CRP, IL-6, fibrinogen, sICAM1, sE-Selectin, tPA, adiponectin, leptin, MCP-1 levels. By contrast, metformin lead to changes only in tPA, leptin, and sICAM1. Men and African Americans experienced greater decrease in leptin following any kind of intervention as compared to other groups. By contrast, women had a greater E-selectin and tPA change but a smaller leptin change. Durable changes in these biomarkers may have long-term implications on disease risk.

Q: Do you have any data on central obesity rather than BMI?

A: We found similar results for our group using waist circumference versus BMI.

Corporate Symposium: Kidney Disease in Type 2 Diabetes: Disease Progression and Emerging Therapies (Sponsored by Reata Pharmaceuticals)

PATHOGENESIS OF CHRONIC KIDNEY DISEASE IN TYPE 2 DIABETES

Jennifer Marks, MD (University of Miami, Miami, FL)

It is estimated that by the year 2020, there will be over 500,000 people with end stage renal disease (nephropathy), 60% of whom will have diabetes as the primary cause of their disease (diabetic nephropathy). Traditional risk factors for nephropathy include: hyperglycemia, hypertension, duration of diabetes, hypercholesterolemia, smoking, genetic predisposition, ethnicity, and male gender. Microalbuminuria (loss of albumin in the urine) remains a risk factor for diabetic nephropathy, but not all patients with microalbuminuria progress to nephropathy. Earlier studies found that 80-90% of patients with type 1 diabetes progressed, but more recent studies have shown that 30-45% progress and some will regress. The correlation between microalbuminuria and insulin resistance, together with the evidence that podocytes uptake glucose in response to insulin, suggest that insulin signaling pathways are relevant to podocyte function. Cytokines and other inflammatory substances that are altered in diabetes may affect the course of diabetic nephropathy through direct modulation of podocyte function, and may represent a novel therapeutic target for the treatment of diabetic nephropathy. Vitamin D and the renin-angiotensin system may also have an important role in the early stages of diabetic nephropathy.

CURRENT GUIDELINES AND THERAPEUTIC OPTIONS FOR TREATMENT OF CHRONIC KIDNEY DISEASE IN TYPE 2 DIABETES

Jerry Yee, MD (Henry Ford Hospital, Detroit, MI)

Dr. Yee reviewed the guidelines produced by the KDOQI (Kidney Disease Outcomes Quality Initiative) regarding diabetic nephropathy. The five recommendations of the group are: 1) annual screening for nephropathy starting five years post-diagnosis of either type 1 or type 2 diabetes - screening should consist of both albumin-creatinine ratio spot and eGFR estimation, with 2-3 samples given for classification; 2) targeting an A1c of <7.0%; 3) managing hypertension with initial therapy consisting of either an ACE inhibitor or ARB, targeting a blood pressure of <130/80 mmHg; 4) targeting an LDL-C of less than 100 mg/dl, and optimally less than 70 mg/dl; and 5) nutritional management with protein limitation.

EMERGING THERAPIES AND FUTURE MANAGEMENT OF CHRONIC KIDNEY DISEASE IN TYPE 2 DIABETES

Mark Molitch, MD (Northwestern University, Chicago, IL)

Dr. Molitch provided an overview of new therapies for the treatment of diabetic nephropathy, focusing on compounds that reduce various aspects of inflammation in the kidney. The four drugs that were discussed that have been shown to reduce residual albuminuria include: atrasentan (highly selective endothelin-A receptor blocker), paricalcitol (active vitamin D), pirfenidone (TGF-beta inhibitor), and bardoxolone (nuclear Nrf2 activator). Bardoxolone is a derivative of oleanolic acid (a naturally occurring compound found in olive oil) but is approximately 10,000 more potently anti-inflammatory than the parent compound. In a recent 52-week study that will be published in the New England Journal of Medicine, bardoxolone 25 mg, 75 mg, or 150 mg was associated with a 4.7, 9.4, and 8.1 sustained improvements in eGFR, respectively. This compared to a decrease in eGFR of 1.1 in the placebo arm. The most common side effects were muscle spasms (54%), nausea (18%), and hypomagnesemia (26%). For each of the drugs that were discussed, large-scale, prospective, randomized, placebo-controlled, multi- center, multi-year studies are just now getting underway that will evaluate cardiovascular status in addition to long-term benefits on kidney function.

  • Bardoxolone is a first-in-class antioxidant inflammation modulator in development by Reata Pharmaceuticals and Abbott for the treatment of diabetic nephropathy (chronic kidney disease caused by diabetes). It is a derivative of oleanolic acid (a naturally occurring compound found in olive oil) but is approximately 10,000 more potently anti- inflammatory than the parent compound. For more details on bardoxolone’s mechanism, its previous clinical trial results, and other upcoming CKD therapies, please see our January 28, 2011 Closer Look.

  • Yearlong phase 2 results, recently published online, in the New England Journal of Medicine show that bardoxolone causes significant gains in estimated glomerular filtration rate (eGFR, a measure of kidney function), with sustained improvements after treatment is stopped. The phase 2 BEAM study randomized 227 adults with chronic kidney disease (mean GFR ~32 ml) to receive placebo or bardoxolone methyl 25, 75, or 150 mg once daily. At the end of the 52-week study, patients who received placebo experienced an averagedecline in estimated GFR (kidney function) of -1.1 ml, whereas patients receiving bardoxolone had a sustained improvement in kidney function of 4.7, 9.4, and 8.1 ml estimated GFR for the 25, 75, and 150 mg bardoxolone groups, respectively. Remarkably, eGFR in patients on 75 and 150 mg continued to improve throughout the 52-week study, whereas at the lower 25 mg dose there was some loss of eGFR from week 24 to week 52. The benefits in GFR persisted after bardoxolone was discontinued at study end, suggesting that bardoxolone has a disease-modifying effect on inflammation and oxidative stress associated with chronic kidney disease.

  • The most common side effects for all patients treated with bardoxolone were muscle spasms (42%, 61%, 59% for 25, 75, 150 mg, respectively; 18% for placebo), nausea (18%,26%, 21% for 25, 75, 150 mg, respectively; 9% for placebo), and hypomagnesemia (21%, 25%, 32%for 25, 75, 150 mg, respectively; 5% for placebo). There were no appreciable clinical sequelae from the hypomagnesemia, and because magnesium is renally cleared this may be an effect of increased kidney function.

  • Reata Pharmaceuticals and Abbott are currently conducting a 1,600-participant phase 3 trial of bardoxolone called BEACON (Bardoxolone methyl EvAluation in patients with Chronic kidney disease and type 2 diabetes: the Occurrence of renal eveNts). The outcomes- driven phase 3 trial includes only a placebo and a 20 mg amorphous bardoxolone methyl treatment arm, which is pharmacologically equivalent to the 75mg crystalline bardoxolone methyl dose used in the phase 2 study. The primary efficacy endpoint is time-to-first-event of either end- stage renal disease (defined as need for chronic dialysis or renal transplant) or cardiovascular death. Reata management has previously said that the trial will end when 235 patients have reached this composite endpoint, and a recent press release from Reata and Abbott forecasts that BEACON will complete in 2013.

  • Interestingly, although bardoxolone is in development specifically for chronic kidney disease, its anti-inflammatory effects may have other positive effects such as improved insulin sensitivity and a cardiovascular benefit. Although there was no change in A1c in the phase 2 study, concomitant anti-hyperglycemic medications were not controlled, and in animal models, insulin-mediated glucose uptake in animal is significantly increased with bardoxolone. To investigate whether bardoxolone provides a cardiovascular benefit, cardiovascular endpoints have been included as secondary endpoints in BEACON trial.

PANEL DISCUSSION

Jennifer Marks, MD (University of Miami, Miami, FL), Jerry Yee, MD (Henry Ford Hospital, Detroit, MI), and Mark Molitch, MD (Northwestern University, Chicago, IL)

Q: What is the ideal vitamin D level?

A: Dr. Mark: Nobody knows. Probably 30 is at the lower limit of what is acceptable.

Q: Paricalcitol in cardiovascular mortality?

A: Dr. Molitch: Prospective randomized trials are in progress. The early studies did show a difference.

Q: What is your guidance regarding the use of aldosterone receptor antagonists?

A: Dr. Yee: I think there is underutilization of mineralicorticoid receptor antagonists – spironolactone and eplerinone. There is going to be an increase in potassium, but people overestimate this risk. Actually, depression of bicarb is a bigger risk.

Q: Can metformin be continued in a patient with a creatinine greater than 1.5?

A: Dr. Yee: My council is that I would continue metformin above 1.5. Most of that literature is flawed. 1.4 in females, 1.5 in males, is too hard a stop point. One meta-analysis showed that lactic acidosis doesn’t even exist on this drug.

Q: Any trials on non-active vitamin D – i.e., D3?

A: Dr. Molitch: Not that I’m aware of.

Q: Can Aliskiren be used to reduce proteinuria?

A: Dr. Yee: Yes, any RAAS blockade can be used.

Corporate Symposium: Issues in the Management of Hyperglycemia in Patients with Concurrent Kidney Disease (Sponsored by Boehringer-Ingelheim and Eli Lilly)

CKD PATHOPHYSIOLOGY AND PROGRESSION IN PATIENTS WITH DIABETES: IMPLICATIONS FOR THE ENDOCRINOLOGIST

George L. Bakris, MD (University of Chicago Pritzker School of Medicine, Chicago, IL)

Dr. Bakris discussed the clinical management of kidney disease in people with diabetes. He posed questions to the audience as he worked through case studies. One question that many attendants answered incorrectly involved the interpretation of an increase in serum creatinine following initiation of angiotensin receptor blocker and ACE inhibitor therapy: a significant serum creatinine increase is a normal response to the therapy and a significant increase should be expected without worry of arterial disease. This answer was met with some consternation by the audience, and he clarified in Q&A that an increase in serum creatinine will only be seen in people with long-standing severe CKD, not moderate or early CKD. He also spent a significant portion of his talk the role of statins in the treatment of patients with CKD. Two large trials (4D and AURORA) showed no benefit of statins in patients with CKD. Theories explaining this effect included “statin resistance” and the idea that CKD involves a different etiology of cardiovascular disease than that of the metabolic syndrome. Dr. Bakris believes that this was because they were underpowered and excluded the highest-risk patients. Supporting his argument, he presented data from the SHARP trial, which demonstrated a 17% risk reduction for atherosclerotic events in patients with CKD on statins.

THE CHALLENGE OF HYPOGLYCEMIA IN THE CKD PATIENT WITH DIABETES: SETTING GOALS, MODIFYING TREATMENT

W. Timothy Garvey, MD (University of Alabama at Birmingham, Birmingham, AL)

Dr. Garvey discussed the maintenance of tight glycemic control in diabetic patients with chronic kidney disease (CKD). Glycemic control is an independent predictive factor for the progression of chronic kidney disease and other nephropathy, but in general the targets for glycemic control for patients with chronic kidney disease are similar to those for patients without kidney disease. However, he argued that kidney disease raises the risk of hypoglycemia due to decreased renal glucose production, reduced gluconeogenesis in the liver due to gluconeogenic substrates, decreased glycogen reserves, reduced systemic response to epinephrine and glucagon, and decreased insulin degradation. Because of this, he suggested that a higher target might be appropriate for these patients, particularly older patients. He also pointed out that when monitoring glycemia in patients with kidney disease, A1c can be artificially lowered as a result of the disease.

GLYCEMIC CONTROL AND OTHER CONSIDERATIONS IN ADVANCED KIDNEY DISEASE

Richard E. Pratley, MD (Sanford Burnham Medical Research Institute, Orlando, FL)

Dr. Pratley focused on the use of incretins and DPP-4 inhibitors in patients with chronic kidney disease. He noted that the ADA consensus statement recommendations are often insufficient in guiding the treatment of patients with chronic kidney disease. In particular, he highlighted that glipizides, metformin, and TZDs would be contraindicated in “little old ladies” with chronic kidney disease, which leaves basal insulin, DPP-4 inhibitors, and incretins for therapeutic options to complement lifestyle intervention. He noted that, of the DPP-4 inhibitors, linagliptin was a particularly good choice for patients with kidney disease, since its excretion is nearly entirely hepatic rather than renal. Sitagliptin is cleared almost entirely renally and saxagliptin is cleared both hepatically and renally. Likewise, he recommended liraglutide over exenatide since a significant portion of exenatide is cleared renally. Liraglutide is eliminated almost entirely by enzymatic degradation in the tissue. Renally excreted drugs do not necessarily harm the kidneys in chronic kidney disease, but their decreased clearance rate necessitates dose adjustment for these medications.

Corporate Symposium: HDL & The Dyslipidemia of Diabetes: Examining the Increased risk of Atherosclerotic Events in Type 2 Diabetes (Sponsored by Genentech)

EMERGING TARGETS FOR LIPID LOWERING THERAPY IN T2DM

Theodore Mazzone, MD, FACP (University of Illinois at Chicago, Chicago, IL)

Dr. Mazzone reviewed the importance of addressing lipoprotein abnormalities in type 2 diabetes, citing his belief that hyperlipidemia is the primary mechanism behind the increased risk of cardiovascular disease seen in patients. He called statins the “great story” of the past decade and stated that the vast majority of patients with type 2 diabetes should be on statin therapy. Dr. Mazzone closed cautiously, urging that intervention studies are needed in patients on statins.

HDL AS TARGETS FOR T2DM AND CVD: WHAT HAVE WE LEARNED?

Henry Ginsberg, MD (Columbia University, New York, NY)

Dr. Ginsburg discussed all facets of HDL, mainly focusing on its cardioprotective benefits. He began by reviewing the production and metabolism of HDL as well as the causes of low HDL levels: hypertriglyceridemia, obesity, insulin resistance, and family history. Turning to cardiovascular risk, Dr. Ginsberg discussed recent data suggesting protective effects of HDL independent of its level in the blood (i.e., anti-oxidant effects, effects on insulin secretion from beta cells, inhibition of endothelial adhesion molecules, prostacyclin stabilization, promotion of NO production, and modulation of monocyte production in BM). He closed by emphasizing the complicated structure of HDL, noting that more data is needed on the molecule.

--by Adam Brown, Karthik Chandran, Eric Chang, Shereef Elnahal, Benjamin Kozak, Lisa Rotenstein, Kyle Rudolph, Joseph Shivers, Jessica Swienckowski, Melissa Tjota, Sanjay Trehan, Lisa Vance, Nick Wilkie, Vincent Wu, Mark Yarchoan, David Zhang, and John and Kelly Close