Journal of Diabetes NEWS

International Diabetes Federation (IDF) Highlights Growing Global Impact of Diabetes in 5th Edition of the Diabetes Atlas

The International Diabetes Federation (IDF) marked World Diabetes Day on November 14, 2011 by releasing the fifth edition of the Diabetes Atlas, a global report covering a wide range of aspects in the diabetes field from health economics to education to epidemiology. The goal of this unique resource is to help governments, public health authorities, healthcare professionals, pharmaceutical industries and international health organizations understand the global impact of diabetes. More importantly, the IDF wants to highlight the serious complications diabetes could have on each country if preventative measures are not instituted early on in the disease course.

The newest publication estimates that 366 million people worldwide have diabetes.¹ Leading the list of countries with the number of people with diabetes (20–79 years) in 2011 is China with 90.0 million people, followed by India with 61.3 million people. South-East Asia, a region with seven countries, currently accounts for almost one-fifth of all adults living with diabetes. Furthermore, the Western Pacific Region alone accounts for 36% of the total number people in the world with diabetes. The Pacific Island nations in this region have some of the highest prevalence rates of diabetes: Marshall Islands (22.2%), Nauru (20.7%), Tuvalu (19.5%), and Vanuatu (16.4%). As a comparison, among the European countries, the Russian Federation has the highest prevalence at 10.0% and Moldova has the lowest prevalence at 2.8%. North America and the Caribbean Region have a prevalence rate of 10.7%. The Diabetes Atlas also published results on type 1 diabetes rates with an estimated 30 700 children being affected in the Western Pacific Region and 111 500 children with the disease in South-East Asia.

Worldwide it is estimated that 552 million people will have diabetes by 2030, which equates to diagnosing three individuals with diabetes every 10 seconds from now until 2030. The number of people with diabetes in China and India is estimated to balloon to 129.7 million and 101.2 million people, respectively, by 2030. Strikingly, the 2030 estimates are substantially higher than those presented in earlier editions of the Diabetes Atlas. According to the IDF, this discrepancy is largely due to new data from China, the Middle East and Africa. We compiled a table of estimates from the early edition of the Diabetes Atlas to emphasize how diabetes has become one of the most challenging health problems of the 21st century.

A substantial proportion of individuals with diabetes remain undiagnosed. The IDF estimates that as many as 183 million people are unaware that they have diabetes, accounting for 50% of the estimated 2011 worldwide cases. However, the proportion of people with undiagnosed diabetes does vary by region, with some countries having up to 90% of their diabetes population undiagnosed due to a lack of resources and initiative to screen for diabetes. Over 60% of all people with undiagnosed diabetes are in the South-East Asia Region (36.2 million) and the Western Pacific Region (73.5 million).

Diabetes caused an estimated US$465 billion in global healthcare expenditures in 2011, representing 11% of total worldwide healthcare expenditures in adults. However, this spending was highly inequitable: only 20% of global healthcare expenditures due to diabetes were made in low- or middle-income countries, where 80% of people with diabetes live. In the Western Pacific Region, 2011 expenditures on healthcare due to diabetes accounted for US$72.2 billion or 15.5% of the global total. China, the country with the most people living with diabetes, spent just US$17 billion, or less than 4% of the global total. The mean healthcare cost per person with diabetes ranged from up to US$3000 in Australia and Japan to less than US$20 in Myanmar and the People’s Democratic Republic of Korea. Surprisingly, although South-East Asia has one-fifth of all the people with diabetes, 2011 healthcare expenditures due to diabetes were approximately US$4.5 billion, or <1% of the global total.

Diabetes caused 4.6 million deaths in 2011, a 13.3% increase over the estimates for the year 2010. The Western Pacific region accounted for the highest number of diabetes-associated deaths with 1.7 million deaths among adults, about 15% of all deaths. South-East Asia led the region with 1.16 million deaths due to diabetes in 2011, with 55% of all deaths occurring in people under the age of 60. Following close behind was China with 1.1 million deaths.

The IDF Atlas gives three objectives to address the diabetes epidemic: improving the health outcomes of people with diabetes, preventing the development of type 2 diabetes and stopping discrimination against people with diabetes. The first objective emphasizes diagnosing, treating and monitoring individuals with diabetes, as well as providing individuals with self-management education. The second objective focuses on preventative measures through healthy nutrition and physical activity. Lastly, the IDF wants to encourage the implementation of legal and policy frameworks to help with diabetes-related healthcare delivery and outcomes. In order to achieve these goals, the IDF proposes that every government develops and implements a National Diabetes Program to improve diabetes prevention and care.

47th Annual Meeting of the European Association for the Study of Diabetes

Between September 12 and 16, 2011, we had the opportunity to travel to Lisbon, Portugal, for the 47th Annual Meeting of the European Association for the Study of Diabetes (EASD). Over 18 000 attendees packed the Feira Internacional de Lisboa for the 6-day conference, up from 17 301 attendees last year in Stockholm, 16 577 delegates in Vienna in 2009, and just ahead of the 17 600 attendees at this year’s American Diabetes Association meeting in San Diego. This year’s EASD showcased six tracks, 361 oral presentations, and 1152 posters, roughly on par with last year’s 360 oral presentations and 1222 posters, although the number of exhibitors fell slightly from 81 to 70. The conference featured research and debate in many areas of interest in diabetes of late, including incretin therapies, SGLT-2s, obesity, and diabetes technology. Below, we include some of our top highlights from the meeting (http://www.abstractsonline.com/plan/AdvancedSearch.aspx? mkey = %7BBAFB2746-B0DD-4110-8588-E385FAF957B7%7D).

Clifford Bailey, PhD (Aston University, Birmingham, UK), presented 102-week results from an extension of a 24-week study evaluating the safety and efficacy of the SGLT-2 inhibitor dapagliflozin (DAPA) in type 2 diabetes patients inadequately controlled on metformin monotherapy. In the study, patients were randomized to receive placebo (n = 137), 2.5 mg (n = 137), 5.0 mg (n = 137), or 10 mg (n = 135) DAPA. At baseline, patients had an average Hb_A1c of 8.0–8.1% and body mass index (BMI) of 31–32 kg/m². At week 102, patients in the placebo, 2.5 mg, 5.0 mg, and 10 mg DAPA arms experienced dose-dependent Hb_A1c reductions of 0.02%, 0.48%, 0.58%, and 0.78%, respectively. Dr Bailey noted that twice as many patients in the 10 mg DAPA achieved an Hb_A1c of <7.0% by the 102-week mark, compared to placebo (30 vs 15%). Patients in the placebo, 2.5 mg, 5.0 mg, and 10 mg DAPA arms experienced average reductions in fasting plasma glucose of 10.4 mg/dL, 19.3 mg/dL, 26.4 mg/dL, and 24.5 mg/dL from baseline, respectively. While some weight regain occurred between week 24 and week 102, patients receiving DAPA still maintained weight loss beyond the placebo group; patients in the placebo arm ended with an average 1.36 kg (3.0 lbs) increase in weight, while those who received 2.5 mg, 5.0 mg, or 10 mg DAPA had an average weight loss of 1.1 kg (2.4 lbs), 1.7 kg (3.7 lbs), and 1.74 kg (3.8 lbs), respectively. In terms of adverse events, those receiving DAPA treatment experienced a higher rate of urinary and genital tract infections than those receiving placebo. In the placebo, 2.5 mg, 5.0 mg and 10 mg DAPA arms, 8.0%, 8.0%, 8.8%, and 13.3% of patients experienced adverse events suggestive of a urinary tract infection and 5.1%, 11.7%, 14.6%, and 12.6% experienced events suggestive of a genital infection. Regarding other adverse events, 5.8%, 3.6%, 5.1%, and 5.2% of patients experienced at least one hypoglycemic event and 1.5%, 0.0%, 2.2%, and 1.5% experienced hypotension, syncope, dehydration, or hypovolemia. Notably, there was only one discontinuation from repeated genital infection.

Samuel Engel, MD (Merck, Rahway, NJ, USA), presented results from a phase 2 study for Merck’s glucagon receptor antagonist MK-0893. The safety and efficacy of the compound was evaluated when used in combination with metformin and sitagliptin over a 4-week period in people with type 2 diabetes. Participants were randomized to receive 40 mg MK-0893 plus 2000 mg metformin (n = 49), 40 mg MK-0893 plus 40 mg sitagliptin (n = 48), or 100 mg sitagliptin plus 2000 mg metformin (n = 49) after a 3-week washout period. At baseline, the average BMI was 31 kg/m² and Hb_A1c was 8.6%. Change in 24-hour weighted mean glucose was the primary efficacy endpoint. The MK-0893 plus metformin regimen provided statistically significantly greater reductions in weighted mean glucose (−6.5 mmol/L [117 mg/dL]) over both other regimens (P < 0.01). The sitagliptin plus metformin regimen (−5.5 mmol/L [99 mg/dL]) provided statistically significantly greater reductions in weighted mean glucose over the MK-0893 plus sitagliptin regimen (−4.8 mmol/dL [86.4 mg/dL]) (P < 0.05). It was hypothesized that the lowest glycemic efficacy was observed in the MK-0893 plus sitagliptin regimen because blockade of the glucagon receptor attenuated the component of sitagliptin efficacy related to suppression of plasma glucagon levels. No statistically significant differences between the arms were observed for fasting insulin, fasting c-peptide, 2-hour postprandial insulin, or homeostasis model assessment (HOMA)-B. Interestingly, while both regimens provided statistically significant increases in total glucagon-like peptide (GLP)-1 levels over the sitagliptin plus metformin regimen, the later regimen provided statistically significantly greater increases in active GLP-1 levels over both MK-0893 regimens, suggesting that the measured increase in total GLP-1 with MK-0893 reflected an inactive form of GLP-1. MK-0893 was associated with greater weight gain than sitagliptin plus metformin (0.4–1.0 kg [0.88–2.2 lbs]) as well as small increases in ALT/AST levels (3–7 IU/L). Drug related adverse events were seen in 14% of the sitagliptin plus metformin arm, 18% in the MK-0893 plus metformin arm, and 10% in the MK-0893 plus sitagliptin arm. The incidence of hypoglycemia was low with all the treatments (0–2%).

In front of a packed and attentive audience, John Buse, MD, PhD (University of North Carolina, Chapel Hill, NC, USA), presented the long-anticipated results from DURATION-6, the 26-week exenatide once weekly (EQW) versus liraglutide head-to-head trial. From respective baseline Hb_A1c values of 8.4% and 8.5%, those receiving EQW and liraglutide experienced average decreases in Hb_A1c of 1.28% and 1.48%, respectively. Hence, EQW did not demonstrate non-inferiority to liraglutide; in addition, more subjects taking liraglutide achieved an Hb_A1c of <7.0% (60.2%) than those receiving EQW (52.3%) (P = 0.008). Over the course of 26 weeks, liraglutide treatment also resulted in significantly more weight loss than EQW treatment (3.58 kg [7.9 lbs] vs 2.68 kg [5.9 lbs]; P < 0.05). Those receiving liraglutide experienced gastrointestinal-related adverse events approximately twice as often as those receiving EQW. Lastly, no major hypoglycemic events occurred during the study; minor and nocturnal hypoglycemic events were balanced between treatment arms. Dr Buse noted that there still are no concrete explanations for the lower-than-previously observed efficacy of EQW seen in DURATION-6; he mentioned that perhaps better renal function of patients in the trial could have reduced their exposure to the medication and thus reduced efficacy (patients were required to have a glomerular filtration rate of ≥60 mL/minute/1.73 m² in DURATION-6, whereas no such inclusion criterion was applied in the other DURATION trials).

Urd Kielgast, MD (Hvidovre Hospital, Copenhagen, Denmark), presented a 4-week study (n = 19) that examined the safety and efficacy of liraglutide in people with type 1 diabetes. Overall, treatment with liraglutide statistically significantly reduced insulin dose requirements in people who were both C-peptide positive and C-peptide negative. The insulin dose reductions achieved were demonstrated to be positively correlated with baseline C-peptide levels. Liraglutide treatment also provided: (i) statistically significant reductions in the overall duration of daily hypoglycemia for people who were C-peptide positive only; (ii) no effect on β-cell function (as measured by C-peptide levels); (iii) reductions in plasma glucagon levels (specific data not provided); and (iv) statistically significantly greater reductions in weight in contrast to insulin only therapy. As expected, the majority of people treated with liraglutide (18 out of 19) experienced gastrointestinal (GI) side effects, while no participants treated with insulin therapy only experienced GI side effects. To Dr Kielgast, these results highlight the potential of liraglutide to be an effective therapy for the management of type 1 diabetes. She concluded with a call for larger and longer trials to help further substantiate these results.

Stephen Bain, MD (Swansea University, Swansea, UK), weighed the pros and cons of using GLP-1 agonists prior to basal insulins in the treatment of type 2 diabetes and concluded with a discussion on the benefits of combination GLP-1/basal insulin therapy. He noted that the current American Diabetes Association (ADA)/EASD guidelines recommend the use of basal insulins in people with type 2 diabetes once metformin and lifestyle interventions fail. However, he argued that it may be advantageous to delay the use of basal insulins in favor of GLP-1 agonists, given the potential of GLP-1 agonists (at least in animals) to preserve β-cells and slow the progression of type 2 diabetes. Moreover, in clinical trials comparing GLP-1 agonists to basal insulins, both liraglutide (LEAD-5) and exenatide have provided comparable reductions in Hb_A1c with the added benefits of weight loss (versus weight gain with insulin) and lower hypoglycemic risk. Turning to combination GLP-1/basal insulin therapy, Dr Bain reviewed results from a 26-week trial comparing combination liraglutide/insulin detemir therapy with liraglutide therapy in patients on a background of metformin. After a 12-week run-in period with 1.8 mg liraglutide, individuals who were unable to achieve an Hb_A1c of <7.0% (average Hb_A1c was 7.6%) were randomized to intensify treatment with the addition of insulin detemir or to continue with liraglutide plus metformin. At 26 weeks, the detemir arm achieved greater additional reductions in Hb_A1c (−0.51% vs +0.02%; baseline post run-in was 7.6%), similar weight loss (−4.0 kg [–8.8 lbs] vs −4.7 kg [−10.4 lbs]), and a similar rate of hypoglycemia (approximately 0.2 events/patient-year) as compared with the non-detemir arm. New in this presentation, Dr Bain revealed that Hb_A1c levels remained stable throughout the 26-week extension period. To Dr Bain, these results provided strong support for the early use of GLP-1 agonists in the treatment of diabetes and the use of basal insulin to intensify GLP-1 treatment when needed.

Mark Mortellaro, PhD (Sensors for Medicine and Science, Germantown, MD, USA), presented data on a novel implantable continuous glucose monitoring (CGM) device. The sensor (about one-fourth the size of a 2€ coin) is wireless, fluorescence-based, and powered by a wristwatch that doubles as a typical CGM receiver. Two sensors were implanted for 1 month in nine subjects (18 sensors in total). Notably, the sensor recorded a mean absolute relative difference (MARD) of 12.2%, with 77% of points in Zone A and 19% in Zone B of the Clarke
Error Grid. This early accuracy data looks strong, especially for an implantable sensor, although the sensor presents a number of challenges to overcome before commercialization. These include: testing longer implant times, overcoming biocompatibility and photobleaching, user-friendly calibration, and additional clinical trial requirements for regulatory authorities.

Olga Kordonouri, MD (Auf der Bult, Hannover, Germany), gave an encouraging presentation on the benefits of initiating continuous subcutaneous insulin infusions (CSII) and continuous glucose monitoring (CGM) at diagnosis. She began by summarizing the results of the multicenter Pediatric ONSET Trial.² As a reminder, the study compared conventional pump therapy plus self-monitoring of blood glucose (SMBG) to sensor-augmented pump (SAP) therapy in patients who were diagnosed with type 1 diabetes within 4 weeks of trial initiation. The 1-year data showed no difference in Hb_A1c between the groups, although the mean amplitude of glycemic excursions (MAGE), a measure of variability, the fasting C-peptide (among 12–16 year olds), and the rate of severe hypoglycemia were better in the SAP group. She concluded by presenting 2-year follow-up data. At 2 years, there was again no significant difference in Hb_A1c between the SAP and conventional pump therapy groups. By sensor use, those who had used more than one sensor per week during the 1-year study period had an Hb_A1c of 7.4% at the 2-year mark, compared to 7.7% in the less frequent sensor group (not significantly different). Overall, the Hb_A1c levels continued an upward trend observed from the middle of the study to the 1-year mark. At 24 months, however, the loss in fasting C-peptide was significantly lower in those who had used more than one sensor per week compared to the lower-use SAP group and the conventional pump group. Among those in the SAP group using more than one sensor per week, the C-peptide decline between diagnosis and 2 years was 0.02 nmol/L (0.16–0.14 nmol/L), compared to 0.06 nmol/L (0.19–0.13 nmol/L) in the SAP group with less frequent sensor wear and 0.07 nmol/L (0.16–0.09 nmol/L) in the conventional pump group (P = 0.046). In the SAP group overall, fasting C-peptide at the 2-year mark was 0.13 nmol/L versus 0.09 nmol/L in the conventional pump therapy group (P = 0.121). Once again, this study seems to confirm what has been seen in other major CGM trials, including STAR (Sensor-augmented pump Therapy for A1_C Reduction)-3 and the Juvenile Diabetes Research Foundation CGM trial: the sensor has the greatest benefit in those who use it the most.

In one of the conference’s most anticipated debates, Peter Butler, MD (University of California, Los Angeles, Los Angeles, CA, USA), delivered a compelling and carefully crafted argument in support of the claim that GLP-1 based therapies increase the risk of developing certain cancers. Focusing on the risk for pancreatitis and pancreatic cancer, Dr Butler’s argument centered around the following findings: (i) in rodents, GLP-1 therapy induces pancreatic duct proliferative changes most notable in the head of the pancreas; (ii) human pancreatic duct cells are competent to GLP-1 pro-proliferative signaling; (iii) GLP-1 receptors are expressed in pancreatic ductal glands; and (iv) pancreatitis and pancreatic cancer are reported in excess with exenatide and sitagliptin therapy in the FDA Advanced Events Reporting System in comparison to other diabetes medications.³ Still, Dr Butler noted that he was “more in the ‘maybe’ camp rather than the ‘yes’ camp” on this issue. Moving forward, he highlighted the need for retrospective analyses on cancer safety for GLP-1 based therapies and urged pharmaceutical companies to publicly share any data or animal tissue samples they have collected from the development programs for these drugs. In conclusion, Dr Butler advised that until a better understanding of the risk for cancer associated with GLP-1 based therapies is gained, all GLP-1 based therapies should be used in combination with metformin, given the latter drug’s potential ability to suppress cancer development.

Debating Dr Butler, Michael Nauck, MD, PhD (Diabeteszentrum Bad Lauterberg, Bad Lauterberg, Germany), argued that the current information we have suggests GLP-1-based therapies do not increase the cancer risk. Regarding pancreatitis (as a precursor of pancreatic cancer), he stated that: (i) animal studies (mechanisms, safety) and epidemiological data do not convincingly demonstrate an elevated risk for pancreatitis with exenatide and sitagliptin; (ii) if at all, GLP-1-based therapies are related to acute (not chronic) pancreatitis; and (iii) studies demonstrating increased risk for chronic pancreatitis and pre-neoplastic lesions with GLP-1 agonists and dipeptidyl peptidase (DPP)-4 inhibitors are absolutely lacking. As for pancreatic cancer, he emphasized that: (i) animal studies with GLP-1 agonists and DPP-4 inhibitors (long duration, high dose) have not shown an increased risk of pancreatic cancer; (ii) chronic pancreatitis and pre-neoplastic lesions have not been described in such studies; and (iii) while the US Food and Drug Aadministration (FDA) Adverse Event Reporting
System (AERS) analysis conducted by Elashoff et al.³ found an increased risk of pancreatic cancer with exenatide and sitagliptin exposure for up to 4 years, it is biologically implausible given the time course of cancer. Dr Nauck proceeded to discuss GLP-1-based therapies and the risk for thyroid cancer, noting that while rodent studies show a high spontaneous rate of C-cell disease, these effects have not been confirmed in human populations. Unlike in rodents, GLP-1 receptors are sparse in human C-cell lines and exposure to GLP-1 agonists has not led to elevations in cyclic adenosine monophosphate (cAMP) and calcitonin levels.⁴ Even though Dr Nauck thought the FDA AERS finding of more thyroid cancer in general is implausible, he suggested that further studies are needed to exclude some influence of incretin-based drugs on medullary C-cell carcinoma. Finally, Dr Nauck highlighted that some forms of cancer/malignancies were actually reduced based on the FDA AERS database.³ Given the remaining uncertainties, Dr Nauck emphasized the need for further investigation; unfortunately, there will likely never be a randomized controlled trial of sufficient size to definitively characterize the risk of cancer with GLP-1 therapies, as it would likely require 80 000 patients in each arm – and would take, to boot, quite a while. During the following discussion, Dr Nauck emphasized that he does not see a reason to change clinical practice as far as the prescription of GLP-1 agonists and DPP-4 inhibitors, but does acknowledge that they are better used in combination with metformin.

In the 46th Minkowski Lecture, Naveed Sattar, MD (University of Glasgow, Glasgow, UK), delivered a fascinating presentation on the use of biomarkers to predict diabetes, as well as on the use of large datasets to provide novel insights into the pathology of diabetes and cardiovascular disease (CVD). Focusing first on diabetes prediction, Dr Sattar argued that a perfect diabetes prediction tool was highly unlikely to ever materialize, and, to boot, is unnecessary anyway. Stressing that our ability to predict diabetes development today is quite good, he pointed out that simple models incorporating weight, BMI, age, waist circumference, and age have an accuracy of 76%, which can be improved to 81% if blood glucose measures are added. While biomarkers (including adiponectin and C-reactive protein) thought to be associated with the development of diabetes have been identified, he underscored that they have yet to provide substantial improvements in prediction accuracy over these simple models. Still, Dr Sattar stressed that diabetes biomarker research was still in its infancy, and that he was particularly excited about the potential for biomarkers in predicting the development of diabetes complications and responses to diabetes treatments – paving the way for a greater individualization of diabetes clinical management. In the second half of his talk, Dr Sattar challenged several pieces of “conventional wisdom” and addressed several oft-discussed questions about diabetes and CVD. As we have heard in the last couple of years from others, Dr Sattar contended that type 2 diabetes is not CVD risk-equivalent at diagnosis, statins and blood pressure medications are much more effective than diabetes medications at lowering CVD risk, statins may increase the risk of diabetes, and attempts to develop criteria that will help predict both diabetes and CVD are unhelpful.

One final highlight from the meeting was the exhibit hall. Indeed, EASD was our first chance to see Amylin/Lilly/Alkermes’ once weekly GLP-1 agonist Bydureon since it received approval in the European Union in June 2011. “Connect, Shake, Inject” was the theme of the injection process, which looked quite simple in demonstrations, despite the four-piece kit and six-step process – it will be interesting to see how clinical trial patient success translates to success in the “real world”. The 23 gauge, 7.8 mm long needle is no doubt larger than the needles recommended for Victoza (5 or 6 mm; 32 gauge) and Byetta (4, 5 or 8 mm; 29, 30 or 31 gauge), but the once-weekly dosing and significantly lower nausea rates will certainly be welcomed by many patients. On the technology front, Ypsomed’s busy booth was front and center at the entrance to this year’s exhibition – the focus was the new second-generation OmniPod, which is 35% smaller and 25% lighter than the previous generation version. Nearby competitor Cellnovo’s corner booth was also notable at this year’s meeting: the company’s pump received European approval during EASD, with FDA submission reportedly expected by the end of the year. Additionally, Cellnovo announced a major alliance with LifeScan, meaning that Verio test strips will be integrated with Cellnovo’s handheld device. Abbott’s FreeStyle InsuLinx, LifeScan’s OneTouch Verio Pro, Roche’s second-generation Accu-Chek Mobile, and Sanofi’s iBGStar were other notable technologies present in the exhibit hall, reminders of the fierce competition, incremental innovation in the field, and a slow-moving US FDA. None of these products are approved in the US. Finally, DPP-4 inhibitors were among the more popular classes in the exhibit hall, with Merck’s Januvia–Janumet franchise, Boehringer Ingelheim/Lilly’s Trajenta, Novartis’ Galvus and Eucreas, and Bristol–Myers Squibb/AstraZeneca’s Onglyza all attracting interested attendees.