Best of ADA 2020

February 12-14, 2020; Cairo Egypt; Highlights Report Days #1-2 – Draft

Executive Highlights

  • The first Best of ADA 2020 has officially kicked off at the Hilton Heliopolis in Cairo, Egypt! See below for our top six days #1-2 highlights.

  • We were pleased to hear from Joslin’s Dr. Osama Hamdy on a new virtual diabetes clinic called Joslin Home. Joslin Home offers remote coaching from a diverse care team and onboards patients with a connected BGM, blood pressure cuff, and connected scale. The goal of this new virtual clinic is to increase patient visits, reduce patient volume burdens on healthcare providers, simplify data management, and make visits more time-efficient. A pilot study found six-month A1c reductions of ~1.2% from baseline 9.5%.

  • We also caught an eye-opening session about diabetes epidemiology and outcomes in Egypt, Lebanon, and Saudi Arabia. Diabetes rates in the region are among the highest in the world, ranging from 11% to 24%. See below for some staggering statistics demonstrating the need for urgent action.

  • The renowned Dr. Robert Ratner gave a valuable overview on the state of microbiome research, claiming that roadmaps identifying the role of specific bacteria and how they interact with each other will be the key for improving diabetes outcomes. Overall, there has been an explosion of activity surrounding the microbiome in recent years (see JAMA Network paper, Milken Institute Conference 2019, ATTD 2018), but the best evidence today from animal studies suggest that passive fecal transplantation has the highest therapeutic potential. Can we combine systems approaches that synthesize genetics, environmental, and social factors to personalize care? We remain hopeful after today’s talk. 

The first ever Best of ADA 2020 is officially underway as over 100 representatives from all over the Middle East have descended on “The City of Thousand Minaret,” Cairo, Egypt. The conference is being held in the magnificent Hilton Heliopolis around the outskirts of the city. See below for our top six highlights from days #1-2!

Top Six Highlights

1. New Joslin Home Virtual Diabetes Clinic Delivers 1.2% A1c Reduction from 9.5% Baseline in Six-Month (n=17) US Pilot Study; Plans to Scale in the US, Singapore, and Middle East

In a fascinating talk about the future of diabetes technology, Joslin’s Dr. Osama Hamdy provided the first glimpse to the new Joslin Home Diabetes Clinic, a virtual diabetes clinic aimed at increasing patient engagement, shortening care visits, and delivering A1c reductions. In a six-month pilot study within the US (n=17 type 1 and type 2 patients, baseline A1c of 9.5%, 12 participants on insulin therapy), patients were onboarded with a connected blood glucose meter, connected blood pressure cuff and connected scale. Patients were also given access to a diverse care team including an endocrinologist, nurse practitioner, exercise physiologist, dietitian, nurse educator, and behavior therapist, with whom they could communicate with on a mobile application. Appointments were geared to be short (5-15 minutes), occur at least once per week, and with any combination of care team members. Patients could schedule all appointments on the application and were given no penalty for cancellations. To maximize the quality of the visit, the care team scheduled for the visit would send five questions to the patient before the visit to see if the patient was actually taking prescribed medications, determine if blood glucose levels had exceeded 200 mg/dl or gone below 70 mg/dl, assess the patient’s meal plan or nutrition concerns, clarify physical activity levels, and understand patient goals for managing diabetes. After six-months of pilot testing, Dr. Hamdy mentioned that participants on average saw an A1c reduction of 1.2%, with 70% of visits conducted in under 20 minutes, ~76% of participants having 6-20 visits in five months (average of one visit per week), appointment cancellation rates the lowest with exercise physiologists and registered dietitians (10-18%), and physicians/NPs able to finish their consultations in ~11 minutes on average. Additional data on the pilot is expected to be shared at the Digital Orlando 2020 conference in April.

  • Dr. Hamdy attributed the initial success of the telehealth-based clinic to five key principles: (i) short visits; (ii) multiple visits; (iii) dynamic schedule; (iv) focused EHR; and (v) easy billing. Joslin prioritized these concepts based on the staggering shortage of endocrinologists in the United States, the long times patients go without seeing their healthcare providers, the short face-to-face time patients have with their providers during visits, the administrative complexities of data logging, and the hassles of dealing with insurance. For example, he illustrated that of 6,300 trained endocrinologists in the country, only 1,000 are actually practicing once those who are focused on academic research, in other positions (e.g. industry, government, non-profits), or retired are excluded. Furthermore, with 29 million people with diabetes and 86 million people with prediabetes in the US, that totals one physician per 29,000 patients and one physician for 3.6 hospitals. Of course, most of the 86 million with pre-diabetes do not know they have pre-diabetes – still, the shortage is acute. Dr. Hamdy also explained that despite overwhelming data showing the association between patient visits and improved outcomes (e.g. A1c drops, reduced time in hypoglycemia, DKA decreases), patients on average only get eight minutes of true contact with their physician during visits, with the majority being devoted to entering data in electronic health records. Although he did not mention specifics, Dr. Hamdy claimed that there are plans to scale across the United States and that the military is especially interested in utilizing the clinic for veterans impacted by diabetes. There are also plans of running the pilot in Singapore and a “small” Middle-Eastern country. While the model certainly has potential, information on how much it might cost and what the outcomes would be in larger populations are unknown. We believe the model has great potential, particularly given potential for education surrounding Time in Range and CGM that has the power to enable various positive interventions (therapeutic, nutritional, etc.).

  • With blood glucose data synced to the cloud, Dr. Hamdy also shared that in the future, advances in artificial intelligence and machine learning could facilitate real-time clinical data transfer into population databases. This would enable physicians in a hypothetical population of 1,000 patients to filter results and see real-time which patients tested their blood glucose, when they performed the tests, and received prior visits, enabling opportunities for intervention through text message or email. It could also have valuable applications for governments in developing countries where robust data collection systems lack to quickly identify population-level trends like average blood glucose on one day, in the past week, in the last month, or even in the last year. Dr. Hamdy mentioned that this model is not yet a replacement for in-person visits but an example of the potential of technology to close gaps in care delivery and increase returns on investment.

2. Dr. Robert Ratner on the Microbiome: “We Need to Crate [Roadmaps] Seeing the Impact What Drugs We Use in Diabetes Affect the Microbiome”

Dr. Robert Ratner kicked off day two with a lecture on the microbiome, stating that the field’s future lies in developing maps identifying bacteria that impact specific signals and using these maps to determine how diabetes drugs affect the microbiome’s ecology. This task will certainly take time given its size and scale. For context, humans on average hold 10-100 times the number of bacterial than human genes, and the type and composition drastically change depending on the depth of the GI tract being examined. Nonetheless, this possibility only exists because of the explosion of work that has recently emerged that attempts to link microbiome imbalances to type 2 diabetes, impaired glucose control, and other metabolic issues. For example, one recent paper in EBioMedicine that Dr. Ratner cited nicely synthesizes 42 human studies reporting associations with the microbiome and disease. These studies have generally found that increases of Bifidobacterium, Bacteroides, Faecalibacterium, Akkermansia, and Roseburia have correlated less with type 2 diabetes. However, the presence of Ruminococcus, Fusobacterium, and Blautia tend to more frequently associate with type 2 diabetes, while Lactobacillus has historically yielded discrepant results. From a therapeutic standpoint, the only information that the field currently has from animal studies is that passive fecal transfer of microbiota conveys clear health benefits (probiotics might have some influence but there are currently no quality controls and the field lacks information on what specific probiotics to use).

  • To make his point about the potential of passive transplantation of microbiota, Dr. Ratner displayed 2013 data from a paper in Science and Medicine regarding the impacts of transferring the microbiota of mice that had undergone gastric bypass compared to those that either had a sham procedure or were germ-free. Thirteen days after colonization, mice that had undergone routine gastric bypass surgery lost ~5% of their body weight while those with the sham surgery displayed no statistically significant differences in weight. Despite the meaningful weight loss, mice that had received routine surgery were consuming the same amounts of food as germ-free mice, while those that had undergone the sham procedure were consuming less food. Dr. Ratner explained that this discrepancy is due to the fact that caloric expenditure in mice that received gastric bypass was much higher than that in the germ-free mice and those that obtained a sham procedure. These findings appear to suggest that the microbiome influences important signaling pathways for host regulation of energy balances, appetite, food intake, and energy expenditure. From a diabetes standpoint, Dr. Ratner also mentioned that commonly taken drugs like metformin tend to have beneficial impacts on the microbiome. Whether or not this is part of their natural mechanism of action still remains to be determined. Nonetheless, with respect to fecal transplantation of microbiota today, additional research and clinical trials are required. The FDA has approved this procedure for treating Clostridiodes Diffcile, but even the requirements for getting this procedure are quite stringent. Even with this, FDA seems wary about this procedure and even convened a meeting in November 2019 after two bacterial infections that resulted from the procedure were identified in June earlier that year. Interestingly, one can even find DIY fecal transplantation videos on YouTube, highlighting the growing interest in this area of work.

  • Beyond fecal transplantation, diet and environment also appear to have critical associations with microbial composition. On this front, at AADE 2018, Dr. Eran Segal (Weizmann Institute of Science) presented data from The Personalized Nutrition Project (n=800, without diabetes or prediabetes) to assess how people have various postprandial responses to the same food. Despite each participant demonstrating consistent responses to identical meals across different days, responses to the same meals varied widely, suggesting that a single diet will have limited impact in a large population. Additionally, in a separate 2018 Nature publication, Dr. Segal also demonstrated based on data from over 1,000 people that the microbiome is not significantly associated with ancestry but is strongly related to environmental circumstances. Ultimately, combining data from genetics, environmental factors, social determinants of health, and diet might be the key for creating the ecological roadmaps Dr. Ratner envisions that are critical for understanding diabetes prognosis and treatments.

3. Alarming Session on Diabetes Epidemiology in the Middle East Finds Prevalence in Egypt, Lebanon, and Saudi Arabia Ranging from 11% to 24%

Wednesday afternoon featured a session on prevalence, epidemiology, and care delivery gaps in Egypt, Lebanon, and Saudi Arabia – overall, the data was quite alarming. Data presented from the IDF Diabetes Atlas 2019 underscores that the highest rates of diabetes today are in the Middle East (after India and China) and that this is expected to carry into 2035. In this session, all three countries had a representative from the region discussing relevant statistics and explaining circumstances unique to each place that have led to poor outcomes. The majority of the data discussed below was individually collected by each presenter’s research group. 

  • Professor Samir Helmy Assaad-Khalil (Alexandria Faculty of Medicine) showcased data finding an age-adjusted prevalence of diabetes of 16.8% (men, 12.7%; women 19.1%), prediabetes of 14.6% (men, 13.5%; women, 15.2%), and newly diagnosed case rates of 5.5% (men, 3.3%; women 6.6%) in Egypt. Interestingly, the factor most associated with diabetes in Egypt is waist circumference, and despite a loose classification cutoff point for abdominally obesity (100.5 cm for males vs. 94 cm in Europids; 96.25 cm for females vs. 80 cm in Europids), over half of the population has abdominal obesity. Many of these factors can be associated with lifestyle: (i) 1/3 of the population has obesity; (ii) 1/4 smokes; (iii) 1/3 have sedentary lives; and (iv) there are no standardized national policies for nutrition and exercise. Not surprisingly, these conditions have paved the way for complications which account for 65% of direct medical costs in Egypt, according to the International Society of Pharmacoeconomics & Outcomes Research. Based on a sample of patients Prof. Assaad-Khalil analyzed (size unspecified), peripheral neuropathy was detected in 20%, kidney disease in over 33%, and retinopathy in nearly 35%. According to 2016 data from the International Diabetes Practices Management Study, the largest observational study of adults living with diabetes in developing countries, only 12.2% of type 1 and 17.8% of type 2 patients in Egypt were achieving an A1c of <7%, respectively. There is not yet official info on Time in Range data, but we imagine it is quite poor.

    • Despite these outcomes, Egypt spends on average only ~$219 per patient on diabetes care, which is low even compared to ~$488 in the Middle East and ~$1,622 worldwide. Notably, almost 80% of this amount is spent out-of-pocket, representing an enormous burden for patients. Of special note, Egypt was ranked as ninth in the world with respect to diabetes prevalence according to the IDF Diabetes Atlas 2019, with nine million people in 2019 and an 89% rise expected to 17 million by 2045. During Q&A, one audience member mentioned that the prevalence could be as high as 16% currently and questioned IDF’s methodology. We see this as plausible, considering that IDF has continued to underestimate multiple diabetes rates over time. See our coverage of the IDF Diabetes 2019 Atlas for a glance at disparities in healthcare spending by region.

  • Dr. Sami Azar (American University of Beirut Medical Center) cited IDF data in Lebanon, showing massive problems associated with type 2 diabetes and prediabetes. Data from Lebanon’s 2016 WHO Country Profile demonstrates an adult obesity prevalence of 31% (if anything, this seems like an underestimate), and a 2012 study published found a prevalence of 11% in 6-19 year-olds. Type 2 diabetes accounts for 11% of the adult population, and a cross-sectional study done by his research group (n=15,016) found only 30% were achieving an A1c < 7%. Diabetes burnout was a common theme throughout the presentation. Data presented (source not specified) illustrates that half of all people with diabetes are afraid of the illness progressing, one-third are tired of complying with medication, and one-third feel stressed because of their diabetes. Doctors claim that 1/6 of type 2 patients have depression in Lebanon. Compared to other places in the Middle East (e.g. Bahrain, Ihrain, Iraq, Kuwait, Qatar, Jordan), Africa, North America, and Europe, Lebanon has more physicians compared to nurses per 10,000 of the population. Lebanon also serves as safe-haven for Syrian and Palestinian refugees, which may further complicate the interpretation of these data. In a passing comment, Dr. Azar mentioned that outcomes among these populations are worse, with one study finding diabetes rates among Syrian refugees around 19%.

    • Dr. Azar also presented some alarming statistics about diabetes rates in other parts of the world, demonstrating that every region has much work to do in advancing care and preventing the illness: (i) 78% of people with diabetes are undiagnosed in Africa; (ii) Europe has the highest prevalence of type 1 diabetes in children; (iii) the Middle East and North Africa contain six of the top 10 countries with the highest rates of diabetes prevalence; (iv) 12.3% of all deaths were due to diabetes in South and Central America; (v) almost 20% of the world’s diabetes population live in seven Southeast Asian countries; and (vi) 132 million adults have diabetes in the Western Pacific region. 

  • Dr. Saud Alsifri Alhada (Taif Armed Forces Hospitals) stated abnormal glucose metabolism in Saudi Arabia has reached “an epidemic level.” He showed data collected from his research group highlighting a prevalence of ~25.4% with more than 40% being unaware of the disease. Renal complications account for nearly 75% of the total cost, with nearly 85% of patients managed by general practitioners and 90% of medical services coming from the government.

4. Dr. Michelle Magee Provides Overview of CGM and AID Landscape, Observes “Incremental Advances Each Year” With More Data Required on Type 2 Populations

Georgetown’s Dr. Michelle Magee provided a valuable overview of the CGM and automated insulin delivery landscape, highlighting current devices on the market, similarities and differences among each one, improvements coming in the pipeline, and barriers preventing adoption. Dr. Magee kicked off her presentation with a slide from Close Concerns’ Abbott 2Q19 report demonstrating increased penetration of the FreeStyle Libre when it crossed the 1.5 million-user base mark. During JPM 2020 and Abbott’s most recent earnings call, we learned that the user base was “approximately two million,” with Mr. Ford adding at JPM a potential market size of “80 to 100 million” people. Despite increased CGM usage, Dr. Magee continued onwards with barriers that need to addressed, citing a 2014 paper which found that 41% of patients discontinued CGM usage after year (n=17,317) because of discomfort, difficulties with insertion, alarm fatigue, and skin reactions – see more below on how some of these have been addressed. Additionally, physicians and other healthcare providers today must learn how synthesize large quantities of CGM data never seen before and tailor that information to unique treatment courses. At the recent ADA Postgraduate course, Drs. Irl Hirsch and Diana Isaacs both highlighted the difficult challenge faced by healthcare teams in using the powerful data generated by CGM to make decisions. Notably, she also expressed that more data is needed on type 2 populations, illustrating an unmet need in this population. On this front, Dexcom shared data at JPM 2020 which garnered some attention: in a small, randomized pilot with Intermountain Healthcare (n=99 type 2s), the 49 patients randomized to CGM showed total annualized cost savings of “~$5,000” per member, excluding the cost of CGM, compared to SMBG.

  • Dr. Magee highlighted improvements in CGM cost, form factor, wear-time, and design that have been made and will continue to occur. For example, Dexcom’s G6 has a much more comfortable insertion than previous generations did, and Dexcom’s G7 will feature a one-piece fully disposable wearable (integrated sensor/transmitter), a significantly lower manufacturing cost reduction, iCGM accuracy and factory calibration, 14-15 day wear, Bluetooth to app connectivity, and possibly an accelerometer. Abbott’s CE-marked FreeStyle Libre 2 has customizable thresholds and optional hypoglycemia/hyperglycemia alarms at the same cost as the original FreeStyle Libre. Medtronic’s pivotal trial for Zeus iCGM (seven-day wear time, reusable form factor, 1 calibration) is still ongoing and there is not as much information on this one. Senseonics secured non-adjunctive labeling for its 90-day Eversense, with the 180-day US study (n=181) expected to finish in 1Q20. iCGM submission for 90-day Eversense is expected in 1H20, with iCGM indication for 180-day Eversense XL in “late” 2020. In the distant future, Senseonics could even pursue an Eversense-365 product, featuring a sensor life of one year and calibration just 1x/week. For now, the table below provides a valuable comparison of today’s CGMs. While it may seem trivial to some, we are glad to see the ADA investing in bringing together clinicians from across the world to learn about diabetes devices, as fast-changing landscape of CGM continues to see rapid uptake across the world.

 

Dexcom G6

Medtronic Guardian 3

Senseonics Eversense

Abbott FreeStyle Libre

Age (FDA Approved)

2+

2+

2+

2+

Surgical Implantation

No

No

Yes

No

Sensor Life

10 days

7 days

90 Days (US)

180 Days (CE)

14 days

Sensor Calibration

None

2/day

2/day

None

Non-adjunctive Use

Yes

No

Yes

Yes

Alarms

Yes

Yes

Yes

No

Ability to Share Data

Yes

Indirectly

Indirectly

Yes

 

5. Dr. Ratner Reaffirms Type 1 Diagnosis with Antibodies and Type 2 as a Beta Cell Disorder; Highlights Therapy Interventions Among Five Type 2 Phenotypes

The renowned Dr. Robert Ratner (George Washington University School of Medicine) kicked off the conference with a reaffirmation of diabetes diagnoses that reflect the underlying pathophysiology and progression of the disease instead of fixed classifications.  Much of Dr. Ratner’s talk touched on similar themes as his Keystone 2019 presentation. For example, with respect to type 1 diabetes, the number of antibodies, their persistence, and type are a strong predictor of an individual developing type 1. Data illustrates that patients that test positive for autoantibodies for six months have essentially a 100% chance of developing type 1. As these individuals can be diagnosed without assessing blood sugars, they are prime candidates for preventive care. Dr. Ratner also highlighted ADA and JDRF’s new classification system for type 1 diabetes, which underscores a three-step transition and variable phenotypes within the condition: (i) multiple islet autoantibodies with normoglycemia and no symptoms; (ii) multiple islet autoantibodies with dysglycemia and no symptoms; and (iii) symptomatic type 1 diabetes. Pivoting to type 2, Dr. Ratner mentioned that there are over 700 genes that have been implicated in its pathogenesis, but only four of them (PPARG, FTO, IRS-1, and HMGA2) are not beta cell genes. To emphasize his point, he presented an interesting case study with the Pima Indian population, a group that sees profound levels of insulin resistance and a type 2 prevalence greater than 50%. Notably, what fundamentally distinguishes this population is not the level of insulin resistance but the amount of insulin secretion. Furthermore, even in the presence of high insulin resistance, only those whose beta cells cannot compensate develop type 2. Likewise, among individuals with obesity who have insulin resistance, only those who get type 2 are unable to compensate. Ultimately, at the cellular level, this can be traced to differences in beta cell apoptosis (cell death). 

  • Dr. Ratner concluded his talk with data from the Andia Cohort, which identified a 5-group classification system for type 2 diabetes: (i) Severe Autoimmune Diabetes (LADA); (ii) Severe Insulin-Deficiency Diabetes; (iii) Severe Insulin Resistance Diabetes; (iv) Mild Obesity-Related Diabetes; and (v) Mild-Age Related Diabetes (MARD). These findings were consistent with other populations based on a separate cluster study performed in the ADOPT study. When it comes to treatment, Dr. Ratner emphasized intensive insulin treatment early on with patients having LADA and Severe Insulin-Deficiency Diabetes, interventions promoting weight loss among Severe Insulin Resistance Diabetes and Mild-Obesity Related diabetes (pharmacologics, surgery, nutritional therapy), and insulin avoidance among MARD patients.

6. Type 2 Diabetes Risk Among College Students in Egypt at 48%

Dr. El Sayed Eid (Delta University) presented data from a study published in Diabetes highlighting that a shocking 48% of college students in Egypt are at risk for developing type 2 diabetes. Dr. Eid performed a cross-sectional study (n=377 of people aged 19-21) that stratified participants according to the Finnish Diabetes Risk Score, a questionnaire geared for assessing the risk of developing type 2. Scores range from 0-20: (i) “low-risk” for <7; (ii) “slightly elevated risk” for 7-11; (iii) “moderately elevated risk” for 12-14; and (iv) “highly elevated risk” for 15-20. Interestingly, the study found that 68% of female participants shared some risk level for type 2. Rates of fast food consumption and social stress in the total cohort were 49.1% and 60.5%, respectively.

  • Dr. Eid also displayed a separate study in Diabetes assessing the prevalence of NAFLD in young Egyptians, finding that in a cohort of 120 undergraduate students, 48% had mild, moderate, or severe hepatic steatosis while 57% had either mild or moderate liver fibrosis! The authors conclude that the rising prevalence of NAFLD in Western nations parallel the obesity epidemic and that additional data is needed in Egyptian populations to understand the true public health impact.

 

 

--by Ani Gururaj, Albert Cai, Martin Kurian, and Kelly Close