Rachmiel Levine 2015 Diabetes and Obesity Symposium

March 1-4, 2015; San Diego, CA – Full Report – Draft

Executive Highlights

This report features our full coverage from the City of Hope’s 2015 Rachmiel Levine Diabetes and Obesity Symposium held in San Diego, CA from March 1-4. This year’s agenda featured a slew of valuable sessions on cell-based therapies, biomarkers, diabetes technology, and more. It brought us in touch with some of the most cutting-edge basic science research going on in diabetes, although there was no lack of commentary on clinical topics as well. Our report below includes a synthesis of some of the meeting’s overarching themes, followed by full coverage of presentations organized into two categories: (i) Diabetes Drugs; and (ii) Additional Topics.

Table of Contents 

Diabetes Drugs

Oral Presentations

Metformin Increases in Vivo 11-Beta HSD1 Activity in Obese Men With and Without Type 2 Diabetes Mellitus

Anna Anderson, MD (University of Edinburgh, UK)

Dr. Anderson presented evidence suggesting that metformin up-regulates 11-beta HSD1 activity in obese patients. This implies that metformin could potentially enhance the activity of 11-beta HSD1 inhibitors, which have been investigated by several companies as a treatment for type 2 diabetes and metabolic syndrome. Companies including Roche, Lilly, and presumably BMS have halted development of 11-beta HSD1 inhibitors in recent years, most likely due to less-than-impressive efficacy in clinical trials; Vitae Pharmaceuticals currently has a candidate, VTP-34072, in phase 2. Dr. Anderson originally hypothesized that the use of metformin as a background therapy in many of those trials may have masked the drug’s efficacy by down-regulating 11-beta HSD1; based on the results of this study, the opposite (if anything) actually appears to be the case. In response to a question about the potential appeal of a metformin/11-beta HSD1 inhibitor combination, she said that many other issues, including tissue specificity, would need to be overcome. We are eager to see whether Vitae’s candidate produces compelling efficacy results in the ongoing phase 2 trial (data expected in 1H15) – we believe the class has potential for more wide-ranging effects than many existing diabetes drugs, but we assume it would need to at least equal the A1c-lowering efficacy of other classes in order to progress.

Methylation-Specific PCR Assay for Detection of Beta Cell Death in Newly Onset Type 1 Diabetes

Mohamed Huseeiny Elsayed, MD (City of Hope, Duarte, CA)

Dr. Elsayed presented on one JDRF-funded biomarker project: a new methylation-specific PCR (MSP) assay that detects the plasma levels of beta cell DNA following their death during the onset of type 1 diabetes. A study in ten newly diagnosed type 1 diabetes patients found that the assay predicted subsequent drops in C-peptide secretion – in our view the data looked pretty good, although we would be interested to see if it could pick up beta cell death even before diagnosis. Interestingly, in some patients the decline in C-peptide was fairly continuous, while in others it was more episodic (tying into the idea of the onset of type 1 diabetes as a potentially relapsing-remitting pattern).

Rachmiel Levine Award Lecture

Evolving Concepts of Incretin Action Inform the Treatment of Metabolic Disorders

Daniel Drucker, MD (Lunenfeld Tanenbaum Research Institute, Toronto, Canada)

In his acceptance speech for the Rachmiel Levine Award, Dr. Daniel Drucker (Lunenfeld-Tanenbaum Research Institute, Toronto, Canada) covered a mix of emerging science and more clinically current points related to incretin-based therapies. Although Dr. Drucker was characteristically optimistic about the future for GLP-1 agonists, which he suggested are at the “end of the beginning” in terms of maturity, he shared brand-new preclinical findings that could raise some concerns down the road. Specifically, preclinical data from Dr. Drucker’s lab published just today in Cell Metabolism shows that GLP-1 agonists promote growth of both the large and small intestine, which could conceivably increase colorectal cancer risk. Dr. Drucker framed the data in a balanced manner. On the worrying side, he noted that although bariatric surgery (which generally increases GLP-1 secretion) reduces the incidence of most cancers, the one outlier with increased incidence in the Swedish Obese Subjects Study was colorectal cancer. However, he was clear that preclinical data could only be seen as hypothesis generating. Even if GLP-1 agonists do stimulate bowel growth in a way that could increase cancer risk, that would need to be weighed against the beneficial effects of weight loss on cancer. Ultimately, large outcomes studies or maybe even larger registry studies may be the only way to know for sure given the low incidence of colorectal cancer. In the meantime, we will be interested to hear how scientists and clinicians perceive these data. 

  • Dr. Drucker’s data on gut growth with GLP-1 agonists were part of a broader discussion of the class’ non-glycemic effects, which are only now beginning to be better understood. We summarize Dr. Drucker’s points in a number of other areas below.
    • Inflammation and innate immunity: Turning again to truly cutting-edge work, Dr. Drucker pointed to emerging preclinical data that GLP-1 may have a substantial impact on innate immune signaling via a population of gut immune cells called intestinal intraepithelial lymphocytes (IELs). This signaling may play a role in mediating inflammation in the GI tract, as mice with a knockout of the GLP-1 receptor in IELs were more susceptible to a model of colitis. These knockouts also appeared to have significant perturbations in their gut microbiome. This work is very early stage, but Dr. Drucker suggested that there could be broad implications on the regulation of immune system function and inflammation.
    • Cardiovascular health: There was no single takeaway message in Dr. Drucker’s presentation about GLP-1 agonists overall cardiovascular impact – the data has simply not yet been conclusive. Although Dr. Drucker used the term “cardioprotective” at multiple points to refer to data showing a beneficial effect of GLP-1 or GLP-1 agonists in acute care settings (such as post-MI), he is looking towards ongoing CVOTs to integrate the benefits vs. the possible negative impact of the increase in heart rate.
    • Other areas: In a rapid-fire slide, Dr. Drucker listed the broader range of applications in which GLP-1 agonists are being used or investigated, including type 1 diabetes, obesity, prediabetes, neurodegeneration, anti-inflammation, and fatty liver disease.
  • Finally, Dr. Drucker briefly touched upon a few pragmatic points regarding incretin-based therapies on the market. He was extremely enthusiastic about the combination of GLP-1 agonists with basal insulin, speculating that in ten to fifteen years, out of the type 2 diabetes patient population needing injectables, the vast majority will be getting the GLP-1 agonist/basal insulin combination; he stated it is the most effective way to treat type 2 diabetes, “bar none.” He presented a more nuanced view on GLP-1 agonists as single agents, suggesting that the immense heterogeneity between agents on the market in terms of molecular structure, efficacy, tolerability, PK/PD, weight loss, and the penetration of different body compartments, have been delaying the class’ penetration. He argued that more work needs to be done to educate providers on which drugs to use in which patients, although he thinks the delay in the class’ penetration will be alleviated by the new longer-acting agents – though he didn’t list any specific products, we see agents like Lilly’s Trulicity (dulaglutide) as an example of a relatively patient-friendly and provider-friendly product.  
  • Turning to DPP-4 inhibitors, Dr. Drucker suggested that the worry about DPP-4’s effects on substrates other than GLP-1 is possibly overblown. Although in vitro studies have found a large number of pharmacologic substrates for the DPP-4 enzyme, a much smaller number are true physiological clinically relevant substrates. Regarding emerging questions about a DPP-4 inhibitor class effect on heart failure, Dr. Drucker simply wished anyone who wanted to investigate the mechanism “good luck” – in his view, the incidence of heart failure is too low and impacted by too many variables to conclusively tie any one mechanism to the effect seen, for example, in SAVOR.

Type 1 Diabetes Biomarkers

Validation Efforts With the JDRF Biomarker Working Group

Jared Odegard, PhD (Benaroya Research Institute, Seattle, WA)

Dr. Odegard provided an overview of efforts by the JDRF Biomarker Working Group to identify biomarkers that predict the rate of progression of recent onset type 1 diabetes. The broad goals of the working group are to validate the technical quality of new biomarker assays and eventually to create clinically meaningful predictive models, most likely involving multiple markers. The group’s first project is focused on identifying biomarkers that predict the rate of C-peptide decline in patients with newly diagnosed type 1 diabetes. The process consists of (i) tests to evaluate an assay’s precision; (ii) a cross-sectional study to characterize the sort of clinical data produced; and (iii) a longitudinal study (two years; n=40) to evaluate the biomarkers’ predictive value. The ultimate goal is to produce a set of validated, FDA-endorsed tools that investigators can use to stratify patients and conduct more targeted, efficient clinical trials. Currently, four assays have advanced to the second stage and seven are undergoing precision testing. More streamlined clinical trial paradigms and better predictive models would both represent enormous advancements toward disease-modifying therapies for type 1 diabetes. As a reminder, JDRF and Janssen recently announced the formation of a partnership focused on even earlier, asymptomatic stages of the disease; specific research projects will hopefully be announced later this year.   


Virus? What Virus?

A debate on whether viruses can act as a trigger for type 1 diabetes illustrated the lack of consensus in the field around the early pathogenesis of the disease. Dr. Bart Roep (Leiden University Medical Center, Leiden, The Netherlands), despite co-authoring a 2007 study that considered enteroviruses as a potential trigger, argued that the many experiments conducted since then have been able to find only “the shadow of a virus, if not the shadow of something different.” For his part, Dr. Matthias von Herrath (La Jolla Institute for Allergy and Immunology, La Jolla, CA) argued that while there is no conclusive evidence supporting a virus as the trigger, there is also currently no other plausible mechanism to explain the relapsing-remitting nature of the early asymptomatic stages of type 1 diabetes. He believes the search for a contributing virus has not yet run its course and even suggested that a trial of antiviral therapy in at-risk patients could be a potentially intriguing study. In his rebuttal, Dr. Roep took issue with the characterization of type 1 diabetes as a relapsing-remitting disease, arguing that the available evidence suggests a more chronic progressive pattern – he noted that conference organizers may have found their debate topic for next year. As Dr. Carla Greenbaum (Benaroya Research Institute, Seattle, WA) pointed out, continued results from the TEDDY study should offer more insight into this area, though results reported in 2013 found no link between viral infections and rapid-onset type 1 diabetes in young children. The longer the search for a trigger remains elusive, the more likely it seems that the pathophysiology of type 1 diabetes is rooted in a complex mix of susceptibilities and environmental influences rather than one or two specific triggers, but such thoughts remain highly speculative while the search goes on.

Plenary Lecture

30 Year History of the DCCT: Patient Pursuit of a Chronic Disease

David Nathan, MD (Massachusetts General Hospital, Boston, MA)

Dr. David Nathan delivered a plenary lecture celebrating the accomplishments of DCCT/EDIC, which crossed the 30-year mark since patient enrollment began with a resounding 95% retention rate. Among the many points that Dr. Nathan emphasized was how unique that level of retention was for any clinical trial, let alone one that has been going on for decades. This retention rate, he noted, bolsters the confidence clinicians and scientists have in the study’s results. Covering key takeaways from the trial, Dr. Nathan recommended that glucose lowering therapy should be implemented early to get the most back for the buck – a notion that the designers of today’s CVOTs (which have generally been enrolling patients with more advanced diabetes) could well take heed of. Dr. Nathan noted that the DCCT/EDIC investigators are always on the lookout for new collaborators; he positioned this by noting that hundreds of millions of taxpayer dollars have gone into the trial and so the huge bank of data should be treated as a shared resource for the broader scientific community. He ended his presentation on a forward-looking note, highlighting the Bionic Pancreas project as one of multiple possible closed loop projects that could provide safer, easier, and better glucose control for many types of patients, including children. 

  • We covered the long-awaited DCCT/EDIC mortality analysis back in January – the data showed a modest but statistically significant reduction in mortality with early intensive glycemic control after 27 years of follow-up (HR = 0.67; 95% CI = 0.46-0.99; p=0.045). We got a peek at the cardiovascular risk analysis at 25 year mark following the end of randomized treatment at an NIH workshop in October – there was a continued (though slightly more modest) legacy effect on a cardiovascular disease composite.

Outcomes from the CIT Trial

Ali Naji, MD, PhD (University of Pennsylvania, Philadelphia, PA)

Dr. Naji provided a peek at results from the CIT-07 islet transplantation trial – publication of the full results is expected in mid-2015. For background, the trial is being conducted by the NIH’s Clinical Islet Transplantation Consortium (CIT), which ultimately aims to obtain licensure from the FDA for an islet product. CIT-07 was a prospective, single-arm phase 3 study that enrolled 48 patients with type 1 diabetes who were followed for 24 months post-transplant; many of the patients are also currently enrolled in the CIT-08 long-term follow-up study. CIT-07’s primary endpoint was the percentage of patients with an A1c <7% at one year and the percentage free from severe hypoglycemia from day 28-365 post-transplant. Dr. Naji did not share any of the primary results but did present data showing significant reductions in insulin dependence, hypoglycemia, and glycemic lability one year post-transplant. Approximately 60% of participants achieved complete insulin independence after one year, though the proportion had dropped to ~30% after two years. We look forward to seeing the full results later this year, though unfortunately the shortage of suitable human islet cells and the need for long-term aggressive immunosuppressants will likely limit the extent to which islet transplantation can become a standard therapeutic option for type 1 diabetes.

Metabolism and Inflammation

Targeting Neuroinflammation for Diabetes Treatment

Matthias Tschöp, MD (Helmholtz Diabetes Center, Munich, Germany)

Dr. Tschöp discussed the potential of GLP-1-based co-agonists. He opened his presentation labeling obesity as a “brain disease,” as he highlighted that the interactions of neurons, astrocytes, and microglia are more relevant to the condition that previously thought. Dr. Tschöp then noted that the effects of bariatric surgery on diabetes are likely due to gut-brain signaling rather than the surgery’s mechanical intervention; therefore, it would be optimal to simulate such mechanisms through pharmacotherapies rather than have patients resort to the high-risk procedures of surgery. In his eyes, such treatments can be found in GLP-1-based co-agonists, based off of his work in collaboration with Dr. Daniel Drucker and Dr. Richard DiMarchi (Indiana University, Bloomington, IN). Specifically, Dr. Tschöp discussed GLP-1/GIP/glucagon triagonists’ benefits on glucose control, body weight, and more in mouse models – please see our coverage of this research recently published in Nature as well as our interview with Dr. DiMarchi where he characterized triagonists as “spectacular molecules.” In addition, Dr. Tschöp discussed the novel strategy of using GLP-1 or glucagon as peptide carriers that can target attached steroids or small molecule drugs to specific tissues such as the pancreas or liver, minimizing off-target effects.

Epigenetics of Diabetes

Panel Discussion

Rama Natarajan, PhD (Beckman Research Institute of City of Hope, Duarte, CA), Evan Rosen (Beth Israel Deaconess Medical Center, Boston, MA), and Maike Sander, MD (University of California San Diego, La Jolla, CA)

This session on the up-and-coming field of epigenetics offered intriguing insights into several aspects of diabetes pathophysiology, including the mysterious phenomenon of metabolic memory. Dr. Rama Natarajan presented data suggesting that the diabetic/obese state increases pathological gene expression, particularly of genes linked to inflammation, by altering the balance of histone modifications (namely, increasing the number of markers for activation and decreasing markers for repression). Excitingly, Dr. Natarajan has conducted studies using cells from DCCT/EDIC participants that demonstrate variations between the intensive and conventional groups in histone acetylation near genes linked to inflammation and diabetes complications; these patterns were significantly correlated with A1c history. While further research is needed to fully characterize these findings – a larger (n=500) study is underway looking at DNA methylation patterns – the results suggest that the reduced long-term risk of complications in the intensive group (the “legacy effect” or “metabolic memory”) may be explained at least in part by epigenetic differences. The session also included data presented by Dr. Rosen suggesting that epigenetic mechanisms may offer a clue to a “common denominator” underlying insulin resistance. We were also treated to insights from Dr. Sander into the role of “pioneer transcription factors” in islet cell differentiation.

Keynote Lecture

Type 1 Diabetes: Adults vs. Kids

Carla Greenbaum, MD (Benaroya Research Institute, Seattle, WA)

In this lecture on the differences between adult and pediatric type 1 diabetes, Dr. Carla Greenbaum suggested that members of industry may occasionally overestimate the difficulty of enrolling pediatric patients in clinical trials. Much of Dr. Greenbaum’s talk was a recap of the recent Consensus Conference on pathways for developing disease-modifying therapies for children with type 1 diabetes, which concluded that the current approach of simply extrapolating clinical trial results from adults to children is likely not appropriate given the fundamental differences in disease pathophysiology between the two groups. During the talk, Dr. Greenbaum noted that the high enrollment rates in pediatric clinical trials was one of the main points of discussion with industry representatives at the conference, many of whom had assumed that enrollment would constitute a major barrier to conducting studies in children. On the contrary, Dr. Greenbaum pointed to several examples (such as a few TrialNet studies) indicating that in type 1 diabetes, enrollment can occur quite quickly in pediatric trials. While there are many unanswered questions regarding the best regulatory pathways for novel therapies targeted toward children, these sorts of misunderstandings about the feasibility of clinical trials should certainly be easier to address than most. We also believe this is an excellent illustration of the value of convening multiple stakeholders (the Consensus Conference was attended by approximately 100 people from the research, industry, nonprofit, and regulatory sectors) to discuss such complex topics.

Additional Topics

Novel Approaches to Understanding and Treating Hypoglycemia

Panel discussion

Robert Vigersky, MD (Walter Reed National Military Medical Center, Washington, DC), Rory McCrimmon, MD (University of Dundee, Dundee, UK), Simon Fisher, MD, PhD (University of Utah, Salt Lake City, UT), and Simon Heller, MD (University of Sheffield, Sheffield, UK)

Turning to football for an analogy, Dr. Vigersky discussed his recently published approach for a composite measures (such as those used to rate quarterbacks) to evaluate diabetes therapies. He argued that relying solely on A1c “insufficiently captures the quality of glycemic control;” in particular, he believes that a therapy’s impact on hypoglycemia must not be overlooked. To illustrate the point, he presented data demonstrating unacceptably high rates of severe hypoglycemia in type 1 and type 2 diabetes along with an analysis estimating the costs of severe hypoglycemia at $24.5 billion per year by 2020. He offered one potential composite approach called the hypoglycemia-A1c score (HAS), outlined in a very recent opinion piece he authored in the Journal of Diabetes Science and Technology, that would score therapies according to A1c reductions, hypoglycemia (events per year or change from baseline), and weight. Clinicians already use this sort of thought process when selecting therapies for patients. Logically, Dr. Vigersky believes it should be codified into a system that can (and should) be more accepted by regulators and payers. We strongly agree that the field’s current A1c-centric view often does not fully capture the relative effectiveness of various therapies and that the use of such composite endpoints would likely paint a more complete picture; this is also why we have frequently argued for greater use of time in range as a standard endpoint in clinical trials.

  • It was clear during Q&A that the interplay between A1c and hypoglycemia remains the subject of intense debate – one attendee went so far as to state that “the right conclusion [from ACCORD] was that if you treat middle-aged people with diabetes with insulin and sulfonylureas, you kill them.” See the appendix for a transcription of the relevant discussion.

Preventing Hypoglycemia-induced Morbidity and Mortality

Simon Fisher, MD (University of Utah, Salt Lake City, UT)

Dr. Fisher reviewed preclinical data suggesting that recurrent moderate hypoglycemia might have a positive preconditioning effect. Of course, nobody at the meeting would argue that hypoglycemia is a good thing, plain and simple. Rather, Dr. Fisher described a recurrent hypoglycemia “paradox” in which certain adaptive effects such as a degree of protection against neurodegeneration, seizure, and death are weighed against the well-known maladaptive effects such as hypoglycemia unawareness and impaired counter-regulation. The rodent data (with all the caveats of preclinical data) showed that recurrent moderate hypoglycemia reduced cognitive decline and mortality following episodes of very severe hypoglycemia. We do not see this theory being tested in humans any time soon for obvious ethical reasons, but the work could still provide valuable insight on the factors that underlie currently mysterious phenomena such as hypoglycemia unawareness. Additionally, we were intrigued by the finding that beta-blockers appeared to drastically reduce mortality in rats following insulin-induced severe hypoglycemia.

Panel Discussion

Dr. Carla Greenbaum (Benaroya Research Institute, Seattle, WA): I like the idea of a composite score because it’s something we wrestle with. A1c was agreed upon because it was associated with mortality. The concern with regulatory authorities is that everything’s a correlate of a correlate of a correlate. Do you have any insights into how they would look at this kind of a score?

Dr. Robert Vigersky (Walter Reed National Military Medical Center, Washington, DC): I have no insights; I do have theories. I think they would be very receptive. They’ve been hearing from all of us that A1c is insufficient and that the reduction in hypoglycemia could be as or more important given certain circumstances. They’re receptive, but first of all, we need to challenge all of us to do the research, to look at outcomes and relate outcomes to these composite scores. Then the FDA might be much more convinced. However we do this, we need them to be active participants in setting up metrics and what we’re looking for to have the score be meaningful.

Dr. Simon Fisher (University of Utah, Salt Lake City, UT): The problem is we’re going to get scores, and then the insurance companies will say this drug had a better score, so you have to use this first, why didn’t you do this first, etc. Sometimes we’re limited by cost, and that should be part of the equation. Another part of it is resources. The DAFNE study is great, but not everyone can do that in practice, so they’re more likely to give a cheaper drug. I worry about having the government step in and telling us what to do.

Dr. Vigersky: That’s a good point. That’s already being done by the payers. We’re already stuck with it, but they’re only using part of the equation. This is an effort to help them make decisions on a broader base. I agree with all those pitfalls, but we’re in a world where we’re judged by outcomes, and so interventions should be judged by outcomes, and composite metrics will help us show whether a particular intervention is more beneficial. It won’t make things worse and it will hopefully make things better.

Q: Dr. Fisher said that study after study showed if you subject a population to tight control, you get worse mortality, and that ACCORD was the most recent study. I would quibble with ACCORD because the investigators didn’t analyze for how they achieved tight control. There are only two classes out of ten that we use that cause hypoglycemia – insulin and sulfonylureas – but they didn’t bother to analyze that. I think the right conclusion is that if you treat middle-aged people with diabetes with insulin and sulfonylureas, you kill them, not that tight control kills them. Every other branch of medicine restores physiology to normal. The target A1c of 7% came out of the DCCT where the curves crossed and it showed that once you got to 7% there was more hypoglycemia, so that was pulled out of a hat. Right now there’s a swing in opinion saying don’t try tight control because you’ll kill people with hypoglycemia, and that’s having a bad effect. Those studies should be reanalyzed, at least ACCORD for what it really showed.

Dr. Simon Heller (University of Sheffield, Sheffield, UK): I’m not arguing that you shouldn’t have tight control. If you speak to the ACCORD investigators, which I have done, they would say that they have looked at medications and have not been able to see an influence. Whether it’s the specific analysis you asked for, I don’t know. No one is saying tight control shouldn’t matter. Hypoglycemia is bad, and if you have a medication that causes it in people who don’t warrant tight control, treatment can be individualized.

Dr. Fisher: The take home message is that drugs that cause hypoglycemia, specifically insulin and sulfonylureas, can cause adverse problems. With newer technology, CGM, newer insulins, and drugs for type 2 diabetes that don’t cause hypoglycemia, those studies will be very interesting to see if we can achieve tight control without hypoglycemia.

Dr. Vigersky: What are the best targets for an intervention to prevent hypoglycemia? Do we know enough to pick something protective?

Dr. Rory McCrimmon (University of Dundee, Dundee, UK): I would suggest structured education, maximize therapy with insulin analogs or pumps, and do frequent monitoring with or without CGM. Hypo COMPASS said it didn’t matter what you used as long as there was structured education. It’s habituation, meaning there should be elements of cognitive behavioral therapy. Down the line, if you look at the adaptations in the brain, it’s better able to use ketones. You could use a ketogenic diet – it sounds bizarre, but the brain is habituated, so it might do better. Diazoxide would be short term while they got structured education.

Dr. Fisher: Technology will be the biggest help in sorting things out. The biggest thing we’ve done in the last hundred years of using insulin is simple sugar checking. That’s the biggest intervention; people are aware of what their sugar is. We have new devices, dual-chambered pumps, novel insulins that are faster acting or quicker onset or longer lasting, insulin analogs, pumps, CGM. We have different drugs that don’t cause hypoglycemia, that are glucose-sensing. Technology is the answer, and it will unfold.

Dr. Heller: Teaching patients to use insulin safely is most important. It’s horrific what a terrible job we do. We’re giving people a potentially lethal drug and not teaching them how to use it. If you’re talking about closed loop, you can use it safely; you can check on people about whether they’re going hypoglycemic overnight and adjust their therapy. We should abolish nocturnal hypoglycemia in the next few years.

Diabetes Technology Updates

Long-term Implanted Glucose Sensors

David Gough (University of California San Diego, La Jolla, CA

In the conference’s primary tech-related session, glucose sensor expert Dr. Gough discussed GlySens’ long-term glucose sensor project. For background, Dr. Gough is one of GlySens’ co-founders (see our most recent coverage of Dr. Gough for more background). GlySens’ long-term fully implanted glucose sensor uses differential oxygen detection at a pair of electrodes to achieve preserve long-term signal stability with minimal need for calibration – in the most recently completed human study the average calibration was -1.7% on average each month. Dr. Gough showed that signals from the implanted sensor corresponded closely with YSI. Dr. Gough ended by reviewing the advantages of GlySens’ sensor: it does not appear to require frequent calibration, it uses a solid pore-free membrane, and the differential oxygen sensing accounts for variable oxygen, perfusion, or foreign body response. Based on the results of small ongoing safety studies, GlySens could be poised to expand into larger clinical trials in the not-too-distant future.

Panel Discussion

Michael Weiss, MD, PhD (Case Western Reserve University, Cleveland, OH), Natalie Wisniewski, PhD (PROFUSA, Inc., San Francisco, CA), Jeffrey Joseph, DO (Thomas Jefferson University, Philadelphia, PA), and David Gough, PhD (UCSD, La Jolla, CA)

During the panel discussion, Dr. Joseph commented frankly on the difficulty in drawing investors to novel glucose sensor projects. “Right now,” he stated, “ what’s hot is apps and IT, which can get to market quickly and provide a high return on investment.” By comparison, long-term implantable sensor and other daring ventures in physical diabetes technology involve technical risks, possible conservatism from the FDA and other regulatory agencies, and uncertainty regarding reimbursement, making it hard to get projects funded. Current efforts are largely getting by with the support of the major foundations, but Dr. Joseph noted that investors of some sort will need to get in the game when it comes time to conduct pivotal clinical trials that will require many tens of millions of dollars. Also during the panel, Dr. Gough suggested that it may still be quite a while before truly safe and reliable pump-based closed loop systems can be developed, at least to the point where patients can completely give up control to the system. His vision for a artificial-pancreas-like system that would represent a significant advance for patients while being more attainable in a reasonable time frame involves a very reliable implanted glucose sensor and something along the lines of a smart insulin pen that would guide patients on insulin dosing. These words aligned with thoughts we heard at ATTD during a workshop on the psychosocial aspects of the artificial pancreas, emphasizing the need to create systems that respond to real patient needs rather than going directly to the closed loop’s most fully automated end point and completely wresting control from the patient.

Personalized Medicine in Future Diabetes Care

Personalized Medicine: The Future is Now

Erwin Bottinger, MD (Mount Sinai Icahn School of Medicine, New York, NY)

Dr. Bottinger discussed the opportunities and practical challenges of applying personalized medicine to diabetes care – this topic got a recent boost in relevance when President Obama mentioned diabetes in the context of precision medicine during his State of the Union Address. He highlighted Mount Sinai’s BioMe Biobank program as an example of how large patient registries can be used to facilitate the creation of individualized risk profiles for patients. The program currently involves over 32,000 patients recruited from primary care offices (the goal is 100,000) and has already produced several published papers, including a recent Nature Communications paper identifying rare genetic variants associated with type 2 diabetes susceptibility. Another paper on defining distinct clinical subtypes of type 2 diabetes is currently under review – we look forward to seeing these results, as this could potentially be a very helpful way to target future therapies to the subsets of patients most likely to experience a benefit. Practically speaking, Dr. Bottinger noted that it is unrealistic to expect to “throw genetic markers and biomarkers at every case of diabetes in a busy primary care practice.” Rather, he believes personalized medicine will be most useful as a tool to stratify patients and alert providers to those at highest risk for a given complication like diabetic nephropathy (the focus of an ongoing Biobank-supported study). He also noted that one of the key challenges will be transmitting information to clinicians in a useful format; he highlighted several efforts (including Mount Sinai’s CLIPMERGE program) that would address this by integrating information into existing electronic medical record (EMR) systems. Particularly given the widespread perception of EMRs as a frustrating bureaucratic hurdle rather than a means to streamline care (so far), we agree that effective implementation of such communication tools will be an incredibly important component of the integration of personalized medicine into clinical practice.

Using Public Big-Data to Discover New Therapeutic and Genetic Insights in Diabetes

Atul Butte, MD, PhD (UCSF, San Francisco, CA)

Dr. Butte pushed for the use of public big-data, emphasizing that finding research commonalities in such data can be a great way to find the “low hanging fruit” in therapeutic targets in diabetes. Specifically, he pointed to the example in which the integration of 130 open-access gene expression experiments yielded the receptor CD44, which has proven to be a top target in diabetes as its inhibition can improve insulin sensitivity, decrease hepatic steatosis, and slow weight gain.

Obesity and Inflammation

Obesity and Immune System

Matthias Bluher, MD (University Hospital Leipzig, Leipzig, Germany)

In this session, Dr. Bluher called for the targeting of adipose tissue in the treatment of obesity-related diseases. Dr. Bluher illustrated that the immune system becomes activated in obesity as cytokines, macrophages, and other immune cells make their way into adipose tissue. The following dysfunction of adipocytes can then lead to inflammation, insulin resistance, and other obesity-related comorbidities. Dr. Bluher argued that it is the composition of adipose tissue rather than its mass that has the greater impact on metabolic health for individuals with obesity – we wonder if tools to measure adipose tissue composition could represent the best way to identify the “healthy obese.” As few anti-obesity strategies focus on adipose tissue – which until recently was viewed as a passive energy storage depot rather than a metabolically active organ – he emphasized a need to identify novel drug targets within this area.


--by Melissa An, Emily Regier, Manu Venkat, and Kelly Close