May 14-18, 2014; Las Vegas, NV; Day #3 Highlights – Draft

Executive Highlights

We’re back with Day #3 of AACE 2014, taking place in always-busy Las Vegas, Nevada. Here we outline our top ten highlights of the day, followed by detailed reports.

1. Dr. W. Timothy Garvey (University of Alabama at Birmingham, AL) presented a new “Advanced Framework” for the diagnosis and treatment of obesity, which includes staging based on both BMI and complications.

2. We learned that the AACE/ACE task force is sharing the draft framework with the stakeholder pillar for input; the task force targets publication of a final framework in late summer in Endocrine Practice, followed by a second implementation-focused consensus conference in December.

3. Dr. Eric Finkelstein, PhD (Duke-NUS Graduate School, Singapore) encouraged industry and obesity advocates to measure the cost-effectiveness of their treatment strategies. We anticipate more and more movement toward cost-effectiveness measurement.

4. The session on AACE/ACE’s new obesity framework concluded with a panel discussion that solicited input from audience members on questions such as whether we need an alternative medical term to refer to “obesity” (to reduce stigma), what improvements can be made in the diagnosis of obesity, and how (and whether we should) discriminate between classes or stages of obesity.

5. Dr. Scott Lee (Medtronic Diabetes, Northridge, CA) shared new data from a 48-patient, 15-day home study of the MiniMed Duo, a three-day wear combined insulin infusion-CGM sensor set. Overall MARD was 15.5% vs. fingersticks.

6. AACE released a consensus statement outlining recommended clinical approaches to pump therapy for patients with type 1 and type 2 diabetes – details below, and see Saturday’s press release here and the full statement here.

7. Dr. Xavier Pi-Sunyer MD, PhD (Columbia University, NYC, NY) presented full results from Novo Nordisk’s phase 3 study of liraglutide 3.0 mg in people who are obese or overweight, with or without prediabetes (SCALE Obesity and Prediabetes). We had a fantastic chance to sit down with Dr. PI-Sunyer – we left the interview knowing that he had very high hopes for lira for obesity – this is very telling and says a lot, given how renowned a researcher Dr. Pi-Sunyer is.

8. Dr. Nitesh Kuhadiya (University of Buffalo, NY) presented data from the first prospective, randomized, double blind, placebo controlled clinical trial of liraglutide in type 1 diabetes. The 12-week, 72-patient study demonstrated a reduction in A1c (0.4-0.7%) with ~19% less insulin and ~5% body weight loss. We found “time in zone” results confusing, however, and are following up.

9. A new meta-analysis of 20 phase 2b and 3b clinical trials (excluding SAVOR) on AZ’s Onglyza (saxagliptin) showed no trend towards CV harm for MACE, its components, or (non-adjudicated) heart failure.

10. Libertarian economist Dr. Jayson Lusk (Oklahoma State University, Stillwater, OK) argued against many common proposals for government intervention in the obesity epidemic.

Detailed Discussion and Commentary

The AACE/ACE Consensus Conference on Obesity: A New Medically-Actionable Diagnosis of Obesity

An Advanced Framework for a New Diagnosis of Obesity as a Chronic Disease

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

In the moment many had been waiting for at AACE, Dr. Timothy Garvey presented the outcome of the Consensus Conference and the writing committee’s hard work since: a new staged “advanced framework” for the diagnosis of obesity (read the AACE press release). The diagnosis of obesity under the new system (see the table below) depends on both an anthropometric component (BMI) and a clinical component (complications staging). Perhaps the most notable aspect of the framework is an “obesity stage 0,” which covers patients with a BMI above 30 kg/m² but that do not have obesity-related complications, for which lifestyle and dietary changes are the primary recommendations. Interestingly, going purely by this approach, an individual with a BMI of 45 kg/m² but no complications might not be recommended for weight loss medication. Obesity stages 1 and 2 apply to any patient with a BMI over 25 kg/m² with mild/moderate or severe obesity-related complications, respectively. Treatment for stage 1 obesity would involve lifestyle and possibly weight loss medication, while treatment for stage 2 would include lifestyle, weight loss medication, and possibly bariatric surgery. Dr. Garvey characterized the new framework as a medical model for obesity that focuses on complications for both diagnosis and treatment. He shared that the AACE Board of Directors had unanimously approved the framework earlier in the week – it was to be sent out to the stakeholder pillars for comment immediately following the conference, and the writing committee hopes to publish the document in Endocrine Practice later in the summer. We learned later that AACE/ACE hope to hold a second Consensus Conference, focusing on the implementation of the framework, in December.  

Table 1: AACE/ACE Advanced Framework for the diagnosis of obesity

AACE/ACE Framework: Key diagnosis and treatment steps

• Before diving into the specifics of the framework, Dr. Garvey covered a number of the writing committee’s key objectives and assumptions. The overall objective was to create an actionable and medically meaningful diagnosis of obesity, which added value, utility, and functionality for all stakeholders (we had head previously that the lack of a single strong definition of obesity was a major barrier to organizing concerted action). The writing committee believed that primary, secondary, and tertiary prevention (the prevention of obesity, the prevention of obesity-related complications, and the treatment of those complications, respectively) all had a role to play in the framework.
• From the presentation, it became increasingly clear that this framework is largely directed towards primary care physicians. Dr. Garvey emphasized that the framework is largely a reflection of what obesity medicine specialists and endocrinologists already do, and that it will have value in entraining primary care physicians to provide higher-quality obesity care, especially for those who do not have much experience or training in obesity management medicine. Given the appropriately recognized emergent concept that current medical education in obesity therapy is insufficient, we would agree on this point, although we hope that AACE/ACE is able to overcome the BMI-centricity in many sectors of the medical world.
• A four-step treatment process accompanies the new framework: 1) Providers will first screen patients using BMI and (when appropriate) waist circumference; 2) Providers will evaluate the presence or absence of obesity-related complications (cardiometabolic, biomechanical, and others); 3) Providers will stage the severity of any complications that exist; and 4) Providers will recommend treatment based on clinical judgment.
  
  • During Q&A, many audience members asked for further clarity on step 3 (staging complications), as they were unsure exactly how to stage many complications. The panel shared that AACE/ACE is consulting with relevant professional organizations to draw up specific criteria for the staging of individual complications. Many of the specific criteria for complications staging are already included in the AACE/ACE draft framework.
• The AACE/ACE framework differs in some key ways from the existing BMI-based definitions. With the new framework, a patient with a BMI of 25 kg/m² (or an Asian patient with a BMI as low as 23 kg/m²) with a severe weight-related complication would be categorized as stage 2 obese, and would be recommended for weight loss medicine. Going by the framework alone, such patients might also be candidates for bariatric surgery, although the AACE/ACE algorithm and provider discretion would probably rule out surgery if the patient’s BMI was that low. Conversely, a patient with a much higher BMI would be categorized as obese stage 0 if they did not have any weight-related complications, and (according to the framework) would not be recommended for interventions beyond lifestyle and diet changes. Dr. Garvey did make it clear that the framework should be seen as a recommendation and not a rule, and that individualization of therapy and clinician judgment are important factors to keep in mind.  
  
  • Notably, metabolic syndrome, type 2 diabetes, and prediabetes are considered sufficient to move someone into the stage 2 obesity category if their BMI is above 25. We were surprised to see this, as this family of conditions will vastly expand the number of patients who are included in the stage 2 group. We cannot help but wonder if an individual with a BMI of 25 with prediabetes should be included in the same stage as an individual with a BMI of 45 with end-stage organ disease from obesity.
• The new framework is intended to complement the existing AACE/ACE algorithm, rather than to replace it. The document expresses support for existing complications-centric algorithms such as the AACE/ACE algorithm (specifically mentioned). It will be interesting to see which of the two documents will take precedence when elements conflict – for example, the AACE/ACE 2013 algorithm sets a BMI of 35 as a threshold to consider bariatric surgery, while the advanced framework seems to suggest a more complications-based threshold rather than a BMI-based threshold. We imagine these details will be ironed out in the coming months, and at the next consensus conference.
  
  • It will be interesting to see how the final published framework will bear similarity to existing staging systems for obesity, such as the Edmonton staging system. During the panel discussion, Dr. Arya Sharma (University of Alberta, Edmonton, Canada), an architect of the Edmonton system, suggested that three stages may not provide enough segmentation for the range of severity of obesity and its complications, and suggested that the system should have perhaps four or five stages – we found Dr. Garber’s response to this suggestion slightly dismissive (see the panel discussion section below).

Summary of the Consensus Conference

Daniel Einhorn, MD (UC San Diego, CA)

ACE President Dr. Daniel Einhorn provided background on the proceedings of the AACE/ACE Consensus Conference on Obesity to the audience, the vast majority of which was not present at that meeting (we, however, were – read our full coverage). The meeting organizers felt strongly that a comprehensive solution to the obesity epidemic would require the engagement and participation of a wide range of groups, and so they brought together a set of four stakeholder “pillars” during the meeting: i) Biomedical; ii) Government & Regulatory; iii) Health Industry & Economics; and iv) Organizations, Education, and Research. These four pillars, first individually but then together, were asked to deliberate over the following five key questions:

• What is obesity?
• What options are available for obesity management?
• What is the optimal use of therapeutic modalities?
• Can the optimal framework be cost-effective?
• What are the knowledge gaps and how can they be filled?

The resulting discussion was used to generate a series of affirmed and emergent concepts, which Dr. Jeffrey Mechanick presented next.

Affirmed and Emergent Concepts

Jeffrey Mechanick, MD (AACE, Jacksonville, FL)

AACE President Dr. Jeffrey Mechanick provided an overview of the affirmed and emergent concepts that came out of the AACE/ACE Consensus Conference, including a few new concepts that were not mentioned in the press release that had followed the conference. He described how the writing committee came up with the lists of concepts, which involved a 5x4 matrix that plotted the four pillars against the five discussion questions. Ideas that were broadly agreed upon and were based on an existing concept were added to the list of affirmed concepts. Ideas that emerged only when the four pillars came together on the second day of the meeting were included as emergent concepts. Dr. Mechanick went through each of the affirmed and emergent concepts, providing some context and potential future directions for many of them. As an overarching idea, Dr. Mechanick noted that obesity is a complex disease, one that must be addressed with complexity. He also suggested that AACE took a risk in pursuing a multi-stakeholder conference in the fashion that it did. The effort was well worth it – the final list of concepts was remarkably multidimensional, and the three new emergent concepts added since the conference (covering topics such as medical training and obesity in the elderly) enhanced that depth. 

• The AACE/ACE writing committee summarized the key findings of the Consensus Conference as follows: i) Obesity is a chronic disease, and once diagnosed, it should be managed using a complications-centric approach such as the AACE/ACE algorithm; ii) A preventative medicine paradigm consisting of structured lifestyle interventions, behavior changes, and alteration in the built environment is necessary to improve outcomes in overweight/obesity; and iii) Comprehensive overweight/obesity interventions producing improved outcomes require demonstration of value in a combined biomedical and public health model. Below, we list the five affirmed concepts (ACs) and eight emergent concepts (ECs), along with Dr. Mechanick’s comments on each.
• AC1: “Obesity is a chronic disease.” There was consensus during the conference that this precept was scientifically justified, and (importantly) was a necessary assumption to engage a broad range of stakeholders in a fight against obesity.
• AC2: “The AACE/ACE Obesity Algorithm should be implemented in patients with obesity.” A complications-centric approach was also broadly accepted during the conference, as well as in existing literature.
• AC3: “Lifestyle intervention is critical to an obesity comprehensive care plan.” Although this idea (as with the other affirmed concepts) was supported by broad consensus, Dr. Mechanick noted that there is a great deal of work that remains to be done to investigate and implement effective lifestyle intervention strategies.
• AC4: “The obesogenic factors in the environment need to be reduced.” Dr. Mechanick spoke particularly passionately on this point, noting that America’s most prolific export is also one of its most harmful: the Western diet and lifestyle. He suggested that by simply living in the American obesogenic environment, we are all at risk for obesity, and therefore should all be targets of obesity primary prevention.
• AC5: “Primary and secondary prevention strategies are critically important.” While our medical system is keyed towards treating patients with advanced complications (tertiary prevention), Dr. Mechanick highlighted that it is critical to focus more on primary and secondary prevention in the patients who have not yet developed complications, and who might not yet have even developed obesity.
• EC1: “The definition of obesity needs to be improved.” Dr. Mechanick highlighted this concept as the single most important concept on the list, and the one that the AACE/ACE writing committee is focusing on now. It became clear during the Consensus Conference that the lack of a quality definition of obesity beyond BMI alone was preventing concerted action against obesity. Dr. Mechanick shared that the AACE/ACE writing committee moved forward with discussions on a new obesity definition within 24 hours of the end of the Consensus Conference. Dr. W. Timothy Garvey (University of Alabama at Birmingham, AB), who followed Dr. Mechanick, presented the results of those discussions (see below).
• EC2: “Regulatory, governmental, and insurance organizations require different thresholds of evidence based on specific mandates and decision processes.” This was one of the findings that the AACE/ACE leaders seemed to be most excited about, as it affirmed the importance of a conference model that brought together a diverse group of stakeholders. Dr. Mechanick provided examples of the lack of communication between the different stakeholder groups: the NIH has money to fund studies but simply hasn’t gotten quality proposals; The FDA needs rigorous clinical trial evidence for drug approval, while CMS (another government body) wants specific data in the elderly and disabled. Insurance companies are not particularly driven by randomized trial data, and would prefer three to five year real-world outcomes data. There is certainly room to align stakeholders more closely – we have heard more recently about the potential benefits if registrational trials could also have a real-world element that would appeal more to payers.
• EC3: “Public awareness can change private insurance carriers’ reimbursement strategies and health care coverage provided by employers.” This was another concept that largely came from the non-scientist pillars at the Consensus Conference. Although insurance companies frequently draw the ire of providers and pharmaceutical companies for poor coverage of obesity therapies, payer representatives at the Consensus Conference countered that it is ultimately up to employers to determine what they want to cover for their employees. Employers, in turn, shared that their employees have not been vocal in demanding greater coverage for obesity therapy. Dr. Mechanick believes that education and health literacy efforts to improve patient understanding of the value of obesity therapy will be a critical goal, as people are willing to pay extra for things that are seen as having real value.
• EC4: “Intergenerational obesity must be prevented through intervention in children, including children 0-24 months old, in pregnancy to manage excessive maternal weight gain, and in reproductive age females.” Metabolic disorders during pregnancy can affect the grandchildren of the pregnant individual, as the progeny’s germ cells are already developing while they are in utero. This effect causes a dilation of the timescale of the obesity epidemic, and (according to Dr. Mechanick) could explain why obesity public health interventions have not shown substantial effects in the space of a few years.
• EC5: “Understanding the value of obesity care is important for patients, physicians, payers, and employers.” This concept was closely linked with EC3, and also included the importance for stakeholders to recognize the difference between analyses of cost-savings, cost-effectiveness, and other ways of evaluating value. Diabetes economist Dr. Eric Finkelstein (Duke-NUS Graduate School, Singapore) discussed the finer points of that distinction in a presentation later in the session (see below).
• EC6: “Greater emphasis on education and training regarding obesity medicine.” This was the first of three new emergent concepts that the writing committee added to the list since the post-Consensus Conference press announcement, and we were very glad to see this critical issue included in the list. The under-representation of obesity (and also diabetes) care in medical school curricula is a major impediment to the recruiting of future endocrinologists. Possible solutions that AACE could drive include initiatives for web-based obesity education, certifying exams, and training programs.
• EC7: “Standardize core elements of different lifestyle intervention programs.” Dr. Mechanick stated that the design of effective and efficient lifestyle intervention programs is a major knowledge gap that needs to be filled with rigorous research. He also listed poor reimbursement for lifestyle interventions as a challenge for providers and patients.
• EC8: “The need for additional data addressing optimal obesity management in elderly patients.” This concept, according to Dr. Mechanick, was the result of an enthusiastic request from CMS. There are elements to obesity care in the elderly, such as sarcopenic obesity (the simultaneous increase in adiposity and decrease in muscle mass), that are unique and might require a slightly different approach than might be needed in younger obese patients. 

Panel Discussion

The panel discussion was organized around seven questions that the committee wanted to pose to AACE constituents attending the session. These questions were:

1. Are there improvements needed in the diagnosis [of obesity] to better indicate what we are treating and why we are treating it?
2. Do we need an alternative medical term, other than “obesity,” for referring to obesity as a chronic disease?
3. What is the optimal role of waist circumference in the diagnosis?
4. Do we need to consistently discriminate between overweight and obesity, or classes of obesity (I-III) in the diagnosis?
5. Is a term or designation for “pre-obesity” needed?
6. How can the diagnosis be altered to better facilitate outcomes, benefit/risk, and cost effectiveness?
7. How should diagnostic and treatment paradigms be altered in elderly patients (e.g., >70 years)?

Selected panel comments are presented here. The strongest agreements that emerged included that there needs to be a new term for “obesity” to reduce the stigma associated with the disease and that better anthropometric or predictive measures are needed for better diagnosis/risk stratification. Most panelists also seemed to agree that a pre-obesity designation could one day be useful but disagreed on how to define it, especially since we have not yet even defined obesity.

Dr. Alan Garber (Baylor College of Medicine, Houston, TX): It’s clear that many of us have been struggling with the diagnosis and management of obesity for decades. AACE has recently come to view these struggles as fundamental to try and produce a paradigm shift in how we deal with the obese patient. Initial steps are such things as having obesity defined as a disease. But we have certain questions that we’d like to know the opinions of our membership.

The first question for discussion is “Are there improvements needed in the diagnosis to better indicate what we are treating and why we are treating it?” We would like input from our membership. Are you happy with the NIH guidelines that on define obesity by BMI, or do you need more information? The Europeans like a measure of insulin resistance or a measure of adiposity. Would that help you refine your diagnostic acumen?

Audience comment: In how I would treat and approach the patient, yes.

Audience Q: Tim, you mentioned that there are two stages for obesity in the framework. There is stage 1, where you have mild to moderate complications, and stage 2, where you have severe complications. I’m scratching my head trying to figure out how to make the distinction between what is a mild complication and what is a severe complication.

Dr. Tim Garvey (University of Alabama, Birmingham, AB): I think that’s a complications-specific designation. I tried to illustrate that for cardiometabolic conditions, sleep apnea, and for early histological staging of NAFLD. I think sometimes we’ll need input from experts in each of these complication areas. For people who treat osteoarthritis, there’s validated questionnaires that assess the impact of disease on quality of life and functionality and grade patients into mild and moderate and severe. I think it’ll be complication-specific is the answer to your question. You really have to rely on those specific criteria.

Dr. Arya Sharma (University of Alberta, Edmonton, Canada): Much of this discussion is very much along the lines of the discussion we’ve had in Canada regarding our staging system. It definitely is very much a question when obesity becomes obesity. There is the distinction that some of our European colleagues make between adiposity and obesity. You can be fat and healthy, but with the current definition of obesity, a metabolically healthy person can be labeled as obese. I think a lot of pushback we’re having is that primary care physicians are not comfortable with this labeling, where we’re potentially labeling people as diseased who are actually pretty healthy. I don’t have better terms to use. What has to come up in discussion is that there is a need for a more etiologically-based subdivision of obesity. It is a very heterogeneous condition. There are lots of reasons why people may put on weight. In every study of obesity, we simply use BMI. If you have someone who’s obese because of an eating disorder, it’s treated the same as someone who has gained weight due to a new medication, or an injury. That information may or may not be important for treatment, but it is something you would want your provider to take into consideration. You want to foster that kind of thinking, and move away from thinking that this is a homogeneous condition.

Another thing I’d like to mention is that in the Canadian obesity staging system we’ve adopted over the past three years, there is more granularity in that endogenous organ damage is categorized as stage 3. We’ve worked with David Allison on NHANES datasets, and have demonstrated the value of differentiation with a bit more granularity. I don’t think that two stages are enough – I think there should be three or four.

Dr. Garber: I think our goal is to try and diagnose and mange this before you get end organ damage. But I appreciate your comments on that.

Dr. Garber: We’ll move on to our second question, which is quite straightforward. Do we need another diagnostic term aside from “obesity” and if we do does that improve our ability to label this new entity as a chronic illness to get away from all the stereotyping, all of the stigmata that come with the word obesity, which is generally rather negative in our society despite its very high prevalence rate?

Audience: It’s a no brainer we need a new term. What would you suggest?

Dr. Garber: When I was on the NECP we tried to rename it. We called them patients with the metabolic syndrome, and you know amount of flack we got over that.

Dr. Jeffrey Mechanick (AACE, Jacksonville, FL): One of the terms that has been put forth is adiposity-based chronic disease, a chronic disease based on the presence of adiposity, which you cannot have if you are underweight. The acronym would be ABCD. That was a working phraseology or nomenclature for this chronic disease. I think it was the sentiment of the entire working group that we need to address this. It would be a secondary aim of the vetting process in the next few months, to get away from this antiquated term that has a lot of negative stigma.

Dr. Garvey: I think the advantage of this is to get away from the term obesity, which really means different things to different people. It is an imprecise term, a sloppy term, and it entails some stigmatization. I am in favor of a term like adiposity-based chronic disease, which is a medical diagnosis of what we’re treating and why we’re treating it. In our discussions about the diagnosis Dr. George Gray kept using the term unhealthy weight, “It’s unhealthy weight.” That kind of simple term brings to fore this weight adversely affecting health of the individual. That’s the idea. Would this help with advancing the cause of medical treatment of patients and allowing our partners to work together in a more concerted way? That’s something we wanted feedback on.

Audience: I like weight-related chronic disease.

Audience: If you really want to get rid of the word “obesity,” you could use the term overweight class one, two, three, etc. If it is frustrating to treat obesity in adults, where you have a few tools, imagine how frustrating it must be for our pediatric colleagues who are trying to treat children, who have no tools whatsoever. Parents are faced with utter frustration when they go from doctor to doctor looking for help, and there is nothing out there. We should include pediatric colleagues in this process and get their input.

Audience: Look what the term “erectile dysfunction” did to impotence. It totally destigmatized it.

Dr. Garber: Onto the next question. What is the optimal use of waist circumference in the diagnosis of obesity? Does it advance our understanding of the diagnosis of the disease?

Dr. Arya Sharma: Where it adds value is in the skinny person. The lower your BMI, the more meaning that the waist circumference measurements has. It becomes completely unreliable at higher BMIs. When I see providers struggling to figure out where to place the tape on a person with a BMI of 45, it is not only embarrassing, it is meaningless. The problem with the way waist circumference cutoffs were designed is that, at high BMI, they correspond to BMI cutoffs. Once you have the BMI you can guess the waist circumference at the high BMI range, but not in the overweight range. I think it has value as a screening tool for skinny people. It reminds us of the importance of ectopic fat. Once you get to a BMI of above 35 kg/m², don’t waste your time measuring waist circumference, it doesn’t add value, and doesn’t point you in the direction of treatment decisions.

Dr. Daniel Einhorn (Scripps Memorial Hospital, La Jolla, CA): How many people in the room have measured waist circumference in patients? [About 5-10 people raise their hands]. How often is it embarrassing ? [About the same number raise their hands again]. So it cuts both ways. It’s non-standardized, it’s difficult to do, but it creates a teachable moment.  I’ve really always thought of that embarrassment as an impediment because it’s uncomfortable, but the more I think about it, the more I realize it creates that teachable moment without having to put words on it, but just seeing that number is generally all one needs to do to make a point. I think there’s value, and I agree with Dr. Sharma that in terms of epidemiology and collecting data and making treatment decisions, it starts to lose meaning in higher BMIs, but separate from that, from classification, it is a clinical tool.

Audience Q: We now have a relatively easy way of measuring % body fat. Would that add anything of value to the assessment of patients?

Dr. Mechanick: Bioelectrical impedence is now FDA approved for resistance, but not reactance, so it’s great and validated for body water. For body fat, it doesn’t really have the type of accuracy we would need to base a therapeutic or diagnostic strategy on it. However, the concept is correct. We need a more robust way to determine an anthropometric component of this diagnostic strategy to identify a patient at risk. And then the diagnostic component is the assessment of complications, and as Dr. Sharma said may want to extend to more granularity at stage four so that predictive value is higher. But the anthropometric component, right now we have height and weight. Numerous different measures have been used and validated in different populations. There are also transcultural factors, sarcopenic obesity – those are all variables that have not been baked in yet to an anthropometric system. That’s fine though because that is the purpose of the vetting process that will ensue. As we convene these pillars, we’ll see where the evidence is. And based on the evidence we can come up with some recommendations.

Dr. Garber: Do we need to consistently discriminate between overweight or obesity, or can we call everyone overweight, regardless whether they fall into overweight or obese categories by BMI, and stage it?

Dr. Mechanick: Let’s reframe the question. Does everyone with a BMI over 25 kg/m² deserve the disease designation, or should there be a pre-disease designation? Perhaps we could view obesity on a continuum. For example, some people think that prediabetes is just early diabetes. Is overweight really just pre-obesity? If someone is in an obesogenic environment and is gaining fat, even if they are not obese yet, is that still a disease?

Dr. Sharma: Again this is a great question. Again, it comes back to clinical utility. The clinical utility as we are now learning – there is very little, surprisingly little, correlation between the size of a patient and the sickness of that patient. You can find a lot of weight-related morbidities even in the overweight range. So given that the whole staging classes are completely arbitrary, these are all made up numbers. Clinically they’re irrelevant. What really matters is how sick is your patient. You don’t’ really care how big your patient is except to make sure you have the right sized seats in your waiting room, but apart from that it doesn’t matter.

Dr. Eric Finkelstein (Duke University, Durham, NC): Just one comment and I’ve been struggling with this – one thing we know and even I’ve actually published on this is people with BMI 25-30 kg/m² tend to do better from a mortality perspective. Even in the low 30s, it is unclear if there is any morality effect at all. In fact John Dixon has argued that the optimal BMI changes as people age. It may or may not be true, but may be possible. To me I’m having a hard time listening to the discussion we’re using these fixed cutpoints with BMI, and I’m not sure they’re entirely consistent with what we know about mortality effects. And my understanding of the evidence is for older adults, BMI becomes somewhat protective. So how do you reconcile that?

Dr. Garvey: […] We definitely need more data there. There is some epidemiological data suggesting that high BMI is conducive to healthier and longer life in the elderly. We need more data, and our draft statement calls for that.

Dr. Garber: What about this term pre-obesity? If overweight is pre-obesity, but it’s not particularly severe in terms of health outcomes, perhaps we’re unnecessarily alarming the population. Are there data to indicate that early interventions produces better outcomes later?

Dr. Einhorn: Alan every metabolic parameter, without exception, lives on a continuum and lives in the context of other risk factors that modify the importance of that parameter. So if we are considering overweight/obesity as a chronic metabolic disease, then I would at least for now follow the rule that we use for every other chronic metabolic disease and recognize it on a continuum. We recognize that the continuum isn’t absolutely predictive but it’s useful to stratify and it’s certainly useful to monitor where the patient goes on that continuum over time. I think there is value in pre-obesity as a concept. It’s been valuable in other metabolic diseases, for example, pre-hypertension and pre-diabetes. For those too, it’s semantic where on the continuum you draw a line so people can have a one-word term for where you are. But still it’s a useful construct to think of pre-obesity as a term, however you call it, and then stratify along the continuum.

Audience: What if we use the term pre-obesity for risk stratification beyond putting a number on it. For example: when I see an obese mother and an obese dad come in with an 11 year-old that is currently normal weight, I consider that child at very high risk for obesity. Maybe we need to reserve that term as risk stratification, looking at the other factors for that individual rather than putting a number on it, whether it is 95^th percentile on weight chart or a specific BMI.

Dr. Garber: How many of you use the term pre-hypertension in your diagnostic charts? Nobody. We use prediabetes of course but I think the JNC concept of pre-hypertension has had virtually no impact [audience nods].

Audience: It’s not even in my EMR – they took it out. I can’t even use it as a diagnosis.

Dr. Sharma: I totally agree that the value of having the term pre- in there, but I don’t think we’re at a point where we can define a pre-obesity because we haven’t defined obesity yet. We probably have to have a solid definition of obesity before we can think about what pre-obesity could be.

Dr. Garber: Next question. How can we make a diagnosis to better facilitate outcomes, risk/benefit ratios, and cost effectiveness? What tweaking needs to happen to the diagnosis? Is staging enough?

Audience: A couple years ago, it was very popular for people to say I’m obese and proud because I’m healthy. I think this is faulty thinking. I think they’re still obese, and they’re going to be aging prematurely ... wearing out their knees, backs, hips. If they’re obese, they’re aging prematurely. If they have these metabolic abnormalities, that’s obviously an additional bad thing but just in being overweight, I don’t think there’s anything like healthy obesity.

Dr. Karl Nadolsky (Walter Reed National Military Center, Bethesda, MD): I’m a huge advocate of waist circumference. My BMI is 30 kg/m², and my waist circumference is 32” [Dr. Nadolsky is a weight-lifter]. I would think I’m healthy obese by the old guidelines, but we did put in one section about judgment for sarcopenic obesity and muscular athletes. It’s my interpretation, or maybe the way I want to see it, that someone like me should not get diagnosed as obese. As with osteoporosis, a DEXA scan does not diagnosis it. It’s a clinical diagnosis. So why can’t obesity be a clinical diagnosis. I think what you have presented actually is. When you walk into office, you know they’re obese. It’s like pornography – you know it when you see it.

Dr. Garvey: I personally like to think the previous speaker had a good point. He didn’t believe there were metabolically healthy obese. I believe that is weight gain on an insulin sensitive background. Those people do exist. In the NHANES database that’s 16% of the population. But I the point I want to make is that’s not a static category. Those patients still require evaluation over time because they can change, and they can become a stage 1 or stage 2 and require more aggressive therapy. I think Karl made a great point of all of the things the clinicians have to consider as caveats. Muscularity has to be considered and edema. I did my internship and residency at Barnes, and my mentors used to say if you can trace edema, that patient has an extra 30 lbs of water and salt on board. So even mild edema can alter weight. All of those things are in there Karl.

Dr. Mechanick: Obesity is not best labeled as a weight-related chronic disease. It is really indexed to adiposity, because muscularity and edema affect weight but not the underlying pathophysiology of the disease. You could look at a static cell of patients with a BMI of 30 kg/m² and do not have complications. If you are in a disease treatment paradigm, you look at that group as metabolically healthy overweight, and you may say that no intervention is needed. But you look at that group in terms of a prevention paradigm, all those cells of patients are moving from cell to cell. Just because you are in one cell does not mean that you will not move into another.

Dr. Garber: How do you modify this in the elderly patient over 70 years?

Dr. Mechanick: I think we could even expand the framework to how to modify to different ethnicities, taking Asian Indians as an example. You have sarcopenia in the elderly. We would have to have a test with very good performance to detect that sarcopenia. There you need better anthropometric measurements. Waist circumference is probably the best we have at lower BMIs.

In-Depth Symposia: The Path Forward for Obesity Medicine: Consensus for Concerted Action and the AACE/ACE Consensus Conference on Obesity

Health Economics of Obesity Treatment

Eric Finkelstein, PhD (Duke-NUS Graduate School, Singapore)

Dr. Eric Finkelstein, one of our favorite participants at the AACE/ACE Consensus Conference on Obesity, presented some of his critical points at that meeting for the AACE audience, along with some notable updates. Similar to what he said at the Consensus Conference, Dr. Finkelstein stressed that obesity treatments, with the potential exception of bariatric surgery, are not cost saving.  Dr. Finkelstein has estimated that when one discounts for the future, converting a ten-year old boy from obesity to normal weight probably only saves $10,000 to $13,000 worth of medical costs over his lifetime if that weight is maintained. Due to the timing of complications onset, this is a similar amount of savings for if a 20-year old or a 65-year old went from being obese to being normal weight. Since reversal to- and long-term maintenance of normal weight is infrequently seen even with bariatric surgery (which costs ~$25,000), it is very hard for any approach to demonstrate cost-savings in these groups (or most other age groups). However, Dr. Finkelstein thinks that cost-effectiveness not cost saving should be demanded of anti-obesity agents, because cost-saving is not required in other disease areas. Instead, Dr. Finkelstein believes that companies and obesity advocates should highlight the value their agents provide per dollar.

• Dr. Finkelstein underscored the need for drug companies to evaluate the economics of their treatment. He stated that the first step in doing this is determining the economic perspective one wants to take (e.g., societal, payer, employer, participant). The key economic measures to track will vary between these groups. For example, an employer will be very interested in the treatment’s impact on absenteeism and presenteeism, while a payer will likely care more about the approaches’ impact on medical costs. Dr. Finkelstein explained that randomized control trials provide the best source of data, and encouraged companies to add their important economic endpoints to the protocol. However, Dr. Finkelstein acknowledged that many studies would need to be too long and too large to be powered for economic measures. Thus, he (and payers at the consensus conference) said that models can be used.
• Dr. Finkelstein presented blinded results from a study that has not yet been published, indicating that two of the most cost-effective approaches (based on their incremental cost-effectiveness ratio) for treating obesity are an un-named lifestyle intervention program, and a pharmacotherapy. At the Consensus Conference, Dr. Finkelstein presented interim unblinded results from this study.
• Dr. Finkelstein suggested that the most cost-effective manner to treat obesity might be with a stepped-care intervention. In a study (Jakicic et al., JAMA 2012) people (n=363) were randomized either into a standard life style intervention, or, initially, a low cost education program. Depending on how well people responded in the education program, they were stepped up to more intense programs, including the receipt of free meal replacements. While people in the step program on average lost less weight (6.9% vs. 8.1%; p<0.001), it appeared to be a better deal than the standard program, costing only ~$125 per kg, as compared to ~$175 per kg for the standard program.

Oral Presentations

Insulin Infusion and Glucose Sensing by Minimed Duo, A New Combination Device

Scott Lee, MD (Medical Director, Medtronic Diabetes, Northridge, CA)

Dr. Scott Lee shared new data from a 48-patient home study of the MiniMed Duo, a three-day wear combined insulin infusion-CGM sensor set (a “snake bite” configuration under a single patch, separated by 11 mm) – see a picture on twitter here. Patients wore five MiniMed Duo devices during the 15-day study and sensor accuracy was compared to SMBG values. The sensor’s overall MARD was 15.5% and 77% of points fell in Zone A of the Consensus Error Grid (n=5,056 paired points) – the accuracy was not particularly groundbreaking, especially considering there were an average of 4.7 calibrations per day. Of course, the combination form factor is a win for patients (especially pediatrics), and sensor accuracy is unquestionably a challenge given the three-day length of wear. Indeed, day one MARD of 18.7% was really the stumbling point, as days two and three saw much improved accuracy (13.1% and 14.1%) – we wonder if the warm-up time could be extended to allow for improved day one accuracy. There were 44 total adverse events, with 98% classified as mild in severity and skin-related (e.g., “bruising and bleeding”); one patient had an abscess that was treated with antibiotics. Overall, this is the longest-term data we’ve seen on this device, and it certainly demonstrates its real-world feasibility. We look forward to hearing about future improvements and whether Medtronic plans to commercialize this device.

• Overall, 75% of paired CGM-SMBG points were within 20% of one another, meeting the study’s “primary success criteria” (>60% was the goal). This struck us as an unconventional primary success criteria, as CGM accuracy is more typically described in MARD terms.
• Dr. Lee shared the Duo’s sensitivity and specificity, noting that the rate of false hypoglycemia alerts was somewhat high. At a CGM alert setting of 70 mg/dl, 92% of hypoglycemia events were correctly detected within 30 minutes, though 57% of alerts were false. The rate of false alarms has been a complaint of the MiniMed 530G as well, and is something we expect will improve with Medtronic’s next-gen sensors.
• The study enrolled current pump + CGM users (94% on Enlite), and the acceptability ratings looked encouraging: 75% of patients indicated “no pain” at the time of insertion; 74% of patients preferred the Duo device to the two-site configuration; and 89% agreed that the overall device experience was acceptable.    
• The MiniMed Duo combines a glucose sensing electrode and a steel insulin infusion cannula under a single adhesive pad. The electrode and the cannula are separated by 11 mm, an “important point” – Dr. Lee said that insulin boluses typically go out ~7 mm into the subcutaneous tissue.
• Three prior studies paved the way for this study: O’Neal et al., JDST 7(2); Frid et al., Pediatr Diabetes 14(S18); Norgaard et al., Diabetes 61(S1). These studies showed 1) an 11 mm separation between the insulin cannula and sensor electrode result in acceptable sensor accuracy; 2) sensor performance and longevity are not affected when recurrent large boluses are delivered during device use; 3) there was a reliable correlation between sensor and meter blood glucose values in a three-day outpatient study.
• BD is also working on a combined CGM-insulin infusion set – this three-year collaboration was announced on June 18, 2013. There is no official timeline, though we imagine this wouldn’t come until after BD’s CGM and insulin infusion sets come out – both individual components are still in the works. A clinical trial of the CGM is expected to be completed in fall 2014, with results expected in early 2015.
• Insulet is attempting to integrate CGM into the OmniPod as well – as of the company’s 1Q14 call, an insertion study was complete, and data was being compiled. This project is still in early stages and an update is expected later this year.
• We are aware of three other groups that were at one time working on combined CGM-insulin infusion:
  
  • Pacific Diabetes Technologies (PDT) and Oregon State University were using a flex circuit to dispose glucose sensors on a flat surface that can be wrapped into a catheter (i.e., each catheter is encased by a glucose sensor); as of the FDA-JDRF-NIH Workshop on Innovation Towards an Artificial Pancreas in April 2013, the device was in pre-animal testing. See page six here.
  • Sensile Medical was developing a single port sensor in which a five-millimeter porous cannula simultaneously measures glucose and delivers insulin by changing the size of its pores in response to glucose concentration. The last update we heard on this came at ATTD 2012. See page 10 here.
  • Medical University of Graz is working to integrate commercially available sensors into an insulin cannula. The last update we heard on this came at ATTD 2012. For more detail, see page 11 here.

Efficacy and Safety of Liraglutide 3.0 mg for Weight Mangement in Overweight and Obese Adults: the SCALE Obesity and Prediabetes, a Randomized, DOuble-Blind, and Placebo-Controlled Trial

Xavier Pi-Sunyer, MD, PhD (Columbia University, NYC, NY)

Dr. Xavier Pi-Sunyer presented results from Novo Nordisk’s phase 3 trial of liraglutide 3.0 mg in overweight and obese adults, with and without prediabetes, SCALE Obesity and Prediabetes. Topline results from the study were announced in May 2013. The trial enrolled 3,731 people (baseline BMI 36 kg/m²) who were obese or overweight and did not include type 2 diabetes; 2,283 (61%) had prediabetes. The trial separately randomized and tracked people with and without prediabetes; however, Dr. Pi-Sunyer specified that he would only present pooled results for people with and without prediabetes (we hope that the prediabetes results will be presented soon, at this time, we do not see them on the ADA program). The biggest news of the presentation was that the incidences of gallbladder disorders and acute pancreatitis while low, were more frequent with liraglutide 3.0 mg – a finding in line with Novo Nordisk’s comments on its 4Q13 call that “imbalances” were seen in these areas in the liraglutide 3.0 mg development program. There were 2.7 events of gallbladder disorders per 100 patient-years of exposure (PYE), the experimental arm versus 1 in the placebo (n=54 vs. 9). Dr. Pi-Sunyer hypothesized that this is due to the liraglutide arm’s greater weight loss, since weight loss is known to cause gallstones. There were 0.3 events of acute pancreatitis per 100 PYE in the liraglutide arm compared to 0.1 event per 100 PYE in the placebo (n=7 vs. 1). There were two people (one in each arm) who had both a gallbladder and a pancreatitis event. The majority of the acute pancreatitis events were mild, and improved quickly upon discontinuation of the drug; there was no consistent mode of presentation or latency period. Novo Nordisk is further investigating this pancreatitis finding.

• As a reminder, participants on liraglutide achieved a mean 8% weight loss, compared to 2.6% for those on placebo – slightly better than the FDA's continuous benchmark for obesity drugs (a mean >5% placebo-adjusted weight loss). Categorically, 63% of people on liraglutide achieved at least 5% weight loss (vs. 27% for placebo) and 33% achieved weight loss of at least 10% (vs. 10% for placebo). As a reminder, the FDA's categorical requirement for obesity drugs is that the percentage of on-drug participants who achieve >5% weight loss must be at least 35% and twice the percentage of people on placebo achieving >5% weight loss.
• Concomitant with the weight loss were improvements in waist circumference, BMI, fasting plasma glucose level, A1c, systolic blood pressure, and diastolic blood pressure. The only baseline provided was for BMI: 36 kg/m² and average baseline weight of 106 kg.

• Dr. Pi-Sunyer stated that the safety profile of liraglutide 3.0 mg in this trial was consistent with that of the agent in other phase 3 studies, and liraglutide 1.8 mg (Victoza) for people with type 2 diabetes. The 56-week completion rate was 72% for liraglutide and 64% for placebo. Withdrawals due to adverse events occurred in 9.9% of the liraglutide arm and 3.8% of the control. The total event rate was 92.1% in the liraglutide arm, as compared to 84.0% in the placebo. The break down of severe adverse events (event interrupted the participant’s normal daily activities) was 12.3% in the liraglutide arm versus 9.1% in the placebo; of serious adverse events (generally defined as an event whose outcome was death, life-threatening, hospitalization, disability, congenital anomaly, or requires intervention to prevent permanent damage) was 6.2% versus 5.0%; and of fatal events was 0.0% versus 0.2%. The most common adverse events for liraglutide were GI-related (i.e., nausea, diarrhea, and constipation). People generally began to feel nauseous during weeks 1-4, the nausea was usually mild or moderate, and the feeling tended to diminish over time.
• During the topline results announcement, Novo Nordisk indicated that liraglutide 3.0 mg appears to improve and prevent prediabetes a finding we expect will increase uptake of the drug among payers, if it is approved. At week 56, 69% of the prediabetes subgroup (61% of all study participants had prediabetes) treated with liraglutide no longer showed signs of prediabetes, compared to 33% of the placebo-treated group. Among people without prediabetes at randomization, 7% of those on liraglutide and 21% of those on placebo developed prediabetes. We are curious what impact liraglutide use had on people transitioning from prediabetes to type 2 diabetes.

Cardiovascular Safety of Saxagliptin in Patients with Type 2 Diabetes: Pooled Analysis of 20 Clinical Trials

Robert Frederich, MD, PhD (Bristol-Myers Squibb, Princeton, NJ)

Dr. Robert Frederich presented the results of a post-hoc meta-analysis of 20 phase 2b and 3b trials of saxagliptin, excluding the SAVOR-TIMI 53 cardiovascular outcomes trial. The 9,156 patients included in the 20-trial pool could be seen as more representative of the type 2 diabetes patient population than patients enrolled in SAVOR, and were (on average) younger, had lower diabetes duration (45-49% less than three years), and at much lower CV risk at baseline. As a result, the event rate for MACE (the primary outcome in both SAVOR and this meta-analysis) was approximately three-fold higher in SAVOR than in the 20-study pool. The meta-analysis of the full 20-study pool found a MACE incident rate ratio (IRR) of 0.74 trending in favor of saxagliptin, although it was not statistically significant (95% CI: 0.45-1.25). Sub-analyses by individual components of MACE were also not statistically significant but may have trended slightly in favor of saxagliptin. Notably, heart failure (which was not a component of the primary MACE endpoint but was also analyzed) trended more strongly towards saxagliptin (IRR = 0.55; 95% CI: 0.27 – 1.12), although heart failure was not adjudicated in these 20 studies as it was in SAVOR. Nevertheless, that results stands in contrast to the statistically significant 27% increase in hospitalization for heart failure seen in SAVOR. It is worth noting that the mean exposure time for patients in the meta-analysis was just under one year, which is substantially less than SAVOR’s median follow-up of 2.1 years. However, on the whole, the results from this analysis were reassuring, especially on the heart failure front.

• In a sub-analysis of 11 trials from within the 20-trial pool that tested saxagliptin as an add-on to metformin, the IRR for MACE with saxagliptin was 0.93 (95% CI: 0.44 – 1.99). These results did not deviate substantially from the result of the overall meta-analysis.
• IRRs for the individual components of MACE in the full 20-study pool were: 0.61 for CV death (95% CI: 0.28 – 1.36); 0.87 for myocardial infarction (95% CI: 0.39 -2.02); and 0.75 for stroke (95% CI: 0.31 – 1.90).
• We also saw reassuring retrospective analyses on DPP-4 inhibitors at this year’s American College of Cardiology Scientific Sessions – there, the data was from the EXAMINE CVOT for Takeda’s Nesina (alogliptin). One sub-analysis suggested a CV death benefit with alogliptin in women and those with shorter duration diabetes and good renal function, while another provided reassurance on the impact of alogliptin on heart failure. Diabetes KOLs have struggled to propose convincing theoretical mechanisms to explain a connection between DPP-4 inhibitors (which generally have a very clean safety profile) and cardiovascular harm, and this data (along with the Onglyza data presented at AACE) provide valuable reassurance, although presumably the debate (especially on heart failure) is far from over.

AACE Individual Press Release

AACE Consensus Statement on Pump Therapy for Patients with Type 1 and Type 2 Diabetes

In the May 2014 issue of Endocrine Practice, AACE has released a consensus statement outlining recommended clinical approaches to pump therapy for patients with type 1 and type 2 diabetes – see Saturday’s press release here and the full statement here. The impressive list of authors includes Drs. George Grunberger, Timothy Bailey, Bruce Bode, Lois Jovanovič, Philip Raskin, and William Tamborlane. The document is quite comprehensive and could best be described as a how-to-guide to pump therapy for HCPs – it offers patient selection criteria, a comparison table of current pumps on the market (Accu-Chek Combo and Spirit, Asante Snap, Medtronic MiniMed Paradigm and 530G, Insulet OmniPod, Animas OneTouch Ping, and Tandem t:slim; there is a separate dedicated table for the Valeritas V-Go), a review of the literature comparing pumps to MDI, a guide to calculating initial pump settings, education and training guidelines, a review of patient safety issues, a summary of coding and reimbursement issues, and a list of future needs (more dose precision, overcoming clinical inertia around pump settings, easier downloading, better education, driving towards closed-loop insulin delivery). The latter notes that predictive low glucose suspend will launch in Europe in 2014 (i.e., Medtronic’s MiniMed 640G – Medtronic’s last call estimated launch by October 31), followed by overnight closed-loop and control-to-range. (This is of course an open debate, as many have been calling for overnight control to be paired with daytime treat-to-range.) The piece concludes that after three decades of clinical pump use, “many critical questions remain.” We certainly agree, and hope that statements like these raise the level of conversation around pumps and drive towards answering those questions.

Late Breaking Clinical Trials

Liraglutide As Additional Treatment To Insulin In Patients With Type 1 Diabetes Mellitus: A Randomized Clinical Trial

Nitesh Kuhadiya, MD, MPH (University of Buffalo, NY)

Dr. Nitesh Kuhadiya presented data from the “first prospective, randomized, double-blind, placebo controlled clinical trial of liraglutide in type 1 diabetes.” Patients (n=72) in the 12-week study were randomized to placebo or one of three liraglutide doses (0.6, 1.2, or 1.8 mg). Though there was not much benefit from the 0.6 mg dose, the data on the 1.2 and 1.8 mg doses of liraglutide was universally strong across the board – a reduction in A1c (declines of 0.4% and 0.7% from baselines of 7.4% and 7.8%), total insulin dose (~ -19%), body weight (~ -5%), daily carb intake (~ -28%), blood pressure (-3 mmHg from 120 mmHg at baseline; only in the 1.8 mg group), and even C-reactive protein (~ -17%; a marker of inflammation). We were particularly elated to see patients wore a Dexcom Seven Plus throughout the 12-week study, and it was used in tandem with fingersticks to titrate insulin. The time-in-range benefit of liraglutide was seen on hyperglycemia side, with patients spending significantly less time >160 mg/dl. Notably, patients on liraglutide even saw a significant improvement in all aspects of diabetes related quality of life. Dr. Kuhadiya concluded that the findings have “significant implications for the future treatment of type 1 diabetes.” We certainly agree, and look forward to results from Novo Nordisk’s ongoing phase 3 trials of liraglutide in type 1 – phase 3 was initiated in 4Q13 and management is “hoping” for a 2015 launch.

• This prospective, randomized, double-blind, placebo-controlled clinical trial of liraglutide randomized 72 patients with type 1 diabetes to placebo or one of three liraglutide doses (0.6, 1.2, or 1.8 mg). Dropout was fairly low, with 63 patients (88%) completing the study. The treatment period was 12 weeks long, and patients wore a Dexcom Seven Plus over the entire period. Insulin doses were titrated based on CGM and fingerstick readings to target a blood glucose of 70-180 mg/dl. If A1c was >7.5%, no reduction was made in pre-prandial insulin or basal insulin. If A1c was 7-7.5%, pre-prandial and basal insulin were decreased by 10%. If A1c was <7%, patients decreased preprandial and basal insulin by 25%.
  
  • At baseline, patients had a combined mean A1c of 7.6%, a mean BMI of 29 kg/m², a mean age of 44 years, and a mean diabetes duration of 20 years. All had no detectable C-peptide and type 1 diabetes for at least one year on insulin therapy.
• Glycemic control results: the 1.2 mg liraglutide group experienced a significant 0.7% decline (baseline 7.8%) (p<0.05 vs. placebo); the reductions in the 0.6 and 1.8 mg groups (-0.2% and -0.4%) were not significantly different from placebo (-0.3%). The complete data, along with time in range results, are below. We thought several things were notable from this data – first, the 1.2 mg dose achieved the most impressive reduction in A1c, and it occurred in patients that were fairly well controlled at baseline. Second, the 1.8 mg dose did not seem to offer additional improvement over the 1.2 mg dose. Third, the 0.6 mg dose did not seem to be that effective, a finding carried throughout the study. Fourth, improvements in glycemia did not come at the expense of additional hypoglycemia. Fifth, the big benefit of liraglutide 1.2 and 1.8 mg seemed to be on cutting down on hyperglycemia >160 mg/dl and >240 mg/dl. We assume the latter stems from the delayed gastric emptying, suppression of postprandial hyperglucagonemia, and the reduction in appetite. Sixth, the “time in zone” reductions didn’t appear to “match” the A1c drops – we are looking into this.

*p<0.05 vs. placebo

• Total insulin dose declined significantly by 17-21% in the 1.2 and 1.8 mg liraglutide groups (p<0.05 vs. placebo). This was a strong result when considered in tandem with the glycemic improvements noted above.

*p<0.05 vs. placebo

• Patients in all three liraglutide groups saw “remarkable” weight loss, noted Dr. Kuhadiya. Participants in this trial were slightly overweight at baseline (mean BMI of 29 kg/m²), which certainly aided showing weight loss efficacy. Many have said that the concept of the “thin type 1” is now no longer the case, and this trial’s population lends credence to that case – we wonder if the BMI distribution in the type 1 population mirrors that of the general population (~2/3 overweight, ~1/3 obese), or if type 1s skew differently.

*p<0.05 vs. placebo

Questions and Answers

Dr. George Grunberger (Grunberger Diabetes Institute, Bloomfield Hills, MI): Was there a way to monitor whether patients were actually taking the dose of liraglutide they were randomized to?

A: We had them bring back the empty pens and we did those calculators. We knew that each pen would last a certain amount of time, so we could figure it out. Four patients in the 0.6 mg group were originally in the 1.2 mg group, but they could not tolerate the 1.2 mg dose – we clumped them into the 0.6 mg group.

Plenary Sessions

America’s Expanding Waistline: Should Government Intervene and If So, How?

Jayson Lusk, PhD (Professor, Oklahoma State University, Stillwater, OK)

Libertarian economist Dr. Jayson Lusk argued that the most appropriate way for the government to intervene in the obesity epidemic is for it to i) fund research on the causes of and treatments for obesity, and ii) ensure that results from these studies are accurately and effectively communicated to the public. Dr. Lusk argued that many other interventions into the obesity epidemic, while motivated on compassionate grounds, have unintended consequences and are not well justified. One of the linchpins for Dr. Lusk’s argument was his belief that most of the costs of obesity are born by the individual, not society. Furthermore, Dr. Lusk thinks that the best way to address obesity’s limited negative externalities (a negative cost of obesity borne on members of society, other than the obese person; e.g., higher premiums), is to have the obese person face more of these costs (Dr. Lusk did not detail how this could be done, however, one possibility would be increasing the premiums on obese people, so that other people’s premiums do not have to go up). Dr. Lusk closed his presentation by calling for humility amongst policy makers and proponents. He pressed that it is not policy makers’ place to determine what is in other people’s interest – for example, some people might prefer to drink soda and be obese, and others should not say that it is in their interest not to. In this vein, Dr. Lusk suggested that some food policy advocates (he referenced author Michael Pollan) have glamorized the lifestyle and diet of the 1950s and 1960s. Dr. Lusk hypothesized that most people would not want to be living in that period as opposed to today’s (women were spending many additional hours cooking and cleaning, etc.), and that higher obesity rates might be a necessary trade-off for these conveniences.

• Dr. Lusk’s argument that the government should largely not intervene in the obesity epidemic was founded on a study suggesting that >95% of the costs of obesity impact the individual not society (Bhattacharya and Sood, JEP 2011). The authors’ found that the marginal cost of a 50-year old American becoming obese is ~$350,000. Dr. Lusk highlighted the paper’s other finding that ~98% of this marginal cost is due to a shortening of the person’s life expectancy: obesity has been found to on average reduce the life-expectancy of such a person by 1.65 years, and the study values a life-year at $200,000 (following statistical precedent). In contrast, on average, this person’s medical care spending would increase by about $15,000. Dr. Lusk concluded that this means most of the cost of obesity is borne by the individual not society. Thus, he believes that it is generally up to the individual how they behave (i.e., what foods they eat, how active they are), in the face of these consequences.
  
  • Based on our reading of the paper cited, the analysis conducted only considered obesity’s impact on life expectancy and on medical costs. Thus, it appears to not account for many of the substantial indirect affects obesity can have, including on a person’s productivity (via absenteeism and presenteeism), which is largely felt by their employers and therefore society.
  • We also note that since one-third of US adults are obese (~110 million people), even a 5% per person cost is substantial. If the average obese adult were 50 years old, by the study’s estimates, the medical cost of obesity would be ~$1.65 trillion over the course of the people’s lifetimes. Indeed, the estimated annual healthcare costs of obesity-related illness are a staggering $190 billion.
  • During a media event with Dr. Lusk we asked him if his judgment on the need for government intervention would differ if he did not think people who are obese are in control of their disease state (e.g., had an unpreventable cancer). Much of Dr. Lusk’s argument to this point on individual’s facing the costs of their lifestyle decisions in coming obese, implied that it was a choice people have, while much of the data suggests that genetics and neural biochemistry play a huge role. Dr. Lusk responded by drawing an analogy: Dr. Lusk is a very tall and lanky man, with very long arms; as a result, he needs to buy specialty shirts, which are more expensive. However, Dr. Lusk does not expect other people to pay this difference for him. The lesson Dr. Lusk drew from this example, was that life and our genetic lot are not fair, and people might have to pay more as a result. We understand and respect Dr. Lusk’s sentiment, yet we note that the additional cost we expect obese people to pay (years of their life, lower quality of life, etc.) is much steeper than that of an extra shirt, meaning that the effort to aid an obese person would like have a higher return on investment.
• Dr. Lusk then explained why he does not think anti-obesity policies like a fat or sugar tax, or modification of farm subsidies would have the large positive effect many obesity advocates expect them to have.
  
  • Taxation – On a possible soda tax, Dr. Lusk warned that research suggests it would increase the consumption of sugary juices, and potentially alcohol (we think this latter assertion would be highly dependent on the state’s culture and liquor laws). More broadly, Dr. Lusk made an argument similar to the one economist Dr. Julian Alston (University of California Davis, Davis, CA) made at The Obesity Society (TOS) 2012. Dr. Lusk stressed that a sugar or fat tax would not be particularly effective. The literature suggests that a fat tax (i.e., $5/kg fat in the food) would cause the average American adult to lose 5.61 lbs (vs. 5.5 lbs via a calorie tax and 5.50 lbs via a sugar tax); however, a calorie tax would be the most cost-effective, costing $0.90/lb. (vs. $1.34/lb for a fat tax and $0.93/lb for a sugar tax). Dr. Lusk warned that a calorie tax would be regressive (negatively impacting poor people more than rich people), since poor Americans spend a larger portion of their income on food than wealthier people. We wonder if this serious concern could be mitigated by using the revenue from such a tax to increase funding for nutrition-safety-net programs (i.e., Supplemental Nutrition Assistance Program [SNAP; previously called Food Stamps], and The Special Supplemental Nutrition Program for Women, Infants, and Children [WIC]). 
  • Farm Subsidies – Also like Dr. Alston, Dr. Lusk explained that removing farm subsidies (e.g., for corn) would not have a large impact on people’s diet or the obesity epidemic. Dr. Lusk explained that only ~10% of the cost of most foods are due to the price of the raw materials. Studies suggest that removing the corn subsidy would only cause a 9-10% drop in the crop’s production, meaning the price would not change substantially (Dr. Lusk did not provide a specific number on how much the price would be impacted; however, Dr. Alston indicated that the retail price of meat would only go up 0.3%). Dr. Alston showed that without current grain and oilseed policies the average American adult would consume 1,200 less calories a year. We would like to see a policy put in place such that healthy fruits and vegetables are the more affordable choice for families, and would love to hear ideas on how this could be effectively achieved.

The Expanding Human Biome: Metabolism and Microbes

Moderator: Robert Zimmerman, MD

Speaker: Stanley Hazen, MD (Cellular and Molecular Medicine Vice Chair, Lerner Research Institute, Cleveland Clinic, Cleveland, OH)

In a packed morning plenary session, Dr. Hazen urged the audience to “take a step back and look at the whole metaorganismal pathway, not just human metabolism; perhaps the microbiome is the next big druggable target.” We’ve heard Dr. Hazen present on the microbiome’s role in human metabolism and cardiovascular risk before at Day #3 of the Cleveland Clinic Innovation Summit. His work focuses on the production and clinical effect of TMAO (trimethylamine N-oxide), a microbial byproduct of intestinal bacteria from the digestion of choline and carnitine (nutrients commonly found in animal products). TMAO has been shown to be very strongly correlated with CV disease and atherosclerosis as well as predicting up to a 2.5-fold increase in risk for death, non-fatal myocardial infarction, and stroke. Interestingly, when TMAO is given to germ-free animals, TMA (the precursor to TMAO) and TMAO are not produced. His findings open up a novel field of pharmacotherapy; instead of treating the patient’s metabolism, future drugs may target the metabolism of the patient’s gut bacteria. Dr. Hazen was optimistic about gut flora enzyme inhibitors (GFEIs) as one such future therapy. GFEIs block TMA production rather than killing gut microbes, and thus effectively halt TMAO production, thus reducing atherosclerosis risk and suppressing a pro-thrombotic phenotype. We were also glad to hear Dr. Hazen add, “Let’s not just look at CVD; this is really just the beginning. We should start looking at diabetes and obesity therapies similarly.”

• We were eager to hear updates on Dr. Hazen’s currently unpublished data, notably that elevated TMAO levels are not only strongly linked to atherosclerosis, but also thrombotic events like myocardial infarction and stroke (p<0.001). Dr. Hazen found that brief exposure to TMAO enhances human platelet aggregation. Surprisingly, within very brief incubation, platelets seem to be more hyper-reactive in the presence of high TMAO. This platelet activation suggests that enhanced intracellular calcium was mobilized, which Dr. Hazen’s group found: increased TMAO dose dependently enhanced thrombin-induced platelet cytosolic Ca²⁺ mobilization.
• Dr. Hazen’s current work also showed the success of transplant susceptibility for pro-thrombotic phenotype by shifting the gut bacteria composition. Transplanting gut microbes of high TMAO, athero-prone donor mice into germ free recipients on high choline diets resulted in high atherosclerosis compared to germ free recipients with transplants from low TMAO, athero-resistant donor mice.
  
  • Similarly, in preclinical human studies, choline supplementation promoted a pro-thrombotic effect. Enrolled subjects were either vegan (baseline of 2.6 TMAO uM) or omnivores (baseline of 2.8 TMAO uM). After just one month on choline supplementation, mean vegan TMAO levels increased 10 fold to 27.3 uM and omnivore TMAO levels increased 13 fold to 36.4 uM. This data suggests that for those chronically exposed to choline-rich foods (mainly animal products), microbial environments shift to make more TMAO, and thus increase susceptibility to atherosclerosis.
• In terms of new diagnostic and therapeutic approaches for the detection/treatment of CVD, Dr. Hazen suggests inhibiting gut flora specifically with gut flora enzyme inhibitors (GFEI). Rather than killing gut flora (which current approaches like statin treatment do), GFEI’s block TMA production via small molecule inhibition of microbioal choline TMA-lyase activity. Thus, GFEI’s effectively prevent the formation of the harmful byproduct TMAO. Dr. Hazen’s recent preclinical work on GFEI’s show that even on a high choline or high carnitine diet, animal models exhibit substantially reduced plasma levels of TMAO and diet-dependent increase in atherosclerosis is suppressed. The clinical implication of this novel therapy would mean the attenuation of a pro-thrombotic phenotype.
• We’re eager to continue following the growing body of microbiome research in the coming years, particularly Dr. Hazen’s work at Cleveland Clinic. Dr. Steve Nissen of the Cleveland Clinic illustrated the magnitude of this work’s potential at the 2013 Cleveland Clinic Innovation Summit by saying, “If you look to the right and to the left of you, one of the three of you will die of cardiovascular disease. So when we come up with a new biomarker for CVD, it has a very big implication,” even projecting that this discovery could lead to a Nobel Prize for Dr. Hazen. The gut microbiome continues to be a growing research field, and the NIH has offered a “couple hundred-million dollars” worth of research grants to 30 institutions studying the microbiome.
• Dr. Hazen also shared his previous background research which showed that dietary choline induced atherosclerosis in the presence of intact gut flora along with formation of TMAO. If apoE-KO mice were fed dietary choline (found in red meat, egg yolks, and dairy products), they had above normal rates of atherosclerosis. However, when comparing animals with intact gut flora and those on antibiotics (both on high choline diets), mice with intact gut flora had increased atherosclerosis (p=0.04). Additionally, comparing control mice with intact gut flora on normal CHOW diet to mice with intact flora on high choline diet, the mice on high choline diet again had increased atherosclerosis (p=0.01).
  
  • Dr. Hazen also noted carnitine, an abundant nutrient in red meat, is also pro-atherogenic and is converted to TMAO in the gut. In a clinical study of human carnitine tolerance (which Dr. Hazen laughingly noted was his fastest clinical trial completed), people were given grilled filet mignons and a d3-carnitine supplement after which TMAO production consistently spiked.  Afterwards, participants were given antibiotics to suppress gut flora and TMAO levels immediately dropped.
  • In an interesting aside, Dr. Hazen stepped onto a self-proclaimed “soap box” to disparage the beverage industry saying, “Carnitine is the single most abundant additive, and it undoubtedly has long-term health effects on cardiovascular health. Truly, human studies have failed to ever demonstrate athletic, energy, or weight-loss benefits of carnitine. Nonetheless, many energy drinks add it and kids are drinking these daily. I don’t think microbes care where carnitine comes from, whether it’s red meat or energy drinks. We are seeding a generation of increased risk for CVD.”

In-depth Symposia: Inflammation in Diabetes: The New Frontier of Intervention

An Aspirin a Day Keeps the Diabetes Away or is Aspirin the Wrong Salicylate?

Steven E. Shoelson, MD (Joslin Diabetes Center, Boston, MA)

In a two-part symposium on salsalate led by Dr. Steven Shoelson and Dr. Allison Goldfine, Dr. Shoelson gave the introductory half on basic background of salicylates and the Stage 1 clinical trial data from the TINSAL-T2D (Targeting Inflammation Using Salsalate in Type 2 Diabetes) study. Stage 1 of TINSAL-T2D evaluated varying doses of salsalate in 108 participants with type 2 diabetes for 14 weeks and found slight A1c reduction from baseline mean A1c of 7.7 in each of three salsalate dosage groups vs. placebo. His presentation was followed by a robust Q&A discussion, even inciting commentary from Dr. Edward Norton (Joslin Diabetes Center, Boston, MA) on the next 10 yrs of the Diabetes Prevention Program.

• Stage 1 of TINSAL-T2D showed mean A1c changes of -0.36% (p=0.02) at 3.0 g/day, -0.34% (p=0.02) at 3.5 g/day, and -0.49 (p=0.001) at 4.o g/day. Other markers of glycemic control also improved in the three salsalate groups as well as circulating triglyceride (p<0.02) and adiponectin concentrations. Despite this minor glycemic improvement, mild hypoglycemia was more common in the salsalate groups. However the documented hypoglycemic events occurred only in patients taking sulfonylureas in their concurrent background medication. Urine albumin concentrations also increased in all salsalate groups vs. placebo. Additionally, salsalate was well tolerated with high compliance and subjects reported improvements in sense of health and well-being including improvement in both physical and social behavior.
• As salicylates have historically provided robust and sustained lowering of blood glucose, Dr. Shoelson framed these compounds with clinical potential questions: 1) Can salicylates offer clues to better understanding the molecular pathogenesis in insulin resistance, T2D, or CVD? 2) Would that knowledge allow us to identify pharmacological targets or new treatment strategies? Dr. Shoelson commented that the NIH has actually sponsored a trial on the latter question, but did not provide details or start dates. For background, Joslin became interested in studying salsalate, an anti-inflammatory drug, after Dr. Schoelson identified inflammation as a factor in the development of type 2 diabetes.

Questions and Answers

Comment: You alluded that metformin may act on AMPK activation. Have you or others looked in vitro? Surprisingly, there are quite a host of papers that target the NF-kB protein complex, and there is nice repetition.

Q: Does that suggest a potential overlap of activation of these pathways in individuals on metformin that were then receiving salsalate? Does that minimize the effect of salsalate as you tease out the background medications in your studies? Or are there differences in terms of other methods/pathways or foresee this as an additive effect?

A: Great question; we are learning more. Two molecules may be activating, but does that mean that they’re less effective because they don’t directly target the same thing? Would they enhance each other’s effect in a synergistic or additive manner? We simply don’t know yet. It’s very difficult to study because almost all patients are already on metformin. We have plans to do more exhaustive combination studies in animals, but we’d like to do these in humans too. It’s easy to conceive these studies, but hard to actually carry out.

Q: Fundamentally, where do you see salicylate working? For example, do you see it working in post-prandial glucose? Could explain the physiology, any cardiomics, DNA fingerprinting, etc.? Lastly, what broader impact do these drugs have?

A: First of all, how broadly do these work? We do have some information about broadness of study. AMPK is driving a lot of things, so it’s difficult to pinpoint any particular pathway involved other than NF-kB. We have been investigating though and have data from people who have done AICAR studies. And yes we’re also interested in genomic effects.

As for the physiology, we haven’t nailed it down yet. We saw some effects on insulin sensitivity, but others showed less effect on insulin sensitivity. We might suspect that this might work in similar ways as metformin, i.e. inhibiting hepatic glucose production.

Dr. Peter Reaven: To add, we looked at patients with impaired glucose tolerance and we didn’t find improved sensitivity. However, fasting glucose did drop and that is a common thread that we see in all our studies – stronger effects of glucose lowering – and that’s consistent with the AMPK/NF-kB pathway we hypothesize is in effect for inflammation.

Q: A broader question since we’re focusing on inflammation and there are many different pathways that involved, are we overconfident that we can understand such a complicated process? Do we believe we can find a drug that will block these appropriate pathways and decrease complications without inadvertently increasing other complications? It makes me think of where we were 10-15 yrs ago when we thought all diseases could be explained by oxidative stress. We went after anti-oxidants thinking it would fix our situation. Is this pathway going to require multiple inflammatory drugs? I think looking beyond single agents is the direction that we’re going in many of the big inflammation trials.

A: Yes, some of what you’re saying is not even speculation. We’re suggesting that metformin has intrinsic anti-inflammatory properties. We would expect that an anti diabetic drug would have macro- and microvascular disease effects. In particular, microvascular effects aren’t studied sufficiently. To be approved, we need to have a large clinical outcomes trial. Metformin was approved long ago and salicylate has never been tested before our clinical trial. We don’t have data to know clinical outcomes for these simple drugs if they’re safe or beneficial effects. If we did better studies on metformin, we might find out a lot. I would expect effects. For microvascular disease, I don’t know; the studies aren’t there I think. The mechanism seems reasonable to test.

Dr. Reaven: I’d add that there is a UKPDS study showing that metformin has some beneficial effect on reducing CVD, and there is one starting in the US looking at patients with diabetes.

Comment: Linking inflammation, CVD, and diabetes seems to focus on reducing inflammation. My sister is a dentist and we’re interested in the oral systemic connection and where mouth is a reservoir for chronic inflammation. The dental literature seems to be looking more carefully than we are. I’m surprised that we’re not hearing more about this [inflammation] in endocrinology conferences. My sister and I both traveled to India regularly and one thing that struck us is that the people developing diabetes have much lower BMI in the face of a lot of food insecurity. However they have developed insulin-requiring type 2 diabetes very early on with pervasive nephropathy. Strikingly, the state of oral health is very poor. What do you think of this connection?

A: One of the prevailing thought processes is that Indians are historically much thinner so small amounts of added weight and a BMI of even 23 or 24 might be too much for these individuals. That might be metabolically synonymous for them and cause of inflammation that is promoting diabetes. The dental health component is very interesting, and though all the connections aren’t in, people are suggesting that it is promoting CVD. We don’t have the data just yet for a diabetes connection. When we think about gastrointestinal flora, those byproducts have an important role on vasculature. When you think about the many different ways bacterial flora contribute to all these processes, it is a very broad picture.

Dr. Edward Horton: I wanted to mention for the audience that we’re [Joslin Diabetes Center] now planning the next 10 yrs of the Diabetes Prevention Program, and one group was randomized to metformin early on. Two key questions: First, what is the effect of metformin on CVD? Secondly, what is its effect on other malignancies? With the DPP, we have a unique population that has now been studied for 18 years, and further study could provide much more detailed information. We have slightly more than 1000 people in the group that has been on metformin since the beginning of DPP and they have excellent retention (over 85%). Though it’s not a huge N, we are looking to get some answers.

From Salicylate to Methotrexate – What is the New Pharmacopeia?

Allison B. Goldfine, MD (Joslin Diabetes Center, Boston, MA)

Dr. Goldfine picked up where Dr. Shoelson left off, presenting on Stage 2 TINSAL-T2D results, showing that after 48 weeks of salsalate treatment, the mean A1c level was 37% lower compared to the placebo group. Additionally, the decrease in fasting glucose concentration was 15 mg/dl greater in the salsalate group vs. placebo. We note that while positive, the glycemic improvement was clinically minimal, and we remain hesitant about the widespread success of the drug as those on salsalate therapy also had increased hypoglycemic events (before adjusting concurrent background treatment), slight weight gain (+1.3 kg/2.9 lbs on average), and modest increases in total cholesterol (+8.6 mg/dl) and LDL (+11.2 mg/dl). However, Dr. Goldfine emphasized that while the estimated glycemic effects were clinically modest, those results could have be attenuated by changes in concomitant diabetes medications. Additionally, she stressed that a longer-term, risk-benefit profile need to be established before widespread use for diabetes patients, citing her current work on the TINSAL-CVD to evaluate salsalate’s effect on coronary artery plaque volume.

• TINSAL T2-D Stage 2 was a 48-week multicenter, double-masked, placebo-controlled phase III trial conducted in 268 participants with type 2 diabetes whose blood glucose was inadequately controlled on current diabetes medications. Salsalate was administered at 3.0 g/day for two weeks, then escalated to 3.5 g/day, as tolerated, divided into three daily doses, or a matching placebo. Baseline characteristics included mean A1c of 7.7%, fasting glucose levels of 151 mg/dl, total cholesterol levels of 165 mg/dl, and triglyceride levels of 157 mg/dl. Dr. Goldfine also noted that 88% of the participants were already on metformin, 5% were on lifestyle alone, and a vast majority was already on mono or dual therapy (41% and 49%, respectively).
• The primary endpoint found a mean A1c reduction of 37% vs. placebo (p<0.001) and average reduction of 15.1 mg/dl over the 48 weeks. However, Dr. Goldfine tempered this slight success by noting that hypoglycemia was more common in the salsalate treatment group. Hypoglycemic events reached ~55 events within the first 10 weeks of the trial for the salsalate group compared to <10 for the placebo group. However, re-adjusting background medication did reduce the number of hypoglycemic events to <10 at end of trial (matching placebo group). Dr. Goldfine noted that these changes in concurrent background therapy may have reduced the magnitude of difference in the primary endpoint of A1c change between the placebo and salsalate groups.
  
  • The salsalate group also showed improvement in markers associated with coronary risk: a 9% reduction in triglycerides, 27% increase in adiponectin (potentially cardioprotective protein), and 18% decrease in uric acid (associated with cardiometabolic conditions and renal disease progression).
  • Conversely, salsalate therapy did have some negative effects including an average 8.6 mg/dl increase in total cholesterol, 11.2 mg/dl increase in LDL, and slight 1.3 kg weight gain from baseline. While mean increase of 2.9 lbs is small, we believe weight gain remains a sticking point for most patients and providers to adopt a new type 2 diabetes drug.
  • Overall the drug was well tolerated, with few adverse events. Specific adverse side effects that occurred with greater frequency in the salsalate treatment group included tinnitus (n=16 events vs. 7 in placebo), weakness/fatigues (n=13 vs. 9 in placebo), and muscle stiffness (n=14 vs. 9 in placebo).
• Dr. Goldfine is currently leading a study, TINSAL-CVD, to evaluate salsalate’s effect on coronary artery plaque volume in participants with established coronary artery disease (n=264). According to ClinicalTrials.gov (identifier: NCT00624923), the primary completion date is estimated for January 2015. We will be eager to see results of this study as it might tip the scales in salsalate’s favor in terms of the benefit-risk ratio. Dr. Goldfine offered some color, stating, “Looking at MDCTA [multidetector computed tomographic angiography] pictures, you can see quite beautifully the areas of non-calcified plaque and whether or not plaque volume changes in patients after taking salsalate.”
• Dr. Goldfine also noted that other interleukin-1 (IL-1) pathway and general disease-modifying antirheumatic drugs (DMARDs) are being studied, including the investigation of chloroquine and methotrexate. In specific, she reviewed the ongoing NHLBI Canakinumab Anti-inflammatory Thrombosis Outcomes Study (CANTOS) (ClinicalTrials.gov identifier: NCT01327846) which will enroll 17,200 adults with post myocardial infarction and hsCRP ³ 2 mg/l despite the stable use of standard secondary prevention therapies, including statins. This study will primarily assess nonfatal MI, nonfatal stroke, and cardiovascular death; and secondarily, mortality, new onset diabetes, and other cardiovascular events.
  
  • Additionally, there is an ongoing study for low-dose methotrexate (LDM) – the Cardiovascular Inflammation Reduction Trial (CIRT) that will directly test the inflammatory hypothesis of atherothrombosis. This randomized, double blind, placebo-controlled trial will evaluate whether LDM given at target dose of 15 to 20 mg weekly will reduce rates of recurrent myocardial infarction, stroke, or CV death among patients with prior history of myocardial infarction and either type 2 diabetes or metabolic syndrome. NHLBI-sponsored enrollment for N=7,000 started in March 2013 at 350 US and Canadian sites. We will be eager to see these results as methotrexate is already being used by millions of patients as a first-line, inexpensive therapy for rheumatoid arthritis and psoriasis. Additionally, it has an enviable safety record, and guidelines for safe use already exist.

Questions and Answers

Q: You do have two peaks of hypoglycemia in the Stage 2 TINSAL-T2D trial, and you explained what happens after the first spike in terms of adjusting concomitant therapy, but I’m curious, what’s causing the second spike in hypoglycemic events?

A: The concomitant medication definitely did play a role in the first peak in hypoglycemic events. In fact, the events were increased 6-fold in those individuals on sulfonylureas. There may be potential benefits that occur over time and there may be some individuals who change lifestyle over the course of the study and thus may require less drugs, perhaps causing this second spike.

Q: Does the dosing regimen allow for any drops in salsalate?

A: Yes, we did allow for drops, but we tried to keep them in the maximum tolerable dose.

Q: If you eat a big mac, you get a postprandial rise in glycemic markers. Does salsalate prevent such a postprandial rise?

A: we don’t’ measure all postprandial biomarkers, but we could look in smaller cohorts. The greatest effect seems to be in fasting blood glucose. If you look at the delta area under the curve its very comparable, but the total area is somewhat less.

Q: Something that occurs frequently, you have a substantial rise in adiponectin. What is the potential mechanism required for this? Is it the AMPK pathway? Have you or Dr. Shoelson tried looking at salsalate in adiponectin receptor knockout animals? Could this determine one or more of these metabolic effects?

A: There is a close relationship between adiponectin and LDL, so yes you may be correct.

Q: You described in your manuscript that there is modest elevation in LDL cholesterol with salsalate and it appears to be relatively consistent, but there is a drop in triglyceride levels. In terms of that inverse relationship, have you been able to look at the drop in TG closely, whether it’s inversely related to LDL or something else?

A: We do find that when Apo-B goes down, Apo-A goes up. So the profile does shift favorably. However, when we measured LDL, it was actually a direct measure and not based on an inverse formulary.

Q: Why would statins, if they are anti-inflammatory have the opposite effect on type 2 diabetes?

A: The mechanism that statins induce in type 2 diabetes is unclear, and there are a number of hypotheses. Some suggest that there is an impairment of insulin secretion. It is not clear completely why those cause dm, but patients that do develop diabetes often have highest fasting blood glucose and metabolic syndrome closest to this conversion. It doesn’t look like there is mounting risk over time when you look at data from the JUPITER study. So why that happens is not completely clear.

Dr. Reaven: So many of our patients are on statins, and that underlying background therapy does complicate study results. In a sub-analysis, when you analyze individuals not on statin, you may see some epithelial function. This is always the problem. When we move forward to larger CV trials, everyone will be on statins in their background therapy. You might wonder whether it is going to be harder and harder to show potential benefits of new agents when looking at people who are in advanced disease stages.

-- by Adam Brown, Jessica Dong, Hannah Deming, Manu Venkat, Jenny Tan, and Kelly Close