American Association of Clinical Endocrinologists 24th Annual Scientific & Clinical Congress (AACE 2015)

May 13-17, 2015; Nashville, TN – Full Report – Draft

Executive Highlights

This report features our full coverage from the American Association of Clinical Endocrinologists’ 24th annual conference held in Nashville, TN from May 13-17. Despite the broad focus on endocrinology topics beyond diabetes and obesity, these two areas commanded a good deal of attention throughout the conference. We consistently noticed that talks in these fields featured packed auditoriums, sometimes standing three rows deep at the back of the hall – just another indication that diabetes and obesity are becoming increasingly top of mind.

Our final report includes in-depth commentary on corporate symposia, meet-the-expert sessions, posters and more. Highlights from this year’s agenda included an update on diabetes technology from Dr. Bruce Bode (Atlanta Diabetes Associates, GA), who discussed Medtronic’s ongoing work to develop patch pumps for both type 1 and type 2 diabetes. The renowned Dr. Irl Hirsch (University of Washington, Seattle, WA) spoke on the practical topic of pump and CGM downloads, while Dr. Ed Damiano (Boston University, Boston, MA) provided a comprehensive overview of the Bionic Pancreas from conception through planned commercialization.

On the drug front, we heard LOTS on SGLT-2 inhibitors – everything from perspectives on euglycemic ketoacidosis to optimism about the potential for cardioprotection. We also heard Dr. John Buse (UNC, Chapel Hill, NC) talk about the cardiovascular safety of incretin-based therapies – noting that CVOTs must be longer and enroll lower-risk patients to best explore possible cardioprotection – while Dr. Ralph DeFronzo (University of Texas Health Science Center, San Antonio, TX) noted candidly that he is optimistic about GLP-1/PYY dual agonists and less positive about glucagon receptor antagonists due to potential safety issues.

Our report below includes a synthesis of some of the meeting’s overarching themes, followed by full coverage of AACE presentations organized into five categories: (i) Diabetes Technolgy; (ii) Diabetes Drugs; (iii) Obesity; and (iv) Exhibit Hall.

Table of Contents 

Diabetes Technology

Diabetes Technology Update for Your Practice

Current Pump and CGM

Bruce Bode, MD (Atlanta Diabetes Associates, GA)

Dr. Bruce Bode provided a phenomenally comprehensive overview of current and future diabetes technology offerings, sharing new details on Medtronic’s, Animas’, Tandem’s and Abbott’s pipelines. Most notably, he revealed that Abbott plans to file FreeStyle Libre with the FDA in June. As he did last year, Dr. Bode shared his view on the advantages and disadvantages of various pump and CGM ofxferings – see our detailed coverage for his tables on the pros/cons of current pump and CGM products (Medtronic, Dexcom, Animas, Roche, Tandem, Asante, Insulet, Valeritas).

  • Dr. Bode shared that Abbott hopes to file its FreeStyle Libre with the FDA in June 2015 – this is the first public timeline we have heard on this front. Abbott recently completed its pivotal trial of FreeStyle Libre at six sites in the US, though study results have not yet been posted. Notably, Dr. Bode suggested that Abbott will not pursue an insulin-dosing claim; however, he also questioned whether that would actually matter – “Who is going to prick really if you don’t have to calibrate?” We wonder if the FDA will mandate some meter calibrations, or if the product embodiment will remain calibration free but with an adjunctive disclaimer on the label – the latter would be a huge boon for patients and for Abbott.
  • Dr. Bode disclosed new details on Medtronic’s pipeline, revealing that the company is working to develop patch pumps for both the type 1 and type 2 spaces. This is the first update we have heard on Medtronic’s interest in a patch pump product since it was revealed at the 2014 Analyst Day that the technology was no longer in the company’s pipeline. We assumed that Medtronic had compromised with its MiniMed Flex “hybrid pump” (a durable pump that can be worn on or off the body), though judging from Dr. Bode’s slide deck, it seems like these are separate projects. Details were not shared, though it is good to hear there is at least some movement within Medtronic on this front. Timing, of course, is a big question mark.
  • The interest in a type 2 patch pump is not altogether surprising, given Medtronic’s increasing move into type 2. The business model is certainly different from the company’s traditional pump business, though the new management team is clearly focused on trying new things – the slew of recent announcements have included: DreaMed artificial pancreas algorithm licensing; Diabeter clinic acquisition; Glooko investment; and a new partnership with IBM. Of course, these announcements make headlines, but the cadence of products is ultimately a test, and we look forward to hearing specific timelines and product launches.  Dr. Bode revealed that Animas’ Calibra Finesse bolus-only insulin delivery device is going into clinical trials across the US. In J&J’s 1Q15 call, management shared that the Finesse is still “a couple years out,” which would put commercialization in 2017 at the earliest. Given management’s conservative commentary on the technology of late, it is good to hear that J&J continues to prioritize the device on some level. Of course, this project has moved quite slowly, given the Calibra acquisition occurred in July 2012.
    • Though he did not share details, Dr. Bode also hinted that LifeScan is planning to come out with a new web-based, data-downloading platform called the “Reveal.” This is the first we have heard of this system, and we don’t know what it will look like. As a reminder, J&J is one company that has not agreed to partner with Tidepool, and the news makes us wonder how the launch of J&J’s own web-based proprietary software might play into that strategic decision.
  • Tandem has reportedly faced reservoir compatibility problems with glucagon, a challenge that has stunted the development of a dual-chamber pump. In its 1Q15 call, management was pretty clear that the JDRF-partnered dual chamber pump is definitely not a near-term project, especially relative to insulin-only closed-loop systems (possible late 2017 launch). Interestingly, the BU/MGH group have not reported similar issues with the Tandem pumps they use, though their academic research studies swap out the glucagon at least daily – perhaps Tandem has been trying out three-day use, or is using the Xeris formulation instead of the reconstituted Lilly/Novo Nordisk glucagon. Given all the questions surrounding glucagon’s cost, as well as Tandem’s limited R&D funds, we do think it’s smart to prioritize insulin-only at this point.
    • On Tandem’s pipeline, Dr. Bode commented that the 480-unit t:flex “will be available very quickly”; this is consistent with management’s 1Q15 commentary putting shipment in June 2015. According to Dr. Bode, the current challenge is establishing capacity to meet demand.
  • Dr. Bode closed his presentation by offering stirring commentary on the utility of patch pumps in the type 2 space – “you take any of your patients failing MDI and put them on this, you’ll be shocked.” We have certainly been saying for years that there have been many many happy patients with Valeritas. He suggested that patients fail MDI because of non-adherence to insulin regimens, an obstacle that around-the-clock patch pump therapy addresses directly. We agree with Dr. Bode that this area has tremendous potential to help many patients out there, though it has suffered from not enough industry focus – Valeritas’ V-Go is the only product available in the US, and the company has only rolled it out regionally. As a reminder, Valeritas filed for an IPO earlier this year, but postponed the offering in March. Still, the interest of big players like Medtronic and BD suggest the type 2 simple patch pump landscape has upside.

Questions and Answers

Q: I have heard Animas reps claim their pumps are more accurate in terms of basal and bolus delivery. What do you think of their data?

A: When you get into accuracy, they are probably slightly more accurate. They have a publication that just came out. It looks at 0.1 increments of insulin doses and how often they are above that. Does that mean much? In the big picture of things, if most people are off by a half unit of insulin, it doesn’t matter. If you are at five units a day, it matters. But if you are at 30-40 units a day, it doesn’t matter. Some people are talking about 0.01 units increments that do not make a difference except in unique individuals. There are so many issues with tubing and length of tubing – so many variables involved – that it’s not so clinically relevant.

Q: Can you talk about active insulin? Is there a lot of value in pre-bolusing?

A: I set active insulin at the true duration of insulin, which is 4 hours. Pediatric providers say that when you give smaller dose it doesn’t last for four hours. But in a clamp study, it does. Pediatric patients do sometimes take a partial dose up front. So to do that, providers might put the active insulin at two hours. We’ll have good data at ADA that if you use faster-acting insulin, there is a significant decrease in postprandial spikes with a significant decrease in hypoglycemia. That’s also been shown with inhaled insulin very clearly.

Q: Can you talk about the possibility of DKA with SGLT-2 inhibitors? What is the mechanism of action?

A: SGLT-2 inhibitors activate the ketogenic pathway. You spill glucose into the urine and that masks the ketoacidosis. Even at 150-200 mg/dl, we are seeing patients going down a ketogenic pathway.

[Comment]: I just wanted to clarify. I’m on a t:slim pump. It’s nice and easy to use. They do have a new algorithm. You used to have to fill the cartridge after you put it into the device, but now you can put the cartridge in after you fill it. And off-label, you can go about five refills without changing it.

Q: How do you look at data in office? What is your process?

A: Once you get comfortable with one software, you tend to stick with that. And once you use that system, you get good at it. If you have five systems, you want to pull your hair out. Ultimately, you need smart software. Tidepool will hopefully work for everybody. They have to be good though. They are working for everybody. Right now, manufacturers have different systems. Overall, I can’t live without a download. I’m wasting time without one.

Q: How do you get patient to change dosage in between visits?

A: With pediatric patients, I try to teach the parents how to make changes on their own. And they do it. But how does a patient do it? If they are comfortable, I have them do it themselves. Otherwise, I have them upload the and then we go into the web-based server and educate them. I provide that for free. As a note, I’m starting to charge for looking at real-time CGM remotely. If you don’t charge for it, you’re not going to get paid.

Case Studies: Interpreting Pump and CGM Downloads

Irl Hirsch, MD (University of Washington, Seattle, WA)

Dr. Irl Hirsch (University of Washington, Seattle, WA) spent most of his presentation discussing CGM and insulin pump downloads, and again expressed his belief that within five years, CGM penetration will outstrip that of insulin pumps. An informal poll of the audience revealed that ~50% of the ~200 attendees routinely download CGM and pump data. Dr. Hirsch noted ruefully that this figure is significantly higher than that seen among pediatric endocrinologists he has surveyed. [Of course, the high percentage may reflect some degree of sampling bias from asking AACE-attending endocrinologists themselves. After all, his informal survey of educators at AADE 2014 revealed that only 5% of the ~400 attendees worked with providers who routinely download CGM and pump data.] Ultimately, Dr. Hirsch believes that there are two reasons that providers do not download technology: (i) there is no infrastructure in the office for downloads leading to misconceptions of a lack of time; and (ii) downloading is a new idea – providers never received downloading education in their training. [To that we would add software/cables that are hard to use – of course, this is getting better as offerings like Diasend, Glooko, and Tidepool improve/come to market.] However, Dr. Hirsch emphasized the difficultly in rationalizing glycemic targets without referring to downloaded data – as Dr. Bruce Bode affirmed: “I can’t live without a download. I’m wasting time without one!” Dr. Hirsch concluded that patient education on data downloads is even more preferable, as it enhances self-management and allows patients to begin to own their diabetes.

  • Dr. Hirsch was very optimistic on the future of CGM – he again expressed his belief that within five years, CGM penetration will outstrip that of insulin pumps. In this lecture, Dr. Hirsch focused on what he sees as the most common patient mistakes in using CGM: (i) over- and under-calibration of the sensor; (ii) ignoring the sensor readout and only responding to alarms; (iii) overreaction to alarms and not taking into account lag times (e.g., insulin stacking); and (iv) not using enough SMBG to confirm a final dosing decision.
  • One of the first components Dr. Hirsch looks at in patients’ downloads is standard deviation (SD). Though it is not perfect, the metric provides an indicator of glucose variability and a quick assessment of a patient’s “diabetes fingerprint.” That said, he emphasized the importance of thinking about SD in the context of the mean (i.e., coefficient of variation, the variable of choice in his FLAT-SUGAR pilot study, which he confirmed will be presented at ADA 2015). Several quick and dirty calculations that Dr. Hirsch uses to get a sense of patients’ management include:
    • For SMBG: SD x 3 < mean glucose. This first-line metric allows clinicians to see if patients are giving their insulin at the correct time and regularly matching their food with insulin. That said, Dr. Hirsch noted that patients who can get to this point are often still making some endogenous insulin. SD x 2 < mean glucose is the minimum amount of variability without having excessive hypo or hyperglycemia, independent of HbA1c. This rule only works for means between 120-180.
    • For CGM: SD x 3 < mean. Dr. Hirsch was forward-looking, noting that current CGM downloads often have 3,000 data points, and we need better metrics. In his opinion, coefficient of variation (CV), TIR (time in range), TBR (time below range), and TAR (time above range) all need to be correlated with outcomes, as they are important pertaining to markers of inflammation, oxidative stress, and severe hypoglycemia. His FLAT-SUGAR study is looking at all of these parameters, though coefficient of variation is the metric of choice.
  • “This is much more complicated than A1c itself!” Dr. Hirsch presented his wish list for downloads. For pumps this included basic insulin statistics, TDD, % basal, % use of bolus calculators, time-specific bolus calculator over-rides and under-rides, and a daily summary to better understand ICR, ISF, basal rates. For CGMs, Dr. Hirsch requested basic blood glucose statistics, overall patterns, and daily decision making to best understand how patients think through challenges.
  • Dr. Hirsch shared that the “timing of insulin is the most important part of dosing.” He said that in the context of the many variables that influence the size of a dose – carb counting, insulin on board, sensitivity factor, etc. – and the current speed of insulin, it is not the dose that usually matters as much as when it is given – “If you give insulin 10-20 minutes after you eat, you’re toast!”
  • Dr. Hirsch believes that the biggest limitation of insulin bolus calculators is that they always assume glucose is unchanging at the time of advice – indeed, he believes that “trends trump insulin on board.” Dr. Hirsch finds that patients who over- and under-ride their bolus calculators ~20-25% of the time tend to be in the best control; those who never over- or under-ride are often not adjusting for glucose trends – of course, this requires patients are on CGM, which is still not usually the case at this stage of penetration. From a clinical perspective, he emphasized that a patient with a blood glucose of 200 mg/dl with “two arrows down” is very different from one with “flat arrows” or “two arrows up.” We would certainly agree, and this was a major point of our November 13 presentation at the FDA’s Workshop on Bolus Calculators and Interoperability.
  • Dr. Hirsch – along with his co-presenters – cited concerns over reimbursement as the most formidable barrier to wider adoption of data downloading. There was no consensus on a solution, but it seemed like nothing short of structural reimbursement reform will provide a long-term solution. We agree, though the mechanics of making this happen continue to allude us. Below, we have included some of the some notable quotes from Dr. Hirsch’s presentation and the resulting Q&A:
    • “The issue is that diabetes is a very time-intensive disease. I’m not convinced that the way we do it is the most efficient way. I don’t think we use the web enough. We should have web classes. We should have Skype and Facetime classes. However, where I live in particular, we can’t get reimbursed for that kind of therapy. We need to though. We have that kind of technology available.” – Dr. Hirsch
    • “We have to mobilize the online patient community. We talk about legislative approaches and ways to get reimbursed. This is something we can’t ignore. If we sit on our butts, then legislators and governments will determine our payments.” – Dr. George Grunberger
    • “I can’t live without a download. I’m wasting time without one!” – Dr. Bruce Bode

Questions and Answers

Q: I’ve got a lot of patients who are pumpers. I also have patients who refuse to buy into that. I don’t know what to tell those people. Do you think patients should sign a contract? How do you deal with that? I’d say that’s 30% of my insulin-treated patients.

A: I do the best I can. We’ve had people die from hypoglycemia. We are learning more about the mechanisms of this. All I can do is be a partner. Yelling at them and scolding them does not work. I sleep at night knowing that some patients aren’t going to play the program.

Q: With respect to exercise physiology, how do you deal with the fact that 45-60 minutes of aerobic exercise is different from 30 minutes of anaerobic exercise?

A: The good news on exercise is that most patients tend to repeat exercise patterns day after day, or week after week. So we can individualize our treatment. If someone is on MDI or pump and is going to bolus within 3-5 hours of exercise, I have them cut down on their bolus dose. Patients on pumps, I’ll have them cut their basal dose. And with patients on MDI, that’s why I like to use glargine or detemir two times a day, because then you can cut one of them down even if they are not on a pump.

Q: How do you set your basal rates on a pump with patients? Do these vary?

A: With pump patients, one of the problems is that absorption on day 1 is different from that on day 3. It’s difficult.

 [Comment]: I had a similar population with at least 30% of patients that didn’t want to be on pumps and would even fake their blood sugars. I think there is a gene that all of us diabetics inherit for lying and non-compliance. The point is that education is a big factor here. We’ve been doing team education in our office for 15 years. It is basic to what these patients do. You cannot yell at them. If you teach them why you’re doing certain things, that helps. If they understand why, it can help. Unfortunately, governments and insurance companies do not pay for education anymore. That’s so unfortunate because that’s the basis of therapy. We have data that shows that patients who get education do better. We also know that lower socio-economic groups tend to do worse, and we have cultures that tend to do worse. We need studies that look at those cultural issues and get governments to appreciate those challenges.

A: The issue is that diabetes is a very time-intensive disease. I’m not convinced that the way we do it is the most efficient way. Having said that, I don’t think we use the web enough. We should have web classes. We should have Skype and Facetime classes. However, where I live in particular, we can’t get reimbursed for that kind of therapy. We need to though. We have that kind of technology available. I agree with everything you said. Yelling and screaming makes no sense at all.

Panel Discussion

Timothy Bailey, MD (UCSD, San Diego, CA), Bruce Bode, MD (Atlanta Diabetes Associates, GA), and Irl Hirsch, MD (University of Washington, Seattle, WA)

A panel discussion with Drs. Timothy Bailey, Bruce Bode, and Irl Hirsch shared highly valuable and very opinionated insights on real-world obstacles to closing the loop, the off-label use of SGLT-2 inhibitors in type 1, the misconception of data downloading as time consuming, and more.

  • Euglycemic ketoacidosis was again raised as a safety issue with use of SGLT-2 inhibitors in type 1 diabetes. We heard consensus that such off-label use should be limited until the field has better adverse event data on the use of the drug class in the type 1 population. Dr. Bode was adamant in advising providers to “hold of using [SGLT-2s] in type 1s.” Indeed, while education and regular ketone checks may mitigate the problem, Dr. Hirsch suggested that this is still an area of serious concern that is developing; he was relatively reserved in his commentary as he noted that he is involved in a publication on the topic that should be coming out shortly (presumably for communication to the FDA and/or via publication to the clinical community).
    • Although little is known about the mechanism behind the effect, Dr. Hirsch suggested that a key contributor may be the fact that SGLT-2 inhibitors allow patients to lower their insulin doses significantly. Dr. Hirsch did acknowledge that most cases of euglycemic ketoacidosis in his experience have been linked to Invokana (canagliflozin), though this is most likely a result of its first-in-class status. We expect this to be a very closely-watched issue by HCPs (especially the KOLs who are now well-attuned to it) for SGLT-2 inhibitors in type 1, especially given that multiple influential thought leaders appear to be seeing similar things (e.g., Dr. Anne Peters at ENDO 2015). Certainly, the ongoing phase 3 studies testing SGLT-2s in type 1 should bring out clearer data. It is early to say if this issue will keep the entire drug class away from type 1 diabetes patients; on the plus side, Dr. Bailey shared that the class has yielded some strikingly positive improvements in A1c and glycemic variability in his type 1s.
  • In response to a question from the highly regarded Dr. Robert Vigersky, all three panelists emphasized that the notion of data downloading as time consuming is partially a misconception. Said Dr. Hirsch, “The time isn’t that big an issue because we all know what we want to look at.” All three panelists estimated that the time required to analyze a data output is typically in the single digits except in extreme cases when a “pattern” is not immediately evident (usually because of a lack of data). Instead, Dr. Bailey suggested that the rate-limiting part of the process is that data downloading is “synchronous” with appointments – patients do not download data before coming into the clinic, which means precious in-clinic time is wasted on uploading the data while the actual analysis is quick. We heard widespread agreement that developing a system (e.g., taking a patients CGM/pump from them while they are in the waiting room) is a key to expediting this process, though we imagine it is not easy to push forward with such structural changes in all healthcare settings.
  • Dr. Irl Hirsch shared that his biggest concern with the artificial pancreas is the scarring and fibrosis associated with insulin infusion. Dr. Hirsch spoke to the very real and frustrating phenomenon of diminishing insulin absorption at infusion sites, resulting over time in little usable real estate. Importantly, he attributed the lipohypertrophy and scarring to the infusion of the foreign protein itself, not the set. He emphasized that next-gen infusion sets – including BD’s recently FDA cleared design – will help with kinking and comfort though he does not believe they will reduce the incidence of lipohypertrophy. With patients getting on pump therapy earlier and earlier, he stressed that this physiologic issue needs attention – “Every patient that uses a pump, you have to look at skin every visit. It’s more important than looking at their feet every visit.”
    • While we would point out that the efficacy side of closing the loop is a critical question, clearly practical questions on the human physiology side should be at the fore as well. They deserve attention in the near future (especially considering how quickly the loop is closing) and we are glad to hear Dr. Hirsch raising the level of conversation.
  • Panelists concurred that a significant upshot of the artificial pancreas on clinical practice will be the enabling of telemedicine. Dr. Bailey suggested that closed-loop systems will change the therapeutic model from one in which providers “tune the patient” to one in which they tune algorithms. With the availability of data (e.g., insulin usage, glucagon usage, 24/7 blood glucose), we heard consensus that closed-loop systems should facilitate remote monitoring and wide-scale population tracking. This is a benefit of closed-loop not often discussed, but certainly one with interesting potential to scale an increasingly limited number of providers. We remain somewhat skeptical, however, that providers have the bandwidth to deal with remote data/algorithm management, particularly in a US reimbursement environment that doesn’t support it.

Panel Discussion

Q: I use a lot of SGLT-2s off-label in type 1s that are overweight. Have you noticed a reduction in insulin in pump patients on SGLT-2s. I have noticed about a 30% reduction.

Dr. Tim Bailey (UCSD, La Jolla, CA): I try not to prescribe SGLT-2s to type 1s at this point because of concerns with diabetic ketoacidosis out there. I have seen two cases and Dr. Anne Peters [USC, Los Angeles, CA] had three cases. But patients love these drugs. There’s a weight benefit; it helps postprandials; weight comes down. But now we know personally or know of patients with DKA. It’s nothing new in patients, but it’s new for patients on this drug. It’s an education effort that we’re going to have to tell them a few times.

Dr. Irl Hirsch (University of Washington, Seattle, WA): I have to choose my words carefully because our report will hopefully be coming online soon. The patients I am concerned about are those in surgery. What we are realizing is that measuring glucose may not reflect what is happening in the liver. It’s tough because where I live with a lot of Asian patients, they are insulin deficient and you see reduction in insulin usage, but you don’t know if they’re getting enough insulin in the liver to reduce ketogenesis. There’s got to be labeling for these drugs in type 1s. What happens is that we tend to give someone a lower dose, they get the flu, and then they drop their insulin further – that results in DKA. I think the bottom line is that all patients should be doing ketone testing. Anyone on SGLT-2 inhibitors has to have the ability to check their ketones at home.

Dr. Bruce Bode (Atlanta Diabetes Associates, Atlanta, GA): I think the rule of thumb is that you should hold of on using SGLT-2 inhibitors in type 1s. Only phase 2 trials have been finished. Phase 3 trials are starting. This is an issue of prescribing a drug when we don’t have safety data on it. Phase 3 will give us safety data. If you are going to use it, never stop your insulin even if you have someone with a GI virus. You have to be very careful.

Q: Do you notice any dosage or brand differences?

Dr. Hirsch: What I can tell you is that we see the most euglycemic DKA with the compound that has been on the market the longest. But the mechanism is the same for all of these.

Dr. Bode: I would just note that there have been cases of DKA in type 2s with normal glucose. These are patients who aren’t eating anything. If you aren’t eating, bad things are going to happen.

Q: There is a concern about overuse of glucagon and physiological use of glucagon. What safety data do we have?

Dr. Bode: There is very little data out there on the continuous use of glucagon. It’s a question to be answered. I asked Irl if he’ll wear an artificial pancreas. He says he’s out of real estate.

Dr. Hirsch: This is an important generic comment. What do we see with long-term physiological glucagon use in glucagonoma. You get rashes and diarrhea. This is the problem. We have patients that have been on pumps for 20-30 years, and they have literally run out of real estate for the pumps. Am I the only one seeing this? [50% of the audience raises their hands]. We try to use the hip and other sites. I don’t believe it’s the catheter. I think it’s the fact that we’re infusing this foreign protein, and I don’t know if part of the problem is the fact that when they started they were using animal insulin. I had a patient earlier this week – I’ve been trying to get him to take a pump holiday, and I finally got him onto MDI. He came back and told me that “this is the first time in five years that I’ve really got my insulin like I thought I was.” I’ve got to tell you; this is a giant experiment. I’ll give you another example. I saw a lady who started on a pump in 1982 and in she is in her 70s, who suffered from DKA because of the scarring and fibrosis. Every patient that uses a pump, you have to look at skin every visit. It’s more important than looking at their feet every visit.

Dr. Bode: Some patients don’t have issues. Some patients have tremendous issues. We do have new sets coming out, and we’re going to see newer infusion sites that will be more comfortable for patients. But lipohypertrophy is a real issue.

Dr. Bailey: There are combination infusion sets too in Europe. But anything to reduce lipohypertrophy is important. Checking site is very important.

Dr. George Grunberger (Grunberger Diabetes Institute, Bloomfield Hills, MI): We talked about importance of downloads, but on downloads you don’t see the frequency of site changes. That’s another variable.

Dr. Hirsch: You’re right. Changing the infusion site also influences the amount getting absorbed. But on CareLink, you can see it. The bottom line is that with the scar tissue and fibrosis, we need to do skin biopsies to really understand the mechanism of what is going on. I don’t think the new infusion sets are going to have as big of an impact. You’ll see less kinking but it’s the foreign protein– the insulin – that causes the problems we see.

Dr. Robert Vigersky (Medtronic Diabetes, Los Angeles, CA): How much time does it take you to review downloads?

Dr. Hirsch: We have a system. We have now been on Epic for roughly a year. From my point of view, it has been an epic disaster. It has not been good. There have been some good things though. Whether it is a CliniPro download or Diasend, it’s put into a PDF format and it’s put into Epic immediately. So the download can occur at the end of the hall 100 feet away, and I have their data immediately on my computer. On the issue of the time I have with patients and how long it takes to download: Diasend and Animas are extremely slow. The quality is a different issue. The time isn’t that big an issue because we all know what we want to look at. There are certain things I will look at and the main thing is that I look at the last five days of the individual detailed data. Individually, I go through that time period with patients and can spend less than ten minutes just doing that and can figure out quickly if we need to make changes. Because we have system that we have running, it is quick. Now if I was the only endocrinologist in a bigger clinic, it could not go that quickly.

Dr. Grunberger: This is a critical question. Because the more experienced, the faster we get at it. However, most professionals are paid by the hour. So the more experienced we get, the less we get paid. We have to figure out how to sell our brains.

Dr. Bailey: We would love for Tidepool or some other system to get in here. Getting data to the cloud is the rate-limiting step. Medtronic does have remote downloading capability. The cloud is important. The thing is that downloading needs to be synchronous. It takes fifteen minutes to download pump and sensor data. We want to make it more asynchronous – patient uploading (preferably automatic) to cloud so that the data is already there when they come for a clinic visit. In my practice, I take three steps to a download. First, I generate a hypothesis. Then, I query the patient, getting more data and filling in the missing data. Finally, I fill in the blanks and give advice for the next step. The really hard step is figuring out the advice. The longer a patient has been with me, the shorter it takes. Newer patients can take an exorbitant amount of time.

Dr. Bode: It takes me 3-10 minutes to go through a download. If you don’t see a pattern, it’s going to take longer. If it’s all over the place, then you need lot more time. If you see a pattern, it’s going to be easy. Then you have to ask them to go home and do stuff. Our educators typically spend about ten minutes going over the reports. Then, it takes three minutes to go over their data. I analyze it separately. We bill for it. If you don’t bill for it, you won’t get paid.

Dr. Grunberger: To go back for a moment, this asynchronous element is important. We have 500 patients in our practice and only one a year actually downloads the data before they come in. That takes time to do in clinic. How do you get covered for this time? We need to figure out how to reimburse the time and our brains. Those are the resources.

Dr. Bode: You have to develop a system. We download from a lot of devices a day, and we have a system. We download 100 meters a day and 50 pumps a day. Ultimately, I think the end goal is getting smart software. After all, how many people read an EKG today?

Dr. Vigersky: Can you take about integrating this software with an EMR? Many systems are impenetrable from the outside.

Dr. Hirsch: I’m not an IT guy, but our IT guy figured out how to put the data in a PDF format to get it into Epic. Now, I can pull this stuff up instantaneously with Epic. You have to be able to do this integration.

[Comment]: We have a sign in the waiting room that says if you have a pump, CGM, or BGM, then you have to hand it over to an assistant before they begin to fill out forms. While the patient is in the waiting room doing their paperwork, they get downloaded. The actual interpretation takes under five minutes, especially if there is a pattern. Once a system is established, it’s pretty easy. Once you get the system down, it’s just a matter of getting everybody coordinated.

Q: Going back to cannula issue, do you notice differences in kinking in patients with different body fat compositions?

Dr. Bode: I think I would recommend metallic if you are thin. They will crimp more easily if hits muscle.

Dr. Hirsch: The metal cannulas do better in terms of crimping. I think over the years we’ve seen more DKA than we’re willing to acknowledge because of the crimping. People that have serious scarring and fibrosis, they don’t do well with anything we try.

Q: How do you cover a high fat meal? Like a bacon cheeseburger?

Dr. Hirsch: There’s no way to do it well. Dr. Howard Wolpert [Joslin Diabetes Center, Boston, MA] has the data on this. I’ve got to tell you something though; we’re just throwing darts. We don’t want to hit the bull-eyes; we just want to hit the dartboard. This is hard. There is no good quantitative way to get the protein and fat to get the carbs for that. The biggest concern is when you are doing this at night. If you are too aggressive, it is dangerous. That is why the CGM is so helpful. It is just really difficult to do.

Dr. Grunberger: In general, how do you deal with high fat and high protein meals?

Dr. Bode: High fat causes insulin resistance, so you’ll need more insulin. So I tell my patients to deliver 30% more insulin, but I tell them to do it over two hours. However, CGM helps you a lot. You can look back and perfect it. We have few proactive studies looking at this though.

Dr. Bailey: Bolus calculators are very simplistic. It’s really a tool to get patients to enter their carbs. It’s not really that helpful. I think the only thing that is going to solve this is the artificial pancreas. Basal-bolus therapy can’t do it. The artificial pancreas is all bolus. It’s very frequent and small doses that are being constantly adjusted.

Q: How do you deal with patients that are undercounting carbs or not telling the truth about how many carbs they are eating?

Dr. Bode: I think you’ve got to ask those patients what they are truly doing. Some of them might say, “If I put my full carbs in, I will go low.” You have to assure them that if that’s the case, then you’ll change it. People will rig the system. They’ve really got to put in the data accurately.

Q: How might the artificial pancreas modify clinical practice?

Dr. Bailey: It’s going to change the model because instead of tuning the patient, we’re going to be tuning the model. It’s going to be a very different visit. We’ll have all their insulin data and glucose data. We can guess what they are eating. It will enable more telemedicine – think what pacemakers did for the cardiology world. I’m hoping we’ll get some coverage for it, because right now, when we use telemedicine, we don’t get paid for it. The biggest change is it will enable telemedicine.

Dr. Grunberger: I think it will transform how we have these visits. It will define our practice. We don’t want people who are not in our shoes to define how we get paid.

Dr. Bode: It is going to change what you do. You’re now going to help with site selection. You’re now going to have to examine their skin and suggest other areas, especially in long-term pump users. You’re now going to have to decide – if we get better and faster insulin coming out – how to manage that. Day-to-day management of glucose won’t be there. What you do, you’ll be able to do remotely.

Dr. Hirsch: We presented the CGM JDRF study in October 2008. Only 8 weeks ago did my patients get coverage from the two major diabetes carriers in Washington. My biggest concern about the artificial pancreas and any new technology in diabetes is that even if it gets covered for patient, I don’t think we’re doing enough to cover the technology assistance patients deserve. We have failed miserably with CGM.

Dr. Grunberger: We have to mobile the online patient community. We talk about legislative approaches and ways to get reimbursed. The way is that we mobile the patient community and mobilize the consumers of our expertise. This is something we can’t ignore. If we sit on our butts, then legislators and governments will determine our payments.

Dr. Hirsch: One last thing. Someone earlier made a comment about patients not being compliant. That comment really hurt me. These patients have never been taught how to use the technology. It doesn’t mean they are not compliant. That really hurt me.


Bionic Pancreas

Ed Damiano, MD (Boston University, Boston, MA)

Dr. Ed Damiano’s plenary lecture (attended by ~1,800 endocrinologists!) provided a comprehensive overview of the Bionic Pancreas from conception through planned commercialization. The majority of his talk was delivered at a high level – geared toward the broader endocrinology audience – though was headlined by our first-ever look at the group’s new pivotal trial design (12 months, n=480) and an update on the timing of the planned FDA submission (nine months after the start of the pivotal and in parallel with the chronic glucagon arm of the pivotal study). The team is still on track to conduct its pivotal in 2016-2017 with commercial launch in 2018 – we’ve been impressed with the team’s ability to stay very nearly on a schedule originally laid out in 2012 (We say “nearly” because the original plan estimated a commercial launch in late 2017, though the pivotal trial was expected to be shorter; the need for chronic glucagon exposure has now lengthened the timing a bit). Notably, he concluded by sharing updated interim results from the recently completed Bionic Pancreas 11-day multicenter study (from 35 of the 38 total patients), first seen at the GTC Bio Diabetes Summit 2015. He shared for the first time details on glucagon usage in the Bionic Pancreas arm (~0.51 mg/day), without a corresponding increase in insulin dose. We continue to follow glucagon and insulin usage closely in light of recent criticisms that the Bionic Pancreas algorithm is “aggressive.”

  • The Bionic Pancreas pivotal trial will assess the system’s efficacy/safety and chronic glucagon exposure over 12 months. The first portion of the trial will randomize 320 patients to the Bionic Pancreas for six months and 160 patients to usual care (either MDI or CSII) monitored with a blinded CGM. Primary outcomes are A1c and the percentage of time spent < 60 mg/dl during the last three months of each arm. These findings will be complemented by an investigation of a randomized selection of patients (n=107) on the Bionic Pancreas who will continue for an additional six months (“Continued Access Study”) to examine the effect of chronic glucagon in order to inform the chronic use indication. The 160 patients randomized to the original control arm will get to follow their six months of usual care with a three-month “Incentive Study” on the Bionic Pancreas (as a way to motivate their participation). We like the idea, since many will be disappointed to end up in this study’s control group – it should indeed motivate participation and help gather even more data in a streamlined way.
  • Dr. Damiano shared that the Bionic Pancreas PMA will be submitted to the FDA as soon as the final patient completes the Bionic Pancreas arm (~nine months after the start of study) and in parallel with the Continued Access arm. It is great to see the FDA’s forward-thinking approach here and willingness to expedite the process. The device side of the Agency has been increasingly patient-centered in the past year (e.g., declassification of secondary CGM data) and it is great to see similar appreciation for the value of the Bionic Pancreas from the FDA’s drug wing.
  • Dr. Damiano presented the latest interim results from the recently completed Bionic Pancreas 11-day multicenter study. Patients had a mean blood glucose of 141 mg/dl (estimated A1c of 6.5%) on the Bionic Pancreas vs. 162 mg/dl (7.3% estimated A1c) on usual care. In our view, “usual care” would be better in this trial than in “real life.” Dr. Damiano also made a point of calling out the narrow inter-patient variability on Bionic Pancreas (SD: 10 mg/dl) vs. on usual care (SD: 29 mg/dl). This is consistent with previous trials where we have seen the Bionic Pancreas’ ability to bring patients into a narrow range despite patients’ wildly varying day-to-day insulin demands. Time spent in hypoglycemia (<60 mg/dl) was reduced by more than two-thirds: 0.6% on the Bionic Pancreas vs. 1.9% on usual care. Notably, these are motivated patients in superb control to begin with, and we are impressed to see that the Bionic Pancreas is able to bring them safely into even tighter control (indeed, one thing we have heard is that someone with a 5.5% A1c went into the trial – and did better!).
    • Dr. Damiano stressed that the Bionic Pancreas has not delivered more insulin than usual care [0.63 units/kg/day usual care vs. 0.66 units/kg/day BP] nor has there been excessive glucagon usage (~0.51 mg/day bionic pancreas). This is notable in light of recent criticisms that the Bionic Pancreas’ use of insulin and glucagon is “aggressive.” At ATTD 2015, Dr. Roman Hovorka reiterated concerns that in the Beacon Hill study, participants had a 32% increase in their daily insulin dose during closed-loop. He implied that the increase results from a “stop-and-go” approach to hormone administration that, in his eyes, aggressively doses insulin and depends on glucagon to counteract that “excessive” administration. So far, it appears that the multicenter trial has differed from the data gathered in Beacon Hill, and perhaps these criticisms will be less of a concern going forward.
  • During a follow-up press conference, Dr. Damiano shared that the group’s current plan is to pursue an indication for the Bionic Pancreas in patients 10 years and older, though he is hoping this can be pushed lower. Apparently, the FDA is actually encouraging the group to aim for a younger indication as they have acknowledged that the Bionic Pancreas will likely be used off-label in the younger population if not indicated. The Agency would rather have the safety data and official indication than have patients use the device off-label – a very pragmatic take in our view. As a reminder, the Bionic Pancreas has been tested in patients as young as six years old (Summer Camp Study 2014).

Q&A from Dr. Damiano’s Press Conference

Q: Can you talk about plans for establishing secure cybersecurity?

A: This is something that the medical device industry hasn’t handle well until now. We have a person dedicated to working on that. We are working on ways to minimize risk. We are going to have as secure system as we can.

Q: What is on your possible list of things that may go wrong?

A: Let me be clear – we’re never going to build a system where we obviate every challenge. There will be times when things go wrong. Things go wrong for airplanes. When you put your life in the hands of a device, there will be cases when things go wrong. You try to make a list and mitigate these things as best you can. But circumstances will happen. However, I would remind you that you’re dealing with an at-risk population to begin with where things are already going wrong all the time. We want to reduce that to zero. Obviously, and we’ll do our best.

Q: When might this be available?

A: I would estimate 2018. I was hoping for the fall of 2017, but because of certain extenuating circumstances, it’s looking like 2018. The pivotal trials were going to be six months, but now they are one year because we are looking for a new indication for glucagon. We need a chronic use indication. We have to have some subset exposed for an additional six months. The FDA really wants to see people on this device for a year. However, to expedite the process of getting this into patient hands, the FDA has offered to let us submit the PMA after the six-month RCT study is over. The chronic glucagon patients will carry on, while the application is looked at. The FDA is really looking to help, but that extra six months is going to push us into 2018.

Q: What ages are you looking to get this indicated in?

A: We’ve proposed ages 10 and above. We may go lower than that. The FDA is really encouraging us to push that as low as we can. They know that this is going to be used off label, so they are looking to push this as low as possible.

Q: Is this dependent on funding from the NIH?

A: Yes. I have a proposal in front of them. Ask me after June 3rd .

Q: Can you talk more about the age indication? What would you like to propose?

A: I would like to get the Bionic Pancreas indicated for ages six and above. Our proposal says ten, but we’d like to get to 6. The FDA has said that they want us to test this in patients as young as possible. After all, I have managmed a child with type 1 diabetes when he was one year old. It’s challenging. As a parent, you are going to use this because it works. The FDA knows that so they’re pushing us to get it down.

Q: Can you talk about your commercial development? What companies are you involved with?

A: I can’t answer that fully. We’re working with investigative collaborators. The companies I  can name are Dexcom, Tandem, and Xeris.

Q: Can you explain the secrecy?

A: I think there are times when it’s approrpriate to reveal what the platform looks like, and there are times when it’s not appropriate. We have relationships with partners who do not want their involvement revealed at this time.

Q: Are there other group working on closed-loop systems?

A: Yes. There are other academic groups working on it, though we don’t know about their industry collaborators either. There are industry companies working on proprietary systems too.

Q: Can you estimate the cost of the Bionic Pancreas?

A: I can give you an estimate. The components of the Bionic Pancreas are similar to what is available on the market. It’s a two-chambered pump not a single chamber, so it will be a bit more expensive than that. That’s a ballpark, and we’re looking to get it down. The real cost is in the consumables. You’ll have to change CGM sensor weekly. Fortunately, private payers do reimburse for CGM. I’ll be clear – this does not work without a reimbursement model. That’s a very big discussion that needs to happen. The reimbursement part is not a trivial problem.

Meet the Expert

Emerging Therapies and Technologies for the Treatment of Diabetes

Irl Hirsch, MD (University of Washington, Seattle, WA)

Dr. Irl Hirsch called Abbott’s FreeStyle Libre a “game-changer” and Dexcom’s new Share Receiver “totally cool.” On the former, he suggested that Libre’s factory-calibration and accessible pricing (he estimated the cost at ~$5 a day) makes it a very attractive option across the board. Interestingly, Dr. Hirsch does not expect the technology to generate as much excitement in the US as it has in the EU due to the wider accessibility of traditional CGM stateside, though he still expects it to appeal to both type 1 patients and insulin-treated type 2s – we’d note that the CGM market is still very underpenetrated right now in the US, especially in type 2, so we believe Libre does have plenty of upside. We also believe that there would be broad interest in it from virtually any patient with type 2 – obviously we are unsure if it would be reimbursed, but given how poorly so many type 2 patients are doing, we think this would be a boon for many. We also see it driving therapeutic change for so many patients with type 2 and making doctors’ jobs much easier. An informal poll of the audience revealed that only ~10% had previously heard of Libre – this was not too surprising, since Abbott has done a good job of limiting the marketing, awareness, and webshop access to European audiences. Dr. Hirsch also offered a positive take on Dexcom’s new Share receiver with built-in Bluetooth (“this is totally cool!”), noting that the use of the cloud may have a profound impact on type 1 diabetes, especially in preventing hypoglycemia. Dexcom has pushed the boundaries on connectivity, and we have high hopes that Gen 5 will be equally well received (currently under FDA review, with approval expected by end of year). More broadly on CGM, Dr. Hirsch suggested that MARDs have improved in recent generations, but need to be better and solidly in the “single digits” to enable the success of closed-loop systems – we’d note that Dexcom’s Software 505/G4AP algorithm has a MARD of 9.0%, so it’s definitely in Dr. Hirsch’s ballpark.

  • Touching on data downloading, Dr. Hirsch said that he is a big fan of Tidepool. He believes that Medtronic’s CareLink is the current “gold standard” in the field, though he forecast that this could easily change in the near future. Indeed, while a poll of the audience revealed that 50% download data from pumps/CGM/BGM, Dr. Hirsch predicted that “all of you will raise your hand in five years!” (Less than ~5% had heard of Tidepool specifically, though Blip is still in a closed beta.) Dr. Hirsch emphasized that Tidepool’s open-source, non-profit model is compelling, not to mention the recent grant from the Helmsley Charitable Trust as a strong vote of confidence in the organization and its “impressive” founder and CEO, Mr. Howard Look.


Evaluating the Effect of V-Go Therapy in Sub-Optimally Controlled Patients with Diabetes: A Retrospective Cohort Analysis in a Large Specialized Diabetes System

Dr. Rosmarie Lajara (President, Diabetes America, Plano, TX) presented the results from a retrospective study sponsored by Valeritas and Diabetes America investigating the effects of switching sub-optimally controlled patients (average baseline A1c: 9.7%) on their existing regimen to V-Go. Patients were either previously on multiple daily insulin injections (MDI), long-acting basal insulin therapy, or were naïve to insulin. The analysis was retrospective, uncontrolled, and conducted in a relatively small sample (n=153); still, the results were very encouraging. All patients experienced a significant improvement in A1c after switching to V-Go for three months – the mean three-month A1c reduction was 1.7%, and the benefits were maintained at six months (-1.8%). The 22 patients naïve to insulin at baseline saw striking 3.1% and 3.3% A1c reductions at the three-month and six-month follow-ups, respectively (baseline A1c: 11.2% – a particularly tough group!). Baseline MDI and basal-only users experienced a 31%-41% reduction in total daily insulin dose along with smaller improvements in A1c – see below. The results are certainly directionally interesting and consistent with prior findings, and we look forward to comparative studies that demonstrate the V-Go’s value and control for the “study effect” – at this point, it’s hard to read too deeply into the headline data. As background, Valeritas performed a health and economic study of V-Go in a subset of this same patient population that was disclosed recently at the Academy of Managed Care Pharmacy’s 27th Annual Meeting and Expo in San Diego, CA – results showed direct pharmacy savings of $1,080 per patient annually, which factored in the cost of insulin, the cost of V-Go therapy, and changes to concomitant anti-hyperglycemic medications. See our detailed coverage here.

  • Data showed that all cohorts experienced a significant improvement in A1c after switching to V-Go therapy at both three and six months.

Group (baseline A1c)

Mean A1c reduction at Three Months

Mean A1c reduction at Six Months

Insulin Naïve (11.2%)



Basal Only (9.6%)



MDI (9.4%)



All Groups (9.7%)



  • Dr. Lajara suggested that patients do well on V-Go because it enables easier adherence to insulin regimens. Anecdotally, she noted that she has had patients taking over 100 units of Lantus daily who switch to the patch delivery device, use nearly half as much insulin, and still achieve better A1c. The pattern certainly speaks to what is so difficult about MDI (the hassles of injections, not to mention the stigma) and convenient about the V-Go (discrete, basal insulin, 24/7 access, one daily stick).
  • Only 16% of the original population (25/153 patients) discontinued V-Go between follow-up visits. Primary reasons included skin irritations (7 patients) and lacking insurance coverage/cost (6 patients).

Experience Using V-Go in Patients with Latent Autoimmune Diabetes of the Adult (LADA) or Type 1 Diabetes

Valeritas and Diabetes America explored the safety and efficacy of V-Go in 21 patients with type 1 diabetes or latent autoimmune diabetes in adults (LADA). All participants switched from a standard multiple daily insulin injection regimens to V-Go. Data showed that all cohorts experienced a significant improvement in A1c over the course of 103 days (-1.2%) that was maintained at 200 days (-1.4%) and 297 days (-1.4%) [baseline A1c: 9.6%]. Notably, all patients also saw a reduction in total insulin utilization from baseline of 18%. We caution again that the analysis was retrospective and uncontrolled, but find the results in type 1 diabetes highly notable, as it is not the core population for which V-Go is designed. Certainly, there is significant potential to expand the market of type 1s on more physiologic insulin delivery, and this poster suggests the V-Go has potential to do so. Though the V-Go’s pre-defined basal (20, 30, or 40 units per 24 hours) and bolus delivery (2-unit increments) don’t allow as much titration for type 1 diabetes, it may not matter as much in these patients with very poor glycemic control. Certainly, the V-Go’s simplicity and cost advantage could appeal to many patients with type 1.

Utilization of Regular Insulin in V-Go for Patients Uncontrolled with Type 2 Diabetes Mellitus (T2DM): A Case Series

Sponsored by Diabetes America, this poster featured two case studies investigating the off-label use of regular insulin delivered via V-Go. (V-Go is indicated only for rapid-acting insulin.) The analysis sought to determine whether regular insulin could provide effective glycemic control, as insurance coverage/cost of rapid-acting formulations is a key factor that drives patients off the V-Go. In one case, the patient experienced a 3.5% reduction in A1c (baseline A1c: 12.3%) and reduced total daily insulin requirements (not reported). In the second case, the patient experienced a 0.5% A1c reduction on regular insulin in the V-Go (baseline: 8.3%), building upon a 1.3% reduction in A1c achieved during a prior three-month period on rapid-acting insulin with the V-Go (original baseline: 9.6%). We find the analysis thought-provoking, especially from a cost perspective, and wonder if Valeritas or other investigators will pursue larger, comparative studies with regular insulin.

Improved Glycemic Control Utilizing Basal Bolus Insulin Delivery with the V- Go Disposable Insulin Delivery Device In the Long-Term Care Setting

The purpose of this study was to determine whether V-Go therapy in patients with type 2 diabetes requiring insulin in the long-term care setting could improve glycemic control and reduce glycemic fluctuations. All patients (n=8) were on existing insulin therapy regimens prior to switching to V-Go (baseline A1c unreported). Findings demonstrated favorable changes in glycemic control and percentage of BGM readings in target range associated with V-Go – mean blood glucose improved from 202 mg/dl to 180 mg/dl and percentage of readings in-range during V-Go therapy increased to 54% from 43% pre-V-Go. [The poster erroneously labeled this as “time in-range,” though CGM was not used.] Consistent with previous findings, the data also indicated a ~9% reduction in total daily insulin dose. The long-term care setting is a very challenging area, and we are glad to see V-Go has potential to improve care.

Corporate Symposium: Road to the Artificial Pancreas

Animas Dinner Event

Thomas McCann (J&J, New Brunswick, NJ)

At an Animas dinner event, Senior R&D manager Mr. Thomas McCann (J&J, New Brunswick, NJ) characterized the artificial pancreas “a priority for J&J Diabetes Care.” Though there were no timing updates or highly notable specifics, Mr. McCann suggested that a hypoglycemia-hyperglycemia minimizer (HHM) is indeed still in the works. The company presented a small closed-loop feasibility study at ADA 2013 (n=20) on this front; an even smaller trial presented at ADA 2014 (n=12) tested only predictive low glucose suspend. Given that the ADA 2014 study was less ambitious, we had been losing hope that Animas was really moving this project forward. This evening’s dinner gave us renewed confidence, and perhaps things will move forward faster now that the Animas Vibe is out in the US and Dexcom is pushing pump partners harder on the automated insulin delivery front (see our Dexcom 1Q15 coverage). We hope fervently that J&J CEO Alex Gorsky (who is always talking on Twitter about topics such like J&J giving lots of money to broad causes like “Operation Smile”) sees how well the Invokana franchise is doing and diverts some of the profits to medical devices. (This area is technically overseen by Consumer Head Sandra Peters though the P&L goes through Medical Devices – a nontraditional arrangement, but we hope that Ms. Peters also lobbies for more diabetes medical devices funding – she used to run blood glucose monitoring at Bayer.) From an outside perspective, this artificial pancreas project has moved at a glacial pace, considering Animas originally signed the partnership with JDRF in 2010. We hope JDRF CEO Derek Rapp can also add some pressure – look at the great outcomes at BD stemming from this JDRF and Helmsley Charitable Trust work! Undoubtedly, Animas certainly must do something in automated insulin delivery to stay competitive with Medtronic, Tandem, and Insulet in the near-term, who have all publicly committed to active R&D projects on this front – Medtronic in particular threw down the gauntlet with expectations to launch its MiniMed 670G in the US by April 2017 (we are on record doubting this timeline can be met though we appreciate the company being ambitious). Meanwhile, the Bionic Pancreas team expects a 2018 commercialization of its system (pivotal trial in 2016-17), and Bigfoot Biomedical plans to be in a pivotal trial by the end of 2016. Exciting times continue, and of course we hope very much the trial designs can be standardized.

Diabetes Drugs

Meet the Expert

Update on PCKS9 Inhibitors and New Therapies

Evan Stein, MD (University of Cincinnati Medical Center, OH)

Dr. Evan Stein some key commentary on PCSK9 inhibitors in the main conference session on the exciting new drug class for LDL lowering. Ever since the presentation of post-hoc phase 3 analyses on Sanofi/Regeneron’s Praluent (alirocumab) and Amgen’s Repatha (evolocumab) showing roughly 50% reductions in CV events (although based on few events), we’ve been interested to hear where the conference commentary would fall on the spectrum between excitement and caution. Dr. Stein was not afraid to be optimistic, noting that the Kaplan-Meier curves for both Praluent and Repatha were nearly superimposable and the divergence from the placebo group was found after just a few years, whereas it took nearly ten years to get similar data in the early days of statins. However, in his conclusion he was slightly more guarded with his words, suggesting only that the early CV analyses have been “moving in the right direction.” The outcomes trials for the first three PCSK9 inhibitors alone plan to collectively enroll 71,5000 patients, meaning that good answers on outcomes are on their way. During Q&A, he shared that his best guess is that the drugs will cost around $10,000-$12,000 per year (we wouldn’t hold him to this but it is helpful to hear guesses) and that in the coming era of multiple generic statins and generic ezetimibe, payers will put hurdles in place to make sure that providers exhaust other options before going straight to PCSK9 inhibitors.

The Year in Diabetes

Claresa Levetan, MD (Chestnut Hill Hospital, Philadelphia, PA)

Dr. Claresa Levetan’s compelling presentation (complete with an introductory poem and rhyming slide titles) on the current state of diabetes brimmed with emotions ranging from frustration at the skyrocketing costs for patients to excitement about recent advances to cautious optimism about the potential for truly curative therapies. She ran through a litany of obstacles her patients face due to the rising prices and confusing reimbursement policies associated with diabetes drugs, explaining that a substantial number of patients simply stop taking needed medications due to cost. On the more positive side, she reviewed a number of exciting advancements in diabetes therapies over the past year, including the arrival of new insulin formulations like Sanofi’s Toujeo (insulin glargine U300) and Sanofi/MannKind’s Afrezza (inhaled insulin), progress in the artificial pancreas arena, the introduction of sensor-augmented pumps, and the publication of Dr. Doug Melton’s (Harvard Stem Cell Institute, Cambridge, MA) beta cell differentiation protocol. However, Dr. Levetan stressed that none of these new available therapies addresses the true underlying pathophysiology of diabetes, which she described as an “ominous singlet” of beta cell dysfunction that produces a “systemic tsunami” of effects. She devoted the latter portion of her talk to her company Perle Bioscience’s efforts to address this gap by developing methods to generate functional islets (not just beta cells, which she believes is insufficient) from progenitor cells in the pancreatic duct. 

  • Dr. Levetan spoke frankly about the challenges of managing diabetes in an era of skyrocketing costs. She lamented the rising prices of diabetes drugs, particularly insulin, noting that some of her patients have stopped taking their medications and even ended up in the hospital with DKA because of prohibitive costs. She also expressed frustration with insurance policies such as insulin supply limits and exclusive formularies (citing the preference given to Lilly’s Humalog [insulin lispro] over Novo Nordisk’s NovoLog [insulin aspart] as one example) that limit patients’ options. In addition, Dr. Levetan noted that even the savings cards offered for most newer drugs do not guarantee smooth sailing, as prior authorizations and other bureaucratic hurdles are often required.
  • Dr. Levetan highlighted several exciting developments in the diabetes drug and device arenas over the past year, though she stressed that all existing therapies come with significant limitations. On the technology side, Dr. Levetan described the Bionic Pancreas as “phenomenal” (please see our coverage from day #2 for details on its pivotal trial) and celebrated the arrival of sensor-augmented pumps like the J&J Animas Vibe-Dexcom G4 Platinum system and Medtronic’s MiniMed 530G. However, she noted that such devices still do not fully restore normal glucose homeostasis, which she believes is necessary to prevent complications based on the DCCT and UKPDS results (though others have interpreted those results to mean that an A1c threshold of ~7% is where meaningful increases in risk begin to emerge). On the drug front, she expressed some excitement about the reductions in nocturnal hypoglycemia seen in type 2 diabetes with Toujeo but characterized the totality of the data as fairly confusing. Notably, she shared that many of her patients are excited about Afrezza – while the drug’s launch has gotten off to a bit of a slow start, this is consistent with our sense that there is a lot of patient enthusiasm for this product once access becomes easier. Nevertheless, she stressed again that no existing pharmacological options address the true pathophysiology of diabetes and emphasized the need for continued efforts in search of a cure.
  • Dr. Levetan believes research into islet neogenesis from pancreatic ductal progenitors offers the best hope for achieving insulin independence, as her company, Perle Biosciences, plans to start a phase 3 trial in this area this month. Perle Biosciences is currently investigating two oral combination approaches with the goal of restoring functional islets in patients with new onset type 1 diabetes. A randomized, double-blind, placebo-controlled phase 3 trial ( Identifier: NCT01762644) scheduled to start this month will evaluate a combination of lansoprazole (a proton pump inhibitor that promotes islet regeneration by increasing gastrin production) and low dose cyclosporine (an immunosuppressant) in 130 patients with recently diagnosed type 1 diabetes. The primary endpoint is stimulated C-peptide area under the curve after 24 weeks and the secondary endpoint is insulin independence. Primary completion is expected in November 2015 according to, though this may have been delayed as the trial’s start date was originally listed as February 2015. The company also has a combination in preclinical development consisting of cyclosporine A and a therapy that targets the Reg gene, which has been shown to promote islet neogenesis in the pancreatic duct in response to gastrin. The hope is that this more direct approach will offer greater potency than a proton pump inhibitor; Dr. Levetan expressed hope that clinical trials will begin within the next six months.
    • By contrast, Dr. Levetan expressed skepticism about stem cell-based beta cell regeneration approaches. While she highlighted the publication of Dr. Melton’s beta cell differentiation protocol as a key event of the year, she suggested that developing adequate encapsulation techniques will be very difficult. She also expressed doubt with regards to the effectiveness of beta cell-only approaches given the extensive vascularization and communication between different cell types in normal human islets. Perle Biosciences has operated under the radar in recent years compared to others in the type 1 cure arena like Dr. Melton’s group and ViaCyte; we are intrigued by this approach and would like to see data that reproduces her work.

Questions and Answers

Q: Can you bring us inside the work that you’re doing and where you are?

A: We’re working with Yale University on new Reg gene therapies, figuring out where is the bioactive region of the gene, how do you give it, and working on new therapies. On the flip side, there’s older work on gastrin in type 1 diabetes resulting in islet neogenesis. Donald Bergman has shared his early work with us demonstrating new islet and beta cell hyperplasia with increased gastrin. Gastrin hits the Reg receptor. We’re doing similar work with type 1 diabetes using a proton pump inhibitor, which increases gastrin, with low dose cyclosporine. With about 1,000 new onset patients with type 1 diabetes on cyclosporine, we had a 50% insulin independence rate with no significant toxicity. It didn’t last forever because there was no islet neogenesis and beta cell regeneration. We’re starting with two FDA approved agents, omeprazole and low dose cyclosporine A, in new-onset type 1 diabetes. We hope to have phase 1/2 trials within the next six months with low dose cyclosporine A and Reg therapy, which is more potent than a proton pump inhibitor. I’m optimistic and excited about combination therapy for type 1 diabetes. In HIV and cancer, we use combination therapy. Immune therapy alone and regeneration alone in type 1 diabetes won’t work. Rabinovich showed in a study that using a proton pump inhibitor doesn’t work alone.

Q: What do you see as the future of stem cells in type 1 diabetes?

A: I remember in the 1980s when I was a fellow and they were looking at a big hockey puck that they inserted to protect islets. I don’t see stem cells from the outside being able to be transplanted in. I don’t think islets or beta cells from a dish will land where they need to. Our team believes and we have data to suggest that you have those progenitor cells that are ready to be turned on and can be turned on. The Vinik study showed that in animals, if you give a label to the Reg gene and get it into a hamster, the only place it goes is the pancreatic ducts. I believe the future is making your own new islets from ductal progenitors. It’s hard to understand how we’ll inject billions of beta cells from a dish or how to package them. The packaging hasn’t seemed to work. It’s just like when people got islets from transplants and needed to be on immune medications, but the islets didn’t last long in the liver. My excitement is for progenitors that are already there.

Q: In the early days of the TZDs, we saw beta cell regeneration in mice and increased granules. Have you looked into that?

A: I’m not a big fan of TZDs. I’m not focused on the beta cell regeneration part. I’m not looking to make more beta cells from the few that are left. I want new islets: five-celled islets with big giant blood vessels inside. But I’ll think about it.

Q: Incretins turn on PDX. Have you conceptualized using incretins in type 1 diabetes?

A: I do use incretins in type 1 diabetes. I try a lot of things. There was a Novo Nordisk study in new-onset type 1 diabetes using GLP-1 where a couple people got off insulin. I will use GLP-1 in combination with a proton pump inhibitor off label. I’m not using low dose cyclosporine A until we get FDA approval for the study. My guess is the people who get off insulin are more like LADA than true type 1s, but you’re right. It’s a matter of getting in there soon enough and whether you can regenerate faster than the attack?

Q: Fifteen percent of my patients have positive autoantibodies but normoglycemia. As a clinician, what do you do for these patients?

A: I keep vitamin D high because it’s an associated risk factor. There’s one study that’s not that good but showed that it could be helpful. I look at the A1c and depending on age, I might do something. Do I put them on low dose insulin? Probably not. A proton pump inhibitor? Maybe. Would I now use Saxenda, which costs $1,068 a month? I might. Everyone’s different. I wouldn’t recommend it, but I’ve had people come in with DKA and I’ve gotten them off insulin with a proton pump inhibitor.

2015 Diabetes Update

Cardiovascular Safety of Incretin Based Therapy

John Buse, MD (University of North Carolina, Chapel Hill, NC)

There is more to GLP-1 agonists than their effects on the GLP-1 receptor. That was a key point in Dr. John Buse’s presentation on the cardiovascular safety of incretin-based therapies. He highlighted the distinction between data on GLP-1 agonists vs. native GLP-1. Native GLP-1 is metabolized into compounds that themselves appear to have beneficial downstream effects on targets such as vascular epithelial cells. However, GLP-1 agonists are designed to resist metabolism by DPP-4 and do not yield the same types of metabolites. As a result, the preclinical and early clinical cardiovascular benefits seen with native GLP-1s cannot necessarily be assumed to exist for GLP-1 agonists, although Dr. Buse did seem generally optimistic that the sum of GLP-1 agonists’ cardiovascular effects should be beneficial. On a closing note, Dr. Buse pointed out that longer outcome trials enrolling patients with less longstanding diabetes are needed to test the hypothesis of cardioprotection. Along these lines, we have been hearing more and more that CVOTs as they are currently designed can do little more than check the box for basic safety – this is so incredibly disappointing from a patient perspective that we can barely think straight about it.

What an Endocrinologists Should Know About Eye Care and Retinopathy

Michael Tolentino, MD (Center for Retina & Macular Disease, Lakeland, FL)

Retinopathy expert Dr. Michael Tolentino endorsed Bayer/Regeneron’s Eylea (intravitreal aflibercept) over other treatments for diabetic macular edema (DME). His opinion was based on the results of an NIH-sponsored study comparing Eylea vs. Roche/Novartis’s Lucentis (intravitreal ranibizumab) vs. Genentech’s Avastin (bevacizumab) that showed significantly greater improvements in visual acuity overall with Eylea vs. either comparator. However, full results showed that the difference was driven entirely by the subgroup with a baseline visual-acuity letter score <69 (~20/50 vision); based on this, we have wondered to what extent the medical community will view the results as a clear-cut win for Eylea. For its part, Roche has suggested that the broader population of patients enrolled in the diabetic retinopathy trials for Lucentis (those with both proliferative and non-proliferative retinopathy) could give the drug an edge over Eylea. Time will tell – as Dr. Tolentino acknowledged, it is the opinions of payers that will ultimately carry the most weight.

Product Theater: SGLT2 and DPP‐4 Inhibition – Multiple Options for Improved Glycemic Control in Adults with Type 2 Diabetes (Sponsored by Lilly/BI)

James Gavin, MD (Emory University, Atlanta, GA)

Dr. Gavin led one of the first product theaters we’ve seen for Lilly/BI’s Glyxambi (empagliflozin/linagliptin). He stuck fairly close to the prescribing information but was still clearly enthusiastic about the first-in-class SGLT-2/DPP-4 inhibitor fixed-dose combination (FDC). His commentary positioned Glyxambi as an ideal second-line agent after metformin – instead of adding another single agent, Dr. Gavin argued, why not add a more complete package of mechanisms to treat the underlying pathophysiology of the disease. Lilly and BI would prefer to have Glyxambi compete for patients not at control on metformin rather than drawing patients already on Tradjenta (linagliptin) or Jardiance (empagliflozin), although we imagine the patient pool that moves to Glyxambi will have a bit of both.

Product Theater: A Case Study in Treating Adult Patients with Type 2 Diabetes (Sponsored by Lilly)


Farhad Zangeneh, MD (Endocrine, Diabetes, and Osteoporosis Clinic, Sterling, VA)

Dr. Farhad Zangeneh cautioned against excessive use of insulin to treat type 2 diabetes at a product theater on Lilly’s Trulicity (dulaglutide). In addition to the standard overview of Trulicity’s label and clinical trial data, Dr. Zangeneh shared his personal commandments of type 2 diabetes management: (i) don’t make obesity worse; (ii) if the pancreas works, use it; (iii) don’t cause hypoglycemia; and (iv) don’t create side effects in asymptomatic patients. This struck us as, among other things, an anti-SFU mantra. From his perspective, moving to insulin therapy in patients who are still producing endogenous insulin violates all of these commandments and often significantly decreases patients’ quality of life. He positioned GLP-1 agonists like Trulicity as a much more favorable alternative, suggesting that they should be the preferred second-line option after metformin unless contraindicated. Results from Lilly’s AWARD-2 and AWARD-4 trials, which studied Trulicity in type 2 diabetes populations vs. basal insulin, back Dr. Zangeneh up: they demonstrated comparable or better glycemic control with Trulicity along with better weight and hypoglycemia profiles. We increasingly see long-acting GLP-1 agonists as a more beginner-friendly injectable for many type 2 diabetes patients, although basal insulin does have a higher efficacy ceiling. The most exciting option, however, may well be co-formulated combinations of basal insulin and GLP-1 agonists, which combine the best elements of both drug classes. 

Corporate Symposium: Combination Therapy in Type 2 Diabetes Management – Charting Courses for Success (Sponsored by AZ)

Best Practices with Long-Acting GLP-1 RAs, Short-Acting GLP-1RAs, and DPP-4 Inhibitors

Vivian Fonseca, MD (Tulane University, New Orleans, LA)

Dr. Vivian Fonseca gave a broad presentation on incretin-based therapies, including some of the newer options in the GLP-1 agonist class. He saved some of his most positive commentary for Intarcia’s implantable exenatide mini-pump ITCA-650, stating enthusiastically that taking the challenge of adherence out of patients’ hands is a great way to treat chronic disease. The greater efficacy and benign side effect profile relative to Byetta (the same exenatide molecule administered twice daily) suggests the benefits of slow release GLP-1 agonist administration. Although he presented the results of a number of GLP-1 agonist head-to-head studies, he argued against choosing agents based on efficacy alone. Using AZ’s DURATION-6 trial as an example, he noted that although Novo Nordisk’s Victoza (liraglutide) came out ahead in terms of A1c, Bydureon caused less nausea, which could be the deciding factor for patients for whom tolerability is more important. Switching over to DPP-4 inhibitors, Dr. Fonseca characterized Lilly/BI’s Tradjenta (linagliptin) as the “drug of choice” for patients with CKD due to its non-renal route of excretion. In line with a wave of comments we’ve heard recently, he is very positive on the combination of GLP-1 agonists and SGLT-2 inhibitors.

Corporate Symposium: Overcoming Barriers in Diabetes Management with Individualized Treatment Strategies (Sponsored by Sanofi)


Speakers at a Sanofi-sponsored breakfast symposium presented a balanced view of Sanofi’s new basal insulin formulation Toujeo (insulin glargine U300). None of the panelists suggested that they are rushing to use Toujeo in all their patients initiating basal insulin or currently on Lantus – this is undoubtedly partly an issue of some patients doing well on Lantus currently and some HCPs not wanting to deal with more administrative hassles and burdens (even if Toujeo on balance would be better for some patients). We don’t have a clear sense of whether patients new to insulin are being charged an arm and a leg (over Lantus) or not. Dr. Lisa Tannock (University of Kentucky, Lexington, KY) noted that at this point, all we know about Toujeo (based on fairly short-term clinical studies) is that there is a nocturnal hypoglycemia benefit and more borderline overall hypoglycemia and weight benefits vs. Lantus. Although she does see Toujeo as an improvement over Lantus, she suggested that if a patient is doing well on Lantus there is little reason to switch to Toujeo. Dr. Zachary Bloomgarden (Mount Sinai School of Medicine, New York, NY) expressed some uncertainty over how useful Toujeo will be relative to other alternatives for patients not at control on Lantus alone, including GLP-1 agonists. Dr. Tannock did note that Toujeo represents a good option for highly insulin resistant patients as the dosing is more straightforward with Toujeo’s pen than with Humulin U500 vials. On access, Dr. Tannock noted wryly that the choice between Toujeo and Lantus is more a matter of patients’ insurance; Dr. Bloomgarden wondered aloud whether the future introduction of biosimilar insulin glargine may further constrain access to premium options like Toujeo. We are eager to see how the conversation on Toujeo vs. Lantus develops in symposia and exhibit halls at coming meetings like ADA.

Meeting Targets with New and Emerging Therapies Added to Basal Insulin

Lawrence Blonde, MD (Ochsner Medical Center, New Orleans, LA)

In Dr. Larry Blonde’s presentation on possible options for patients not at control with basal insulin, GLP-1 agonists came out as a favorite. In fact, Dr. Blonde noted that for many type 2 diabetes patients GLP-1 agonists may be the better choice as a first injectable – after all, head-to-head phase 3 trials show that basal insulin is no better in terms of A1c (or weight or hypo) and some trials even show GLP-1 agonists coming out ahead. The common wisdom is that for patients not at control on orals, insulin may be best for patients with very high A1cs due to its higher efficacy ceiling. Dr. Blonde agreed with that notion but suggested that the A1c threshold may be higher than some might expect: post-hoc analyses of head-to-head trials of basal insulin vs. GLP-1 agonists show that even patients with starting A1c as high as 9%-9.5% did at least as well with a GLP-1 agonist. It isn’t just Dr. Blonde who views GLP-1 agonists favorably as a starting injectable: during a show-of-hands poll, audience members favored GLP-1 agonists over basal insulin by at least a three-to-one margin. At the very end of his talk, Dr. Blonde put in a good word for Sanofi’s new inhaled insulin Afrezza. He stuck pretty closely to the script of the product label but did put Afrezza squarely in the win column for weight and hypoglycemia in a trial (based on a trial vs. premixed Biaspart (70/30)).

Panel Discussion – Selected Highlights

Q: So which patients do you start on insulin glargine U300?

Dr. Tannock: I have had a few patients I have switched to U300. I use it in my patients who report frequent nocturnal hypos because that’s where I think the data is most robust. With any new product on the market, the decision is largely in payers’ control. But it is individuals having hypos with U100 where I’d consider U300.

Q: When you need to start a patient on injectable therapy, by a show of hands, who goes to insulin first, and who uses GLP-1 agonists first?

[Attendees preferring GLP-1 agonists outnumbered those preferring insulin by around three to one]

Dr. Blonde: Whenever you ask an endocrinologist a question, the most frequent answer is it depends. All other things equal, GLP-1 agonists are the best first injectable.

Dr. Bloomgarden: We fail to realize how much better endogenous insulin is than exogenous. The beta cells deliver insulin primarily to the liver. Portal insulin levels are two to four-fold higher than in the periphery. When we give exogenous insulin, we alter that balance in a deleterious fashion, which increases adipogenesis and changes lipid levels. Sulfonylureas allow more endogenous insulin but they’re glucose insensitive which makes the benefit more tenuous. With the onset of incretin therapy, you get endogenous insulin to the liver first, so it’s working in the right fashion. That’s an underappreciated reason why these are better drugs.

Q: Do you use C-peptide to decide whether to use basal insulin vs. GLP-1 agonists?

Dr. Bloomgarden: I think that as a clinician, measuring insulin and C-peptide is rarely helpful. We really have a good sense of the degree of a patient’s insulin secretory defect based on their clinical and glycemic characteristics. Although conceptually you could say that if someone has low C-peptide then they would be more in need of exogenous insulin, as Larry said, for this audience, I think that all of you will know that well before the time you get the result of the C-peptide back from the lab.

Dr. Blonde: The other reason I agree with you is that C-peptide is only helpful if the value is high. If a patient has hyperglycemia, C-peptide may be low, but if you lower the hyperglycemia then it may come back up. It’s not that helpful clinically.

Q: We saw in the EDITION trials that there was a need for a greater insulin dose with Toujeo? Why, and how much of a dose increase was needed?

Dr. Tannock: I don’t remember the difference, but despite the difference there was less nocturnal hypoglycemia and less weight gain. The flip side is increased cost with more units.

Dr. Moghissi: It is about a 10%-15% increase, I think. It is because you have a more concentrated depot.

Q: Is dulaglutide approved for use with basal insulin?

Dr. Blonde: The data is good when used with rapid-acting analogs, but at present dulaglutide has not been studied with basal insulin.

Q: Could you talk about bone health with SGLT-2 inhibitors?

Dr. Bloomgarden: I think that there is interesting data on bone health with these agents, but nothing is conclusive in the way that TZDs were shown mechanistically and in clinical trials to adversely impact bone health. We need to keep aware of this, but I don’t think that this truly constitutes something that is at the level of a clinical issue at the present time.

Q: We are getting lots of questions about Toujeo’s clinical applications, because it is longer acting and has a flatter profile. When should we switch from U100 to U300? Is the increment of the benefit the same as from NPH to insulin glargine?

Dr. Tannock: One of the issues is that we have relatively short-term studies only. It’s the same insulin as Lantus but more concentrated. We have no outcomes data on long-term effects. All we have to go on is a decrease in nocturnal hypoglycemia. If a patient is doing well on U100, why change? As we get more data on U300 that may change. Right now we see fewer nocturnal hypoglycemia and trends towards less daytime hypoglycemia that bordered on significant. Also, there was less weight gain. That latter factor is something we may want to consider for patients who are struggling with their weight. However, often the decision to try Toujeo is not my choice – it’s up to payers.

Dr. Blonde: I am glad to see the field moving away from the term “minor hypoglycemia,” because I have seldom seen a patient who has had a hypoglycemia episode that they consider minor. Often times, clinicians may not be aware that their patients are experiencing these. Now that there may be options that can further improve the benefits of basal insulin analogs, with new analogs that have even lower hypoglycemia risk, it behooves us to ask in more detail whether our patients have been experiencing hypos that they have not been bringing to our attention.

Dr. Bloomgarden: It is important that if patients are having nocturnal hypos, you have to reassess and see if you are overusing basal insulin. If so, you can try workarounds like going to basal bolus or a basal plus a GLP-1 agonist. I don’t know yet how useful insulin glargine U300 is going to be within the overall set of choices, but it may offer a bit of an option. With payers I don’t know – with biosimilar insulins coming we may be forced to never use it. 

Q: How do you divide the dose when starting U500?

Dr. Tannock: There’s no study other than that single study. I individualize it based on when patients eat.

Q: What is the cheapest insulin?

Dr. Tannock: It depends on where you practice. Insulin costs differ at different institutions. We pass on NPH regular 70/30 at a wholesale cost of $9 a bottle. That’s by far the cheapest until you get up to very significant doses. On a per unit basis, there was a time when U500 was by far the cheapest. Now it’s about 9 to 10 cents per unit, which is on par with most analogs. It varies pharmacy to pharmacy, so I tell patients to shop around. For many the break point is 300 units per day where U500 becomes cost-effective.

Dr. Bloomgarden: In people who are insulin sensitive, pens become less expensive if someone uses ~15-20 units a day. They’re wasting a huge amount if they buy a bottle. That’s an important effect for patients with type 1 diabetes. It’s not relevant to most patients with type 2 diabetes.

Q: At what level do you consider the insulin receptors to be saturated?

Dr. Tannock: The simple answer is no. You keep titrating until you get there. Some patients have insulin receptor defects, but that’s very rare. For most patients, you’ll get an effect when you hit the dose you need.

Dr. Blonde: I showed a study where adding exenatide to U500 didn’t make a difference. There’s another study where liraglutide did make a difference, but it was small and poorly controlled. There should be more studies in populations using U500 to see whether other agents like GLP-1 agonists might be helpful.

Corporate Symposium: Applying the Fundamentals of Insulin Physiology to Modernize Exogenous Insulin Therapy: An Expert Panel Exchange and Interactive Town Hall Discussion (Sponsored by Lilly)

Breakfast Symposium

Speakers at a Lilly-sponsored breakfast symposium on insulin physiology appeared fairly positive on Lilly’s novel basal insulin peglispro (BIL). The renowned Dr. Yehuda Handelsman (Metabolic Institute of America, Tarzana, CA) characterized peglispro as a potential game-changer in the use of insulin, primarily due to the benefits on weight and hypoglycemia in the context of superior glycemic control. Dr. Sunder Mudaliar (VA Medical Center, San Diego, CA) highlighted the potential for weight loss and hypoglycemia reductions due to the product’s hepatoselectivity. In addition, he framed the concerns about liver safety and elevated lipids primarily as unanswered scientific questions rather than definitive warning signs. Topline phase 3 results in type 1 and type 2 for peglispro were exciting for their ability to demonstrate consistent A1c superiority vs. the market leader Lantus, which made the delay in peglispro’s timeline to better understand the changes in liver fat particularly disappointing.

Corporate Symposium: LDL-C Reduction in the High-Risk Patient: How Low Should We Go? (Sponsored by sponsored by Sanofi and Regeneron)

Evening Symposium

An evening symposium sponsored by Sanofi and Regeneron provided a packed room of around 350 attendees with background on the exciting new PCSK9 inhibitor class for LDL lowering. The two furthest-along agents in the class, Sanofi/Regeneron’s Praluent (alirocumab) and Amgen’s Repatha (evolocumab), have demonstrated ~40%-70% reductions from baseline in LDL cholesterol in a wide range of patients, including those with statin intolerance. Additionally, phase 3 post-hoc analyses appear to show roughly 50% reductions in CV events, although these are not outcomes trials and involve fairly few events (see our ACC 2015 Report for more background on this). During the dinner, diabeto-cardiologist Dr. Robert Eckel (University of Colorado, Aurora, CO) declined to predict whether the FDA will approve Repatha and Praluent for broad indications or stick with targeted subpopulations (i.e.: familial hypercholesterolemia) until outcomes data arrives in 2018. However he did note that no matter how clinically compelling PCSK9 inhibitors are, the drugs will not be used if they are not covered by payers and must therefore be at least somewhat affordable. The panel agreed that most patients should try proven therapies like statins (and possibly even ezetimibe) before going to PCSK9 inhibitors.

Corporate Symposium: The Effect of a Once-Weekly Therapy on A1C and Weight Over 3 Years (Sponsored by AstraZeneca)


Carolina Solis-Herrera (University of Texas Health Science Center, San Antonio, TX)

Dr. Carolina Solis-Herrera presented expanded data suggesting that AZ’s Farxiga (dapagliflozin) improves beta cell function and insulin sensitivity. The study involved 24 men with type 2 diabetes on background metformin who were randomized 2:1 to receive daily treatment with Farxiga or placebo for two weeks; participants underwent a euglycemic insulin clamp and an oral glucose tolerance test (OGTT) at baseline and at the end of the study. As background, previous results from the same study with 18 participants were published in the Journal of Clinical Investigation in early 2014. As expected, results showed significantly greater reductions in fasting glucose and two-hour postprandial glucose with Farxiga vs. placebo. Notably, Farxiga also led to significant increases in fasting insulin and C-peptide levels, insulin secretion (measured by change in C-peptide/change in glucose), and measures of beta cell function (C-peptide/glucose adjusted for glucose disposal). In addition, the drug produced significantly greater improvements in total body and tissue glucose disposal and the insulin secretion/insulin resistance index compared to placebo, suggesting improved insulin sensitivity. As Dr. Ralph DeFronzo (the principal investigator for the study) noted recently at GTC Bio, this data illustrates the important role of glucotoxicity in the pathophysiology of type 2 diabetes by demonstrating that lowering glucose via a completely insulin-independent mechanism can lead to improved beta cell function. In the same talk, Dr. DeFronzo referenced an intriguing study he is conducting to investigate the use of SGLT-2 inhibitors to correct glucotoxicity in prediabetes – we are very excited to learn more about this.

Novel Therapeutic Approaches to Type 2 Diabetes and Complications

The Kidney, Intestine, or Both as Therapeutic Targets in Diabetes

James Gavin, MD (Emory University, Atlanta, GA)

 Dr. James Gavin expressed optimism about the potential for cardioprotection with SGLT-2 inhibitors. While emphasizing that there is not yet any definitive outcomes data on the class, he presented an intriguing meta-analysis of 19 phase 2b/3 randomized controlled trials which showed a fairly early and sustained reduction in cardiovascular adverse events with AZ’s Farxiga (dapagliflozin) vs. placebo. Based on this data and the improvements in risk factors (e.g., systolic blood pressure, HDL cholesterol, etc.) produced by SGLT-2 inhibitors, Dr. Gavin suggested that “there may be some promise” of cardiovascular benefit that will hopefully be revealed in ongoing outcomes trials. As a reminder, we learned on day #2 of AACE that the first CVOT for an SGLT-2 inhibitor – EMPA-REG OUTCOME for Lilly/BI’s Jardiance (empagliflozin) – is now complete and will report results at this upcoming EASD – this will be a draw for Stockholm, big-time! While we agree that SGLT-2 inhibitors, along with GLP-1 agonists, are one of the most likely diabetes drug classes to show cardioprotection, we are somewhat concerned that the design of ongoing CVOTs (i.e., short duration, high-risk population) may limit their ability to demonstrate the true extent of the benefits.

  • Dr. Gavin also highlighted dual SGLT-1/2 inhibitors, like Lexicon’s sotagliflozin, as a promising option on the horizon for type 2 diabetes. Referring to data on both sotagliflozin and J&J’s Invokana (canagliflozin), which causes a small amount of SGLT-1 inhibition, he explained that targeting both receptors should enable greater reductions in postprandial glucose compared to inhibiting SGLT-2 alone. He stressed that there is much more to learn with these agents, particularly regarding their effects on the GI tract, but suggested that the effects could be quite beneficial if the right balance between the two components is achieved. Getting the dose ratio right will certainly be key – too much SGLT-1 action has been shown to lead to prohibitive GI side effects, but as Dr. DeFronzo pointed out at GTC Bio, more tolerable drugs (potentially including sotagliflozin) may not have sufficient potency to produce meaningful effects.

New Insulin and Insulin GLP-1 Combinations in Development: Application and Therapeutic Efficacy

John Buse, MD (University of North Carolina, Chapel Hill, NC)

Dr. John Buse argued that the risk of “euglycemic DKA” with SGLT-2 inhibitors should be manageable as long as people are aware of the risk. Dr. Buse, along with Dr. Hirsch and Dr. Anne Peters (USC, Los Angeles, CA), has led the way in drawing wider attention to this issue with a forthcoming paper detailing several cases of euglycemic DKA. During Q&A at a session on type 2 diabetes therapies, he suggested that while the phenomenon seems to be “moderately frequent” in type 1 diabetes, many patients have been able to manage it on their own as long as they recognize it. Dr. Peters herself had slightly stronger words of caution on day #2 of AACE, when she recommended against prescribing SGLT-2 inhibitors off-label in type 1 diabetes until the risk is better understood. While it is hard to predict at this stage what the effect of euglycemic DKA will be on the class as a whole, Dr. Buse’s words certainly seemed apt, particularly regarding type 2 diabetes, as the risk so far with type 2 appears to be fairly low and linked to specific situations surrounding surgery in this population.

  • We imagine the risk may have a more substantial impact on the chances of SGLT-2 inhibitors gaining approval for type 1 diabetes, though the class benefits on glucose control, glycemic variability, weight, and insulin dosage remain very compelling and we believe patient advocacy for the class related to type 1 will be very high. In the meantime, it seems that the most important goal is to educate patients and providers (including ER physicians) about the phenomenon so that the risk can be managed appropriately – we believe that the FDA’s recent drug safety communication on the issue should go a long way in terms of raising broader awareness. We also believe sales of ketone meters will go up – a check from Walgreen’s in San Francisco this weekend showed that they were sold out (admittedly this could be spurious correlation).


Oral Presentations on Planning Research in Diabetes

Stanley Schwartz, MD (University of Pennsylvania, Philadelphia, PA)

Dr. Stanley Schwartz suggested that people underestimate the true cost of sulfonylureas by failing to consider their long-term consequences. He argued that the initial savings compared to newer drug classes are outweighed by the need for more aggressive treatment once sulfonylureas fail and by the cost of complications associated with hypoglycemia. This is certainly plausible to us, and we hope to see the healthcare community, particularly payers, shift toward such a long-term perspective when evaluating the value of various therapies. However, it is important to note that currently, many patients with limited means and restricted insurance coverage simply cannot access expensive branded medications, no matter how cost-effective they may prove to be down the road – this is why we hope to see the DPP-4 inhibitor class remain popular until it goes generic so that it can replace SFUs as a cheap non-metformin option. Dr. Schwartz also noted that sulfonylureas would not live up to the current FDA CV safety standards if they were being developed today – this is an opinion we have heard repeatedly since the implementation of the 2008 CV Guidance.

Irl Hirsch, MD (University of Washington, Seattle, WA)

Dr. Irl Hirsch suggested that biological “cures” for type 1 diabetes could be considered successful even if they do not lead to complete insulin independence. He explained that his thinking in this area has changed dramatically since his days as a young researcher in the 1980s when the widespread expectation was that type 1 diabetes would be cured within the next five years. (We believe there is some variance here – we certainly know patients who did not anticipate a cure in the 1980s – but some certainly also did receive misinformation on this.) After seeing many patients derive enormous benefit from a small amount of residual insulin production (whether naturally or after a “failed” pancreas transplant), he now believes that even reducing insulin doses by ~50% with a cell replacement therapy could lead to meaningful improvements in quality of life – this would be especially true if the therapy reduced glycemic variability alongside lowering glucose. Setting realistic expectations for success will be very important as cell-based therapies like ViaCyte’s VC-01 move closer to reaching patients, as we imagine there is plenty of room for disappointment if these therapies fall short of being complete cures. We heard similarly pragmatic commentary from Dr. Alexander Fleming (Kinexum, Harpers Ferry, WV) at the GTC Diabetes Summit focused primarily on regulatory expectations for cell-based therapies, and we hope to hear much more as these therapies progress through clinical trials.

Broad Commentary

Euglycemic Ketoacidosis

Euglycemic ketoacidosis continued to be raised as a safety issue with using SGLT-2 inhibitors in type 1 diabetes. We heard consensus that such off-label should be limited until the field has better adverse event data on the use of the drug class in the type 1 population. Although little is known about the mechanism behind the effect, Dr. Irl Hirsch (University of Washington, Seattle, WA) suggested that a key contributor may be the fact that SGLT-2 inhibitors allow patients to lower their insulin doses significantly. During his talk on glucagon, Dr. Ralph DeFronzo (University of Texas Health Science Center, San Antonio, TX) suggested that the elevated glucagon production seen with SGLT-2 inhibitors may contribute by pushing the liver into a ketogenic mode, as elevated glucagon is normally a signal of a fasting state. Dr. Hirsch also noted that most cases of euglycemic ketoacidosis in his experience have been linked to J&J’s Invokana (canagliflozin), though this may be a result of its first-in-class status in the US. Dr. Vivian Fonseca (Tulane University, New Orleans, LA) noted reports from Japan of the phenomenon happening in type 2 diabetes patients. His recommendations were for type 1 diabetes patients using SGLT-2 inhibitors off label to check urine ketones regularly, although he did not urge all patients taking the drugs off label to stop immediately.

Combination Therapy

Dr. Daniel Einhorn (Scripps Health, La Jolla, CA) highlighted the complementary mechanisms and strong efficacy of SGLT-2 inhibitor/DPP-4 inhibitor combinations, which are now finally on the market following the launch of Lilly/BI’s Glyxambi (empagliflozin/linagliptin). During his afternoon talk, Dr. Irl Hirsch (University of Washington, Seattle, WA) discussed the even more exciting prospect of GLP-1 agonist/SGLT-2 inhibitor combinations, which he believes could potentially have additive or even synergistic efficacy and powerful effects on body weight – of course, as Dr. Fonseca noted during the AZ symposium, these combinations also add an additional layer of complexity for patients compared to oral fixed-dose combinations. He highlighted the ongoing trial of combination therapy with Invokana and Novo Nordisk’s Victoza (liraglutide) led by Dr. DeFronzo; AZ is also currently conducting a trial investigating combination therapy with Bydureon (exenatide once weekly) and Farxiga (dapagliflozin). Dr. Hirsch also noted that this combination could potentially eliminate the risk of euglycemic DKA, as the GLP-1 agonist component would hopefully blunt the increased glucagon production induced by the SGLT-2 inhibitor that is believed to be a cause of the phenomenon (see above). On the negative side, he cautioned that cost could be a major barrier with these two very expensive drug classes – the success of a future combination product would certainly depend heavily on the level of reimbursement, though we imagine that such an impressive clinical profile (if the expectations are borne out in clinical trials) would be very appealing to payers.

SGLT-2 Inhibitors

At a product theater for J&J’s Invokana (canagliflozin), Dr. James Gavin (Emory University, Atlanta, GA) noted that his prepared slides heralding the product’s commercial success were already out of date: Invokana has now reached over 4 million prescriptions since its launch and achieved over 80% commercial and Medicare coverage. In a talk later in the afternoon, Dr. Hirsch offered similar sentiments, sharing projections suggesting that Invokana will be a $2-$3 billion drug by 2020 – we see that as a conservative estimate though it depends largely on how ambitious J&J is on the formulary front (it also assumes euglycemic DKA stays manageable – most if not all cases to date have been associated with Invokana – unsurprising given its dominance in the US). Speakers at both sessions characterized the class as very appealing to patients and providers – Dr. Daniel Einhorn (Scripps Health, La Jolla, CA) said they were “one of the most adhered to medications I’ve ever used” – oral administration can’t hurt, and weight benefits are highly motivating for patients.


Meet the Expert

Practicalities in the Use of Pharmacologic Agents

Caroline Apovian, MD (Boston Medical Center, Boston, MA)

In a session on obesity pharmacotherapy, Dr. Caroline Apovian stated that she sees a combination of medications and devices as the future of obesity treatment. Fielding a question on the potential of combinations such as phentermine/lorcaserin, she noted that she foresees not only combinations of medications but also combinations of medications and devices; specifically, she expressed hope that a combination of drugs and devices could produce weight loss comparable to that achieved with bariatric surgery. In her presentation, she noted the upcoming emergence of many obesity devices, including the endoluminal barrier, Gelesis’ Gelesis100 pill (a pill that soaks up water and expands in the stomach, which she commented “may come out shortly”), and intragastric balloons. While we do not see any currently available obesity devices as holding great potential (see our coverage of the approval of EnteroMedics’ VBLOC Therapy), we would agree that combinations that utilize different targets will likely provide more success in weight management in the future – please read our coverage of last week’s European Congress on Obesity (ECO) to see similar sentiments on combination therapies.

  • On the drug front, Dr. Apovian pointed out that if Zafgen’s beloranib does not have deleterious side effects, it may “really be a game changer.” Indeed, beloranib’s mechanism of action (it alters fat metabolism by inhibiting MetAp2) differentiates it from the many centrally acting drugs currently available. As a reminder, beloranib is in phase 2/3 development and has so far shown solid safety and efficacy data in hypothalamic-induced associated obesity and severe obesity.
  • During Q&A in a subsequent session, Dr. Joshua Thaler (University of Washington, Seattle, WA) provided commentary on his clinical experience with obesity drugs. He pointed out that he has seen significant variability in weight loss with liraglutide: some people respond dramatically while others do not experience much weight loss at all – this reflected Dr. Apovian’s earlier point that providers should not expect the average weight loss produced by a drug in clinical trials to predict their patients’ experience, which is heavily dependent on whether or not the individual responds to the drug. On the other hand, Dr. Thaler added that he has had “a lot of luck” with Vivus’ Qsymia (phentermine/topiramate) as he has seen many patients successfully respond to the drug. He described Orexigen/Takeda’s Contrave (naltrexone/bupropion) as an intermediate option and stated that Arena/Eisai’s Belviq (lorcaserin) is not very potent in terms of weight loss but has a benign side effect profile (this is similar to commentary we have heard in the past). Overall, when thinking about providers’ different characterizations of obesity pharmacotherapies, we see the ability to forecast which drug will be effective in which patients as an increasingly important question. Unfortunately, we do not see this being answered anytime soon due to the disappointing progress on genetic analysis in obesity: the loci that have been identified so far contribute to only a very small amount of BMI variance. 

The AACE Module and Tool Kit for Obesity Medicine

The AACE Module and Tool Kit for Obesity Medicine

Timothy Garvey, MD (University of Alabama, Birmingham, AL)

In a special session on obesity, Dr. Timothy Garvey discussed AACE’s future plans, which include the creation of new evidence-based guidelines and a second consensus conference on obesity. Dr. Garvey prefaced by noting the need for comprehensive, evidence-based guidelines in obesity medicine, commenting that some of the current guidelines are not adequately comprehensive and that the multiple existing guidelines can create confusion among providers. AACE is thus planning to fill this gap with guidelines focused on a complications-centric approach; Dr. Garvey stated that the guideline is expected to be finished by late summer or early fall. Notably, Dr. Garvey also announced that this past Tuesday, the AACE board has approved a second consensus conference on obesity, which is to take place early next year. The conference will bring together various professional societies as well as patients with the goal of harmonizing existing guidelines – please see our coverage of AACE’s first consensus conference from last March for more. In addition, AACE is working on creating a white paper and toolkit that will offer practical guidelines and efficient data-gathering methods for providers to simplify the practice of obesity medicine, which will eventually be translated into a CME program – tools in the toolkit include algorithms, guidance on equipping and staffing an office for obesity management, strategies for patient communication, and more. We are excited to see AACE investing in such significant initiatives in obesity and alongside the Endocrine Society’s recent guideline for obesity pharmacotherapy, we hope that these efforts will quell the reluctance among many providers to aggressively treat overweight and obesity.

Step 2: The Clinical Component of the Diagnosis

Felice Caldarella, MD (Hunterdon Medical Center, Flemington, NJ)

In the same session, Dr. Felice Caldarella pointed out that obesity counseling under Medicare is only reimbursed for primary care providers, excluding endocrinologists – this understandably prompted dismay among attendees. Reimbursement codes for obesity screening, nutrition counseling, and intensive behavioral counseling and therapy are thus of little use to endocrinologists; additionally, Dr. Caldarella commented that the national average of reimbursement for such services is only $26. We strongly agree that this loophole in the incentive structure is quite an obstacle for obesity treatment, as primary care providers’ time with patients is overall even more limited than that of endocrinologists – see our coverage of NPR and Kaiser Health’s related article for more on this. In response to attendees’ frustration regarding this restriction during Q&A, Dr. Michael Gonzales-Campoy (Minnesota Center for Obesity, Metabolism, and Endocrinology, Egan, MN) stressed that AACE is indeed concerned about these constraints and is working with federal agencies on the issue. We would also note that these limitations remain in the context of the recent USPSTF “B” recommendation to provide behavioral interventions for overweight and obese adults with CV risk factors (in addition to the older “B” recommendation to screen all adults for obesity and provide behavioral interventions for those with a BMI of at least 30 kg/m2), making the federal government’s current policies a bit paradoxical.

Product Theater: The Risks and Rewards of Bariatric Surgery for the T2DM, Obese Patient (Sponsored by Ethicon)

Bariatric Surgery

David Cummings, MD (University of Washington, Seattle, WA)

Dr. David Cummings presented positive unpublished follow-up data from the CROSSROADS study on the effectiveness of bariatric surgery in diabetes. Formally known as the Caloric Reduction or Surgery: Seeking Remission for Obesity and Diabetes study (n=40), this randomized controlled trial examined the outcomes of individuals with type 2 diabetes and BMIs of 30-40 kg/m2 (mean baseline BMI of ~38 kg/m2 and A1c of ~7.5%) who were randomized to either standard roux-en-Y gastric bypass and medical care or intensive medical and lifestyle treatment. Results demonstrated that at one year, the surgical group experienced a significantly higher rate of diabetes remission (60% vs. 6%) and a larger drop in A1c (~1.3% vs. ~0.4%). Dr. Cummings commented that the surgical group achieved these outcomes with fewer diabetes medications compared to the medical/lifestyle group. In addition, the surgical group had significantly reduced fasting insulin, insulin resistance (as measured by HOMA-IR), body weight, body fat, and systolic blood pressure compared to the medical/lifestyle group. On the other hand, the two interventions produced similar reductions in diastolic blood pressure, cholesterol, LDL, and triglycerides and a similar increase in HDL. Alongside these findings, Dr. Cummings concluded that there is a growing body of randomized controlled trial evidence demonstrating the effectiveness of bariatric surgery in diabetes. In particular, he noted that such results provide support for the use of surgery in patients with diabetes who have lower BMIs (<35 kg/m2). We hear this point made fairly often by KOLs in obesity, although the holy grail would be to replicate the striking benefits of bariatric surgery through medication. For more on Dr. Cummings’ insights on bariatric surgery, please see our comprehensive interview with him from 2013.


Novo Nordisk SCALE Trial

Novo Nordisk presented a poster on new data from the SCALE Obesity and Prediabetes trial, showing that patients’ early responses to Saxenda (liraglutide 3.0 mg) can help identify individuals who will respond positively to the drug later on. This exploratory analysis compared results at week 56 in early responders (≥4% weight loss at 16 weeks) compared to early non-responders (<4% weight loss at 16 weeks) and found that 84% of early responders lost ≥5% body weight vs. 28% of early non-responders; 50% of early responders lost >10% body weight vs. 6% of early non-responders; and 21% of early responders lost >15% body weight vs. 2% of early non-responders. The predictive value of early response extended to secondary effects as well. On average, early responders saw a reduction of 8 mg/dl in fasting plasma glucose and 0.4% in A1c vs. early non-responders’ reduction of 6 mg/dl in fasting plasma glucose and 0.2% in A1c (keep in mind that these were not diabetes patients). In addition, early responders had greater reductions in waist circumference (11 cm vs. 5 cm in early non-responders) as well as greater increases in Impact of Weight on Quality of Life-Lite total scores (13 points vs. 8 points in early non-responders). On the safety front, the two groups generally shared a similar safety profile, with the exception that early responders experienced more gallbladder disorders (3% vs. 1%), which the poster states was likely due to the group’s greater weight loss. Overall, these findings indicate that weight loss at 16 weeks is a relatively effective predictor of which individuals will be successful on Saxenda (supportive of the drug’s stopping rule). Novo Nordisk has highlighted this as a potential selling point for the drug with prescribers and payers, suggesting that being able to target predictors early can help shift the benefit/cost balance in favor of the drug (which has a very costly list price of $1,068 per month). Cynics say that at present, the main commercial impact of Saxenda will be to drive Victoza interest and sales; we certainly hope that in the highest need patients, reimbursement will materialize since for the most costly patients, total costs stemming from obesity are far higher than this monthly fee for Saxenda.

  • Novo Nordisk presented additional posters from the SCALE trials showing similar efficacy and safety with Saxenda across all racial groups and confirming that the weight loss is not explained by GI side effects. In post-hoc analyses using pooled data of all four SCALE trials, findings demonstrated that in both black/African Americans and Hispanics, the Saxenda treatment arm experienced clinically meaningful weight loss compared to the placebo arm. Weight loss, improvement in cardiometabolic risk factors, and safety profiles were also shown to be generally consistent across all racial subgroups. With racial disparities in obesity already so significant, these findings are reassuring, although unequal access to such treatments remains a significant barrier.

Corporate Symposium: Managing Obesity in 2015 – Setting the Framework and Assessing All of the Options (Sponsored by Novo Nordisk)

Panel Discussion – Selected Highlights

Q: These newer medications are all approved for use as individual agents. Why can’t we use these medications in combination?

Dr. Christopher Still (Geisinger Obesity Research Institute, Danville, PA): The answer from the companies is it’s because they haven’t been tested. There have been no official trials in combination. From a clinical standpoint, obesity is a chronic disease, and I don’t think it’s unreasonable if there’s not a contraindication with regard to similarities in the mechanisms of action to use additional medications. It’s similar to addition of hypertension medications. It’s really a matter of how comfortable you are using a combination of medications. It comes with practice style and familiarity with the agents.

Dr. Michael Gonzalez-Campoy (Minnesota Center for Obesity, Metabolism and Endocrinology, Eagan, MN): I would echo those comments. These agents are all approved as individual agents. The labeling states that they haven’t been studied in combination with others, so they’re really meant to be monotherapies. That doesn’t mean you can’t use your clinical acumen if you think a patient can tolerate a combination. I think that’s fair game. We’re dealing with a chronic disease, so applying the same model of management that we apply to other diseases is fair game. From a medical-legal perspective, you need to document that you’ve discussed with the patient that this is off-label use because it’s not in the label and you should protect yourself.

Q: What is the difference in weight loss and glycemic control between 1.8 mg Victoza and 3.0 mg Saxenda? Is there a big difference between the two that would justify Saxenda over Victoza in patients with diabetes?

Dr. Scott Kahan (George Washington University, Washington, DC): It’s a very straight answer. A dose-finding clinical trial found significantly better weight loss with the 3.0 mg dose. The 3.0 mg dose, Saxenda, is FDA approved for chronic treatment of obesity. 1.8 mg is approved for diabetes.

Q: If a patient cannot tolerate liraglutide, we should discontinue the medication. But how about putting the patient at the highest tolerable dose of 1.8 mg?

Dr. Felice Caldarella (Hunterdon Medical Center, Flemington, NJ): Per the label, the effective dose for liraglutide in obesity is 3.0 mg, so that would be totally off-label if you did that.

Q: How about the combination of naltrexone/bupropion? Is that agent effective as an antidepressant?

Dr. Caldarella: Bupropion is an antidepressant on its own. So there may be some qualities of it but it’s not indicated.

Dr. Still: But if you have a depressed smoker, it could be a good choice!

Q: Motivational interviewing is powerful, but how do you justify spending 10-15 minutes doing it? How do you bill for it?

Dr. Kahan: The skills and spirit of motivational interviewing guide every patient interaction. Even if you’re not doing it formally, the basic principles are relevant. Treat patients with respect and non-judgmentally and elicit their input in decision-making. While I can’t code for it, it’s an essential part of most interactions. To be fair, I am an obesity medicine specialist so almost every interaction I have has to do with obesity-related behavior changes.

Q: How does your program bring back real financial value for physician time with patients? Does a third-party free practice solve the problem?

Dr. Kahan: We’ve been doing a lot of work on increasing reimbursement, coverage, and access to medications and surgery. We’ve been making a lot of progress and setting the stage for more progress to come. The undertone of that question is frustration about not having reasonable coverage for things we need to do. We are well in the midst of change from a policy level.

Dr. Still: Dr. Kahan is being modest but he was instrumental, though this doesn’t directly affect endocrinologists, in getting CMS to reimburse behavioral treatment based on BMI for primary care and others. Hopefully subspecialists will be included in the next round.

Dr. Gonzalez-Campoy: For those in the US that provide services and want to get paid, Medicare recognizes mid-level providers like nurse practitioners and physician assistants. Those services can be billed under the umbrella of those providers.

Exhibit Hall


Abbott’s classic yellow signage quickly drew our attention as we approach the middle of the exhibit hall. The relatively large booth focused on the company’s FreeStyle InsuLinx product line, though representatives did offer optimistic commentary on the recently launched Precision Neo BGM. As expected, representatives did not comment on a timeline for FreeStyle Libre in the US. We were interested to hear that the representatives had gotten a handful of visitors to the booth asking about Libre, suggesting that word of the product’s EU popularity is beginning to spread stateside. We believe that at ADA, Abbott Diabetes Care would be smart to show its Precision kits measuring ketones – this is really needed for any type 1 taking SGLT-2s off label – and far more education on ketones in general is needed given that “hundreds” of patients have experienced the dreaded “euglycemic DKA” to date, according to Dr. Anne Peters.


AZ’s booth gave essentially equal attention to all of its diabetes products – somewhat of a contrast from the heavy emphasis on Farxiga (SGLT-2 dapagliflozin) and the new Bydureon (GLP-1 once-weekly exenatide) pen in the company’s display at ENDO. The Bydureon pen did receive some extra promotional attention, with representatives eager to pull attendees aside for demonstrations on their way to the frozen yogurt stand. While the pen is certainly an improvement over the old vial-syringe system, it still does look a tiny bit clunky when compared to the Trulicity pen across the aisle at Lilly’s booth – clearly Lilly learned a lot from Amylin in developing this product. However, there’s a lot of room for success here among multiple companies –the Trulicity launch is going well and there is stronger-than-ever uptake of the Bydureon pen and franchise since its launch. We think one reason why is that popularity of SGLT-2s is actually driving this since the two seem to work well in combination (more satiety coming from GLP-1 contributes to greater weight loss).


The Auto-Shield Duo (launched in August 2014) was the focus of BD’s small booth tucked in the back left corner of the exhibit hall. Friendly representatives provided a demo of the new pen needle product, while commenting on their excitement at the recent FDA 510(k) clearance of the company’s first-generation insulin infusion set. They acknowledged there has been little innovation in infusion sets over the years – “the weak link in the pump industry.” We are elated to see BD doing its part to change this and believe that Convatec will follow. That said, the question of how to break into this market – and cut into Unomedical’s (Convatec’s) near-monopoly – remains to be answered.


Dexcom’s small, understated booth was lightly trafficked near the entrance of the exhibit hall. The company’s Share receiver with built-in Bluetooth was the major focus of the booth with representative demoing the Share and Follow apps on their phones. While the apps were demoed on iPhones, reps emphasized the software was “coming soon” for Android – certainly by the end of the year but most likely much earlier. Anecdotally, they suggested that uptake of Share has been quick and driven by interest from caregivers looking for solutions to help pediatric patients. Ultimately, there appeared to be a high level of interest among attendees despite the broader endocrinology (including “other endo”) audience. Indeed, one representative shared that when she began working for Dexcom seven years ago, the most common question she received in the exhibit booth was, “What is CGM?” Now, the majority of visitors that approach her exclaim: “We love CGM. We love Dexcom.” Indeed, how far the field has come – but we would urge endos to spread the word since despite the enthusiasm, overall uptake is lower than we would have expected by this point.


Eisai had a small pop up booth set up near the back of the exhibit hall, solely focusing on obesity drug Belviq (lorcaserin) in its usual coral/navy blue/white color scheme. The booth’s back wall had three touch screens, which reps used to guide attendees through Belviq’s mechanism of action, clinical data, and patient support.


GSK occupied one of the larger booths on the exhibit hall floor, but it was tucked away on one side of the hall. As a result, foot traffic was quite low at the time we swung by. Consistent with previous conferences, GSK’s booth was fairly sparse, white, carpeted, and entirely focused on the fairly new once weekly GLP-1 agonist Tanzeum (albiglutide). The screens and displays at the booth stuck fairly closely to the prescribing instructions. Given what GSK has invested in booth space at recent conferences, we are somewhat surprised that the company hasn’t pursued more energetic displays for Tanzeum (for all intents and purposes their only major diabetes product on the market since Avandia’s move to generic status), although the once-weekly GLP-1 agonist class has certainly become more competitive with the entry of new options like Lilly’s Trulicity (dulaglutide).


Chocolate Omnipods greeted us at Insulet’s modest booth, tucked away toward the right side of the exhibit hall. We (politely) rebuffed their sugary offers, but took them up on the chance to play with the OmniPod and accompanying PDM. In the company’s classic style, posters of patients tended to display the OmniPod on the tricep, putting a clear emphasis on the product’s small size and wearability. We received confirmation in speaking to reps that the next-gen, Bluetooth-enabled OmniPod PDM will likely be displayed for the first time at ADA 2015. Get ready! The device will be a bit thicker than an iPhone 5 and will feature a color touchscreen and built-in blood glucose meter. We love to see that Insulet is thinking out of the box on this project, since getting Bluetooth, touchscreen, and the Dexcom integration are just a few of the innovations needed to stay ahead of the patient demand curve. No commentary was shared on Insulet’s work in the closed-loop field. C’mon guys, please don’t let the traditional pumps rule this arena …

J&J: Janssen

As at most diabetes conferences this year, J&J’s booth focused almost entirely on SGLT-2 Invokana (canagliflozin), with only a small corner devoted to the LifeScan division. Interestingly, Invokamet (canagliflozin/metformin) received essentially equal billing to standalone Invokana; we wonder if this reflects a growing popularity of the combination product or is perhaps an attempt to compete with other companies that have SGLT-2 inhibitor/DPP-4 inhibitor fixed-dose combinations (FDCs) on the market or in development – a sales representative did acknowledge that “everyone is interested” in this class. J&J management stated that the company had no plans for such a product as of 3Q14, though we learned recently that Mitsubishi Tanabe is currently conducting a trial investigating co-administration of Invokana with its DPP-4 inhibitor Tenelia (teneligliptin) that will clearly lay the groundwork for a FDC if it works. What else? Maybe J&J should buy Intarcia so it also can promote combo SGLT-2 and GLP-1. In terms of differentiating factors between Invokana and other SGLT-2 inhibitors, a representative noted that Invokana is the only product that has demonstrated superiority vs. active comparators in clinical trials, though she cautioned that there have been no head-to-head studies comparing the three SGLT-2 inhibitors to each other. J&J has clearly benefit from its first-mover advantage …

J&J: LifeScan/Animas

As noted, LifeScan/Animas’ occupied a very small portion of J&J’s overall Diabetes Care real estate. As expected, the Vibe with Dexcom G4 Platinum integration was the major focus of the Animas booth, with attendees encouraged to toggle through the interface and get comfortable with the device. Representatives remarked that early uptake has been quick and were optimistic on the future, despite the coming launch of Tandem’s integrated-CGM platform. They acknowledged that the lack of the meter remote to bolus has been a bit of a disadvantage, especially among women who appreciated the discretion. That said, they were quick to turn the conversation back to the positives of the Vibe, such as recently published results of an accuracy study (Capurro et al.; presented in a poster at ATTD 2015) that show improved dose accuracy, precision, and speed of delivery of the Vibe relative to Tandem’s t:slim and Insulet’s second-gen OmniPod pumps. We are not sure that the results are clinically meaningful. Stepping to the left, we arrived at LifeScan’s small, rectangular counter that displayed the company’s meters, including the OneTouch Verio recently launched in the US in 4Q14. This is the first time we can recall seeing the Verio display at a LifeScan booth; representatives highlighted the color-coding based on in-range results and motivational progress notifications. Both Animas and LifeScan reps were unable to comment on either a launch date for the Finesse (at 1Q15, the Finesse was expected to launch in 2017 at the earliest) or launch metrics on the Verio in the US. We personally are really enjoying the Verio (excellent on-meter branding – we see this more and more since Close Concerns’ insurance company Aetna now charges us $350/month for Abbott strips that go with Kelly’s Omnipod and $50/month for Life Scan strips … “so, yeah, no problem, Aetna, I’m glad to carry around yet another medical device even though this formulary decision means that my safety feature insulin-on-board now doesn’t work on my pump …”).


Lilly’s plush carpeted booth was the largest company-sponsored display of the AACE exhibit hall at 50’x50’, located on the front edge of the hall. Given a massive banner for the recently launched first-in-class Glyxambi (empagliflozin/linagliptin) in the conference center lobby, we were a bit surprised to see only a relatively small section of the booth devoted to       it. However, representatives at the adjacent Jardiance (empagliflozin) and Tradjenta (linagliptin) sections were eager to discuss the new combination. According to a sales representative, Lilly/BI are seeing Glyxambi prescribed primarily as a third-line option after metformin and a DPP-4 inhibitor or for patients already taking both an SGLT-2 inhibitor and a DPP-4 inhibitor as separate pills. This is somewhat distinct from Lilly management’s suggestion during the company’s 1Q15 update that Glyxambi should appeal primarily to patients on metformin alone and is unlikely to cannibalize Jardiance or Tradjenta sales to a large degree. Lilly’s other new offering, Trulicity (dulaglutide), occupied a more prominent position in the booth, with an entire wall devoted to its impressive clinical data and convenient delivery device – not a surprise given its strong post-launch trajectory. Lilly’s recent marketing push for Humulin R U500 was also evident, though the company appears to have dispensed with the materials we saw at ENDO featuring a woman trapped inside a giant syringe. Dr. Lisa Tannock (University of Kentucky, Lexington, KY) referred to these recent promotional efforts during her talk on insulin resistant patients at a Sanofi breakfast symposium, describing the product as “resurgent.” Indeed, Lilly stated in 2Q14 that Humulin U500 had been a significant driver of Humulin growth; on the flip side, we have also heard complaints from providers about the way Lilly raised the price for Humulin U500 sharply to achieve per-unit price parity with less concentrated formulations.


Medtronic classic light blue signage greeted visitors near the front of the exhibit hall. The small booth allowed attendees to play with the company’s recently launched 530G/Enlite with MiniMed platform; representatives seemed more enthused to talk about the coming MiniMed 640G/Enlite Enhanced CGM system although there was no pre-marketing. This was not on display though there is clearly a lot of excitement here. As a reminder, Medtronic plans to submit a PMA to the FDA for its MiniMed 640G/Enlite Enhanced CGM later this calendar year. Other pipeline items – such as the MiniMed Flex “hybrid pump,” the Enlite 3 sensor, and the type 2 business – were not mentioned in the booth, unsurprisingly since the products themselves are not approved.


In its typical green and white color scheme, Merck had a decently-sized booth closer to the front of the exhibit hall. Alongside its trusty froyo station, the booth highlighted the company’s Januvia (sitagliptin) franchise, with two boards on Januvia and one on Janumet (sitagliptin/metformin) with clinical data, label information, and images of smiling patients. Unsurprisingly, there was nothing on the FDC product that it has with Pfizer or smart insulin or regular insulin.

Novo Nordisk: Diabetes

The company’s centrally located medium-sized diabetes booth followed a familiar path, with a primary focus on Victoza (liraglutide) and Levemir (insulin detemir). The Victoza displays had screens with a plant and root theme, while Levemir displays largely focused on the fairly new FlexTouch pen. There was a large multiscreen display with an animation of the FlexTouch, which is an ergonomic improvement over the previous FlexPen. There were also desks where patients could get some coffee or sign up for the AACE Fun Run (this will be cool!). Consistent with ENDO, the newest offering in Novo Nordisk’s exhibit hall presence was found a few booths away in the Saxenda display (see below).

Novo Nordisk: Obesity

The company also had two booths devoted to obesity, one focused on the recently launched Saxenda (liraglutide 3.0 mg) and the second focused on general obesity management. Under a large banner with “Saxenda” displayed in purple text, the Saxenda booth was modestly sized but relatively busy. Most notably, we heard new details (just released today, according to Dr. Apovian) on Saxenda’s savings program under which patients in prescription plans that cover Saxenda can pay no more than $30 per month for up to 12 prescriptions and patients who pay out of pocket can save $200 per month (this information can also be found at with a log-in or registration). It will be hard for the smaller companies to compete with this. The booth also gave attendees the opportunity to get a feel for the drug’s purple FlexTouch pen (see our coverage of the approval for more on this), which we learned lasts for six days (each monthly prescription includes five pens). In addition, reps expressed hopes for an improving reimbursement environment and stated that many attendees have asked about the potential overlap between Victoza (liraglutide 1.8 mg) and Saxenda; they have stressed that the distinction will ultimately be based on clinicians’ judgment of what the primary clinical goal is. Novo Nordisk’s second booth, branded “Rethink Obesity” (which offered tea to attendees), focused strictly on general obesity management with reps guiding attendees through education on the basic science behind obesity and providing handouts on resources and tools. While its real estate also remained relatively moderate, reps expressed that the booth has similarly been very active and attracted great interest.


The company had perhaps the best placed booth, right in front of the only main entrance to the exhibit hall. The booth footprint was modest, but all of Sanofi’s US diabetes products were represented. Staring right at you as you entered was a big model of the Toujeo SoloStar pen (a very cool pen) and displays on Toujeo (insulin glargine U300), Sanofi’s new basal insulin option that it hopes will be a successor to market-leading Lantus. The booth seemed to be catching plenty of passers-by due to its placement, and, as a result, a fair number of attendees were receiving introductions to Toujeo’s more stable PK/PD profile and other benefits. A key question in our mind has been whether Sanofi/MannKind’s new inhaled insulin Afrezza might suffer due to Sanofi’s major push on Toujeo. Indeed, and what a missed opportunity in our view, Afrezza was not visible from the front of the booth – similar to the booth layout at ENDO, you had to walk to the inside or around the back to see displays on Afrezza. Reps here framed Afrezza fairly modestly as a “tool” that may be best for some patients rather than a dazzling game-changer – this is probably a wise move to keep expectations at an appropriate level though it certainly is a game-changer for some patients. Booth staff acknowledged that Afrezza is less a product that endos might introduce to every patient and more the type of product that will spread by word of mouth – we’d love to see both. We noticed some updated displays on the lung function test requirements for Afrezza. Very responsibly, the signage made it clear when testing is recommended (before initiation, after six months of use, and annually thereafter) and also mentioned the percentage of patients who smoke or have asthma or COPD to underscore the size of the patient population for whom Afrezza is not the best option. Lantus and Apidra were represented in back corners of the booth and were not receiving much attention since people know what these are.    


Tandem’s modest booth was located on the right side of the product theater area of the exhibit hall. The company did not have any new marketing or products to share information about – as a reminder, FDA approval/launch of the G4 Platinum integrated t:slim is expected in 2H15 while shipments of the 480-unit t:flex are expected in June.


Vivus also had a small, pop-up booth near the back of the hall, which included a medical booth to the side and a few of iPads at the front that featured the Qsymia website, allowing attendees to scroll through the drug’s safety and efficacy information, savings programs, and more. Although Vivus had a small presence in the exhibit hall, it did support a product theater lunch on Qsymia during the day, which focused on the drug’s clinical data and label information along with a focus on best practices for provider/patient communication.


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