Memorandum
At The diaTribe Foundation and TCOYD’s 12th Annual Diabetes Forum on Monday night, Drs. James Gavin, Mikhail Kosiborod, Jeremy Pettus, and Carol Wysham comprised an insightful panel moderated by Dr. Steve Edelman and our own Adam Brown.
-
First tackling the role of CVOTs in the modern era of diabetes, panelists were generally in favor of the requirement continuing. Dr. Kosiborod pointed out that, without CVOTs, our knowledge of diabetes therapies would be hindered at best and inaccurate at worst; HCPs would think DPP-4 inhibitors were cardioprotective (as phase 3 data indicated) and the cardio- and renal-protective effects of SGLT-2s and GLP-1s could have gone unnoticed. While Dr. Edelman, ever the devil’s advocate, argued that some of the safety signals to come out of CVOTs have allowed entire classes to be vilified, Dr. Gavin countered that those safety signals are the purpose of CVOTs – and they’ve also reminded HCPs to be more vigilant about screening for complications. Dr. Wysham, for her part, asserted that CVOTs should become optional, to be conducted when a company thinks their therapy is actually cardioprotective; we see the appeal of this approach, given the CVOT requirement can be a significant barrier to novel drug development. Of course, the most successful, cost-bending diabetes therapies will need to show evidence of cardioprotection, renal protection, or other major short- and long-term cost-savings – meaning companies will need to invest here at some point.
-
In technology, we loved hearing enthusiasm regarding the role of CGM in type 2 – as well as optimism about where reimbursement is going. Dr. Wysham has every patient she sees put on professional CGM before their first visit, which she finds especially useful for patients who need insulin or GLP-1 initiation or intensification. Dr. Gavin was similarly positive and asserted that more CGM use will allow patients to focus on time-in-range instead of chasing A1c, which will both help patients feel better, connect hyper/hypoglycemia to moods/energy, and improve outcomes. Dr. Edelman argued against the use of blinded CGM, as he thinks much of the utility of CGM comes from engaging patients and promoting behavior change with real-time results. Rounding things out, Dr. Pettus reminded the audience of the remarkable innovation that has occurred in CGM in the last few months: FDA approval of Senseonics’ 90-day Eversense and the no-calibration Dexcom G6.
-
All of this said, however, the question of access hung over the entire conversation, and every panelist expressed serious concern about the ability of most patients to access the best drugs and technology. Dr. Kosiborod laid out the inherent difficultly of drug and tech development, in which coming up with lifesaving treatments takes a huge investment and results in a world where patients cannot afford the therapies they need. Added Dr. Gavin, “We can have all the best technology and drugs in the world, but the elephant [in the room] is getting access to those that need it most.” Explaining what this looks like in practice, Dr. Wysham said that she develops a treatment plan then tells her patients, “We’ll see what your insurance says,” illustrating just how nebulous the reimbursement and coverage landscape is, even to providers.
-
Finally, adjunct therapy for type 1 – a definite theme of this meeting – also spent time on center-stage; Dr. Pettus highlighted Friday’s ATTD-sponsored consensus meeting on DKA risk with SGLT inhibitors as his favorite part of ADA 2018, and Dr. Edelman emphasized that while the DKA concern with these agents in type 1 is real, the increase isn’t above and beyond the background rate of DKA in the general type 1 population. While we continue to think this is an interesting point, we also maintain that it’s difficult to make comparisons between real-world observations and clinical trial DKA rates, and it’s very possible that the real-world DKA rates with SGLTs could be higher than those seen in phase 3 trials. On the other hand, Dr. Wysham argued that DKA is extremely easy to prevent with education, and she has had only two cases in her three decades of practice; the problem, she suspects, will arise when these therapies start being prescribed by providers without experience managing diabetes.
-
Below, you’ll find the full transcript of the Forum with some of our favorite moments highlighted in yellow.
Panel Discussion
Panelists: Jim Gavin, MD (Healing Our Village, Atlanta, GA), Mikhail Kosiborod, MD (Saint Luke’s Health System, Kansas City, MO), Jeremy Pettus, MD (University of California, San Diego, CA), Carol Wysham, MD (University of Washington, Spokane, WA)
Moderators: Adam Brown (Close Concerns, San Francisco, CA) and Steve Edelman, MD (University of California, San Diego, CA)
Dr. Steve Edelman: To start things off, I’ll ask the panel to tell us: What was most significant to you at this ADA?
Dr. Mikhail Kosiborod: If you ask me as a cardiologist, the video you just showed was the highlight of ADA. I have to say, when it comes to cardiovascular disease and complications, which is what I’m most passionate about, it has not been a particularly impactful conference. There’s a real dearth of new information, nothing really status-changing. I usually don’t bring up subgroup analyses as highlights, but a CANVAS analysis presented today was quite reassuring if not necessarily unexpected. It found a consistent reduction in CV events in patients with significantly reduced renal function, so there’s no difference in the CV benefit of canagliflozin regardless of baseline renal function. Of course, this is a common issue, particularly given the label restrictions on SGLT-2 inhibitors.
Dr. Edelman: And also OBSERVE-4D?
Dr. Kosiborod: Yes, OBSERVE-4D was presented yesterday, and that’s very useful data. This is a real-world evaluation of a large claims dataset, including several hundreds of thousands of patients in the US, trying to understand how many patients treated with canagliflozin had amputations. It found no signal for increased risk with canagliflozin, but you have to remember that the majority of patients in the study had very low risk vs. those in CANVAS. The methodology was quite solid, and they did their due diligence, looking at the subpopulation with CVD where there was also no increased risk detected. One limitation is that the average duration of follow-up is six months in OBSERVE-4D, which is probably not adequate.
Dr. Carol Wysham: I have to say that I’ve really enjoyed handling the technology, seeing the devices, and talking to the representatives in the exhibit hall – living in Spokane I don’t get to do this often. I also went to a clinical-translational program looking at the cardio-renal axis and how SGLT inhibitors impact sodium-hydrogen exchange, explaining why these agents have benefits at low eGFR. That was great, because I think we’re all wondering how these agents are working even with low renal function.
Dr. Jim Gavin: OBSERVE-4D was going to be mine; this is so important because we now have data-driven information about lower extremity amputations and canagliflozin, alone and vs. other anti-hyperglycemic agents. We can use this data-driven narrative to combat some of the hysteria that has been launched around this issue. This is extremely important: There’s a lot at stake in terms of risk to the class, and now we have a piece of work to help with that. Otherwise, there was nothing big and earth-shattering at this meeting so I got to explore some, and there is this small molecule I looked into covered in three posters: ZGN-1061, [Zafgen’s] methionine aminopeptidase 2 (MetAP2) inhibitor. This is a fascinating molecule originally explored as an anti-angiogenic drug for cancer. It’s been repurposed and is absolutely wonderful in terms of mechanism of action. It can actually reduce weight and glycemia in type 2 diabetes, as it also reduces hepatic liver scores in NAFLD, so this is a wonderful technology that combines well with other things and could be a whole new class.
Dr. Jeremy Pettus: I’ll focus on a couple topics I thought were interesting: The use of adjunctive therapies in type 1 and artificial pancreas technology. First, for SGLT-2 and 1/2 inhibitors, there’s no doubt these drugs work on time-in-range, weight, and blood pressure, but they do also increase risk of DKA and I believe this is a class effect. So now we have to focus on risk/benefit and mitigating DKA. My favorite thing that I did at this meeting was a Friday morning ATTD-sponsored meeting of all the stakeholders, including JDRF, FDA, industry, and KOLs involved in this research. We were all in this really small room, I counted 50 people, with no microphones and we were just hashing this out, focusing really on what type of ketone monitoring should be involved. We didn’t really come to the clearest consensus, but we’re getting there – and everybody was there. Everyone showed up because people are finally getting it that these medications have role in type 1. Even if the A1c reduction is not that great, this idea of outcomes beyond A1c is really coming home to roost. These drugs make a difference in quality of life and help people manage a disease that, frankly, really sucks. Any drug that can help is an improvement, and all these people being there shows that this risk of DKA is worth it, and we want to come to a solution.
Mr. Adam Brown: If you were the manufacturer of an SGLT inhibitor and had to design the perfect way to mitigate risk, what would it be? Should ketone meters be shipped when the drug is prescribed to a type 1?
Dr. Pettus: That could be a six-hour conversation. It’s going to require more awareness first. But whatever solution we come up with will mitigate the risk of people on these drugs, as well as DKA risk, period. DKA rates in general are still unacceptably high, and there’s a lack of awareness about what ketone monitoring is.
Dr. Edelman: In the general type 1 population, the annual rate of DKA is about 5%, and with risk mitigation strategies in these trials it was 2-3% depending on the dose. Jeremy and I did a focus group of six endocrinologists and six patients with type 1. We separated them after seeing this data, the endocrinologists said, “Oh, but it doesn’t really lower A1c that much.” With the patients, the word “A1c” didn’t come up once. Three hours more time in range a day is significant to them!
Dr. Wysham: Especially if the time-in-range increase is while you’re awake. The point I want to make is that DKA is incredibly easy to prevent. With really good education, I’ve had two cases of DKA in my own patients in 30 years.
Dr. Edelman: And how many unnecessary admissions to the hospital happen because of a pump infusion line leaking and the patient doesn’t know what to do? Just general education on DKA would seriously reduce what we spend.
Mr. Brown: Can you prescribe a ketone meter when someone goes on these? Or have it come with a card that says how to measure and what ketone levels should be?
Dr. Wysham: It works if people use it. We have to educate patients on when to use a meter and how to react to the information it gives. It takes a lot of knowledge from prescribers, and those of us at this table can do it. I worry about it when it gets beyond us.
Dr. Edelman: Let’s shift gears a little bit: Do you think we still need the CVOT mandate?
Dr. Kosiborod: If you want to be brief – yes, we still need CVOTs. I think that we wouldn’t be talking about most of what we’re talking about today without them. There may be certain reasons a mandate happened but we’re beyond that. What’s important now is how do we mange type 2 diabetes? And where would the field be going today if these CVOTs never happened? We would be at a tremendous disadvantage, and we would think one drug that lowers A1c is the same as another. Patients are going to dramatically benefit because of these.
Dr. Gavin: CVOTs have served us well, and we still need them, not only for giving us evidence on CV safety, but also because it has given us information on other forms of safety. There is a problem, though, which is this: How are we going to continue these? If you’re going to do a randomized controlled trial against standard-of-care, what do we do when all of these drugs are out there with all of these benefits? We’re going to have a real dilemma with what all of these trials look like going forward.
Dr. Edelman: Do you think the ADA treatment algorithm will change? All these trial protocols are different – not all CVOTs are created equal, and how much money the company has can determine whether it’s powered or superiority or not.
Dr. Wysham: I think it should be optional. If you have a molecule you think is effective in addition to standard of care, companies can have the option. I do think the ADA algorithm should be re-written. If CVD exists, it’s appropriate to examine the individual’s other comorbidities and see which agent has been proven effective and start that concurrently with metformin. I’d like to see the ADA/AHA CV risk score that comes up include duration of diabetes and other comorbidities better.
Dr. Edelman: Some negative things have come out of CVOTs too: amputations, intraocular injections, other imbalances…and then the drug gets vilified. They’re large studies and they find weird stuff.
Dr. Gavin: The incidence of these adverse events is small, though, and what they’ve served to do is remind us to be more diligent about what we’re doing anyway. We ought to look at people’s eyes and assess the competence of the vasculature in their extremities. It has raised the bar in terms of how diligent we need to be.
Dr. Kosiborod: And it’s not about vilifying a drug. We need to make informed treatment decisions about the patients we see and give them adequate information about risks and benefits of different treatment options. If we didn’t have these data, we’d still be telling patients DPP-4 inhibitors reduce CV events by 45% [based on phase 3 data]. That would not be good patient care; we’d be telling them something that is completely false. The amputation story needs to be told; it could certainly be the case that this is a real risk, and with more trials coming down the pike, we’ll eventually know. If you want to inform patients, give them valid info, and let them make informed decisions; we need these trials. False information is worse than no information.
Dr. Edelman: But aren’t we forgetting about blood pressure and cholesterol now? Those have a significant impact on CVD.
Dr. Gavin: I think if you’re forced to be more diligent in terms of monitoring vasculature and eye health, you’ll also be more vigilant about checking blood pressure.
Mr. Brown: It makes me think, if you’re a pharmaceutical company, a $200 million trial is a big investment. How far could that investment go in terms of developing better prescribing tools and titration? For example, using clinical decision support for basal insulin titration. How do you think about that: More investment in long-term studies and new molecules, or technology to better use what we already have?
Dr. Wysham: And also putting systems in place to help patients without access to technology.
Mr. Brown: The drugs we have are great, but should we use them in smarter ways by creating tools to use and prescribe them better?
Dr. Kosiborod: It’s impossible to disagree with a statement like that.
Dr. Gavin: The elephant in the room is access. We can have all the best technology and drugs in the world, but the elephant is getting access to those that need it most.
Dr. Kosiborod: What worries me is that we produce all of this knowledge and we can’t implement it. It’s not just for the reasons you stated – funds do need to be directed elsewhere – but a large part of this is that patients can’t afford lifesaving treatments. Contrast that with the fact that coming up with these treatments requires a big investment. We live in a world where we need to have returns to make sure life-saving treatments are developed. However, I grew up in the Soviet Union where we had price controls, and then you had situations where you would be in line for 3 hours to get a chicken.
Dr. Gavin: Three hours for one chicken?! [Laughter]
Dr. Kosiborod: Just one chicken.
Dr. Edelman: The big thing for me at this meeting was type 1 – sensors and artificial pancreas progression. Dr. Pettus, what’s your impression on progress of artificial pancreas and hybrid closed loop programs?
Dr. Pettus: Artificial pancreas technology is now moving along at a rapid clip. In three years, people will be choosing between hybrid closed loops, if not fully closed loops. What I’ve been really excited about is DIY systems. I use Loop, a DIY system that automates basal insulin delivery. They’re trying to design a study to get baseline metrics on the thousands of people across the country using this grassroots system. FDA is completely on board with this approach, and they’ve realized they can’t put their head in the sand about all these patients and are trying to accelerate the process for commercializing this and getting it into the hands of the masses. Whether you agree with these DIY systems or not, some of these will become kind of mainstream. There will be lots of competition in this space one way or another, and closed loop systems will make huge difference.
Mr. Brown: How many people in this room are on Loop? Looks like seven – that’s pretty striking right there.
Dr. Pettus: FDA does have different regulatory questions, as opposed to taking a new experimental product and getting it approved. All these people are already using it.
Mr. Brown: I also found it notable that this company SOOIL from South Korea plans to submit to FDA with an open-protocol pump, tailored for the DIY community. In Europe they actually market that: “Our pump is compatible with the DIY community.”
Dr. Pettus: The bottom line is that this DIY artificial pancreas technology is moving from a fringe, techie person thing to a more mainstream practice.
Mr. Brown: What are your current thoughts on CGM, especially in type 2 diabetes? How is that progressing?
Dr. Wysham: The coverage is much better. It’s something that’s very impactful. I have a lot of experience of using professional CGM and seeing how patients respond; not everyone wants to sign up for that option; but when patients are on MDI, the expectation now is that CGM is standard-of-care, and we work hard to get it. Libre is actually affordable even if not covered – it’s a huge part of my practice.
Dr. Edelman: If you can really engage patients to look at results, I think that’s the key role of CGMs. I’ve seen some patients really turn around after having a CGM. I do think it’s going to be more accessible and affordable, and I do think it will change the way we practice with type 2 diabetes patients.
Dr. Pettus: We just had Senseonics’ Eversense approved by FDA, so that’ll be another option, and it currently lasts three months. Dexcom G6 was also recently approved and launched with no fingerstick calibration; I’ve used it for a couple months now and have checked my blood glucose with a fingersticks maybe 3 times – that’s amazing. A huge change is happening right now that can lower barriers to providing CGM. People ask me, “How accurate is G6?” I say, “I have no freaking clue. I don’t fingerstick and I don’t want to.” [Laughter] People diagnosed today are in a world where fingersticks can be far less important, and they can even go away. (Read diaTribe’s G6 test drive here.)
Dr. Gavin: Addressing the original question about type 2s, I think that having the engagement that’s now possible through CGM is so important, and it enables a new narrative about what it really means to control diabetes. It’s less about chasing biochemical metrics and more about looking to see how much time-in-range you have. The flatter that line is, the better they feel.
Dr. Edelman: Time-in-range is starting to come out in type 2 studies. People with type 2 hate pricking their fingers. No one likes it. The biggest headache is that people don’t test and they make up numbers, so Libre is making a big difference there.
Mr. Brown: Do you think about a spectrum of patients with type 2, in terms of some people being a great candidate for professional or real-time CGM? Or perhaps a quarterly professional CGM?
Dr. Edelman: It’s like adult diapers – it depends. At one end there are some people with prediabetes, and MDI or pump users on the other. There are different ways to use CGM in the type 2 population. My huge bias is that I’m totally against blinded CGM except for use in trials. People need to see their numbers. They may change their behavior, and some people say you won’t know if a medication made a difference. But if it does change their behavior, that’s diagnosis and treatment. As it gets cheaper, easier to apply, and more accessible, it will make a difference – but you have to engage patients.
Dr. Gavin: You know the high-risk groups. We will get to a point with CGMs where they are more accessible and more patient friendly, and we will be able to stop fooling ourselves that there are people with prediabetes: they have diabetes and we just don’t know when it happened. We will be able to use technology to get real time data to better inform decisions about what their status really is.
Dr. Wysham: I do use the professional version on all of my patients before their first visit. Patients can’t make an appointment with me until they see a diabetes educator, who puts them on professional CGM. When I see them, we’re ready to talk. It’s incredibly helpful for those patients who need insulin, or for intensification of insulin, or to add a GLP-1 agonist. It’s so much easier to do that than it is to urge and irk them on.
Dr. Edelman: That’s impressive.
Mr. Brown: Do you find the reimbursement for professional CGM is solid?
Dr. Wysham: It’s great – much better than for real-time.
Dr. Edelman: This was the year of the alpha cell. We’re seeing a lot of glucagon antagonist and companies coming up with strategies to improve glucagon sensitization. Dr. Pettus has been really interested in glucagon lately.
Dr. Pettus: I think SGLT inhibitors have opened the door for adjunctive therapies in type 1, and I think glucagon receptor antagonists definitely work on improving glycemic control. And I really think they reduce the risk of DKA, because glucagon normally induces ketone formation. There will be some baggage on liver effects and alpha cell hyperplasia, but we’re entering an era where we can address multiple defects of type 1. Don’t forget that people with type 1 have a problem with glucagon, too.
Dr. Edelman: Even in the future we’ll have mini-dose glucagon, and soon we’ll have nasal glucagon. There’s a big advancement coming in that area. I’m looking forward to going to the fridge at night to get a squirt of glucagon.
Mr. Brown: What’s your preferred type 2 treatment algorithm if cost isn’t an issue?
Dr. Edelman: I saw Mikhail’s, and it’s amazing.
Dr. Kosiborod: Maybe I should patent it. My viewpoint really has been that we don’t treat type 2 diabetes to make glucose numbers look better. The goal of treatment is to make people live longer and feel better, and hopefully both. To do that you need to prevent complications. There are significant differences in treatment options to reduce complications and in the timeframe over which they are reduced. Some can do it quickly and some take longer, and others we’re not sure if they ever will. I’m a cardiologist for people with diabetes, and I want to avoid cardiovascular events. So I ask where a patient is on the CVD continuum? I look at ASCVD, HF, stroke, and MI, and I choose treatments based on the best evidence available, much of which is from CVOTs. That tells me the likelihood of people benefitting from a class or several classes to prevent CV events. I tailor it based on the patient in front of me.
Dr. Wysham: I always tell patients what I think they should be on, then I say ‘we’ll see what your insurance says.’ All things being equal, I’m a fan of early intervention with agents shown to have more durability if started early: SGLT-2 inhibitors and GLP-1 agonists. I’d follow Dr. Ralph DeFronzo’s suggestion to get three medications on as early as possible to spare the beta cell. The RISE protocol was announced; they’re trying to see if they can do early intervention with liraglutide and metformin to prolong beta cell function; unfortunately, I think they missed out on including an SGLT-2 inhibitor.
Dr. Gavin: Metformin, GLP-1 agonists, SGLT-2 inhibitors, and maybe under some circumstances a whiff of pioglitazone. I learned well from Dr. DeFronzo.
Dr. Kosiborod: If you look at what would make sense to preserve the beta cell and reduce insulin resistance, the algorithm you just described also has the highest likelihood of reducing cardiovascular and renal disease. This will move all those endpoints in the right direction.
Mr. Brown: If you could have a billboard all over the world with one message to healthcare providers, what would it say?
Dr. Wysham: Don’t delay.
Dr. Gavin: It’s the combination, stupid.
Dr. Kosiborod: Make people with diabetes live longer.
Dr. Pettus: Vote for Steve Edelman. [Laughter]