← Diabetes Core Update
Diabetes Core Update
Special Edition: The Cardiovascular Outcome Trials – Origin and Perspective
00:31:05
In this special series on The Cardiovascular Outcome Trials our host, Dr. Neil Skolnik will discuss the history and importance of the Cardiovascular Outcome Trials for diabetes medications . In Part 1 we discuss an historical perspective with Dr....
Loading summary
Transcript29 lines
- [00:03]
A
Welcome to this special two part series of Diabetes Core Update where we will discuss what is arguably the most important advance in the care of people with diabetes since the discovery of insulin, and that is the development of the cardiovascular outcome trials, the cvots. I'm your host, Dr. Neal Skolnick, professor of Family and Community Medicine at the Sidney Kimmel Medical College of Thomas Jefferson University. This special series of Diabetes Core Update is sponsored by Lilly. Today we're going to put these trials into historical context, how they began and discuss their importance. And we are so fortunate. Joining us to accomplish this is Dr. Steve Nissen. Dr. Nissen is Chief Academic Officer of the Heart and Vascular Institute at the Cleveland Clinic and Professor of Medicine at the Lerner College of Medicine. He served as the 2006-2007 President of the American College of Cardiology and was named one of the world's 100 most influential people by Time magazine in 2007. In 2015, Dr. Nissen was named by Thomson Reuters as one of the world's most highly cited physician scientists. Important for today's discussion is that his work, which I will say was courageous, was probably the single greatest catalyst for the FDA requiring CVOTs beginning in 2008. Steve, welcome to our podcast.
- [01:41]
B
Neal, thank you so much for having me and for doing this because I think it is a good discussion to have.
- [01:47]
A
Let's start back in the early 2000s or even prior. When did you begin to think about A1C not being the best measure of efficacy for, for diabetes medicines and that it might not actually have a direct relationship with diabetes outcomes?
- [02:06]
B
Well, first of all, I have never been a fan of surrogate endpoints. You know, a drug can have biochemical effects that may not in fact predict its true biological effects. And we've seen this over the years with a variety of surrogate outcomes. To me, it did not make sense that a drug could be approved with a pretty small trial simply by showing that it lowered blood sugar. We didn't really have good evidence that control of blood sugar was the principal driver of cardiovascular adverse outcomes. And then I got a wake up call and the wake up call came with a drug that nobody knows about now called Muraglitazar. And Miraglitazar was a dual PPAR agonist. And I was interested in these PPARs because they had promise for potential cardiovascular benefits. And this drug was really powerful for people with the highest A1Cs. It lowered A1C by about 2%. And because of its PPAR alpha effects, it also had favorable effects on lipids, triglycerides went down, that sort of thing. So it looked like it was a dual acting drug that could be a really a breakthrough drug. It came before the FDA panel and I was a member of the cardiorenal panel, this was the endocrine and metabolism panel. And I was looking because I was interested in the class. And what happened was I saw the information that was on the FDA website. Now something very pivotal happened. A public citizen had sued the FDA because the year prior and all previous history, the materials submitted by a drug company for drug approval were considered confidential and not in the public domain, as unbelievable as that may seem. And Public Citizen won the lawsuit and FDA was required to post the briefing materials that were used to consider drugs for approval. And I'm looking through the briefing materials and every trial there were substantially more cardiovascular events in the group that was treated with muraglitazar compared with placebo or other comparators. And so I huddled with my statistician who has been working with me now for more than 20 years, Kathy Wolsky. And I said, let's take all the data that the FDA has posted and let's do our own analysis. And in the meantime, the FDA panel had voted something like 8 to 4 for approval and they were ready to launch the drug. When we analyzed the data, everything, all the cardiovascular events went in the wrong direction and they were statistically significant, literally a doubling of morbidity mortality with Miraglitazar. And so I called up the editor of the JAMA and Kathy DeAngelis, who's a really wonderful person, and I told her what we had, and I said, I really want you to try to expedite review because this drug's coming up, it's going to be approved very shortly. And so JAMA moved it quickly through, published it, and after the publication, the FDA decided that maybe they shouldn't approve it after all. The company kept the trials open, and the further out they went, the worse it looked. And the trial and the drug never saw the light of day. So it went away. So now I knew that it was possible for a drug to profoundly lower blood sugar, but actually harm patients with respect to cardiovascular outcomes. And that sensitized me. And then two more trials were published with another drug, Rosiglitazone. One was the Adopt trial, and that was a glycemic durability trial showing that the drug was more durable. And the other was the Dream trial, showing that the drug could prevent new onset diabetes. And I looked at those two trials and the events all went in the wrong direction. In fact, for dream, there were 75 events with the rosiglitazone group and 55 with placebo. And the hazard ratio was elevated and the P value for harm was 0.08. So I wrote a letter to the editor at the Lancet. It was published in the Lancet. And I said, look, given this directional trend, it rules out benefit. There's no way there's going to be benefit. And it suggests there may be harm, but that's as far as I could take it. And then serendipity arrives. And the long discredited Attorney General of New York, Elliot Spitzer, had sued the maker of Rosiglitazone because the company had suppressed information about suicide and suicidality in young people taking their antidepressant depressant drug. And he thought that was not in the public interest for that data to have been suppressed. And so he got a settlement. And the settlement was that the company, in this case, GlaxoSmithKline, would pay a dollar, but they would agree to put all of their data from all their clinical trials on a website, which they did. And the website was out there. And I was looking for data on Rosiglitazone because I had concerns about it. I thought it might be like Miraglitazar. And quite by serendipity with the search engine, I found the not very publicly available, not very easily found website. And there were, As I recall, 42 trials of rosiglitazone that had been conducted, 35 of which were unpublished.
- [08:45]
A
Wow.
- [08:46]
B
And I downloaded all of those trials and I sent a one word email to my statistician, Kathy Wolsky, entitled Eureka, I have found it. And she and I huddled and she did a meta analysis, and it only took her about 24 hours. And we looked at the data and there was a increased risk of myocardial infarction in the range of around 40% and a similarly sized increase in the risk of cardiovascular death. And these were both? Well, the MI was statistically significant. The death was borderline. It was.0506 for a P value. And so I talked with the editors of the New England Journal and I said, look, you know, I think this belongs in the public domain. And literally locked myself in the office and wrote that manuscript in about 24 hours and submitted it on May 1, 2007. It was ultimately published on May 21. So fast. For New England Journal, it's unprecedented. It's 21 days. But about a week after I submitted the manuscript, I got a call from the chief medical officer of the company gsk, saying he needed to meet with me urgently. And I knew what was going on, and I agreed to have him come and meet with me, with his team. And they came and they said, look, we know you're interested in our drug. We'd like to work with you and do an analysis together. You know, our statisticians can do the analysis. And I said, no, we cannot work in that way. We must do the analysis. And that relationship didn't go very far. I only learned a couple of years later that one of the reviewers for the New England Journal, a very prominent diabetologist, had as soon as he got the manuscript to review, he leaked it to the company so that they knew it was coming and they could plan the public relations campaign to discredit both the manuscript and me personally. Very rough time. Very, very rough time.
- [11:17]
A
I'll bet. The phrase that comes to mind is what John Lewis called good trouble. Doing things for the right reason and having the courage to stick to what you is correct, even if it makes others uncomfortable. And this really, to me, reminds me of that ideal.
- [11:41]
B
Well, what's kind of interesting about this is that some years later, I was me, I won't tell you who with. A very prominent person in the Congress who used that very term good trouble to refer to this manuscript. So that is interesting. And I did get to meet John Lewis at one point, who is a fabulous guy. But the bottom line is I was Persona non grata with most endocrinologists because they were getting calls from all their patients. You know, we're in a panic when the manuscript came out. It was the lead story on the front page of the Wall Street Journal. It was everywhere. And it, you know, caused a huge amount of controversy. And everybody was coming off the bench and criticizing our methodology, our findings. And we knew that the meta analysis that we had done, which was not a patient level, it was a study level meta analysis. We knew that it was not perfect, but we thought it belonged in the public domain. And I was willing to take the heat. The ADA very graciously invited me to come to it. A major plenary at the ADA meeting. And I went on stage with Dr. Holm, who was doing the record trial, and we had kind of a mini debate in front of a huge audience at the ada. I felt like I was in. In a den of lions, because I don't think anybody in there had any love for me at that point. But. But we got it done. And the controversy led to a lot of discussion.
- [13:21]
A
I'll bet it did. And I will say ada, as long as I've been going to their science sessions, really does a great job of airing multiple points of view in debates like that. And all of us in the audience always learn because we were able to then make an assessment based on a range of opinion and data. Now the implications of this publication were enormous because I remember as a practicing clinician when this came out and because of the Adopt trial that had come out roughly six months before, showing Rosiglitazone had much longer glycemic durability. Many people like myself began using Rosiglitazone as a preferred first line agent. And this came out and it was like the world shook. What, what then happened as we move forward in this story?
- [14:19]
B
Yeah, by the way, you were right. At the time that we published this, it was the largest selling diabetes drug in the world, Rosiglitosone. So it was not, you know, it was not something that wasn't being used. It was being used. So FDA and I made one very big mistake. I should have shared the data with FDA before we published it, but they didn't react terribly well to it. Bureaucrats don't like it when they feel like they're being criticized and that's the way they felt. But they warmed up to it and they scheduled an advisory panel meeting for 2008, the next year. And they invited me to present, which again I give FDA credit for. And I did a lot of thinking about what should we do? And I went there to that meeting and was almost exclusively the panel was the endocrine metabolism panel. So it was essentially endocrinologists, diabetologists. And what I thought about was, look, if we require a full blown outcome trial before approval, then it will slow down new drug approval and diabetes. And so I proposed a two stage approval process where you first had to do a modest size trial to rule out a high level of harm, a hazard ratio of 1.8. But then you had to, after the drug you could get to the market if you ruled out 1.8, but then you had to go on and rule out 1.3 for an upper hazard ratio, which is a more substantial trial, in order to stay on the market. The idea was not to put a shell on new drug development, but to make sure that the medical community ultimately got good information about both the benefits and risks of new therapies that came along. Now there were opponents of that.
- [16:22]
A
Now Steve, before you go on, I just want to clarify for our listeners who might not know. So when we're saying that 1.8 and 1.3. These were for non inferiority, that there could not be a greater than that amount of bad outcomes.
- [16:39]
B
That's a very good point you make. Absolutely. However, I will tell you that in my final couple of slides of that presentation, I said if you require non inferiority trials, sooner or later one or more of these drugs is actually going to show benefits. And it was very clear. Now, what's also interesting is there were two more speakers at the external speakers at the panel. I was followed immediately by Rob Califf, who ultimately went on and became FDA Commissioner many years later, and then by Rory Holman, who many of your listeners will know is a legend in diabetes, having formed the UK pds. And both Ruri and Rob supported my proposal. And they hadn't seen it before, but when they saw it, they said it's a sensible, balanced proposal. I thought there was zero chance that the FDA panel would vote in favor of it because I was asking the diabetes community to do something difficult, which is accept that the drugs couldn't get to market simply because they lowered blood sugar, but that they had to show something else. And I will tell you, I feel very strongly about this. The benefits have to be there for these drugs to be valuable.
- [18:03]
A
Yeah. And I mean, what you were proposing was truly a change in paradigm because everyone had been, to use a term that's now in vogue, glucocentric in their approach to medicine approval. And we've now moved so far away from a glucocentric approach, if we look at the current ADA standards of care, that term seems a little bit odd and old, but that's where we were at at the time. Now, I imagine this would have raised some consternation among our industry colleagues because this is creating a high bar for drug approval. But it seemed like you were able to figure out a way to allow medicines to move forward at a reasonable rate and still accomplish what what was proposed.
- [18:56]
B
And that was exactly what I was hoping for. That was my desire. Now, I will tell you that there were some very prominent endocrinologists that also opined during this meeting and during this period of time. And at least one of them, I won't mention them because they are a prominent person in the ADA ABCs, said if you do what Dr. Nissen is proposing, there will never be another diabetes drug. And I pushed back and I said no. And I will tell you, although initially industry wasn't, didn't think this was a terribly good idea, eventually what they learned is that it benefited industry more than anybody else because today with these newer drugs, if we had not shown that, for example, that SVLT2 inhibitors can protect the kidney, can protect against heart failure, that GLP1 agonists can reduce morbidity mortality, these drugs would not be successful in the marketplace. Who's going to spend all that money on a new drug if it's not any better than what we already have? So they actually ended up being beneficiaries and they now will now say that when I talk.
- [20:15]
A
Oh, I have no doubt. You know, it's interesting here and in life in general, adversity often leads to better results than if that adversity had never been encountered. And were this issue never raised, had you never put this forward, we wouldn't have had all of these positive cvots that you just listed.
- [20:39]
B
Yes, absolutely. And you know, so I do think that I could not see the future, but I will tell you that my slides did pressing that the, that there would likely be drugs coming that would be beneficial if we could just stay the course and test them and see what they actually are doing for what we really care about, which is the morbidity and mortality associated with diabetes and of course the kidney failure and all the other things that diabetes.
- [21:14]
A
So as a result of what you just went over the FDA in 2008, can you just encapsulate what their ruling was, what was required of trials then moving forward?
- [21:26]
B
They fundamentally accepted the proposal pretty much exactly as I had made it at the panel. And they said that you could get on the market if you could rule out in a non inferiority trial, as you point out, an upper confidence interval of 1.8. That would take about something like 80 to 100 events to do that. That's a trial of maybe a couple thousand people for a couple of years. It's not trivial, but it's very doable and it's not going to cost a huge amount of money. And they then said you could get to market if everything looked good in that initial trial, but that you then had to do a trial to rule out upper confidence interval of 1.3. And that was precisely what I proposed and I was obviously very pleased and very supportive of that happening.
- [22:31]
A
Yeah, and it really is an amazing story and thanks so much for going over the detail that you have. And that led to really the first large positive CVOT was I guess in September of 2015, right, the EMPA reg trial.
- [22:49]
B
Yes. And there were several early trials that showed this EMPA reg was certainly an interesting one. Now you know what's interesting is, is that the EMPA reg trial was. The initial trial was designed to rule out 1.8. So it wasn't the bigger trial, but it actually showed benefit on mortality. Fascinating. And that was obviously a very important finding. And it held up because again, dapagaflozin, all the other, you know, SGLT2 inhibitors confirmed that that wasn't a one off. That in fact was like class related benefit. But what's interesting about these trials is that although they're not the primary endpoint, the companies were, and the academic leadership of the trial did have the good sense to study kidney outcomes and heart failure outcomes. And the studies were not powered for those outcomes and they were not ones that would result in a regulatory action because they were secondary endpoints. But they showed really strong signals. And so what happens when you do that is you say, well, wait a minute, the kidney function is looking, you know, looking a lot better in these patients. Let's do a dedicated renal trial and gee, we're seeing less heart failure. Let's do a dedicated heart failure trial. And so what happens when you do a comprehensive outcome trial is you can see a broad range of both benefits and risks. And it enables you to then pursue other endpoints that might be of clinical significance and that might be beneficial to patients. And that's how things evolved. Same thing happened with the GLP1 agonists. The first trial leader I was asked to be on the executive committee for that showed a modest benefit. You know, it wasn't the most powerful of the GLP1s, but it was a good drug and it showed a benefit on major adverse cardiovascular events. But then subsequent studies showed with more powerful, longer duration drugs, this incredible range of benefits, including, by the way, the kidney benefit that's been seen and the morbidity mortality benefit, which is really quite profound for the GLP1 agonist. And now, you know, we've now seen a very good result with the GLP GIP1 drug, tirzepatide. And there's much more coming because now we have a triple agonist in development. We have oral GLP1 drugs in development. There is also another fascinating development, and that is FDA took away the requirement for cardiovascular outcome trials. I'm trying to remember what year it was. You probably note this. It was several years ago.
- [25:52]
A
Yeah, 2020, I believe.
- [25:54]
B
And I was okay with that. And I'll tell you why. Because you can't market a new diabetes drug now if you can't show that it has good outcomes on things that are really important to patients, you know, death, stroke, MI kidney function, heart failure, all these other things had to be studied for these drugs to actually be successful in the marketplace. And so the marketplace has now been kind of self correcting and it's now made it impossible to get a drug approved without that. You can get a drug approved without it but you're not going to get payers to pay for it. So that's the.
- [26:35]
A
Yeah, it is fascinating. And yeah, in the standards of care it says to use a. And then it says the class name SGLT2GLP1 with proven cardiovascular benefit or with proven effect on heart failure or renal endpoints depending on the category. But that is now included, that verbiage is included in the standards of care. And I think you're right, correctly demanded by us as, as clinicians. This is Steve. It must be an amazing feeling to have been integral to, to making this all happen.
- [27:17]
B
Well look, I am, I'm proud of what I did. I can't tell you that it was easy because I really did take a lot of criticism and but I felt it was the right thing to do and I was willing to stick by it. You know it's interesting, a lot of the manuscripts I now publish have young people as co authors. So my young faculty, there were only two authors of that New England Journal paper on Rosiglitazone, myself and my statistician because I did not want a young person early in their career to take what I knew was going to likely be a pretty big hit over kind of blowing things up. And so it was just myself and Kathy Wolski, my lead statistician, just the two of us were the only two authors of that paper. You see a lot of New England Journal papers now have 30 authors and we didn't do that in those days. So now we have a new issue. The issue now is with certain classes of drugs where we have such established benefits you can't do placebo controlled trials anymore. And so the latest trial recently presented at the EASD was the surpass CBOT which compared the dual agonist tirzepatide against the GLP1 agonist dulaglutide. And it was designed as a non inferiority trial but also tested for superiority. And so what's going on now is that we're now entering a new era when we will have to study new drugs in comparison to drugs with established cardiovascular, renal and other benefits. And it just raises the bar even further. So now if your new drug is going to be successful, you've got to do at least as well as the first generation of drugs. And so I like the idea that we're now going to keep raising the bar and making it, you know, making these new drugs have to show even further benefits.
- [29:29]
A
That is fantastic. I think this is truly one of the most interesting stories in modern medicine where. Steve, we're about out of time. Any final thoughts you want to share with our listeners?
- [29:42]
B
Yep. I want to tell everyone, and there may be some young people watching this, have the courage of your convictions. You know, stand by your findings, be willing to be a little bit disruptive and be willing to get into good trouble because sometimes good trouble leads to good outcomes and better care for patients. And I hope some other people in the next generation of, of physician scientists will take on these challenging issues.
- [30:15]
A
That's wise advice. That's how we move forward as a medical community, as a society. Dr. Stevenessen, thank you so much for joining us.
- [30:23]
B
Okay.
- [30:24]
A
And most of all, to our listeners, thank you for joining us on this first of a two part series on CVOTs. This special series of Diabetes Core Update is sponsored by by Lilly. We want to thank you for listening and for the American diabetes association, I'm Dr. Neal Skalnick. Till next time, stay safe and keep learning.