Colchicine or MRA Post-MI: No CLEAR Benefit

This transcript has been edited for clarity. 
Michelle L. O’Donoghue, MD, MPH: Hi. This is Dr Michelle O’Donoghue, reporting for Medscape. We’re here in Chicago at the American Heart Association Scientific Sessions, and joining me today is Dr Sanjit Jolly. He’s an interventional cardiologist at McMaster in Hamilton in Canada, and a clinical trialist with Population Health Research Institute. Thanks for joining me. 
You have presented over the past couple of months some very interesting findings that potentially carry important clinical implications as to how we treat our patients. Specifically, this was the CLEAR SYNERGY (OASIS 9) trial, and it was a 2 x 2 factorial design, testing two different hypotheses. One was a question of colchicine, and one was a question of spironolactone in patients who are hospitalized with an acute myocardial infarction (MI). 
Let’s start off with the spironolactone findings. Following the publication of both RALES and EPHESUS, we’ve been using spironolactone to treat our patients post-MI if they have evidence of heart failure and perhaps low ejection fraction in and of itself. Those are the patients that we’ve been typically giving spironolactone. But part of what you were testing was expanding beyond those patients, right? 
Sanjit S. Jolly, MD, MSc: Yes. We know that EPHESUS was primarily a population that had heart failure during their hospitalization. You could get into that trial with low ejection fraction only if you had diabetes. So the vast majority were heart failure patients. What we wanted to see, similar to ACE inhibitors, is whether routine use is beneficial, and can it improve outcomes? That’s what we sought to determine. 
O’Donoghue: And did you have any prior evidence to suggest that spironolactone may be beneficial for those even without any signs or symptoms of heart failure? 
Jolly: There was the ALBATROSS trial, led by Gilles Montalescot, where in STEMI patients in particular, there was a reduction in mortality in those without heart failure with routine administration, but it was a subgroup; it was in need of validation. 
O’Donoghue: There’s been a lot of interest, in general, in the class of mineralocorticoid receptor antagonists (MRAs). And now there’s finerenone. The keen interest in the MRA class is in particular for the heart failure population with preserved ejection fraction, where it now appears that there is some efficacy; but MRAs had not been specifically studied post-MI, of course, right? 
Jolly: No, so we randomized 7062 patients across 14 countries. And this trial took 4 years to complete. As you know, these are challenging trials to do, especially investigator-initiated trials that are supported by grants. 
We actually had to increase our sample size during the trial because event rates were lower than planned. We amended our primary outcomes and made co-primary endpoints. We did a total recurrent event analysis for CV death and new or worsening heart failure, and made that a second co-primary. Bertram Pitt is on our steering committee and is the father, or grandfather, of MRAs, and he had this belief that perhaps MRAs may prevent ischemic events, not just heart failure. So we included CV death, MI, stroke, and heart failure as a second co-primary. 
O’Donoghue: And what did the results show? 
Jolly: We had no significant benefit for either co-primary endpoint. For CV death or new or worsening heart failure, the hazard ratio was 0.89 and crossed unity. It was similar — ie, no effect on CV death, MI, stroke, or heart failure. We did see a reduction in one of the components, which was new or worsening heart failure. 
We did a prespecified on-treatment analysis where we saw about a 20% reduction for both CV death and new or worsening heart failure, or the broader composite. But, of course, this is exploratory, because our primary analysis was intention-to-treat.
O’Donoghue: Do you think any of this is a function of the patient population enrolled? Taking a step back, what proportion were anterior STEMI patients? Did you enroll any patients with signs or symptoms of heart failure at the time of presentation?
Jolly: I’ll answer the second question first. Fewer than 1% of patients had heart failure during their presentation. So 99% didn’t have heart failure. This is primarily a study of patients without heart failure. Anterior MI made up about 39% of the overall study population, so, a large subgroup of the patients treated and randomized.
O’Donoghue: What about the safety of spironolactone? 
Jolly: We did see an increase in hyperkalemia, but hyperkalemia was relatively uncommon: 1% of the spironolactone group got hypokalemia vs 0.5% of the control group. We saw more gynecomastia. Somewhere between 2% and 3% had gynecomastia with spironolactone. It’s an old drug, it does have some side effects. That’s why the new agents were developed. 
O’Donoghue: You had a fairly high rate of drug discontinuation. Do you think that was related to the adverse events, or do you think it was more a factor of trial fatigue from the patients? 
Jolly: It’s probably multifactorial. Running a trial during the pandemic is a challenge, and we had to switch to a lot of virtual follow-up as a result of the pandemic. We also had to ship drugs via FedEx to patients instead of them coming in. The second thing is, when you add a factorial design, you’re asking patients to take not only one pill, but two pills, for years at a time, and that certainly adds a pill burden. And then the third issue is that colchicine caused diarrhea, and we generally noticed that if a patient got bad diarrhea, they stopped both study drugs, and it took a lot of convincing for them to even restart the spironolactone because of that bad experience. 
O’Donoghue: That is one of the challenges of a 2 x 2 factorial design. Before we wrap up the spironolactone component, would you say that this changes anything in terms of who spironolactone is currently indicated for post-MI? Right now, the guidelines would support the use of spironolactone in patients with signs or symptoms of heart failure, with reduced ejection fraction or with diabetes. Does this inform upon that population?
Jolly: One of the most important things is that we’re not doing a very good job of using MRAs. There’s data from the NCDR [National Cardiovascular Data Registry] that show that the vast majority of patients who have heart failure post-MI don’t get an MRA. We’re good at statins and maybe ACE inhibitors, but we’re not very good at prescribing MRAs. So that’s the first message: We need to translate the prior evidence into practice. 
The second issue is, should we be prescribing more MRAs? I’m not sure that this trial suggests that we need to, and certainly, routine use is not beneficial. We’re going to dig into the data further, but Marc Bonaca put a slide up of EPHESUS and our trial; the point estimates are identical on all the outcomes, but the baseline risk is just so much lower. 
If you have a young patient with an inferior STEMI who is undergoing primary PCI, probably there’s very little value, because their risk for heart failure is so low. I think you need to go back to the baseline risk. In different populations, say 75- or 80-year-olds with late-presentation STEMI, they may have a left ventricular diastolic pressure of 30 but not have evidence of heart failure. That’s probably a patient that you’re going to want to treat.
O’Donoghue: As you said, you didn’t really have patients with signs or symptoms of heart failure at presentation. So for those higher-risk patients where you saw at least a directional signal to suggest perhaps a reduction in heart failure events, there could be, in theory, greater benefit 
Jolly: Absolutely. 
O’Donoghue: Now let’s pivot to part two, thinking about colchicine. There is a lot of interest in colchicine, albeit, in most parts of the world, it hasn’t actually been embraced in terms of routine use. That being said, we had a couple of trials to suggest that there was clinical benefit for colchicine in patients with coronary artery disease (CAD) we’ve had data in the post-MI population and in stable CAD, and enough so that the European guidelines now have a Class IIa recommendation for use of low-dose colchicine in patients with stable coronary disease. Whether to use it post-MI has been a little bit more controversial, and that’s what you tested here. Do you want to talk a little bit more about that?
Jolly: Sure. We started this trial before the LoDoCo [low-dose colchicine] trials came out. This trial was substantially larger than those prior trials and had substantially longer follow-up and more events. And so even though this was a post-MI trial, patients were randomized to colchicine vs placebo within 72 hours of their MI. The median treatment duration was 3 years. The maximum was 5 years. So this was actually both an acute- and a long-term–treatment trial. 
We think of acute trials as 30 days’ follow-up. This is much longer. And one of the important things is that we saw a reduction in hs-CRP with colchicine. We saw a significant increase in diarrhea. So we knew that the drug was working. And what we were really surprised to see is that the curves were exactly superimposable for CV death, MI, stroke, or ischemia-driven revascularization. There was no difference at all, and each of the individual outcomes was superimposable. 
We did an on-treatment analysis because we had significant drug discontinuation, and while it was different for the spironolactone arm, for colchicine it was identical. The hazard ratio is nearly 1. I’ve stopped prescribing colchicine. There was a subset of patients I was using it in, but I no longer use it because, obviously, our data with more than 600 outcome events were clearly neutral.
O’Donoghue: And that includes the restenosis cohort. You had an outcome of ischemia-driven revascularization and you didn’t see anything for that, because there has been a lot of interest in whether colchicine may actually modify restenosis of an existing stent.
Jolly: We did not see any effect on further revascularization events.
O’Donoghue: One question that you didn’t specifically look at is whether patients with higher degrees of baseline inflammation may actually derive some treatment benefit. You mentioned the reduction in hs-CRP, but you didn’t enroll patients specifically on the basis of having a high hs-CRP. Could those folks perhaps derive some benefit from a drug like colchicine?
Jolly: It’s interesting that in LoDoCo2, the hs-CRP levels were actually very low. So it is still a conundrum. It makes biologic sense that perhaps that’s the future population we should test. 
O’Donoghue: There had been questions about whether colchicine may increase the risk for non-CV death, which was a surprising signal seen in some prior studies. You didn’t see that at all; if anything, it looked favorable.
Jolly: The truth is probably that there’s no effect on non-CV death, and that was just a spurious finding in the other trials.
O’Donoghue: What about other trials in the anti-inflammatory space? We previously had canakinumab. That did show clinical benefit in CANTOS. This is an IL-1beta drug. And then we also have ziltivekimab which is now in phase 3 testing, targeting IL-6. Do you think that these results should in any way be extrapolated to those different targets?
Jolly: I think the story on inflammation is that inflammation is not this class effect where you’re just going to target inflammation. Certainly with methotrexate, we saw that there was no effect. The canakinumab data is a very modest effect, a 15% reduction in events, but an increase in fatal infections. So, on balance, people are not using canakinumab. 
Now, in terms of IL-6 inhibitors, I think we really need to target different parts of the pathway and do the trials, and I’m very interested to see what those results are. I think we do need to answer this question. People ask me, is the inflammatory hypothesis dead? And I say, absolutely not. We know that inflammation is a marker of outcomes. We don’t know that modifying inflammation can change those outcomes, other than likely the canakinumab data, but it was a modest treatment effect. So I’m really interested to see what the IL-6 story tells us.
O’Donoghue: Ultimately, I think it comes down to what the target is. That remains the big unknown: What should we be going after when it comes to inflammation? 
Thank you again for joining me today to discuss a very interesting topic. We’ll have to see how it shapes future guidelines as well.
And thanks for joining me at an especially difficult time for you, having recently lost your father. I’m sorry. 
Jolly: I think this is special for him, because he was a cardiologist working in Winnipeg for 25 years. Seeing his son presenting at the AHA would make him very happy.
O’Donoghue: Thank you. Signing off for Medscape, this is Dr Michelle O’Donoghue. 
Michelle O’Donoghue is a cardiologist at Brigham and Women’s Hospital and senior investigator with the TIMI Study Group. A strong believer in evidence-based medicine, she relishes discussions about the published literature. A native Canadian, Michelle loves spending time outdoors with her family but admits with shame that she’s never strapped on hockey skates.