Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast, download the Medscape app or subscribe on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only.
In This Week’s Podcast
For the week ending December 01, 2023, John Mandrola, MD comments on the following news and features stories.
DAPA-MI
The SGLT2 inhibitor class of drugs have as positives both tons of empirical data and ease of use. Chronic kidney disease (CKD), diabetes, heart failure (HF). Clear benefits. We can debate cost efficacy, but not benefits.
What about post-myocardial infarction (MI)?
Swedish investigators, first author, Stefan James, tested the idea of adding dapagliflozin to patients’ regimens after an MI that either left the patient with Q-waves or mild left ventricular (LV) dysfunction.
Why not? they thought. Post-MI care has plateaued in the post-reperfusion era, which makes perfect sense, because stopping an MI before it causes much damage is hard to improve on.
But SGLT2 inhibitors are so loved; they should be tried. These investigators presented their findings at the 2023 American Heart Association (AHA) meeting.
Before I say anything more, the trialists have done a nice job. Yet, DAPA-MI will require its consumers — you and me — to wear our thinking hats.
The original trial plan, though, had to be changed. Originally, they planned to look at hard clinical outcomes like cardiovascular (CV) death and hospitalizations for HF (HHF).
Then they discovered a problem. Post MI patients (even with mild LV dysfunction or Q waves) don’t have many bad events. This left the trialists with a few choices — forget the trial, enroll tons more patients (costly), or change the primary endpoint.
So, they changed the endpoints, without looking at the data. They now decided to look at a bunch of cardiometabolic data in addition to the hard endpoints.
Endpoints now included a hierarchical composite of death, HHF, MI, atrial fibrillation (AF), type 2 diabetes, New York Heart Association function class at the last visit, and body weight loss of 5% or greater at the last visit, using the win ratio analysis method.
The key secondary outcome was the same hierarchical composite excluding the body weight component.
For the win ratio analysis, matched pairs are created between two arms, then each strategy is assigned a win or a loss per pair, starting with the worst endpoint, death, then moving down the list sequentially if the previous event did not occur.
Win ratio is a way to use all the data, but its downside is that it’s hard to apply clinically.
The trial enrolled about 2000 patients per arm very soon after MI. These were 63-year-olds. Only 20% female, from the United Kingdom and Sweden; 20% had normal ejection fractions (EF). Most had ST elevation MIs (STEMIs), but non-STEMIs, were allowed; Kidney function was normal.
And, these patients were already on really good therapies, with high percentages of ACE/ARB, beta blockers, and of course antiplatelet drugs.
The bottom line is that DAPA-MI is technically positive:
The analysis of the primary hierarchical composite outcome resulted in significantly more wins for dapagliflozin than for placebo (win ratio, 1.34; 95% confidence interval [CI], 1.20 to 1.50; P < 0.001)
The problem, and reason for the addition of the modifier technically for positive was that hard clinical endpoints such as death, CV death, HHF, MI, stroke, and even all-cause hospitalizations were not different in the two groups.
The primary endpoint was driven by softer endpoints like new diagnoses of diabetes and weight loss, both of which are way down the hierarchical scale of that primary endpoint.
Adverse events were similar. I specifically looked at ketoacidosis, because we have seen a bunch of these cases in our quality review committee. We see them because the non-hyperglycemic ketoacidosis on SGLT2 inhibitors tricks doctors.
My Comments. I like this trial. The post-MI space is too tough to make much of a difference. Revascularization dwarfs everything else.
SGLT2 inhibitors have a role in treating diabetes, preventing progression of kidney disease, and in HF with reduced EF. But the interpretation of DAPA-MI is clear.
There’s no compelling reason to use the drugs for a purely post-MI reason. Yes, if there is diabetes, or CKD, or lowish EF, but not in this population, as a pure post MI drug or guideline directed medical therapy.
We wouldn’t know that if not for these researchers; so I am glad they did the trial.
I also like this trial because it forces us, the consumers of medical evidence to stop and think about using results.
DAPA-MI is emblematic of the main challenge for cardiology in the coming decades: That is, it’s going to be hard to improve on what we have now, at least, in a cost-effective manner.
Blood Transfusion for Anemia in Patients Post-MI: The MINT trial
Speaking of having to stop and think about applying trial results to patients, the MINT trial is a head scratcher. Teaser: The P-value is going to factor into the thinking.
This question comes up often: What to do with anemia in the post-MI setting? Restrict blood or be liberal about transfusing?
On the one hand, having enough red blood cells may help oxygen delivery and limit damage. But on the other hand, giving blood comes with the risk of volume overload, infection, thrombosis, and inflammation.
MINT didn’t get the big press because industry wasn’t involved but if you practice cardiology and like medical evidence, it’s the most interesting trial from AHA.
Previous trials in post-MI patients are interesting.
The REALITY trial, published in JAMA in 2021, found that the restrictive strategy was noninferior when it came to major adverse cardiac events (MACE), though the CI included what may be a clinically important harm. The relative risk of the primary outcome was 0.79 (1-sided 97.5% CI, 0.00-1.19), so that favors the restrictive strategy, but the upper bounds of that hazard ratio (HR) was 1.19, which allows for a 19% worse outcome.
But blood transfusion strategies — be they restrictive or liberal — have been done in many other clinical situations. A Cochrane review of trials of more than 21,000 patients showed a 50% decreased use of blood without differences in morbidity or mortality with the restrictive strategy.
This is the background of the MINT trial – restrictive vs liberal transfusion approaches in anemic patients who are immediately post-MI.
In MINT, the restrictive strategy reserved transfusion for a hemoglobin of 7 to 8 g/dL vs liberal, which allowed transfusion for a hemoglobin less than 10 g/dL.
The primary outcome is strong: MI or death at 30 days.
About 3500 patients were randomly assigned 1:1. Patients were almost half female, age 72 years, EF normal, and median hemoglobin of 8.6 g/dL.
The result: The mean number of red cell units that were transfused was 0.7 in the restrictive-strategy group and 2.5 in the liberal-strategy group.
This resulted in a hemoglobin level that was 1.3 to 1.6 g/dL lower in the restrictive-strategy group than in the liberal-strategy group on days 1 to 3 after randomization.
The main results were tricky, and fun to discuss. Here goes.
Death or MI, the primary outcome, occurred 16.9% of patients in the restrictive strategy group vs 14.5% of patients in the liberal strategy group. Liberal looked better; 2.4% in absolute risk reduction (ARR) with a number needed to treat of 42.
Was this significant? That is the hard part.
The crude risk ratio for restrictive vs liberal was 1.16 or 16% worse. The 95% CI ranged from 1.00 to 1.35. And we usually say if 1.00 or no difference was included it is not statistically significant. But it gets slightly more complicated.
There was incomplete follow-up on 20 patients in the restrictive vs 37 in liberal strategy.
With techniques called multiple imputations, the estimated adjusted risk ratio came out to 1.15 with CI to 0.99-1.34 and a P-value of 0.07.
Death was 19% higher in the restrictive arm but the CIs were wide and this did not reach significant.
MI was also 19% higher and with the same issue — wide CI and non-significance.
Death due to cardiac disease, which was not a primary outcome event, was 74% higher in the restrictive arm; here the CI went from 1.26 to 2.4. So, clearly worse.
Subgroup analysis revealed one possible heterogeneous treatment effect – that is, in type I MI the restrictive arm led to more primary outcome events (HR 1.32; 1.04-1.67).
The frequency of HF and other safety-outcome events was similar in the two transfusion groups.
The NEJM conclusion has a hedge. As they did in the PROTECTED TAVR trial, which had a non-significant primary endpoint, their conclusion included this phrase after the declaration of no difference, “…but on the basis of the 95% confidence interval around this outcome, the results may not rule out a benefit of….”
For the MINT trial, which had a P-value of 0.07, the conclusion read:
In patients with acute MI and anemia, a liberal transfusion strategy did not significantly reduce the risk of recurrent myocardial infarction or death at 30 days. However, potential harms of a restrictive transfusion strategy cannot be excluded.
My Comments. I like this conclusion. After reviewing this paper for a few days, I think MINT is my favorite trial from AHA. It makes you think.
The last phrase adds nuance to the interpretation. Here, I think common sense and strict statistical thresholds (P = 0.05) come into tension.
I am biased about blood. In the setting of an MI, when you want to limit myocardial injury, it makes sense to have enough red corpuscles delivering nutrients.
I know your rebuttal. You’d say, Mandrola, you are the evidence person. Oodles of studies have found that restrictive strategies of blood transfusion save blood products without affecting outcomes.
Well, I get that. But let’s look at this data without the stats.
Here you have more primary outcome events in the restrictive strategy. More MIs. More CV death. Patients with type I MIs especially did worse with restriction. These seem like clinically important differences.
And there were no differences in adverse outcomes, such as HF.
Yes, the lower bound of the CI allows for no difference but for all intents and purposes, nearly the entire span of the CI is worse with the restrictive strategy.
I would point also to other factors that strengthen my argument that this was a positive trial.
It was pragmatic. Unlike, say PARADIGM-HF, with its long run-in periods, longer in the sacubitril/valsartan arm, MINT enrolled a broad group of patients with both STEMI and non-STEMI as well as type I and type 2 MIs.
They enrolled patients with multi-morbidity.
The trial transfusion protocol led to large differences in the number of blood transfusions and there were clinically meaningful differences in hemoglobin levels between the two groups.
One important caveat — one that I learned from being on Twitter and following trial guru, Dr. Sanjay Kaul —if you take a Bayesian view of the MINT trial results and use the REALITY trial results as a prior (and recall that in REALITY, the restrictive arm did better) then MINT results get pulled towards the null.
I have to admit that I can’t see an obvious reason why REALITY and MINT went in opposite directions. REALITY was better for restrictive, MINT, worse with restrictive.
In sum, if I’ve got a post-MI patient with a hemoglobin of 8.5 g/dL, and I think blood would help, MINT supports a transfusion. It doesn’t compel us to transfuse. That’s like the perfect trial: it provides evidence but allows clinical judgement.To principal investigator Jeff Carson and MINT trialists, nice work.
Factor XI Inhibitors Take a Big Hit
There were two big stories on Factor XI inhibitors since the last podcast. First is the phase 2 AZALEA-TIMI-71 trial, which pitted the subcutaneously given Factor XI inhibitor abelacimab vs rivaroxaban in patients with AF and a primary endpoint of bleeding.
Abelacimab is a highly selective fully human monoclonal antibody. As you recall, Factor XI inhibitors are hoped to be the perfect anticoagulant (AC), because they have the potential to uncouple hemostasis from thrombosis; that is, block pathological thrombosis while preserving hemostasis.
Abelacimab actually has some clinical data in venous thromboembolism (VTE) prevention after total knee replacement (TKR). This was published in NEJM in 2021.
Three doses of the drug were used vs 40 mg of enoxaparin, after TKR.
The lower dose of abelacimab was noninferior to enoxaparin and the two higher doses were superior to enoxaparin for the prevention of VTE detected by venography.
AZALEA was a similar dose ranging phase 2 study but looked primarily at bleeding.
The trial was stopped in September because of the substantially greater than expected reduction of bleeding in the abelacimab arms compared with rivaroxaban.
These were 74-year-old patients with a median CHADSVASC score of 5. The HR for major bleeding was 0.23-0.33 depending on dose. Obviously, highly significant.
Here was the problem: The trial cannot assess for stroke. There were too few patients and too few events. But strokes were numerically higher in abelacimab group. The rate of ischemic stroke in the lower dose was 82% higher than rivaroxaban.
Again, you can’t say anything about that because there were too few events and CI are wide. I mention this potential signal because of the next story.
OCEANIC AF
In a stunning announcement on the weekend, we received notice from Bayer, the makers of the oral Factor XI inhibitor, asundexian, and sponsor of the phase-3 OCEANIC AF trial of asundexian vs apixaban in patients with AF, that the trial was being stopped just 11 months after starting enrollment.
The note said termination of the trial was for lack of efficacy. The trial had planned to enroll 18,000 patients. We were randomizing patients at our office, and it was pretty easy.
OCEANIC AF was like the direct oral AC vs warfarin trials. Stroke and systemic embolism were the primary endpoint.
Bayer said the decision was based on a recommendation from the data safety monitoring board (DSMB), which noted inferior efficacy.
That this happened so early into the trial is quite surprising; but in retrospect, some of the earlier trials of asundexian had similar signals. That is, there were impressive reductions of bleeding but numerically higher rates of thrombosis.
The thing about less bleeding is that an espresso, or gelato, or placebo, can lead to less bleeding. An AC has to prevent thrombosis.
If a new drug can’t do that, well, then it can’t win.
We don’t yet have the data. One question I have is that I don’t know whether the DSMB was totally independent. There’s no protocol paper yet.
One worry I have is whether the sponsor got to see the data and decided that it wasn’t worth the trouble to continue.
Nonetheless, the final and most important lesson is, again, how we know what we know in medicine. This is exactly why we should conduct proper large trials. Because, despite the theoretical advantages of Factor XI inhibitors and the decent preliminary data, you need to show benefit in properly powered trials looking at clinical outcomes.
Note that I used the plural for trial(s). We use DOACs because four trials showed similar net benefits. We use SGLT2 inhibitors because numerous trials have shown benefit. The failure of asundexian in OCEANIC AF is a huge teaching lesson.
FDA approves another renal denervation device
A few weeks ago, I reported on the US Food and Drug Administration (FDA) approval of a renal denervation (RDN) device using ultrasound. In my opinion, it was weak approval because the reductions of systolic blood pressure (SBP) were very modest and the placebo-resistant effect was confirmed for only months.
But at least, in this case, the FDA followed the recommendation of its advisory committee, which also recommended approval.
Well, last week, the FDA approved the main competitor to the ultrasound-based RDN system. This is a radiofrequency (RF) energy-based system called the SYMPLICTY Spyral device. And this is an even more dubious approval. First, the advisory committee voted against it on efficacy. That makes perfect sense because, of the two Symplicity Spyral sham controlled trials, only one was positive. The other was totally negative. Totally.
This RF system was tested in two big trials — one in which meds could be stopped, called, SPYRAL HTN OFF MED.
This was published in the Lancet in 2020. It found that the treatment difference between the two groups for 24-hour SBP was −3.9 mmHg. For the in-office SBP the difference was −6.5 mmHg. Both were statistically significant. These numbers were at 3 months.
The second regulatory trial of RDN with Spyral system was called SPYRAL-HTN-ON MED. This was super confusing. It took me about 3 days to sort out.
The first paper was the pilot study, published in the Lancet in 2018. This included 80 patients and found about a 7 mmHg reduction in SBP in 24-hour ambulatory measures.
The final paper of SPYRAL-HTN-ON MED (expansion pilot) was published in the Journal of the American College of Cardiology in 2023.
This was totally negative. Here they randomly assigned 337 patients (including the 80 in the pilot) with about 200 patients in the RDN arm and 131 in the sham arm.
I am copying and pasting the primary endpoint because it is a bit confusing: The primary efficacy endpoint was the baseline-adjusted change in mean 24-hour ambulatory systolic BP at 6 months between groups using a Bayesian trial design and analysis.
This was an important trial because, at least the initial indication for RDN would seemingly be as an adjunct to medicines, presumably for patients with difficult to control BP.
Let me tell you about this pivotal trial. They initially enrolled 1780 patients but ended up randomizing only 337 patients.
Patients were followed at 3 and 6 months. Mean age was young at 55 years; 80% male; baseline office SBP 163 mmHg.
In the primary efficacy analysis, no significant difference was observed across treatment groups for 24-hour ambulatory SBP at 6 months (Bayesian treatment difference: −0.03 mm Hg [95% Bayesian credible interval: −2.82 to 2.77 mmHg].
This is discouraging because remember that in the pilot, which included 80 patients, they had shown a -7 mmHg. Now, essentially nothing.
The curious thing is that a graph of this non-significant, nearly identical BP was published in the Supplemental Figure 2.
Figure 4 is also remarkable, with no difference or change in 24-hour ambulatory BP.
The authors write about changes in medicines. At baseline, there were no differences in numbers of medicines, but over the trial, and before primary endpoint measurement, patients in the RDN group were prescribed fewer BP meds at 6 months; 1.9 vs 2.1, P-value significant.
In summary, I know it’s confusing. The SPRYAL-HTN-OFF MED was positive, albeit with low SBP reductions.
The SPYRAL HTN ON MED was split into a pilot study of 80 patients, which also looked modestly positive, but when SPYRAL-HTN-ON MED was expanded to a larger 300+ recruitment, the reductions of SBP relative to sham disappeared.
This is why the FDA advisory recommended against approval. But the FDA approved it.
So now we’ll have two devices on the market that have extremely modest effects on BP, one, likely little to no effect with meds, and neither device has data beyond 6 months.
Nothing has happened yet; we wait for payers and national coverage decisions, but in my mind, RDN has the potential to be lowest of low value interventions in the history of cardiology.
It doesn’t have to be this way. I believe there is potential here but we need more data. These trials need to be extended — with blinding — to at least a year or possibly 2 years.
RDN could be studied as an adjunct to AF ablation. It could be studied in patients with high-adrenergic tone hypertension. But as it stands now, I am very worried about the low value nature of RDN.
© 2023 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Dec 01, 2023 This Week in Cardiology Podcast - Medscape - Dec 01, 2023.
Comments