May 29 2026 This Week in Cardiology

I would like to first thank the Danish Cardiac Society for inviting me here to speak in Neubourg, which I mispronounced last week as Nyborg. I know that I say it often, but I love Denmark. As I record here in my hotel room overlooking the sea, I had to actually close the window because of the sea noises in the background.

It's no wonder that Danes are rated as one of the happiest people. This country is beautiful. Now it was imposing to speak to more than three hundred Danish cardiologists because I know that they have such a strong foundation in scientific appraisal of evidence. Yet to my surprise there are still a handful of left atrial appendage closure enthusiasts here. Of course, I showed them my appendage closure slides and tried to persuade them.

I also personally thank Dr. Jens Brock Johansson for taking me on an epic 109-kilometer journey to the northern tip of Denmark. And today I ride with Soren Nielsen and I thank you all for the great adventures on the bike.

The first topic today is a potential permanent, permanent lipid lowering therapy. New England Journal of Medicine published on Monday an amazing paper with thirty five patients who had familial hypercholesterolemia.

who underwent a base editing therapy designed to durably inactivate PCS K nine protein in the liver. The basic science here is beautiful. It's long been known that loss of function variants in the gene ProProtein Converte subtilsin Kexin type nine PCSK9 confer lifelong low levels of blood PCSK9 protein and low density lipoprotein L cholesterol without apparent adverse effects. People with this genetic signature also have markedly lower rates of atherosclerotic coronary disease.

Indeed, PCSK9 is now a well-established way to reduce major adverse cardiac events. In addition, these drugs appear to be safe from off-target effects, at least in shorter-term trials of two to five years. Another observation from genetic studies is that the cumulative exposure to low LDL cholesterol also seems important. Yet the logistical problem for prevention

of heart disease is that many patients do not take lipid lowering therapy, or when they do, they do not adhere to it. Observational studies have reported discontinuation rates of thirty to fifty percent. So doctor Sec Cetherason and his team have been working on a base editing infusion that mimics favorable PCSK nine gene variants. The way it works.

VERV 102 consists of two drug substances, a messenger RNA encoding for an adenine-based editor protein, and a guide RNA targeting PCSK9. These two drugs substances are encapsulated in a lipid nanoparticle delivery system that incorporates an N acetogalactosamine targeting ligand to enhance delivery to hepatocells. The guide RNA targets a splice site at the five prime end of intron one of PCSK nine.

it altering the splice site with a single DNA base pair substitution then enables a read-through of a stop codon that prevents translation and expression of PCSK9 protein in the liver. There is a really nice figure in a New England Journal of Medicine paper. Now included patients had to have the diagnosis of heterozygous familial hypercholesterolemia or premature coronary artery disease, which was defined as disease before age fifty five in men or before age sixty five in women.

Participants also had to have a fasting LDL cholesterol level of at least 70 milligrams per deciliter while receiving the maximum tolerated dose of oral lipid lowering therapy, a statin usually with or without azitamite. Infusions of varying doses took four hours, and they used pretreatment steroids and antihistamines. Patients underwent inpatient observation for at least two days and had daily clinical laboratory assessments through day four.

Patients were on average fifty two years old, twenty nine of the thirty five had FH, mean LDL cholesterol was one hundred twenty nine milligrams per deciliter, and most of these patients were on high potency statins. So as far as safety goes, one in five patients had minor infusion events. These were just a few patients with minor liver enzyme elevations, and these were mostly transient.

This is presumably permanent treatment of hypercholesterolemia. Now whether these patients have to remain on statins as well has is remains to be seen. In the 1 mg per kilogram dose of this infusion, the mean reduction was 78 milligrams per deciliter. This would be estimated to reduce atherosclerotic CV death or CV events by 50% for patients with F-HA.

I mean this represents an iteration on a previous attempt from this company. The previous attempt was associated with serious liver inflammation and low platelets. The team made changes in the lipid nanoparticle, and this attempt looked to be safer. Second, the degree of PCSK9 inhibition seems similar to that seen with the sub Q injections and consistent with near complete suppression of at least hepatically produced PCSK9 proteins.

Now the authors caution that hepatocytes turn over every two hundred to three hundred days, so there needs to be longer term follow up to assess durability of this effect. And of course there are plenty of limitations. This is a phase one study. In other words, it's mile two of a marathon. Efficacy and safety are major worries, right? For me, the risk of off-targeting gene editing is very concerning. What effects will there be in 10, 20 years?

Recall that the alternative is a fully reversible therapy with either a sub Q injection or soon a tablet. But I say congratulations to Sek Katharason and his team on this potential, potential breakthrough. Much remains to be known safety wise, but this is a remarkable achievement in biology.

Okay, next topic is a Vesalius substudy on PCSK9 use in patients with previous PCI. Circulation has published a subgroup analysis from the main Vacellius trial of Evi Lochumab versus placebo in 12,000 patients with ASCDD or diabetes, but without an MI previously. New England Journal of Medicine published the main study, the main Vacellius study in twenty twenty five, and this was notable because the two previous PCS K9 trials had been done in patients who had an ACS event.

Odyssey outcomes was alorochumab in patients with ACS and Fourier was emulacumab in patients with a previous MI stroke or PAD. Vesalias included lower risk patients, these without MI, but with CAD or PAD or cardiovascular disease of some sort. Patients in Vesalias also had to have an LDL cholesterol greater than ninety. The main results of the main Vacelius trial was a statistically significant 25% reduction in a three-point mace endpoint of coronary death, MI, or stroke.

The absolute event rates were 6.2% versus 8.0%. That's a 1.8% absolute risk reduction, an NNT of 55%. Now, as it was in all of the PCSK9 trials, the reduction was in nonfatal events. In Vasilius, there was no no reduction in coronary death or all-cause death. So the trial was positive, but the risk reduction is modest. And that is not surprising, because if baseline risk was lower, you'd expect less absolute risk reduction.

The circulation reanalysis focused on patients in the main trial who had had PCI. Importantly, the main trial enrolled 12,000 patients. That's the power that it took to sort out event rates. You need a lot of patients when drugs have small effect size. Now the subgroup of patients who had had PCI is even smaller. It's about 3,600 patients, or 30% of the main trial had undergone PCI.

Now the question is effect size in these patients versus the seventy percent who had not had PCI. Uh in both the placebo and the evolutionab arm, patients who had had prior PCI had higher rates of both primary endpoints to three and four point mace. I'll focus on the more robust three point mace, which is coronary death MI and stroke. Now you'd expect these patients to have higher absolute rates because if you have a PCI, that's gotta be a marker for more significant disease.

In patients with prior PCI, EVlocumab reduced the risk of three-point mace by 30%. Seven point zero versus nine point five percent. That's a hazard ratio of zero point seven zero. Conference intervals were significant, zero point five six to zero point eight nine, p value zero point zero zero four. And that's compared to placebo. Components of the endpoint were mostly driven by MI, but coronary death was directionally consistent.

Here were the numbers. For coronary death, the hazard ratio was 0.69. Conference intervals were 0.43 to 1.09. For MI, the hazard ratio is 0.5. The conference interval 0.36 to 0.70. And for stroke, it was 1.06, but there were very few strokes in wide conference intervals. Now rates of CV death and all cause death were also lower, but these are reported with nominal p values due to multiple testing. So my comments. Now much is being made about this subgroup analysis. I've seen it on

uh social media and there's just a lot of press about it. But I don't find it that shocking at all, right? The main reduction of the three point mace in the main trial was twenty-five percent. In this higher risk group of patients who had had PTI, it was thirty percent. You'd expect that, especially since patients in Vacellias had to have an LDL cholesterol greater than ninety, and the actual LDL cholesterol was as high as 116 milligrams per deciliter.

So no surprises. In patients who have established disease, especially previous PCI, we would all agree that you'd want an LDL cholesterol lower than 116. PCSK9 drugs achieve that. But so does higher potency statins, or so would statins plus azidomide probably also achieve that.

Now, despite C V death and all cause death reductions in this subgroup analysis meeting significant levels, that is, that when you look at the upper bound, it's less than one point zero I would not say that this drug reduces death in this subgroup, and that's because there are simply too many comparisons and the subgroup is too small. Recall that trials are powered for the entire group. When you slice up groups of patients within a main trial,

you lose power and the risk of false positive findings simply by chance is greater. For instance, like the astrologic sign analysis in ISIS two. That said, I would propose this study simply shows that patients with previous PCI benefit to a similar degree than those with CAD who have not had PCI, like the patients that were enrolled in Fasalia.

All right, next topic is terzepatide looking to be the best GLP1 agonist again. The Annals of Internal Medicine has published a study called Surpass Early. This is terzepatide versus intensified conventional care in patients who have obesity and type 2 diabetes. The Lily sponsored trial was carried out in seventy-eight sites in ten countries, and the report this week were the two year results of a planned four year trial.

This was nearly 800 patients were randomized to s to zepatide with dose escalation getting to the goal of fifteen milligrams, or intensified conventional care with a goal of normal glycemia. Now the primary endpoint was a change in hemoglobin A one C from baseline to two years. Initial testing was with non inferiority. Secondary objectives included superiority tests. of the change in hemoglobin A one C as well as weight and waste circumference.

These patients were young, fifty-three years old, nearly half were female, and the BMI was thirty-five at baseline.

The starting trial hemoglobin A one C was elevated at seven point eight and most of these patients had had diabetes for about two point five years. Now before I tell you the results, notable was that at two years, eighty five percent of patients in the intensified conventional care arm were on GLP one agonist with seventy four percent on semiglutide and twelve percent on dualoglutide.

SGOT two inhibitors were used in seventeen percent and in much smaller numbers on sulfono ureas, insulin, and DPP four drugs. Most patients, 88% were on just one glucose lowering agent alongside metformin with only 10% on two and only one percent on three. So in essence, really, and I didn't realize this when I first looked at the trial, this was predominantly a semiglutide versus trzepatide comparison.

So first glycemic control, the hemoglobin A1C. Terceptide reduced the hemoglobin A1C by 1.99 percentage points versus 1.32 with intensified conventional care. Mean hemoglobin A1C landed at five point five percent with Zepatide versus six point three five with uh conventional care. 60% of trzepatide patients reach normal glycemia, defined as a hemoglobin A1C less than 5.7, versus only 24% with conventional care. That's 60% versus 24%. It's a substantial difference.

Weight loss, as you'd expect, trzepatide decreased the weight by thirteen point eight kilograms versus five point nine kilograms in the conventional care. Forty-four percent of trzepatite patients lost more than 15% of their body weight versus just 12% with conventional care. Waist circumference, very same numbers, 11.6% centimeters reduction versus 5.4%.

Other metabolic benefits and I'll comment on this later, there were greater improvements in triglycerides, HDL, blood pressure, beta cell function, and insulin sensitivity with trisepatite. versus the conventional care, which again was mostly semi glutide. Side effects were mostly GI as you'd expect. No severe hypoglycemia was noted in either group. There was one confirmed case of pancreatitis in the trisepatide group.

And treatment continuation due to side effects was low, but higher, which was epitide, four point five percent versus zero point three percent.

Uh one patient stands out, for instance. This is a sixty-five year old man. He's not that heavy at twenty eight BMI. He took GLP ones. and dropped his abnormal triglycerides and LDL cholesterol such that our both are now in the normal range, without a a lipid lowering therapy. He had to stop his valsartan due to low blood pressure with his eight kilogram weight loss. So then just playing around with the PCE pull cohort equation and the ASCVD risk.

His risk went from 10% to 6.6%. So he went from statin eligible or statin indicated to below that 7.5% threshold. AFib episodes, in case you're wondering, decreased. Now, if you're going to use GLP1s, and I think it is time for cardiologists to step up and start using these drugs. we should probably use trisepatite due to its superiority over semaglutide for weight loss, and in this study glycemic parameters. And this isn't the only study showing trzepatite as more potent.

The surpass early study reconfirms the Surmount five trial published in the England Journal of Medicine in twenty twenty five, and I covered that then. Sermon five found that trusepatide was better than semaglutide for weight loss. The w weight loss t was six percent greater with trusepatite. Now one question that I hear is what is causing

these improvements. Whether their improved metabolic factors occur because of weight loss or some other weight loss independent mechanism does not matter that much to me. That patients can lower their ASCVD risk this much makes me wonder whether GLP1s will become a more important drug class than, for instance, statins.

I mean it's a long way to prove that because statin drugs have more than a hundred and seventy five thousand patients randomized in trials. But the changes w I'm seeing with these drugs are amazing. They're stunning. Now, I'm not saying we should be treating young people with simple, uncomplicated overweight or obesity, but abul adults like these patients enrolled in the trials, these patients are heading for trouble soon. I consider a patient in his fifties or eighteen.

with type two diabetes, hypertension, sleep apnea, obesity, this person has a serious illness and they're going to get in trouble. And these drugs seem to treat the serious ill illness and I think it's a good thing.

All right, next topic arrhythmias in cardiac amylidosis. Like you, I see more patients with cardiac amyloidosis. Whether this is due to increased surveillance or increased disease incidence, I don't think is well sorted out, but it sort of doesn't matter. My overall impression is that these patients require lots of effort in surveillance and treatment.

Well Jack has published a small but very nice paper from the UK National Amylo Center and they implanted loop recorders ILRs in one hundred and ten treatment naive patients. with newly diagnosed cardiac amyloidosis. They followed these patients for thirty six months.

Now, while loop recorders admittedly have very marginal value in real world clinical practice, in fact I learned this week that that in the cost-constrained Danish healthcare system, ILRs are rarely used, but in the profit-driven US system, ILRs are like cash machines. Now people ask I do use ILRs uh sparingly, mostly in patients with infrequent but severe syncope, and in the rare patient with AF who has postconversion pauses whom

We ablate and then we want to prove that the AFib is gone, because if the AFib is gone, we know that the pauses will go away as well. Now ILR use, in my opinion, in post-stroke patients is wildly overused. In the U.S., it's especially bad because not only is it lucrative to implant these, we get reimbursed quite well, but patients get a recurring down low bill every 30 days. Uh that said, the ILR can have great use in research.

The many studies coming from the loop investigators, for instance, have shed great light on the natural history of short duration atrial fibrillation. This study, which the authors named Excalibur from the UK group, simply put in ILRs in patients with cardiac amyloidosis to see what happened. In sum, my friends, a lot happened. As you'd expect, with this awful infiltrative disease.

Here's the breakdown of the study as follows. Of the 110 patients, they had 43 patients with wild type ATTR, they had 20 variant ATTRs, and then they had 47 light chain AL cardiac amyloidosis patients. And they found three main arrhythmia types. Brady arrhythmia is first. 17% of the patients developed a class one pacemaker indication, roughly twice the rate reported in prior retrospective studies.

This included complete AV block and 42%, Mobitz 2 and 26%, and Symptomatic Six Cent of Syndrome and 21%. These were significantly more common in ATTR cardiac amyloid, twenty-four percent versus nine percent in AL cardiac amyloid.

And they were associated with higher myocardial amyloid burden, and that was measured by MRI. As far as atrial fibrillation goes, Nuance at atrial fibrillation occurred in twenty eight percent of patients who had without prior AF, again, more frequent in ATTR cardiomyopathy fifty percent versus AL cardiac amyloid twelve percent.

There were strong correlations with the degree of amyloid infiltration. Also interesting was that all but two patients were asymptomatic. All patients were started on anticoagulation and no strokes occurred during the time of this study. Ventricular arrhythmia? Sustained ventricular arrhythmy has occurred exclusively in AL cardiac amyloid, six point four percent, and at an at a stage of moderate rather than advanced amyloid burden. Non sustained VT was common.

It occurred in twelve patients total, that's eleven percent of the overall population. The NSVT incidence was nearly the same in the three subtypes of cardiac gamyloid. None of the patients were symptomatic during these events, and importantly, none of the NSVT patients progressed to develop sustained VT. What about terminal rhythms? All ten ATTR cardiomyopathy deaths uh showed pulseless electrical activity.

Pulses electric electroactivity was also the most common terminal rhythm in AL cardiac amyloidosis. However, two patients did have VTVF as a terminal rhythm in that type of cardiac amyloid. So my comments. Now studies like this don't make the New England Journal of Medicine. They don't win you Nobel Prizes, but I find this extremely useful natural history information. It's a good study and it's important and it's uh informs us.

There is a lot of heart block in these patients. Now we can warn patients to pay attention. This group put in devices for asymptomatic patients. But and we don't usually put in pacemakers for asymptomatic uh brachocardia, but I doubt a patient with heart block with amyloid would stay asymptomatic for very long.

Now as for the AF detection, we know that screening in a general older population, like those in the loop trial, did not end up reducing stroke, but I agree with the use of anticoagulation in these much higher risk patients. Now whether we should have a screening study, perhaps with smart watches or smartphone ECGs in this population of patients with cardiac amyloid is something to consider because the AF uh rate was so high. And of course the stroke risk would be high in these patients.

I also find it fascinating that none of the deaths in the ATTR group were due to VT, and none of the SVT progressed to malignant VT. Now I don't want to make too l strong of conclusions from this 110 patient study, but it's interesting, isn't it? And it it indeed is one strong data point. Note that ILRs are very useful as research tools. They are always on and they're very good at arrhythmia detection.

Another clinical implication from the lack of VTVFS terminal events is the resistance to put in ICDs in these patients. Cardiac amyloid, I think, represents a classic scenario where progression of disease is likely the cause of death and not a stochastic type cardiac arrest. So please, my friends, watch closely for hard block, have a low threshold for anticoagulation, but resist the urge to place ICDs in these patients, especially if you're using non-sustained VT as a marker.

All right, the last topic today is about lipid guidelines. As you all know, I wrote a critique of the new ACC and AHA guidelines. My main points were that the guidelines were too complex. and they were too devoid of RCT level data and that rendered many of the class one indications. Should you and and class one indications are things that you should do or if you don't do, you're in trouble. These were poorly supported with evidence.

I also struggled with the declaration that the prevent score is better than the simple PCE, and I felt that experts with intellectual and financial dua dualities of interest should not be writing guidelines. The disclosure part is not enough in my opinion. This document had to have huge lists of conflict of interest, not only of the authors actually, but also of the peer reviewers. I just ask you, for instance, if a person's entire research area is an imaging,

and he or she has a financial stake in an imaging company, what do you think this person is gonna say about imaging in the guideline? There there are they gonna say that there's no strong data and we should wait to use imaging? It's not likely. Anyways, lead author Roger Blumenthal and some of his co-authors have written a lengthy rebuttal, and Medscape has published it. And as it is my rule, I do not comment on their rebuttal. You can read their rebuttal and decide for yourself.

I will say that they did not change my mind, but you all decide by reading the article and see what you think. So until next week, this is John Mandrola from the Heart.org Medscape Cardiology. And remember, if you like this podcast, please take the time. Write us a review. If you have a question about something I said or you disagree, send me a note and we can do listener feedback.