Cholesterol and The Killer Plaque

LDL, inherited Lp(a) risk and PCSK9 medicines in one prevention story.

John Griffin, MD | The Arising | November 2026

A cholesterol panel, an inherited risk result and a discussion about an injectable medicine can sound like three separate conversations. They belong to the same practical question: what can reduce the chance of a cardiovascular event for this person? The answer requires understanding the particles, the person’s overall risk and the evidence for the exact treatment. A better laboratory number matters most when we understand what it means for life outside the laboratory.

Start with a decision, not a grade

A cholesterol report can look like a school report card: numbers, ranges, and a few results marked high. It is tempting to ask only whether you passed.

A more useful question is what the results change. Do they help explain a family history of early heart disease? Do they support a different eating pattern, a medication discussion, or closer attention to another risk factor? Has someone already had a heart attack, or are we trying to prevent a first one?

Those situations are different. The same laboratory value does not automatically lead to the same decision for every person.

Cholesterol is necessary for normal biology. The concern is how cholesterol-carrying particles contribute to disease in artery walls over time, and how much of that risk can be reduced. The purpose of care is fewer heart attacks, strokes and other important problems, not merely a more attractive laboratory report. [1]

How plaque can cause harm

Atherosclerosis develops when some cholesterol-carrying particles become retained in artery walls and promote inflammation and plaque growth. Some plaques can rupture or erode, prompting a blood clot that can cause a heart attack or an ischemic stroke. Not every plaque ruptures, and not every stroke is caused by this process. [15, 16]

Several measurements, several questions

LDL cholesterol, often written LDL-C, describes the amount of cholesterol carried in LDL particles. It does not directly count those particles. LDL is often called “bad cholesterol,” but that shorthand can hide what the test actually measures.

Non-HDL cholesterol is total cholesterol minus HDL cholesterol. It includes cholesterol carried in several kinds of particles that can contribute to atherosclerosis, the buildup of disease in artery walls. It can add useful information beyond LDL-C alone. [1]

Apolipoprotein B, or apoB, is a protein on LDL and other particles that can contribute to artery plaque. Its concentration helps estimate how many of those particles are circulating. Sometimes the amount of cholesterol and the number of particles give different impressions of risk. In selected circumstances, apoB can help clarify that mismatch. It is not a test that every person needs simply because it is available. [1]

HDL cholesterol is another measurement on the usual panel. A high HDL result does not cancel out the risk associated with high LDL or other risk factors. The report is not a balance sheet in which one favorable number erases every unfavorable one.

Lipoprotein(a), or Lp(a), is different again: a largely inherited LDL-like particle with an additional protein attached. It is not separately measured by the ordinary LDL-C number. Current guidance recommends at least one measurement in adulthood. Its units and interpretation deserve their own discussion below. [1]

These measurements overlap in what they tell us, but they are not interchangeable. More tests are useful when they answer a question that changes care.

The number sits inside a life

A person who has already had a cardiovascular event is considering prevention of another event. This is often called secondary prevention. A person without known cardiovascular disease is considering primary prevention: reducing the chance of a first event.

Age, blood pressure, smoking, diabetes, family history and other circumstances affect the overall picture. Someone's current risk and the consequences of continued exposure over time both matter. A lower short-term estimate does not mean that a markedly elevated LDL level can always be ignored.

The 2026 US dyslipidemia guideline uses the PREVENT risk framework to help guide primary-prevention discussions in adults aged 30–79. A risk calculator is a decision aid, not a prophecy. It cannot know everything about a person, and the result still needs to be interpreted with relevant clinical information. [1]

This is why one universal “perfect cholesterol number” is not a complete plan. The intensity of treatment and the LDL and non-HDL goals depend on the situation. Your clinician should be able to explain what the proposed target is intended to accomplish for you.

What makes this particle different?

Cholesterol travels through blood in particles called lipoproteins. Lp(a) is related to an LDL particle, with an additional protein called apolipoprotein(a) attached. That added protein helps distinguish it from the LDL cholesterol discussed in most routine visits. [5]

Your Lp(a) level is largely inherited. This is why a high result can appear in someone whose daily habits are sensible. It is not evidence that the person has failed to eat properly or exercise enough.

Evidence from genetics and population studies supports Lp(a) as a contributor to atherosclerotic cardiovascular disease, the artery disease behind many heart attacks and strokes. Elevated Lp(a) is also linked causally to calcific aortic valve stenosis, in which the valve that lets blood leave the heart becomes narrowed and stiff. [5]

Those are population-level findings about risk. A blood result does not diagnose blocked arteries or a narrowed valve. Nor does a lower result erase other risks, such as smoking, high blood pressure or an elevated LDL cholesterol level.

A useful way to understand the test is as an added layer of information, not a replacement for the rest of the picture.

Read the number with its units

Lp(a) reports may use milligrams per deciliter, written mg/dL, or nanomoles per liter, written nmol/L. The first describes mass; the second describes particle concentration. Because the size of the attached protein varies, there is no single conversion factor that works reliably for everyone. [5]

Keep the result in the units used by the laboratory. A number copied into a message without its units can create confusion that did not exist in the original report.

Common professional risk thresholds include 50 mg/dL or 125 nmol/L. These are separate thresholds used in guidance, not a promise that multiplying one by a fixed number produces the other. Risk also increases across a range; it does not suddenly appear at one dividing line. [5, 1]

Imagine two people whose results both exceed a guideline threshold. One has already had a heart attack. The other is younger, has no known cardiovascular disease and has several relatives with early heart disease. Both deserve a thoughtful discussion, but their immediate decisions need not be identical. These are illustrative situations, not individual treatment recommendations.

Age, existing disease, LDL cholesterol, blood pressure, diabetes, smoking and family history all affect what an additional risk factor means. The same laboratory number can sit within quite different overall risks.

Why one measurement can be useful

Because Lp(a) is largely genetically determined, repeated frequent testing is often less useful than it is for a measure being actively changed by treatment. “At least once” does not mean “never again”: certain secondary conditions and clinical circumstances can justify reassessment. A clinician can decide whether repeating the measurement would answer a real question. [5]

An elevated result can also open a family conversation. Close relatives may share the inherited tendency, so asking whether they should be tested can be worthwhile. Family testing is an invitation to understand risk, not a way to assign blame or assume that everyone has the same result.

If you have a copy of an old laboratory report, bring it. The exact value, units and date are more useful than remembering that something was “high.”

Lowering LDL has outcome evidence

An important distinction separates cholesterol care from simply chasing a blood marker: randomized trials have shown that lowering LDL with established treatments can reduce important cardiovascular events.

A large Cholesterol Treatment Trialists' analysis combined 27 randomized trials involving 174,149 participants. They compared statins with control treatment, or more-intensive with less-intensive statin treatment. For each reduction of about 1 mmol/L in LDL cholesterol, it found roughly a 21% relative reduction in major vascular events. The comparison included major coronary events, strokes and procedures to restore blood flow in coronary arteries. [2]

“Relative” is the word to notice. It does not mean that 21 out of every 100 people avoided an event. The absolute benefit depends on how likely an event was to occur in the first place.

Among the analysis's lower-risk participants, with an estimated five-year major vascular-event risk below 10%, lowering LDL by 1 mmol/L was associated with about 11 fewer major vascular events per 1,000 people over five years. That is about 1.1 percentage points, not 11 percentage points. It is a finding for those study populations, not a personal forecast for everyone reading this issue. [2]

Someone at higher risk can have a larger absolute benefit from a similar proportional reduction. The treatment burden, potential adverse effects and the person's preferences still belong in the decision.

Why evidence belongs to a treatment

The benefit is not confined to one way of lowering LDL. But evidence from one medicine cannot simply be assigned to every other medicine that changes the same laboratory result.

IMPROVE-IT studied 18,144 people after an acute coronary syndrome, such as a heart attack or unstable symptoms caused by reduced blood supply to the heart. Researchers compared adding ezetimibe to simvastatin with simvastatin alone. Its combined outcome included cardiovascular death, a nonfatal heart attack, unstable angina requiring rehospitalization, coronary revascularization at least 30 days after randomization, or a nonfatal stroke. Revascularization means a procedure to restore blood flow.

The seven-year rates were Kaplan-Meier estimates, which account for differing follow-up rather than simply dividing the number of events by all enrolled participants. Those estimates were 32.7% versus 34.7%, an absolute difference of 2 percentage points. Median follow-up was six years. [3]

That illustrates a clinically meaningful question: does adding a particular treatment improve outcomes beyond the treatment already being used? It does not mean that every healthy person can expect the same benefit, or that every LDL-lowering product has completed the same test.

PCSK9-directed therapies require the same care in interpretation. Antibodies that block the PCSK9 protein and inclisiran, which reduces its production through an RNA-based mechanism, are not identical products with interchangeable evidence. A discussion should name the medicine, the population studied, the outcome measured and the practical burden.

The Lp(a) trial discussed below is a useful reminder. A treatment can markedly change a blood marker without demonstrating the expected reduction in cardiovascular events. That does not invalidate established LDL trial evidence; it shows why the result for each treatment matters.

Statins can cause muscle symptoms and modestly increase diabetes risk. Rare serious muscle injury and interactions with other medicines also matter. Symptoms deserve assessment rather than dismissal; the expected benefit and possible harms should be considered together. [2]

One target, more than one way to reach it

The PCSK9 story makes those distinctions concrete: medicines can share a biological target while differing in how they work, how long their effects last and the outcomes they have demonstrated.

PCSK9 is a protein involved in regulating LDL receptors. These receptors help the liver remove LDL particles from blood. Interfering with PCSK9 can allow more receptors to remain available and lower circulating LDL cholesterol.

That shared target connects a group of medicines. It does not make their evidence interchangeable.

For a person considering treatment, the useful question is not just “Does it lower cholesterol?” It is “Which medicine, for which reason, with what evidence and what practical burden?”

Block the protein or reduce its production

PCSK9 antibodies, such as evolocumab, act on the protein. Inclisiran takes another route: it is a small interfering RNA designed to reduce PCSK9 production in the liver.

RNA interference uses cellular machinery to reduce a particular protein-making message. It should not be confused with an mRNA vaccine, and it is not permanent editing of the person's DNA.

Gene-editing approaches are another category altogether. Sharing a target does not mean that an antibody's safety record or an RNA medicine's cholesterol result validates a gene-editing strategy. Exposure, persistence, reversibility and evidence have to be considered separately.

Inclisiran's longer-lasting action creates the possibility of less frequent administration. Duration can be convenient, but it is not itself proof that a treatment is the best clinical choice.

What the original inclisiran trials measured

ORION-10 and ORION-11 were randomized phase 3 trials in people with elevated LDL cholesterol despite maximally tolerated lipid treatment. ORION-10 included 1,561 participants with established atherosclerotic cardiovascular disease. ORION-11 included 1,617 with that disease or high-risk equivalents. “Maximally tolerated” did not mean that every participant was taking a statin: documented intolerance could mean no statin was being used. [10]

The main endpoints concerned LDL cholesterol. At day 510, placebo-corrected LDL reductions were 52.3% in ORION-10 and 49.9% in ORION-11. In everyday terms, inclisiran lowered LDL by about half relative to the comparison group in these studies. [10]

That is a substantial lipid effect. It is not a finding that heart attacks were cut in half, deaths were cut in half, or everyone's personal risk fell by the same amount.

A percentage belongs to the outcome that was measured. Moving it from a cholesterol result to a clinical event would change the claim into something the trial did not establish.

The selected populations also matter. These were clinical trials in people who qualified because of their disease or risk and lipid levels, not a test of whether everyone with a mildly elevated result should receive the medicine.

A different PCSK9 medicine has cardiovascular outcome evidence

FOURIER tested evolocumab, a PCSK9 antibody, in 27,564 people with established atherosclerotic cardiovascular disease who were receiving statin therapy. Its main combined outcome included cardiovascular death, heart attack, stroke, hospitalization for unstable angina, or coronary revascularization, a procedure to restore blood flow. Over a median follow-up of 2.2 years, its main combined cardiovascular outcome occurred in 9.8% of the evolocumab group and 11.3% of the placebo group. [11]

That is an absolute difference of 1.5 percentage points during the trial's follow-up. The estimated hazard ratio was 0.85, a relative comparison of event rates over time, with a 95% confidence interval of 0.79–0.92. The relative estimate and the absolute difference answer related but different questions.

The finding supports an outcome benefit for this medicine in this studied population. It is not a blanket mortality claim, a promise of the same absolute benefit for a low-risk person, or a completed outcome trial for inclisiran.

Injection-site reactions occurred in 2.1% with evolocumab and 1.6% with placebo. Benefits and harms both belong in the discussion. [11]

This is why naming the medicine is important. A family of treatments can share a mechanism without sharing every piece of evidence.

What has and has not been learned about inclisiran outcomes

A prespecified exploratory pooled analysis of ORION-9, ORION-10 and ORION-11 favored inclisiran on combined cardiovascular events. These observations are worth knowing about, but the analyses were not designed or powered to provide a definitive cardiovascular outcome test, and the events were not formally adjudicated in the way a dedicated outcomes trial would assess them. [12]

“Not adjudicated” means that the event reports did not go through the same standardized independent confirmation process expected for a dedicated clinical endpoint study. It does not mean the observations should be ignored. It means the confidence placed in them should match how they were collected.

ORION-4 is a separate, large cardiovascular outcomes trial. The retrieved official record lists it as active but no longer recruiting, with no results posted. Its design paper anticipated outcome reporting in early 2027; that was a forecast, not a publication of the outcome itself. [13]

The careful position is therefore neither “there are no event observations” nor “the definitive benefit has already been established.” There are exploratory findings, and the dedicated question remains separate.

Harms and burdens belong beside benefits

Injection-site reactions were more frequent with inclisiran in the original trials. The reported rates were 2.6% versus 0.9% in ORION-10 and 4.7% versus 0.5% in ORION-11; none were classified as severe or persistent in that report. [10]

Those trial findings do not establish that every possible reaction is mild. Current prescribing information includes serious hypersensitivity reactions, including anaphylaxis and angioedema. Angioedema is swelling that can involve the face, lips or airway. A previous serious reaction to inclisiran or its ingredients is a contraindication. Severe allergic symptoms require urgent medical care. [14]

Medication selection also requires a discussion of pregnancy and other relevant medical circumstances. A general article cannot determine eligibility or replace the current prescribing information for the exact product.

It is useful to ask about what was common in trials, what serious risks are recognized now, what the monitoring involves and what to do if an unexpected problem appears. A lower LDL number does not remove those questions.

The new result that changes the treatment story

A major question has been whether a drug designed to lower Lp(a) will also prevent cardiovascular events. The HORIZON trial tested pelacarsen in people with established cardiovascular disease and Lp(a) of at least 70 mg/dL. Its results were published in November 2026. [6]

Its main outcome was the first occurrence of cardiovascular death, a nonfatal heart attack, a nonfatal ischemic stroke, or urgent coronary revascularization requiring hospitalization: an urgent procedure to restore blood flow in the heart’s arteries.

The report described 8,350 people randomized to pelacarsen or placebo, with 8,322 included in the efficacy analysis. Over a median follow-up of 4.7 years, the main cardiovascular outcome occurred in 511 of 4,157 participants receiving pelacarsen, or 12.3%, and 503 of 4,165 receiving placebo, or 12.1%. The trial did not demonstrate a cardiovascular benefit despite markedly lowering Lp(a). [6]

The estimated hazard ratio was 1.02. In everyday language, the observed rates over time were very similar. The 95% confidence interval, an uncertainty range from 0.90 to 1.15, allowed for some benefit or some harm; it did not establish an improvement. This was not a result showing that a dramatic laboratory change reliably delivered better health outcomes.

The trial also involved treatment burdens. An investigator-institution account reported more flu-like symptoms and injection-site reactions, with treatment discontinuations due to adverse events in 20.6% versus 7.5%. A drug's ability to lower a number is only one part of deciding whether its use helps people. [7]

This result matters, but its meaning should not be stretched. It does not show that Lp(a) is irrelevant, erase the genetic evidence, or prove that every other approach must fail. It tells us that this drug, in this studied population and follow-up, did not demonstrate the hoped-for cardiovascular benefit.

It is a reason to be precise, not a reason to abandon either curiosity or prevention.

Other approaches still have to answer the outcome question

Other investigational medicines aim to reduce production of Lp(a), including olpasiran and lepodisiran. Their large outcome trials address the question that a striking drop in a biomarker cannot settle on its own: do fewer people experience important cardiovascular events? [8, 9]

The retrieved trial records list both studies as active but not recruiting, with no results posted. Their estimated primary completion dates are in 2028 and 2029, respectively. Those dates are plans, not guarantees, and a registry entry is not a result. [8, 9]

Different medicines may have different effects, harms and practical burdens. An encouraging early change in Lp(a) should therefore be described as a laboratory finding until clinical outcomes support a stronger claim. An investigational treatment is not a reason to postpone appropriate care available now.

What can change now, even if Lp(a) does not?

A common frustration is that healthy habits may have little effect on Lp(a) itself. That does not make those habits pointless. They can improve other parts of cardiovascular risk and daily health while the inherited number remains similar. [5]

Clinicians can act on LDL cholesterol, blood pressure, smoking and diabetes, and discuss physical activity and a workable eating pattern. Someone with established cardiovascular disease may need a different intensity of prevention from someone without it. A high Lp(a) result can strengthen the reason to take the whole picture seriously.

This also explains a seemingly confusing point about statins. They may modestly raise Lp(a) in some people, yet that is not a reason to stop a treatment prescribed to reduce cardiovascular risk. A change in one laboratory measure must be interpreted alongside the treatment's demonstrated benefits and the individual's situation. Medication changes belong in a clinical discussion. [5]

PCSK9-directed medicines are another part of lipid care, but they should not all be treated as interchangeable. The evidence for a particular medicine, its intended use and its effects on LDL and Lp(a) need to be kept separate. A lower Lp(a) measurement during treatment does not by itself prove that this change caused a reduction in heart attacks.

In selected patients, specialists may consider lipoprotein apheresis, a procedure that removes certain lipoproteins from blood. It is not a routine first step for everyone with an elevated test. Eligibility and access depend on the clinical situation and local rules. [5]

There is useful work to do now without promising that every part of inherited risk can be removed.

Food is a pattern, not a single rescue ingredient

An eating pattern can support cardiovascular health without turning every meal into a calculation. The 2026 American Heart Association guidance emphasizes vegetables and fruit, whole grains, healthy protein sources and replacing saturated fats with unsaturated fats. It also supports attention to fiber, added sugars and sodium. [4]

Replacement is important. “Eat less fat” does not say what you will eat instead. Replacing some butter with an unsaturated plant oil is a different change from replacing a meal with refined starch and sugary snacks. The whole pattern is more informative than a single nutrient slogan.

Practical changes might include adding beans to a familiar dish, choosing a whole-grain option you enjoy, or making nuts, vegetables and fruit easier to reach. These are examples of applying a dietary pattern, not promises of a specific LDL reduction for every person.

A useful plan also fits budget, culture, appetite and the ability to shop and cook. A theoretically ideal menu that cannot be sustained may help less than a few workable changes repeated often.

Alcohol is not something to start for heart protection. And food changes need not become a contest between lifestyle and medicine. For some people, both are useful; for others, the next step is a conversation about what is realistic and sufficient. [4]

A lower glucose curve is not the same as lower cholesterol risk

A meal's effect on blood glucose answers a different question from its effect on LDL cholesterol, nutrition or long-term cardiovascular risk. Adding a great deal of saturated fat may change a glucose response without making the overall pattern more heart-healthy.

This matters as continuous glucose monitors become more visible outside diabetes care. A smooth line is not a complete score for a food or a person's health. Glycemic index, glucose response, LDL cholesterol and Lp(a) should help answer their own questions rather than compete to become the only number that matters.

The useful connection is a broader one: a sustainable pattern of food, movement and appropriate clinical care can address several risks together, while each test retains its limits.

Convenience is part of care, not the disappearance of care

Less frequent administration can be attractive to someone who struggles with a daily regimen. But a missed appointment is a different adherence problem from a missed tablet, not the end of adherence as a concern.

Access, cost, scheduling, monitoring and continuity of care all affect whether a treatment can work in everyday life. A plan that looks convenient on paper may be difficult if it requires travel or appointments that are hard to maintain.

Long duration also makes it sensible to ask how long an effect persists and what can be done if a problem occurs. A clinician should be able to explain the proposed schedule, the follow-up plan and the alternatives without treating convenience as the only deciding factor.

Imagine two people with similar LDL results. One has already had a cardiovascular event and has difficulty tolerating several medicines. Another has no known cardiovascular disease and wants fewer daily pills. These are illustrative situations, not treatment recommendations. Their laboratory similarity does not erase the differences in expected benefit, options and personal priorities.

Bring the three conversations together

The useful plan begins with the actual report and the person who brought it. LDL-C, non-HDL-C, apoB and Lp(a) are not interchangeable measurements. A largely inherited result does not erase what can be changed, and a new technology does not inherit the outcome evidence of every other medicine aimed at the same target.

Bring your laboratory report, current medicines and family history. Ask:

• Are we preventing a first cardiovascular event or another event, and how does that change the plan?
• What do LDL-C and non-HDL-C tell us? Would apoB answer a remaining question?
• Have I had Lp(a) measured, in which units, and should close relatives discuss testing?
• Which medicine are we discussing, what outcome has it demonstrated, and in what kind of patient?
• What other treatments have been tried or remain reasonable options?
• Would specialist advice answer a specific question?
• What absolute benefit is realistic over the relevant period, and what harms or interactions matter?
• Which food and activity changes are useful and sustainable for me?
• Can I manage the cost, appointments and monitoring, and when will we reassess?

Do not stop or alter a prescribed medicine because of a general article. A treatment that is difficult to tolerate or maintain deserves a conversation about alternatives and support. The goal is neither a perfect-looking report nor the newest technology. It is a prevention plan whose evidence, practical demands and expected benefit make sense for the life you want to keep living.

Sources and further reading

[1] Blumenthal RS, Morris PB, Gaudino M, et al. 2026 ACC/AHA multisociety guideline on the management of dyslipidemia. Circulation. 2026;153:e1154–e1276. Publication. Official summary.

[2] Cholesterol Treatment Trialists’ Collaborators, Mihaylova B, Emberson J, et al. Effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials. Lancet. 2012;380:581–590. Publication. Full text.

[3] Cannon CP, Blazing MA, Giugliano RP, et al. Ezetimibe added to statin therapy after acute coronary syndromes. New England Journal of Medicine. 2015. Publication.

[4] Lichtenstein AH, Khera A, Anderson CAM, et al. 2026 Dietary Guidance to Improve Cardiovascular Health. Publication. Official summary.

[5] Kronenberg F, Mora S, Stroes ESG, et al. Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement. European Heart Journal. 2022. Full text.

[6] Cho L, Nicholls SJ, Frndak S, et al. Pelacarsen and Cardiovascular Outcomes in Patients with High Lipoprotein(a). New England Journal of Medicine. Published November 7, 2026. Primary report.

[7] Cleveland Clinic. Lipoprotein(a) reduction with pelacarsen has no impact on cardiac events.

[8] OCEAN(a)-Outcomes. Trial record NCT05581303.

[9] ACCLAIM-Lp(a). Trial record NCT06292013.

[10] Ray KK, Wright RS, Kallend D, et al. Two Phase 3 Trials of Inclisiran in Patients with Elevated LDL Cholesterol. New England Journal of Medicine. 2020;382:1507–1519. Publication.

[11] Sabatine MS, Giugliano RP, Keech AC, et al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease. New England Journal of Medicine. 2017;376:1713–1722. Publication.

[12] Inclisiran pooled patient-level cardiovascular event analysis. Publication.

[13] ORION-4. Trial record. Design paper.

[14] Current official inclisiran prescribing information.

[15] Robinson JG, Williams KJ, Gidding S, et al. Eradicating the Burden of Atherosclerotic Cardiovascular Disease by Lowering Apolipoprotein B Lipoproteins Earlier in Life. Journal of the American Heart Association. 2018. Full text.

[16] National Heart, Lung, and Blood Institute. Atherosclerosis.

Funding and evidence limits

The CTT synthesis received public and charitable funding, including support from the British Heart Foundation and Medical Research Council; contributing trials may have received industry funding. IMPROVE-IT was funded by Merck. HORIZON was funded by Novartis. The Cleveland Clinic account comes from the investigators’ institution and is not an independent trial. The original inclisiran trials were funded and designed with the Medicines Company; the pooled analysis was funded by Novartis. FOURIER was funded by Amgen, and ORION-4 is co-sponsored by Oxford and Novartis. Authors disclosed industry relationships.

The HORIZON discussion is based on the publisher-indexed abstract, corroborating investigator-institution report and sponsor announcement; the full manuscript and supplements were not available in the source review. FOURIER and IMPROVE-IT were assessed from their primary abstracts, and the full current guideline and dietary statement were not reviewed. This is a focused explanation, not an exhaustive review of every lipid treatment, safety signal or regulatory status. Trial-status statements reflect records checked November 9, 2026; planned dates do not establish that results have been reported.