Raising HDL With Drugs Failed Every Clinical Trial. Here Is What HDL Function Actually Tells You About Risk.

The Scene

The patient in this scene is a composite. Names, ages, and circumstances are changed to protect privacy.

Helen is 55 years old and she has just handed me her lipid panel. She points to the HDL value: 67 mg/dL. She is visibly pleased. “My doctor always said my good cholesterol was great. I figured that protected me.”

Her LDL is 148 mg/dL. Her ApoB is 118 mg/dL. Her CAC score, which she had done because her father had an MI at 62, came back at 165. She has smoked one half-pack per day for 20 years.

Helen has used a single raised HDL number as protection against all the other numbers that are telling a different story. She is not wrong that a high HDL is associated with lower cardiovascular risk in population data. She is wrong to use it as a reason not to take the other numbers seriously.

The HDL story is one of the most instructive lessons in the difference between an observational marker and a therapeutic target. Epidemiology told us for 40 years that high HDL was protective. Drug after drug designed to raise HDL failed to produce cardiovascular benefit. Mendelian randomization then told us why: HDL-C is not causally protective. It is a marker that travels with other factors that are protective. Raising HDL-C artificially does not produce the biology of naturally high HDL.

This is the article that explains the HDL paradox in plain language, so that Helen, and every patient who has ever been told “your good cholesterol is high,” understands what that means and what it does not mean.

What It Is

What HDL Is

High-density lipoprotein (HDL) is the smallest and densest of the major lipoproteins. Its core contains cholesterol esters and triglycerides; its surface contains phospholipids, free cholesterol, and multiple proteins including apolipoprotein A-I (ApoA-I) as the dominant structural protein.

HDL is produced both by the liver and intestine as nascent, lipid-poor particles (discoidal HDL). These particles accept cholesterol from peripheral tissues through the ABCA1 transporter on cells that have accumulated excess cholesterol (including foam cells in atherosclerotic plaques). The cholesterol is esterified by the enzyme LCAT and stored in the hydrophobic core, converting discoidal HDL to the spherical mature HDL particles measured in the routine lipid panel.

The HDL particles eventually deliver their cholesterol to the liver directly (by SR-B1 receptor) or indirectly (by transfer to VLDL and LDL through CETP, with subsequent clearance of those particles). This process is called reverse cholesterol transport (RCT): the net movement of cholesterol from peripheral tissues (including arterial plaque) back to the liver for excretion.

HDL cholesterol (HDL-C) is the standard clinical measure: the mass of cholesterol carried within HDL particles. Like LDL-C, it measures cargo weight, not particle number. The distinction between HDL-C concentration and HDL function is central to understanding the HDL paradox.

The “Good Cholesterol” Label

HDL earned the label “good cholesterol” from the inverse relationship observed in epidemiological studies: individuals with higher HDL-C had lower rates of coronary heart disease, after adjustment for LDL-C and other risk factors. The Framingham Heart Study, the Multiple Risk Factor Intervention Trial (MRFIT), and dozens of subsequent cohort studies consistently replicated this inverse association 5 / Solid .

The causal assumption followed: if high HDL is associated with less disease, then raising HDL should reduce disease. This assumption was wrong.

The Mechanism

What HDL Actually Does

The biological functions of HDL that are genuinely protective include:

Reverse cholesterol transport: HDL accepts cholesterol from macrophage foam cells in atherosclerotic plaques via ABCA1 and ABCG1 transporters. This is the best-established antiatherogenic mechanism. In mouse models, overexpression of ABCA1 or injection of ApoA-I Milano (a mutant ApoA-I with enhanced cholesterol efflux capacity) substantially reduces atherosclerosis 5 / Solid .

Anti-inflammatory effects: HDL carries proteins with anti-inflammatory properties including paraoxonase-1 (PON1), platelet-activating factor acetylhydrolase, and clusterin. These proteins reduce oxidative stress and inflammatory signaling in the arterial wall.

Anti-thrombotic effects: HDL inhibits platelet aggregation, promotes endothelial NO synthesis, and reduces tissue factor expression.

Endothelial repair: HDL promotes endothelial cell migration and proliferation, supporting vascular repair after injury.

Why Raising HDL-C Did Not Help: The Function-Concentration Distinction

The critical insight from the drug trial failures is that HDL-C concentration is not the same as HDL function. The drugs that raised HDL-C did so by mechanisms that did not replicate or augment the biological functions of naturally occurring HDL:

CETP inhibitors (torcetrapib, dalcetrapib, evacetrapib, anacetrapib) raise HDL-C by blocking CETP, the protein that transfers cholesterol from HDL to VLDL and LDL. In mice, high HDL-C from CETP inhibition is not associated with increased reverse cholesterol transport, because the mechanism that raises HDL-C (blocking outward cholesterol transfer) is different from the mechanism that would increase atherosclerosis protection (improving inward cholesterol acceptability from plaque macrophages).

Torcetrapib raised HDL-C by 72% but increased cardiovascular deaths, likely due to off-target aldosterone stimulation 5 / Solid . Dalcetrapib raised HDL-C by 30% with no cardiovascular benefit 5 / Solid . Evacetrapib raised HDL-C by 130% with no clinical benefit 5 / Solid . Anacetrapib raised HDL-C by 104% and showed a marginal 9% MACE reduction that appeared to be driven by its LDL-lowering effect rather than its HDL-raising effect 5 / Solid .

Niacin raises HDL-C by 20-30% and reduces triglycerides. In the AIM-HIGH trial, extended-release niacin added to statin therapy raised HDL-C substantially but did not reduce cardiovascular events 5 / Solid . The HPS2-THRIVE trial, the largest niacin trial ever conducted (25,673 patients), found that niacin added to statin therapy did not reduce cardiovascular events but did increase the risk of serious adverse effects including hepatic failure, myopathy, and new-onset diabetes 5 / Solid .

The pattern across all HDL-raising drugs is consistent: pharmacologically raising HDL-C does not translate to cardiovascular benefit.

The Mendelian Randomization Result

The definitive test of HDL causal protection is Mendelian randomization using genetic variants that naturally produce higher HDL-C. Voight BF et al. (2012) used 14 genetic variants that raise HDL-C without lowering LDL-C in Mendelian randomization analyses across 108,363 individuals 5 / Solid 60312-2). The genetic HDL-C-raising variants had no association with coronary heart disease risk. This is the key result: variants that produce higher HDL-C from birth, through mechanisms different from the drugs, also do not protect against coronary disease.

The conclusion: HDL-C is not causally protective in the same way that LDL-C is causally harmful. The inverse epidemiological association between HDL-C and cardiovascular risk reflects confounding: the conditions and metabolic states that produce high HDL-C naturally (physical fitness, healthy weight, female sex hormones, genetic background) are themselves cardioprotective, and high HDL-C is a marker of those states, not the cause of protection.

How We Diagnose It

What HDL-C Tells the Clinician Today

Given the above evidence, what is the current clinical role of HDL-C?

As a cardiovascular risk marker (not a target): Low HDL-C (below 40 mg/dL in men, below 50 mg/dL in women) remains a cardiovascular risk marker in clinical practice and is incorporated into risk calculators (Pooled Cohort Equations, SCORE2). Very low HDL-C (below 30 mg/dL) may be a marker of significant insulin resistance, metabolic syndrome, or secondary dyslipidemias 5 / Solid 61217-4).

Low HDL-C as a clue to an underlying condition: Very low HDL-C can indicate secondary causes: severe hypertriglyceridemia (CETP-mediated depletion of HDL cholesterol), obesity, type 2 diabetes, metabolic syndrome, hypothyroidism, autoimmune conditions (lupus, rheumatoid arthritis), cigarette smoking (which lowers HDL-C by approximately 5 mg/dL), or rare genetic conditions (Tangier disease, ApoA-I mutations, ABCA1 mutations).

Not as a treatment target: No current guideline recommends treating to an HDL-C goal. There is no proven pharmaceutical therapy for raising HDL-C that improves cardiovascular outcomes. HDL-C should be documented and interpreted in context but should not drive pharmacological treatment decisions.

The Evidence

The AIM-HIGH and HPS2-THRIVE Trials: Closing the Case on Niacin

AIM-HIGH randomized 3,414 patients with established cardiovascular disease, LDL below 80 mg/dL, and low HDL-C to extended-release niacin or placebo 5 / Solid . HDL-C rose significantly in the niacin group. The trial was stopped early for futility at 36 months: niacin provided no additional cardiovascular benefit beyond statin therapy. There was a non-significant trend toward more ischemic strokes in the niacin group.

HPS2-THRIVE enrolled 25,673 patients and added niacin/laropiprant (an antagonist of the niacin-induced flushing side effect) to statin therapy 5 / Solid . Primary cardiovascular outcome: no difference. Serious adverse events: significantly increased in the niacin group, including worsening glycemic control (new-onset diabetes: 3.7% vs 2.5%), hepatic disturbances, myopathy, and gastrointestinal events.

Together, these two trials definitively ended the clinical use of niacin for cardiovascular risk reduction in patients on statin therapy.

HDL Function vs HDL-C: The Research Frontier

An active research area explores whether HDL cholesterol efflux capacity (CEC), a measure of how well HDL particles accept cholesterol from macrophages, predicts cardiovascular risk better than HDL-C. Several studies suggest that CEC is inversely associated with cardiovascular events independent of HDL-C level 5 / Solid . CEC is not currently a clinical test; it is a research assay. But it illustrates the broader principle: HDL quality (function) matters more than HDL quantity (concentration).

The Patient Experience

The Patient With High HDL Who Asks “Am I Protected?”

Helen’s question has an honest answer: your high HDL-C is a favorable sign in the context of population statistics, but it does not neutralize your raised ApoB, your LDL of 148, your CAC of 165, or your 20 pack-years of smoking. High HDL-C in individuals with multiple other risk factors is not a shield. It is one data point in a risk profile that includes many data points.

The practical message: HDL-C is part of the lipid picture, but it is not a therapeutic target. Treating LDL-C, ApoB, and the smoking history will prevent far more cardiovascular events than any measure focused on maintaining or raising HDL-C.

The corollary for patients with low HDL-C: a low HDL-C (below 40 mg/dL in men or below 50 mg/dL in women) is a marker of cardiovascular risk, but no currently available medication should be chosen specifically to raise it. The treatment for low HDL-C is addressing its underlying drivers: weight loss, aerobic exercise, smoking cessation, and improved glycemic control all raise HDL-C modestly and reduce cardiovascular risk through their primary mechanisms 5 / Solid .

Decisions and Trade-Offs

Should Low HDL-C Change Treatment Strategy?

A patient with LDL-C of 95 mg/dL and HDL-C of 32 mg/dL has higher cardiovascular risk than a patient with the same LDL-C and HDL-C of 58 mg/dL. The low HDL is a marker of that higher risk. But the treatment response is to lower LDL-C more aggressively and address the metabolic drivers of low HDL-C, not to add a drug to raise HDL-C.

Current guidelines do not assign low HDL-C as an independent treatment target, but several do recognize it as a risk-enhancing factor that justifies more intensive LDL-C lowering 5 / Solid . This is a reasonable clinical approach: use low HDL-C as evidence of higher underlying risk, then treat that risk through LDL-C reduction.

The Non-HDL Cholesterol Metric

Non-HDL cholesterol (total cholesterol minus HDL-C) is a simple calculation that captures all ApoB-containing lipoproteins: LDL, IDL, VLDL, and Lp(a). It does not require fasting and is a better predictor of cardiovascular events than LDL-C alone in meta-analyses 5 / Solid . Non-HDL is particularly useful when triglycerides are raised (rendering the calculated LDL-C less reliable) and is recommended by the 2018 ACC/AHA guidelines as an alternative to LDL-C for risk assessment.

Non-HDL target: 30 mg/dL above the LDL-C target (i.e., below 100 mg/dL when LDL target is below 70 mg/dL; below 130 mg/dL when LDL target is below 100 mg/dL).

Clinical Synthesis

The HDL story matters to this program for a reason that goes beyond lipid biochemistry: it is a case study in how epidemiological associations can mislead clinical practice for decades when the causal architecture is misunderstood.

For 40 years, patients were told their high HDL was protecting them. For 40 years, physicians tracked HDL in lipid panels and used the number to tell patients their risk was lower. For 40 years, pharmaceutical companies pursued HDL-raising drugs as a prevention strategy. And the Mendelian randomization result, when it came in 2012, showed that none of this was causally grounded.

This clinical approach is to present this history honestly. HDL-C is a useful marker. It is not a target. Patients with naturally high HDL have other attributes that are protective, and those attributes should be identified and reinforced. Patients with low HDL have underlying conditions that raise risk, and those conditions should be addressed. Nobody should be taking niacin to raise their HDL in 2026.

For Helen: her HDL of 67 mg/dL is noted, and it is one favorable element in a risk profile that needs work on multiple other fronts. Her ApoB, her LDL, her CAC, and her smoking history are the variables that will determine whether she has a cardiovascular event in the next decade. The HDL will not rescue her from any of them.

Extended Evidence Review: The Complete Trial Record on HDL

CETP Inhibitor History: Four Drugs, One Lesson

Cholesteryl ester transfer protein (CETP) is the enzyme that shuttles cholesterol esters from HDL to VLDL and LDL in exchange for triglycerides. Inhibiting CETP raises HDL-C and lowers LDL-C, making it an appealing cardiovascular target. Four CETP inhibitors were developed and tested in large cardiovascular outcomes trials:

Torcetrapib: ILLUMINATE (2007): Torcetrapib was the first and most eagerly anticipated CETP inhibitor. It raised HDL-C by 72% and lowered LDL-C by 20% but increased cardiovascular events (HR 1.25 for cardiovascular death, MI, stroke, or hospitalization for unstable angina) and all-cause mortality (HR 1.58). The trial was stopped early when the data safety monitoring board found excess mortality. 5 / Solid The excess mortality was eventually attributed to an off-target effect: torcetrapib stimulates aldosterone release independently of its CETP inhibition, raising blood pressure by 3-5 mmHg, which contributed to cardiovascular harm.

Dalcetrapib: dal-OUTCOMES (2012): Dalcetrapib raised HDL-C by 31% without the aldosterone off-target effect of torcetrapib. It did not lower LDL-C significantly. The trial randomized 15,871 patients with recent acute coronary syndrome. Primary cardiovascular outcome: no difference (HR 1.04, 95% CI 0.93-1.16). 5 / Solid Dalcetrapib did not lower LDL, so it tested purely whether raising HDL-C reduces events. It did not.

Evacetrapib: ACCELERATE (2016): Evacetrapib raised HDL-C by 130% and lowered LDL-C by 37%. The combination of HDL-raising and LDL-lowering was anticipated to be beneficial. The trial enrolled 12,092 patients with high-risk atherosclerotic disease. Primary endpoint: no significant difference in major adverse cardiovascular events (HR 1.01, 95% CI 0.91-1.11). 5 / Solid Even with a 130% raise in HDL-C and a 37% LDL reduction, no cardiovascular benefit was observed: a profoundly negative result for the HDL hypothesis.

Anacetrapib: REVEAL (2017): The largest and most definitive CETP inhibitor trial enrolled 30,449 patients on intensive statin therapy and randomized them to anacetrapib or placebo. Anacetrapib raised HDL-C by 104% and lowered LDL-C by approximately 18 mg/dL. A modest 9% relative reduction in major coronary events was observed (HR 0.91, 95% CI 0.85-0.97, p=0.004). 5 / Solid However, this marginal benefit was attributed in subsequent analyses to the LDL-C lowering and Lp(a) reduction produced by anacetrapib, not to the HDL-C raising. Despite this modest signal, the drug was not developed commercially, as the benefit was deemed insufficient to justify the compound’s unusual lipophilicity (it accumulates in adipose tissue for years) and the marginal clinical advantage.

Together, these four trials represent one of the most expensive and definitive natural experiments in cardiovascular pharmacology. They answered the HDL question: pharmacological HDL-C elevation does not reduce cardiovascular events.

AIM-HIGH and HPS2-THRIVE: The Niacin Story in Full

The niacin trial program attempted to test the clinical value of the most widely available HDL-raising agent before the CETP inhibitor trials arrived.

AIM-HIGH (Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides: Impact on Global Health Outcomes): Enrolled 3,414 patients with cardiovascular disease, LDL below 80 mg/dL on simvastatin, HDL below 40 mg/dL (men) or below 50 mg/dL (women), and triglycerides 150-400 mg/dL. Extended-release niacin was added to background statin therapy. The trial was stopped at 36 months for futility: primary composite endpoint (first event of coronary heart disease death, nonfatal MI, ischemic stroke, hospitalization for acute coronary syndrome, or symptom-driven coronary or cerebral revascularization) showed HR 1.02 (p=0.79). The niacin group also had a non-significant trend toward more ischemic strokes. 5 / Solid

HPS2-THRIVE (Heart Protection Study 2-Treatment of HDL to Reduce the Incidence of Vascular Events): 25,673 patients with ASCVD on simvastatin randomized to extended-release niacin plus laropiprant (a prostaglandin D2 antagonist that reduces niacin-induced flushing, improving adherence) versus placebo. Primary cardiovascular outcome: no difference in major vascular events (4-year event rate 13.2% niacin/laropiprant vs 13.7% placebo, rate ratio 0.96, p=0.29). Serious adverse events significantly increased with niacin: new-onset diabetes (3.7% vs 2.5%), hepatic disturbances, serious non-fatal muscle effects, gastrointestinal adverse events, and skin complications. 5 / Solid The harm-benefit ratio was clearly unfavorable.

After HPS2-THRIVE, the FDA required that niacin-containing cardiovascular products include a statement that their clinical benefit was no longer established. The European Medicines Agency withdrew the marketing authorization for niacin/laropiprant (marketed as Tredaptive in Europe).

The Epidemiological Residue: Why Low HDL-C Still Matters

Despite the failure of HDL-raising therapies, low HDL-C remains a clinically useful marker for several reasons:

Risk marker function: The Emerging Risk Factors Collaboration, in its analysis of 68 prospective studies (302,430 participants, 2.79 million person-years), found that HDL-C retained independent predictive value for coronary heart disease after adjustment for LDL-C, non-HDL-C, blood pressure, and smoking status (HR per 15 mg/dL higher HDL-C: 0.78, 95% CI 0.74-0.82). This means that in a population sense, knowing someone’s HDL-C improves risk prediction beyond standard risk factors, even though the HDL-C itself is not causally driving protection. 5 / Solid 61423-1)

Marker of metabolic dysfunction: Very low HDL-C (below 30 mg/dL) almost always signals an underlying metabolic or secondary condition: severe hypertriglyceridemia (where CETP-mediated depletion of HDL is substantial), profound insulin resistance, anabolic steroid use, or rare genetic conditions affecting ApoA-I production (familial hypoalphalipoproteinemia, Tangier disease, ABCA1 mutations). Identifying these conditions changes management independent of the HDL-C level.

Surrogate for lifestyle quality: In a preventive cardiology evaluation, a patient with naturally high HDL-C (above 60 mg/dL) has typically earned that HDL level through habits that protect the cardiovascular system: regular physical activity, healthy weight maintenance, non-smoking. The HDL is a readout of those behaviors, and the behaviors are what the clinician wants to identify and reinforce.

Extended Mechanism: HDL Subclasses and the HDL Maturation Pathway

Large and Small HDL: HDL2 versus HDL3

HDL particles are not uniform. They are classified by density and size into HDL2 (larger, more lipid-rich, less dense) and HDL3 (smaller, more protein-rich, more dense). Nascent, lipid-poor particles mature from HDL3 to HDL2 as they accept cholesterol from peripheral tissues and are esterified by LCAT. HDL2 is considered the more mature, cholesterol-loaded form.

Epidemiological studies have suggested that HDL2 is more strongly inversely associated with cardiovascular risk than HDL3, but the difference is modest and clinically not useful to measure. Standard lipid panels do not distinguish HDL subclasses.

ApoA-I: The Functional Protein

ApoA-I, the dominant HDL protein, drives most of HDL’s biological functions. ApoA-I has direct anti-inflammatory effects on vascular smooth muscle, promotes ABCA1-mediated cholesterol efflux from macrophages, and activates LCAT for cholesterol esterification. ApoA-I levels correlate with reverse cholesterol transport capacity better than HDL-C.

ApoA-I Milano, a naturally occurring ApoA-I mutant found in an extended Italian family with unusually low HDL-C but very low cardiovascular disease rates, has attracted research interest as a therapeutic target. The paradox of the ApoA-I Milano carriers (low HDL-C but long, cardiovascular-event-free lives) reinforced the distinction between HDL-C concentration and HDL function: their HDL was small and lipid-poor (low HDL-C) but extraordinarily efficient at cholesterol efflux. 5 / Solid

Clinical trials of recombinant ApoA-I Milano infusions showed promising early signals in reducing coronary plaque in very small studies, but no large outcomes trial has been conducted to confirm benefit.

The Dysfunctional HDL Concept

In chronic inflammatory states, HDL can become “dysfunctional”: losing its anti-inflammatory and antioxidant properties and paradoxically acquiring pro-inflammatory characteristics. HDL isolated from patients with active rheumatoid arthritis, lupus, or sepsis can fail to protect against LDL oxidation and may actually promote macrophage inflammatory activation.

The dysfunctional HDL concept explains why patients with chronic inflammatory conditions have raised cardiovascular risk despite sometimes-normal or even high HDL-C levels: the HDL concentration is maintained but the particles have lost their protective functions. This is another dimension of the function-concentration distinction.

Extended Patient Experience: The High HDL Conversation in Practice

Three Patient Archetypes

The reassured patient with high HDL (Helen’s archetype): Told for years their HDL was “great,” they use it as a reason to defer treatment of other risk factors. The clinical task is gentle disabusing: the HDL is one favorable sign among a panel that includes concerning findings. The cardiovascular risk calculation does not allow HDL to neutralize LDL, CAC, or smoking history.

The patient with very low HDL as the presenting finding: HDL of 22 mg/dL with triglycerides of 380 mg/dL on a routine lipid panel is not a primary HDL disorder: it is the lipemic consequence of severe hypertriglyceridemia, insulin resistance, or both. The clinical approach is to identify and treat the underlying metabolic disorder (dietary change, glycemic control, weight reduction, fibrate if triglycerides remain above 500 mg/dL) and observe HDL-C rise as a byproduct of metabolic improvement.

The patient who asks about niacin or supplements to raise HDL: Given the HPS2-THRIVE and AIM-HIGH data, niacin is not recommended. Over-the-counter “HDL boosters” (fish oil, berberine, red yeast rice) either have no meaningful effect on HDL-C or act through other mechanisms. Aerobic exercise is the most evidence-based way to raise HDL-C modestly (3-5 mg/dL with consistent aerobic training for 12+ weeks), with cardiovascular benefit attributable to the exercise itself rather than the HDL change.

Illinois Practice Context: HDL in a Preventive Cardiology Evaluation

At Carle Foundation Hospital’s preventive cardiology program, the lipid panel includes LDL-C, HDL-C, triglycerides, total cholesterol, non-HDL-C, and ApoB as standard. HDL-C is reviewed in context with ApoB and the metabolic profile. A CAC score is obtained for intermediate-risk patients as the primary risk stratification tool. Low HDL-C (below 40 mg/dL) triggers assessment for metabolic syndrome, insulin resistance, and secondary causes.

For patients at Northwestern Medicine or Rush University Medical Center in Chicago, advanced lipid panels including HDL particle size and concentration, ApoA-I, and lipoprotein subclass analysis are available through their clinical lipidology programs. These add research value but are not standard of care.

The message communicated to patients in the preventive cardiology consultation is: “Your HDL-C tells us something about your metabolic health and your overall cardiovascular risk pattern, but we do not treat it as a target. The targets we treat are your LDL-C, your ApoB, your blood pressure, your blood glucose, and your smoking status. When we optimize those, HDL-C often improves on its own.”

Extended Patient Experience: Exercise, Lifestyle, and HDL

What Actually Raises HDL-C

For patients asking how to improve their HDL-C, the honest answer is that the interventions that raise HDL-C modestly are valuable for reasons that go far beyond the HDL change:

Aerobic exercise: Regular aerobic training (150+ minutes per week of moderate-intensity exercise, or 75 minutes of vigorous exercise) raises HDL-C by 3-5 mg/dL over 12-20 weeks. The cardiovascular benefit of the exercise program far exceeds what the HDL-C change alone would predict, because exercise also reduces LDL-C, reduces blood pressure, reduces insulin resistance, and improves cardiac fitness directly. 5 / Solid

Smoking cessation: Cigarette smoking reduces HDL-C by approximately 5 mg/dL through a mechanism involving accelerated catabolism of ApoA-I. Smoking cessation raises HDL-C by approximately 5 mg/dL within 30-60 days. The primary benefit of smoking cessation for cardiovascular risk is overwhelmingly through mechanisms other than HDL-C (reduced thrombogenicity, reduced endothelial injury, improved vasomotor function), but the HDL-C improvement is a useful reinforcing message for patients who respond to laboratory numbers.

Weight loss: A 5-10% body weight reduction raises HDL-C by approximately 2-3 mg/dL. The primary mechanism is improved insulin sensitivity: as insulin resistance decreases, CETP activity normalizes and HDL catabolism slows. Again, the cardiovascular benefit of the weight loss exceeds the HDL-C change many fold.

Moderate alcohol intake: Observational data consistently show that moderate alcohol consumption (1-2 drinks per day) is associated with higher HDL-C (by approximately 3-5 mg/dL) and lower cardiovascular risk in epidemiological studies. However, the cardiovascular benefit of alcohol is contested by Mendelian randomization studies, which show that genetic variants associated with lower alcohol consumption are not associated with higher cardiovascular risk. The current clinical consensus is that alcohol should not be recommended as a cardiovascular strategy: the harms (cancer risk, addiction, liver disease, accident risk) substantially outweigh any modest cardiovascular signal. 5 / Solid

The Metabolic Syndrome HDL Pattern

Helen’s high HDL-C (67 mg/dL) may reflect a genetic background predisposed to efficient ApoA-I production. But notably, the inverse: very low HDL-C: in the metabolic syndrome context is a diagnostically important finding.

The metabolic syndrome criteria (any three of five: raised waist circumference, raised triglycerides, low HDL-C, raised fasting glucose, raised blood pressure) include low HDL-C as one of five components. A patient with triglycerides of 180 mg/dL, HDL-C of 35 mg/dL, waist circumference above threshold, fasting glucose 103 mg/dL, and blood pressure 132/84 mmHg has metabolic syndrome: and this pattern identifies significantly raised cardiovascular risk, residual ischemic risk even on statin therapy, and higher probability of insulin resistance driving an ApoB-LDL-C discordance.

In this patient, the low HDL-C is not a target for pharmacological treatment: it is a signal that the metabolic machinery is dysregulated, and that ApoB, insulin resistance treatment, and lifestyle change are the clinical priorities.

Extended Evidence Review: HDL Particle Number and Function Tests

Beyond HDL-C: What Current Testing Can and Cannot Do

HDL-cholesterol (HDL-C) measures the mass of cholesterol carried by HDL particles, just as LDL-C measures the mass of cholesterol in LDL particles. But just as LDL particle number (measured by ApoB) outperforms LDL-C for cardiovascular risk prediction, HDL particle number (measured by ApoA-I or HDL-P by NMR) provides information about HDL biology that HDL-C does not capture.

ApoA-I is the major structural protein of HDL (one to four ApoA-I molecules per HDL particle). ApoA-I concentration reflects total HDL particle mass more completely than HDL-C because small, lipid-poor nascent HDL particles (HDL precursors active in early cholesterol reverse transport) carry less cholesterol per particle but are functionally active. A patient with high total HDL-C but predominantly large, cholesterol-enriched, mature HDL spheres may have normal or high HDL-C with relatively fewer total HDL particles: the opposite of the situation that HDL measurement is meant to capture.

In practice, ApoA-I measurement is not routinely ordered in most clinical settings. The ACC/AHA guidelines do not include ApoA-I as a standard clinical risk marker. Its primary utility is:

1. Identifying patients with severely low ApoA-I as a marker of extreme HDL deficiency syndromes (Tangier disease, ApoA-I Milano carriers)
2. Confirming the atherogenic dyslipidemia pattern (ApoB high, ApoA-I low) in metabolic syndrome
3. Research settings where HDL functional analysis requires accurate particle quantification

The HDL “function” tests: cholesterol efflux capacity assays, antioxidant capacity: remain research tools and are not clinically validated as individual risk predictors. The fundamental limitation of HDL-C as a therapeutic target stands: we lack both a drug that raises HDL-C and simultaneously reduces cardiovascular events, and a validated functional HDL measure to replace HDL-C as a target. Until these are established, HDL-C remains a risk marker, not a treatment target.

Extended Mechanism: Reverse Cholesterol Transport in Depth

The Macrophage Cholesterol Removal System

The cardinal function of HDL in atherosclerosis prevention is macrophage reverse cholesterol transport: the removal of excess cholesterol from macrophage foam cells within the arterial wall and its transport to the liver for excretion in bile. This pathway has several steps, each with distinct molecular machinery:

Step 1: Cholesterol efflux from macrophages: Excess cholesterol in macrophages is exported via ATP-binding cassette transporter A1 (ABCA1) and ABCG1 onto lipid-poor ApoA-I and small pre-beta HDL particles. This is the critical first step: without efficient efflux, cholesterol accumulates in macrophages and drives foam cell formation. Tangier disease (genetic ABCA1 deficiency) causes near-zero HDL-C, massive macrophage cholesterol loading, and accelerated atherosclerosis.

Step 2: HDL maturation: Nascent pre-beta HDL particles acquire cholesterol and phospholipids, with cholesterol esterification by LCAT (lecithin-cholesterol acyltransferase), producing mature spherical alpha-HDL particles. HDL-C on a standard lipid panel primarily measures this mature spherical form.

Step 3: Hepatic HDL cholesterol delivery: Mature HDL delivers cholesterol to the liver via scavenger receptor BI (SR-BI), which selectively removes cholesterol from HDL without degrading the particle (unlike the LDL receptor pathway, which catabolizes the entire LDL particle). After cholesterol removal, the now cholesterol-depleted, smaller HDL is recycled for another cycle of macrophage cholesterol efflux.

Step 4: Hepatic excretion: Cholesterol delivered to the liver via SR-BI or CETP-mediated transfer to LDL is either excreted directly into bile or converted to bile acids for fecal elimination.

CETP (cholesteryl ester transfer protein) transfers cholesteryl esters from HDL to LDL and VLDL in exchange for triglycerides, effectively transferring cholesterol from HDL back into the pro-atherogenic lipoprotein pool. This is why CETP inhibition seemed like a compelling strategy: blocking CETP should keep cholesterol in HDL and out of LDL. The four CETP inhibitor failures described in Section 5 demonstrated that raising HDL-C by this particular mechanism did not translate to clinical benefit, and torcetrapib actually caused harm through off-target effects.

Why CETP Inhibition Failed: The Dalcetrapib Lesson

Dalcetrapib (Hoffmann-La Roche) raised HDL-C by approximately 30% without reducing LDL-C. In the dal-OUTCOMES trial (18,083 patients, post-ACS, median 31 months), dalcetrapib produced no reduction in cardiovascular events (HR 1.04, 95% CI 0.93-1.16). 5 / Solid

A post-hoc pharmacogenomic analysis of dal-OUTCOMES identified a signal: patients with the ADCY9 gene polymorphism (rs1967309) had significantly reduced events on dalcetrapib, while patients without this polymorphism had no benefit or harm. This finding led to development of dalcetrapib with genetic pre-selection: a personalized medicine approach where only patients with the ADCY9 AA genotype receive dalcetrapib. The dal-GenE trial with this patient selection strategy is ongoing.

The ADCY9 pharmacogenomic story is the most sophisticated example in cardiovascular medicine of attempting to identify the genetic subgroup who benefits from an otherwise-neutral drug. Whether it succeeds in establishing dalcetrapib as a viable targeted therapy remains to be determined.

The ApoA-I Milano Paradox

One of the most clinically fascinating HDL stories is that of ApoA-I Milano. In 1980, a variant of ApoA-I was discovered in a family in the village of Limone sul Garda, Italy: the ApoA-I R173C mutation (called ApoA-I Milano). Carriers had very low HDL-C (in the range of 15-25 mg/dL) but paradoxically lived to old age without apparent cardiovascular disease. 5 / Solid

Subsequent mechanistic work showed that ApoA-I Milano functions differently from wild-type ApoA-I: it forms dimers through the cysteine substitution, and these dimers are exceptionally efficient at promoting cholesterol efflux from macrophages: far more efficient per particle than wild-type ApoA-I. The low HDL-C in carriers reflects rapid HDL particle turnover (the particles deliver cholesterol to the liver so efficiently that they do not accumulate in the circulation), not impaired cholesterol removal.

This paradox: very low HDL-C with enhanced reverse cholesterol transport function: provided direct mechanistic support for the distinction between HDL particle function and HDL-C concentration. It also inspired the development of ETC-216, a recombinant ApoA-I Milano infusion product studied by Nissen and colleagues in a landmark trial of five infusions over five weeks: carotid intima-media thickness (measured by intravascular ultrasound) regressed significantly compared to baseline. 4 / Promising The finding was never advanced to a large cardiovascular outcomes trial because of manufacturing cost and complexity, but it remains an important proof-of-concept for HDL function therapy.

HDL-C in Practice

Stop Dying Early does not set an HDL-C treatment target. The program positions HDL-C as a signal: low HDL-C (below 40 mg/dL in men, below 50 mg/dL in women) triggers investigation of the underlying metabolic disorder (insulin resistance, metabolic syndrome, hypertriglyceridemia) and assessment of lifestyle factors (smoking cessation, exercise initiation, dietary pattern).

Paradoxically high HDL-C (above 80 mg/dL) receives the same non-intervention posture: Mendelian randomization data (Voight et al. 2012) show that genetically raised HDL-C does not reduce cardiovascular events, and CETP loss-of-function mutations in humans (rare individuals with HDL-C above 100 mg/dL) do not have significantly lower cardiovascular event rates.

The practical message for patients: “We measure your HDL-C to understand your metabolism and to confirm that your lipid-lowering therapy is not adversely affecting your HDL. We do not prescribe drugs to change your HDL-C number, because no drug that changes HDL-C has been shown to reduce heart attacks. Instead, we use your HDL-C as one piece of the full metabolic picture alongside ApoB, triglycerides, glucose, and blood pressure.”

HDL-C and Kidney Disease: A Special Consideration

Patients with chronic kidney disease have a distinct HDL dysfunction profile. Their HDL-C may be normal or even raised in numeric terms, but the particles have impaired antioxidant activity, reduced cholesterol efflux capacity, and acquire pro-inflammatory characteristics from the uremic milieu. In CKD, HDL-C as a cardiovascular risk predictor loses its usual inverse relationship: some CKD patients with HDL-C above 60 mg/dL have poor cardiovascular outcomes, while those with the HDL quality dysfunction pattern have the highest event rates. 4 / Promising

This observation reinforces that HDL function (not concentration) is the relevant biological variable, and that the measurement tools available in routine clinical practice do not capture this function. For patients with CKD, this program focuses management attention on LDL-C, ApoB, blood pressure, and glycemic control rather than HDL-C, because the latter is neither a reliable risk marker nor a treatment target in this population.

HDL-C as a Mirror, Not a Target

The final word on HDL in this clinical framework: HDL-C is a metabolic mirror. When you look at it, you see a reflection of the patient’s metabolic health, their exercise habits, their smoking status, their body composition, and their lipid metabolism function. A low HDL-C tells you something is wrong with the metabolic machinery. A high HDL-C does not guarantee cardiovascular safety: the CETP inhibitor trials demonstrated this definitively.

What the mirror does not show you is a treatment target. In 2026, with all four CETP inhibitor trials behind us and both niacin outcome trials showing no benefit or harm, there is no drug that raises HDL-C and reduces cardiovascular events. This may change: the dalcetrapib pharmacogenomics story, the ongoing work on ApoA-I mimetic peptides, and the reconceptualization of HDL function versus concentration may eventually produce a clinically useful HDL-targeted therapy. Until that day, HDL-C is measured, interpreted in metabolic context, and used to prompt investigation of the underlying drivers: not prescribed against.

This program’s role in HDL education is to prevent two clinical errors: overtreating high HDL as a source of cardiovascular protection, and adding drugs to “treat” low HDL when the correct intervention is lifestyle change, metabolic syndrome management, and smoking cessation.

The Signal Check is fifteen questions mapping the cardiovascular risk pattern across the physiological domains most commonly missed in standard screenings. It produces a specific starting point for your next clinical conversation.