Review

The changing landscape of atherosclerosis

https://doi.org/10.1038/s41586-021-03392-8 Peter Libby1,2 ✉

Received: 29 May 2020

Accepted: 24 February 2021 Emerging evidence has spurred a considerable evolution of concepts relating to
Published online: 21 April 2021 atherosclerosis, and has called into question many previous notions. Here I review this
evidence, and discuss its implications for understanding of atherosclerosis. The risk
Check for updates
of developing atherosclerosis is no longer concentrated in Western countries, and it is
instead involved in the majority of deaths worldwide. Atherosclerosis now affects
younger people, and more women and individuals from a diverse range of ethnic
backgrounds, than was formerly the case. The risk factor profile has shifted as levels of
low-density lipoprotein (LDL) cholesterol, blood pressure and smoking have
decreased. Recent research has challenged the protective effects of high-density
lipoprotein, and now focuses on triglyceride-rich lipoproteins in addition
to low-density lipoprotein as causal in atherosclerosis. Non-traditional drivers of
atherosclerosis—such as disturbed sleep, physical inactivity, the microbiome, air
pollution and environmental stress—have also gained attention. Inflammatory
pathways and leukocytes link traditional and emerging risk factors alike to the altered
behaviour of arterial wall cells. Probing the pathogenesis of atherosclerosis has
highlighted the role of the bone marrow: somatic mutations in stem cells can cause
clonal haematopoiesis, which represents a previously unrecognized but common and
potent age-related contributor to the risk of developing cardiovascular disease.
Characterizations of the mechanisms that underpin thrombotic complications of
atherosclerosis have evolved beyond the ‘vulnerable plaque’ concept. These advances
in our understanding of the biology of atherosclerosis have opened avenues to
therapeutic interventions that promise to improve the prevention and treatment of
now-ubiquitous atherosclerotic diseases.

Although atherosclerotic cardiovascular disease was previously consid- burden. This Review addresses the evolving concepts of atherogen-
ered a problem that was concentrated in the industrialized world, it now esis and the opportunities for preventing and treating atherosclerosis
spans the globe. We have witnessed an ‘epidemiological transition’1,2. afforded by new insights into its pathogenesis.
Increased sanitation, vaccination and the treatment of acute infections
have diminished the prevalence of communicable diseases in develop-
ing countries, and more individuals now survive to experience chronic The changing face of atherosclerosis
diseases such as atherosclerosis (see ref. 3 for an introduction to the The classic candidate for heart attack was a middle-aged, white man
fundamental concepts of atherosclerosis). The adoption of less healthy with hypertension and hypercholesterolaemia, who smoked cigarettes.
dietary patterns may also have contributed to this trend. Many people This picture has evolved considerably in recent decades. We now pos-
now live longer only to suffer the consequences of atherosclerosis, sess effective therapies for the treatment of high blood pressure and
including myocardial infarction, ischaemic cardiomyopathy (which is lipid disorders, and control of hypertension and hypercholesterolaemia
the commonest cause of heart failure), strokes (which often rob people has improved4. A reduction in smoking, accompanied by a decrease
of their independence, mobility, cognition or ability to communicate) in second-hand smoke, has gained a foothold in many societies. We
and peripheral arterial disease, which limits activity and jeopardizes have further witnessed a vast change in people with cardiovascular
limbs. These conditions contribute to ‘morbidity extension’ in the risk and those who suffer from acute coronary syndromes. Although
developing world such that, although many individuals escape early individuals in mid-life do not evade risk, coronary artery disease now
death, they must bear the burden not only of chronic cardiovascular affects an increased number of younger women, and—with the ageing
diseases but also of arthritis, depression and other long-term impedi- of the population in many countries—the very old now account for an
ments to a healthy life. Today, the major pool of risk for developing increasing proportion of patients with cardiac conditions4–6.
cardiovascular disease occurs not in Western countries but in the more An epidemic of obesity has swept across the world7,8. Excess adipos-
populous developing world. Atherosclerotic cardiovascular disease ity (especially that accumulated in the abdomen) and a fatty liver drive
now accounts for the majority of mortality worldwide. This global insulin resistance, which sets the stage for diabetes, and shows links
spread creates an urgent need to understand the genesis of this malady, with hypertension. From a metabolic perspective, individuals of some
advance in its management and develop prospects for alleviating its ethnicities often tolerate the accumulation of visceral adipose tissue

Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, MA, USA. 2Harvard Medical School, Boston, MA, USA. ✉e-mail: [email protected]
1

524 | Nature | Vol 592 | 22 April 2021

particularly poorly9. Asian and South Asian individuals, as well as people Large-scale cohort investigations, such as the Framingham study,
of some ethnicities in Central and South America, can develop dys- revealed risk factors for atherosclerosis that we now regard as ‘tradi-
metabolism manifested as glucose intolerance at lower abdominal tional’25. However, long-term trends have modified risk factors such that
girths than white individuals. Given the large populations in Asia and these traditional factors no longer capture the contemporary reality
Central and South America, increased prosperity—with attendant shifts of atherosclerosis. Genetic risk scores have undergone continuing
from traditional dietary habits—and continuing tobacco use, as well as refinement and incorporate increasingly expanded numbers of inher-
the growing burden of obesity and diabetes (often accompanied by ited variants that influence atherosclerotic events. As these genetic
hypertension) presents an enormous public health challenge and is a panels can predict risk from birth, they may inform the early targeted
contributor to regional risks of atherosclerotic cardiovascular disease. allocation of preventive measures in younger individuals who have an
Rather than elevated low-density lipoprotein (LDL) cholesterol, an augmented genetic predilection to develop atherosclerotic disease26.
elevation in triglyceride-rich lipoproteins (TGRL) and low high-density Indeed, lifestyle measures appear to mitigate risk of cardiovascular
lipoprotein (HDL) now comprise the major pattern of lipid abnormality events across the spectrum of estimated genetic risk. Yet, the ability of
in many patients who are treated for atherosclerotic cardiovascular even the latest generation of genetic risk scores to improve prediction
disease10. Highly effective and now-inexpensive therapies for lowering of atherosclerotic events over more traditional algorithms remains
LDL have contributed to an overall drop in LDL, whereas obesity, its controversial27,28.
attendant insulin resistance and a high-carbohydrate diet favour a rise Recent research has challenged and expanded on the traditional risk
in the prevalence of the cluster of conditions referred to as the ‘meta- factors. With global trends towards a decrease in LDL and the introduc-
bolic syndrome’, which is characterized in part by an elevation in TGRL. tion of highly effective therapies for lowering LDL, as well as inexpensive
The prevalence of this cluster—which also includes increased waist and efficacious antihypertensive therapies, these drivers of chronic
circumference, low HDL cholesterol, high blood pressure and raised risk contribute less today than in previous years. Most markedly and
fasting blood glucose—rose by 35% from 1988–1994 to 2007–2012 in despite decades of belief that HDL protected from atherosclerosis,
the USA11. Women, members of minoritized groups and populations in recent human genetic studies—and the failure of several independent
developing countries thus bear an increasing burden of atherosclerotic pharmacological measures to raise HDL to reduce atherosclerotic
cardiovascular disease. Indeed, despite advances in controlling risk events—have called into question the protective effect of HDL29. Men-
factors in many high-income countries, atherosclerotic cardiovascular delian randomization studies that have corrected for pleiotropy have,
disease now predominates in lower-income areas. The Global Burden of however, provided some support for the protective effect of HDL30.
Disease study shows that world prevalence of ischaemic heart disease Moreover, functional attributes of HDL fractions that are not captured
has increased from about 100 million in 1990 to over 180 million cases by steady-state measurements of total HDL cholesterol concentrations
in 20198. In some regions of the USA and UK, the decline in prevalence (such as the capacity to mediate cholesterol efflux or anti-inflammatory
of ischaemic heart disease—attributed to controlling risk factor—has actions) may yet exert anti-atherosclerotic effects31,32.
slowed or halted in the period 2014 to 20198. Atherosclerotic cardio- The risk of plasma triglyceride concentration (a biomarker of a class
vascular disease has become a global concern, and we may be losing of lipoproteins that include the TGRL) was passed over for many years,
ground in prevention even in higher-income countries. as the belief in the protective effect of HDL rendered it logical to adjust
triglycerides for HDL—a precaution that attenuated the risk attributed
to TGRL33. Triglycerides and HDL tend to vary inversely, and a recent
A reassessment of the lipid risk factors ranking34,35 of the relevant risk factors demotes HDL as a protective
LDL, a particle encircled by its signature apolipoprotein B compo- factor and points to TGRL as a potent predictor of cardiovascular risk.
nent, causes atherosclerosis12,13. If the entire population maintained Moreover, in contrast to the situation with HDL, contemporary and
LDL concentrations akin to those of a neonate (or to those of adults of concordant human genetic studies strongly support a causal role for
most other animal species), atherosclerosis might well be an orphan TGRL in atherosclerosis and its complications36. A variety of inherited
disease14. The duration and extent of exposure to above-ideal concen- sequence variations that affect lipoprotein lipase, or factors that modu-
trations of LDL associate with atherosclerotic disease13,15. However, late the activity of this enzyme, alter the rate of atherosclerotic events,
the treatment of children and adolescents with cholesterol-lowering and these findings furnish strong human genetic evidence for the causal
drugs presents many challenges, and lifelong elevated concentrations role of TGRL in their pathogenesis. Apolipoprotein CIII, ANGPTL3 and
of LDL cholesterol have already sown the seeds of atherosclerosis in ANGPTL4 inhibit the ability of lipoprotein lipase to hydrolyse triglyc-
millions of people, increasing their risk for cardiovascular disease. erides in TGRL, and thus cause accumulation of these particles. By
Despite effective interventions for the control of LDL, blood pressure contrast, apolipoprotein V augments the activity of lipoprotein lipase
and other traditional risk factors, a considerable residual risk remains and enhances TGRL clearance37,38. The activity of lipoprotein lipase thus
for atherosclerotic cardiovascular disease16. For example, recent clinical regulates plasma triglyceride concentrations. Gain- or loss-of-function
trials of novel cardiovascular agents performed in subjects optimally variants in this pathway that raise TGRL track with increased numbers
treated with standard-of-care background therapy found that about of atherosclerotic events, and those that lower TGRL correlate with
1 in 20 will have a recurrent ischaemic event in the year after an acute improved outcomes. The triglyceride component of TGRL does not
coronary syndrome17,18. One in ten individuals who survive an acute appear to account for their atherogenicity10. TGRL, as with LDL, bear
myocardial infarction in the USA will require readmission within one apolipoprotein B; they also contain cholesterol, and can deliver it effec-
month, at considerable personal and societal cost19. Moreover, in addi- tively to macrophages in the atheroma. TGRL provoke inflammation, in
tion to obesity and its associated insulin resistance, rises in air pollution, part owing to their apolipoprotein CIII content. TGRL concentrations
transitions from traditional diets towards those that may aggravate correlate better with inflammatory status than does LDL itself39,40. This
cardiovascular risk and other exposures under intense investigation refocusing on TGRL as a causal risk factor, and lack of actionability of
(ranging from environmental noise to impaired sleep) may mitigate altering HDL thus far, has notable therapeutic implications.
some of the advances that have been made in prevention20–22. Chief Observational epidemiological studies have long associated a
among readily remedied shifts towards unhealthy diets, the consump- special form of LDL, lipoprotein(a), with atherothrombotic risk41.
tion of sugar-sweetened beverages, often high in fructose content, Lipoprotein(a) consists of an LDL particle, the signature apolipoprotein
may contribute to obesity and its adverse metabolic consequences23,24 (apolipoprotein B) of which has bound covalently to apolipoprotein(a).
Indeed, modifiable risk factors contribute enormously to the global Lipoprotein(a) carries oxidized lipids and may inhibit fibrinolysis owing
burden of ischaemic heart disease2. to a structural similarity with plasminogen. Concordant human genetic

Nature | Vol 592 | 22 April 2021 | 525

Review

Leukocyte
adhesion
molecule Cognate
ligands

T lymphocyte
Lipoprotein Endothelium
Monocyte

Intima
Internal elastic
lamina
Foam cell
Smooth
muscle cell Media

External elastic
lamina

Adventitia

Fig. 1 | Initiation of atherosclerosis. The normal artery comprises three arise from blood monocytes that have matured into macrophages. Recent
layers: the innermost intima (in close contact with the bloodstream), the tunica evidence146 in mice indicates that smooth muscle cells can undergo metaplasia,
media, and the outer coat and adventitia. Under homeostatic conditions, the and give rise to foam cells that bear markers in common with those of
endothelial monolayer that lines the intima does not gather blood leukocytes. macrophages. T lymphocytes—although fewer in number than the foam cells—
When activated by proinflammatory cytokines or other irritative stimuli produce mediators that orchestrate many functions of these innate immune
related to cardiovascular risk factors, endothelial cells can express a leukocyte cells. In humans (but not in many of the small animals that are often used
adhesion molecule (such as VCAM-1) that interacts with its cognate ligands experimentally), the intima contains resident smooth muscle cells. Other
(VLA4) to promote the rolling, and eventually adherence, of blood monocytes smooth muscle cells (that are usually located in the media) can penetrate into
and lymphocytes to the endothelial layer. Chemoattractant cytokines can the intima, where they join resident smooth muscle cells to promote the
direct the migration of these bound leukocytes into the intima. Within the accumulation of extracellular matrix that these cells synthesize within the
intima, foam cells form by uptake of lipids. Some of these lipid-laden foam cells expanding intima.

studies provide persuasive evidence for the causality of elevated in mice46,50. Candidate antigens include forms of LDL as discussed in
lipoprotein(a) not only in atherosclerosis but also in calcific aortic ‘Oxidized LDL and the initiation of lesions’. B lymphocytes also can exert
valve disease42–45. a dual role in atherogenesis. Natural IgM antibodies encoded in the
germline and produced by B1 cells mitigate experimental atheroscle-
rosis51,52. B2 cells can produce antibodies that may drive this process.
Inflammation drives atherosclerosis One candidate antigen, the mitochondrial enzyme ALDH4A1, emerged
Beyond dyslipidaemia, a convincing body of experimental and clini- from analysis of B cells isolated from mouse atheromata53.
cal data now indicates that inflammation participates fundamentally An increased number of links between inflammation, immunity and
in atherogenesis and in the pathophysiology of ischaemic events46. intermediary metabolism have recently emerged. Inflammatory acti-
Inflammation does not supplant or demote lipid risk; rather, inflamma- vation of mononuclear phagocytes and endothelial cells tends to shift
tory responses provide a series of pathways that link lipids and other their metabolism towards glycolytic pathways54–56. Altered tryptophan
traditional risk factors to atherosclerosis. For example, concentrations metabolism has also garnered considerable interest in atherosclerosis.
of remnant lipoprotein show links with levels of C-reactive protein, Cytokines induce indolamine dioxygenase (the rate-limiting step in
a biomarker of inflammation40. A large body of evidence implicates tryptophan catabolism), which lowers intracellular tryptophan stores
inflammation in hypertension47. Experimental investigations have and augments the production of kynurenine and its metabolites; this
pinpointed the participation of innate and adaptive immunity in ath- latter pathway may have a counter-regulatory function by muting
erosclerosis (Figs. 1, 2). Human biomarker studies have shown that inflammation and the cellular immune response57–59.
indicators of inflammation predict risk of cardiovascular disease in The applicability to humans of experimental evidence that impli-
a broad swath of individuals with or without manifest cardiovascular cated inflammation in atherosclerosis has engendered considerable
disease, and independently of all traditional risk factors48. The acute controversy60,61. However, recent clinical trials have shown that target-
phase reactant (C-reactive protein), which can be measured with a ing inflammation can reduce cardiovascular events even in individuals
highly sensitive assay (known as hsCRP), is a validated and clinically who have already been treated with a full panel of effective standard
useful gauge of the overall innate immune status of an individual in therapies. The ‘Canakinumab Anti-inflammatory Thrombosis Out-
relation to atherosclerotic risk49. comes Study’ (CANTOS) allocated randomly an antibody that neu-
In addition to innate immunity (which depends largely on cytokines tralizes the proinflammatory cytokine IL-1β to patients with stable
and macrophages), the adaptive arm of the immune response operates coronary artery disease at least one month after a qualifying myocar-
during atherogenesis. T lymphocytes primarily aggravate atherogen- dial infarction62. The enrolled population had signs of inflammation
esis, but T-helper 2 and regulatory T cells can mute this process, at least despite standard-of-care medical therapy, as mandated by prevailing

526 | Nature | Vol 592 | 22 April 2021

 IgM natural
antibody B1 cell
Endothelium

Factors that promote Factors that mute

atherogenesis PDGF Dividing SMC atherogenesis
IL-10 TH2 cell
IL-1, TNF
Intima B2 cell BAFF Foam cell
TGF-β Treg cell
M-CSF
IFNγ Apoptotic
Dividing cell
Media
macrophage Efferocytosis

TH1 cell

Adventitia

Fig. 2 | The progression of atherosclerosis reflects and interplay between cytokine IL-10; and regulatory T (Treg) cells can secrete TGFβ. These mediators
factors that promote or mitigate atherogenesis. This diagram summarizes can antagonize cellular proliferation, promote extracellular matrix synthesis
results from experimental studies in mice, and observations on human and quell inflammation. Mononuclear phagocytes can engulf dying or dead
atherosclerotic plaques. Pathways thought to promote lesion formation cells that arise through apoptosis, through a process known as efferocytosis.
(factors in red) are shown on the left, and mechanisms that may moderate Inefficient efferocytosis favours the accumulation of debris from dead or
atherogenesis (factors in blue) are on the right. Smooth muscle cells and dying cells, and promotes formation of the central lipid core of the
macrophages can proliferate as the intimal lesion grows. PDGF promotes the atherosclerotic plaque. B2 lymphocytes secrete mediators (such as BAFF, a
migration and replication of smooth muscle cells, and then production of member of the TNF family) that can aggravate atherogenesis. This diagram
extracellular matrix. All cells in the atheromatous plaque can secrete shows only a subset of the mediators that have been implicated in promoting or
cytokines, examples of which include IL-1, TNF and M-CSF (also known as CSF1). antagonizing aspects of atherogenesis. Current research suggests an ongoing
Activated T-helper 1 (TH1) lymphocytes produce IFNγ, which can stimulate struggle between proliferation and death, involving proinflammatory,
mononuclear phagocytes and aggravate atherosclerosis. Other types of cell anti-inflammatory and proresolving mediators—generally through a
produce countervailing mediators. B1 lymphocytes can secrete IgM natural prolonged course of many years in the evolution of the human atherosclerotic
antibody; T-helper 2 (TH2) lymphocytes produce the anti-inflammatory plaque.

guidelines (gauged by an hsCRP above 2 mg l−1). The participants had a with host defence. But not all anti-inflammatory interventions have
baseline LDL of approximately 2 mM (81 mg dl−1). The anti-inflammatory yielded clinical benefit66. For example, a trial with low-dose weekly
therapy yielded a 15% relative reduction in risk for recurrent myocar- methotrexate did not improve cardiovascular outcomes nor did it exert
dial infarction, stroke or cardiac death. In an on-treatment analysis, an anti-inflammatory effect in the population studied67.
individuals who responded to the IL-1β neutralization by achieving a Obesity and its attendant dysmetabolism, often manifested by insulin
greater-than-median reduction in hsCRP had a 26% reduction in the resistance and diabetes, now drives an increasing proportion of cardio-
primary end point, and a decrease in all-cause mortality. As IL-1β par- vascular disease risk worldwide. Adipose tissue abounds with inflam-
ticipates in host defences, it was not surprising that CANTOS showed a matory cells, produces proinflammatory mediators and inflammation
small but statistically significant increase in infections (including fatal contributes mechanistically to the link between obesity, insulin resist-
infections) in patients randomized to canakinumab. A highly signifi- ance and atherosclerotic risk68. Moreover, environmental factors, such
cant reduction in incident and fatal lung cancer noted in exploratory as air pollution, noise, disturbed sleep and other stressors have gained
analyses counterbalanced this risk of infection63. increasing recognition as contributors to the risk of atherosclerotic
The natural product colchicine has served as an anti-inflammatory events in part thorough the activation of inflammatory pathways20,21.
therapy for many years, and its use has become standard-of-care in An additional aspect of risk allied with inflammation has recently
the treatment of pericarditis. Two recent large-scale outcome trials emerged. With age, we accumulate somatic mutations in haematopoi-
have indicated efficacy in reducing recurrent cardiovascular events etic stem cells in the bone marrow in genes that drive the development
after the development of acute coronary syndromes. The ‘Colchicine of acute leukaemia69,70. Investigators seeking the origins of leukaemia
Cardiovascular Outcomes Trial’ (COLCOT) showed a 23% reduction in found that individuals who are apparently well and do not have haema-
the composite primary end point, driven primarily by fewer revasculari- tological malignancies can generate clones of leukocytes that circulate
zations in patients treated in the early phase after developing an acute in peripheral blood and that bear mutations in a handful of the known
coronary syndrome (4–30 days)64. The incidence of pneumonia more driver genes for leukaemia. The prevalence of this condition in individu-
than doubled in the group who were treated with colchicine. The ‘Low als aged 70 exceeds 10%, and this burden increases with further ageing.
Dose Colchicine 2’ (LoDoCo2) study administered colchicine (5 mg As the population ages, the number of individuals who bear these clones
d−1) to individuals with stable coronary artery disease, and reported will increase. As expected, those who have this condition—known as
a reduction in recurrent events similar to that seen in the COLCOT65. clonal haematopoiesis of indeterminate potential (CHIP)—have a risk
These recent large-scale clinical outcome trials have bolstered the of developing acute leukaemia of more than tenfold that of unaffected
clinical applicability of decades of fundamental research into inflam- individuals. However, the increase in total mortality in individuals
matory pathways in the pathogenesis of atherosclerosis. The increase in with CHIP by far exceeds that attributable to transformation to acute
infections noted with canakinumab and colchicine indicate an opportu- leukaemia.
nity for the refinement of anti-inflammatory therapy for atherosclerosis Cardiovascular disease accounts for this gap in mortality71. Fully
that retains efficacy in limiting adverse outcomes, while interfering less adjusted for all traditional risk factors, CHIP confers a risk for

Nature | Vol 592 | 22 April 2021 | 527

Review
myocardial infarction and stroke equal to or greater than previously subset of monocytes gives rise to lesional macrophages87,88. Recent
recognized contributors to cardiovascular risk, save for age itself. The lineage-tracking experiments support a smooth muscle origin of many
qualification ‘indeterminant potential’ reflects the lack of symptoms foam cells in mouse atheromata89. The cooperation between these
or consistent laboratory findings in bearers of these mutant clones cellular constituents of innate and adaptive immunity stimulate the
of peripheral leukocytes and our current inability to predict which production of proinflammatory cytokines that sustain and amplify
individuals with CHIP will develop leukaemia or cardiovascular dis- the local inflammatory response.
ease. Indeed, many carriers of CHIP will never know that they have
this condition72,73. Inexorability of atheroma progression
The connection between CHIP and inflammation arises from exper- Many have considered atherosclerosis an inevitable ‘degenerative’
imental work that demonstrated that mice engineered to simulate process that progresses continuously over time (Fig. 2), but current
CHIP with mutations in Dnmt3a or Tet2 have accelerated atheroscle- evidence supports a much more dynamic and discontinuous evolu-
rosis, and demonstrate increased activity of the NLRP inflammasome tion of atheromata90–92. Episodes of systemic inflammation or regional
IL-1β–IL-6 proinflammatory pathway71,74. Mice with myeloid cells that inflammation that is remote from the atherosclerotic plaque itself can
bear Jak2V617F (another mutation associated with CHIP) show activa- provoke ‘crises’ in the evolution of the plaque, and stimulate a round of
tion of the AIM2 inflammasome75. Concentrations of the marker of inflammatory activation that can promote cell migration, proliferation,
inflammation hsCRP do not rise consistently in individuals with CHIP. lesion progression and complication (Fig. 2). The concept of ‘trained
This observation indicates the existence of aspects of inflammation immunity’ raises the possibility that successive encounters with irrita-
that mark augmented cardiovascular risk but that are not captured tive stimuli elicit exaggerated responses93,94. Arterial smooth muscle
by measurements of hsCRP. cells—which normally reside quiescent in the middle layer of the artery
In sum, we have witnessed a considerable change in the ranking of risk (the tunica media)—enter the intimal layer, where they can prolifer-
factors for atherosclerosis: some factors have receded in importance ate and may undergo metaplasia to become macrophage-like cells89.
and relevance given current therapies, and others have rapidly expanded, The application of advanced cell-sorting techniques and of single-cell
in part owing to socioeconomic and behavioural factors2 (Table 1). RNA sequencing has disclosed a high degree of heterogeneity in the
cellular contributors to atherosclerosis95–97. Sorting out the functional
consequences of the newly identified cell types that participate in
Revised concepts of atherogenesis atherogenesis will require considerable work, and holds promise for
Oxidized LDL and the initiation of lesions advancing the identification of new therapeutic targets.
Most reviews of the mechanisms of atherosclerosis posit a pivotal role Atherosclerosis may not proceed continuously, but rather in phases
for oxidized LDL as the prime mover of this disease (Fig. 1). Although of relative quiescence punctuated by periods of rapid growth. Emerging
LDL participates causally in atherogenesis, despite a large body of evidence points to haematopoiesis as a key contributor to lesion evolu-
animal research, scant evidence actually supports the causal role of oxi- tion and as a link between regional inflammation, environmental stimuli
dized LDL in humans. A variety of clinical intervention trials with anti- and atherogenesis98. Mental stress, sleep disturbance and remote injury
oxidant vitamins or one of a highly effective lipophilic antioxidant drug or infection can stimulate haematopoiesis in the bone marrow, fur-
have not reduced atherosclerotic events. Native, rather than oxidized, nishing leukocytes that can populate the plaque98,99. Extramedullary
LDL appears to drive the adaptive immune response in mice76. LDL haematopoiesis, as well as the mobilization of preformed pools of
per se appears to be a relatively weak stimulus to innate immune activa- leukocytes in the spleen, furnish further leukocytes that can home in
tion. Recent work supports the participation of cavelolin-1-dependent on atheromata under stress situations. Indeed, the work that identified
LDL transcytosis through the endothelium in experimental athero- CHIP as a risk factor for atherosclerosis underscores the link between
sclerosis77. ALK1 and SRB1 can also participate in LDL transcytosis78,79. atherosclerosis and haematopoiesis. These observations have opened
Although these results emphasize the causality of LDL in atherogenesis, a window onto the pathogenesis of atherosclerosis, and provide a link
they do not invoke the participation of oxidized LDL in this process. between oncogenesis and atherogenesis that was unsuspected only
How LDL causes atherosclerosis is not understood completely, and a few years ago. The death of mononuclear phagocytes in the lesion,
we should seek explanations beyond the oxidation hypothesis. When and their ineffective clearance (defective efferocytosis), promote the
oxidized lipid binds to plasminogen, they can activate fibrinolysis80. formation of the lipid or necrotic core of the atherosclerotic lesion100.
Thus, oxidized lipids may promote atherogenesis but boost thromboly- Lesion progression can occur silently over many decades. Indeed,
sis—an opposing effect that could contribute to the net lack of benefit many young or middle-aged individuals contain asymptomatic and
in trials of anti-oxidant strategies81. LDL that aggregates in the intima in subclinical atherosclerotic lesions, as shown by autopsy and imaging
association with proteoglycan, or adaptive immune responses to native studies101–103.
LDL, provide alternative mechanisms through which this lipoprotein
promotes atherogenesis. Macrophages in plaques take up aggregated ‘Vulnerable plaques’
LDL82, and the LDL receptor-related protein can mediate the update of The acute events such as myocardial infarctions and ischaemic strokes
aggregated LDL by intimal smooth muscle cells83. that complicate atherosclerosis arise from thrombosis or formation
Regardless of the initial trigger or triggers, experimental and human of blood clots (Fig. 3), a physical disruption of atherosclerotic plaques
observations agree that the recruitment of blood leukocytes mediated provokes most acute thromboses. The so-called ‘vulnerable plaque’
by activation of endothelial cells that line the arterial lumen occurs early has received considerable attention104,105. A fracture of the fibrous cap
in lesion formation (Fig. 1). The resting endothelium resists attachment of the plaque (which overlies the necrotic core) exposes blood and
of blood leukocytes. In an atherogenic environment, endothelial cells its coagulation proteins to thrombogenic substances (such as tissue
can express leukocyte adhesion molecules that mediate the rolling factor) within the plaque, triggering acute thrombosis106 (Fig. 3a). The
and firm attachment of white blood cells to the intimal surface (Fig. 1). fibrous cap owes its tensile strength largely to interstitial collagen.
Chemokines direct the migration of the adherent leukocytes into the The thinning of the fibrous cap arises from an inflammation-related
arterial intima. Mononuclear phagocytes can proliferate within the decrease in collagen synthesis and augmented degradation owing
intimal layer (the site of lesion initiation)84. These cells engulf lipids to overexpression of collagenases by inflammatory cells107. Autopsy
and become foam cells, the hallmark of atherosclerotic lesions. studies104 have implicated rupture of the fibrous cap as the cause of
T lymphocytes, which drive the adaptive immune response, interact the majority of fatal acute coronary syndromes, stimulating focus on
with innate immune cells within the intima85,86. A proinflammatory the thin-capped fibroatheroma as a possible culprit. Yet, post-mortem

528 | Nature | Vol 592 | 22 April 2021

 a b
Thrombus

Fibrous cap

NETs

Lipid core

c d

Buried cap

Fig. 3 | Thrombotic complications of atherosclerosis and evolution of the reactor on the intimal surface. NETs can propagate inflammation and
atherosclerotic plaque. a, Plaque rupture. This involves a fracture or fissure thrombosis. c, d, Plaques can heal, which augments the bulk of the plaque and
of the fibrous cap that overlies the lipid core of the plaque. This physical promotes the formation of flow-limiting stenosis in previously disrupted
disruption permits contact of blood coagulation factors with thrombogenic arteries. During thrombosis, platelets release PDGF and TGFβ, which promote
material (principally the potent procoagulant tissue factor) within the plaque. the synthesis of extracellular matrix proteins that contribute to fibrosis and
The ensuing thrombosis can obstruct blood flow and lead to cardiac ischaemia. plaque growth. c, Ruptured plaques that have healed often show morphologic
This mechanism accounts for about two-thirds of acute myocardial infarction, evidence of the rupture underneath a layer of more recently deposited
but appears to be waning; current preventive therapies lead to a reduction in extracellular matrix (a ‘buried’ fibrous cap). d, Plaques can also grow through
accumulation of lipid within plaques and to the reinforcement of the fibrous incorporation of a thrombus. Lesions can also calcify (not shown), in part owing
cap. b, Superficial erosion. This cause of coronary artery clot formation to cell-derived microvesicles that can nucleate this process. Regions of spotty
involves a sloughing or desquamation of the endothelial monolayer. calcification imaged by computed tomography correlate with an increased risk
Granulocytes trapped in the plaque or adherent to the intimal basement of a thrombotic event. In contrast to smaller deposits of calcium, macroscopic
membrane can form neutrophil extracellular traps (NETs). NETs are strands of plates of calcium may stabilize plaques from mechanical disruption (rather
nuclear DNA that have unwound, present various neutrophil granular proteins than create inhomogeneity in stresses that promotes thrombotic
and bear other proteins that they bind from the blood, forming a solid-state complications due to plaque disruption).

studies such as these lack a denominator for how many lesions with the an imperative to encourage healthy lifestyle choices and a public
characteristics attributed to vulnerability do not cause acute throm- environment that supports them by encouraging physical activity,
botic complications. Recent in vivo imaging studies in humans have discouraging sugar-sweetened beverages that contribute to the obe-
furnished this missing information, and have shown that plaques that sity epidemic and continuing to abate tobacco use. Indeed, a healthy
thin-capped plaques seldom cause clinical events108–110. Thus, current lifestyle can mitigate—in part—genetic risk for atherosclerotic events120.
evidence shows that 'vulnerable plaque' is a misnomer111,112. Such public health measures include promoting pedestrian zones,
In an era of intense lipid lowering, plaques of the classical vulnerable bicycle paths and playgrounds, and providing healthy foods in schools.
morphology are on the wane113. Another mechanism of plaque disruption Second, the drug therapy for atherosclerosis has advanced not only
(known as superficial erosion) currently appears to be on the rise and in the availability of agents that address an expanded array of causal
probably has a distinct pathophysiology114–116 (Fig. 3b). This trigger to targets, but also by harnessing biomarkers and genetic information to
coronary artery stenosis does not involve fissure or rupture of the fibrous deploy therapy more precisely62,121. The application of evidence-based
cap of the plaque, but rather a discontinuity in the intimal endothelial interventions that address the established risk factors provides a firm
lining. The application of an intravascular imaging modality known as foundation for future advances.
optical coherence tomography enables identification of plaque rupture, The introduction of statins (agents that are highly effective in lower-
and has led to the development of criteria for the diagnosis of probable ing LDL and that also quell inflammation independently of effects on
or definite erosion in individuals with acute coronary syndromes117. The lipids) has revolutionized the prevention and treatment of atheroscle-
mechanisms of erosion involve endothelial injury, the participation of rosis, a topic that has been well-reviewed in previous publications122.
polymorphonuclear leukocytes and neutrophil extracellular traps as a The availability of an inhibitor of intestinal absorption of cholesterol
local contributor to thrombus formation and propagation114,118,119. that targets the Niemann Pick C 1-like protein 1 yields further reduc-
tions in LDL and a further decrement in cardiovascular events123. The
identification of mutations in PCSK9 as a cause of autosomal dominant
Clinical implications hypercholesterolaemia led rapidly to the development of therapeutic
From a therapeutic perspective, we have reason for optimism in agents that can improve further cardiovascular outcomes in patients
addressing the growing burden of atherosclerotic risk. First, we face who were already treated with statins17,18,124. PCSK9 conducts the LDL

Nature | Vol 592 | 22 April 2021 | 529

Review
receptor to the lysosome, where it undergoes proteolytic degradation. Table 1 | Changing views on atherosclerosis
Inhibition of PCSK9 favours the recycling of undegraded LDL recep-
tors to the cell surface, where they can capture and internalize LDL, Past Present
and thus lowers the plasma concentration of this highly atherogenic Atherosclerosis predominantly Developing countries now bear the
lipoprotein124. The introduction and recent approval in the USA of affects developed countries greatest burden of atherosclerosis
bempedoic acid (an inhibitor of ATP citrate lyase, which acts upstream Coronary thrombosis affects Women, younger individuals, individuals
of hydroxymethyglutaryl co-enzyme A (the target of statins)) primarily middle-aged white men from a range of ethnic backgrounds and
the very old suffer increasingly from acute
adds to the range of non-statin agents that lower LDL125,126. The avail- coronary syndromes
ability of inclisiran (a small interfering RNA that limits the production
Atherosclerosis is a lipid storage Inflammation links dyslipidaemia and other
of PCSK9) provides a notably long duration of action, and could be disease risk factors to atherogenesis
administered twice a year or even annually127. Large-scale clinical trials
Oxidized LDL drives Native or aggregated LDL drives
in progress will evaluate the ability of these newer LDL-lowering treat- atherosclerosis atherogenesis
ments to improve cardiovascular outcomes. An antisense RNA agent HDL cholesterol protects against TGRL participate causally in
has entered clinical investigation that targets lipoprotein(a), the highly atherosclerosis atherosclerosis
atherogenic cousin of LDL128. Treatment of elevated lipoprotein(a) Thin-capped fibroatheromata are The ‘vulnerable plaque’ is a misnomer;
(which is often familial) has proven an enduring problem in preven- vulnerable plaques superficial erosion is an increasing cause of
tive cardiology129. arterial thrombosis
Beyond LDL, a cardiovascular end-point study is evaluating a selective Atherosclerosis is an inevitable, Atherosclerosis evolves episodically,
PPARα agonist in individuals with high triglycerides and low HDL130. steady and degenerative can regress, and lifestyle and medical
accompaniment to ageing measures can modulate the process
The ‘Reduction of Cardiovascular Events with Icosapent Ethyl–Inter-
vention Trial’ revealed that prescription-grade eicosapentaenoic acid
can reduce events substantially in individuals with hypertriglyceridae-
mia131,132. Part of this benefit (but probably not all of it) results from a non-coding RNAs in atherosclerosis has improved. MicroRNAs and
lowering of blood triglycerides; some of the benefit may accrue from long non-coding RNAs alter the transcription of genes implicated in
an anti-inflammatory action133. atherosclerosis144,145. These advances will doubtless lead to therapies
Recent studies targeting IL-1β (CANTOS) and trials with colchi- to address the unacceptable burden of risk that persists in spite of
cine (COLCOT and LoDoCO2) have demonstrated the ability of current interventions.
anti-inflammatory therapies that do not lower atherogenic lipids to Transforming these scientific advances in therapies has required
reduce cardiovascular events in patients who are already receiving a large-scale clinical investigations, which—because of the success of
full standard-of-care regimen, including high-dose statins62. Beyond standard-of-care therapies—have required increasing ingenuity and
affirming the inflammation hypothesis of atherosclerosis, these studies investment. Placebo-controlled, randomized clinical trials remain the
identify colchicine as a readily actionable anti-inflammatory therapeu- most reliable tool for validating the application to humans of thera-
tic agent for atherosclerosis. pies derived from laboratory discoveries. However, we also need to
There is currently a revolution at the conjunction of diabetes and embrace targeting segments of the larger population of patients to
cardiovascular disease. A ‘glucocentric’ view of diabetic complica- enrich those enrolled in clinical trials for enhanced risk and respon-
tions has long prevailed134. Although microvascular complications siveness to specific interventions. This approach has revolutionized
(such as retinopathy, nephropathy and neuropathy) do respond to the management of cancer, but has barely begun in the cardiovascular
glucose lowering, the heightened atherosclerotic risk in people with arena. The application of polygenic risk scores may identify young
diabetes had—until recently—proven intractable to traditional hypo- people who may benefit particularly from early preventive treatments.
glycaemic interventions. Recent studies with SGLT2 inhibitors and On the one hand, we are witnessing a globalization of cardiovascular
GLP1 receptor agonists indicate that the macrovascular complications disease risk that has increased the overall burden of atherosclerotic
of diabetes involve mechanisms beyond glucose lowering135–141. These disease. On the other hand, progress in laboratory and clinical inves-
late-generation agents promise to make inroads in the prevention of tigation promises to provide us with tools to confront this global epi-
cardiovascular complications of diabetes (including myocardial infarc- demic. Ultimately, making inroads in the control of atherosclerosis
tion, stroke, heart failure, renal disease and premature death). The will require a multidisciplinary partnership of public health measures,
success of these drugs in forestalling cardiovascular events empha- applied behavioural psychology, risk factor control, consistent applica-
sizes protective mechanisms beyond those of glucose lowering; the tion of existing therapies, and the development and validation of new
investigation of these mechanisms promises to open avenues in the therapeutic approaches.
understanding and treatment of atherosclerosis and heart failure, a
common complication of coronary artery disease.
1. Gaziano, T. A., Prabhakaran, D. & Gaziano, J. M. in Braunwald’s Heart Disease (eds Zipes, D.
P. et al.) 1–18 (Saunders, 2018).
2. Dai, H. et al. Global, regional, and national burden of ischemic heart disease and its
Atherosclerosis is a moving target attributable risk factors, 1990-2017: results from the global Burden of Disease Study
A conjunction of fundamental research and clinical investigations 2017. Eur. Heart J. Qual. Care Clin. Outcomes, https://doi.org/10.1093/ehjqcco/qcaa076
(2020).
has markedly altered traditional concepts of atherosclerosis, and has 3. Libby, P. et al. Atherosclerosis. Nat. Rev. Dis. Primers 5, 56 (2019).
informed improvements in our ability to manage atherosclerotic risk 4. Virani, S. S. et al. Heart disease and stroke statistics—2021 update: a report from the
(Table 1). At the same time, the clinical profile of patients with athero- American Heart Association. Circulation 143, e254–e743 (2021).
5. Arora, S. et al. Twenty year trends and sex differences in young adults hospitalized with
sclerosis has evolved considerably from that of classical cohort studies acute myocardial infarction. Circulation 139, 1047–1056 (2019).
that have long furnished the basis of our thinking about this disease. 6. Towfighi, A., Markovic, D. & Ovbiagele, B. National gender-specific trends in myocardial
Research advances have arisen from improvements in human genetics infarction hospitalization rates among patients aged 35 to 64 years. Am. J. Cardiol. 108,
1102–1107 (2011).
studies enabled by next-generation sequencing and other technologi- 7. Blüher, M. Obesity: global epidemiology and pathogenesis. Nat. Rev. Endocrinol. 15,
cal innovations (including bulk and single-cell RNA sequencing), and 288–298 (2019).
the ever-evolving toolkit for genetic manipulation of mice including 8. Roth, G. A. et al. Global burden of cardiovascular diseases and risk factors, 1990–2019:
update from the GBD 2019 study. J. Am. Coll. Cardiol. 76, 2982–3021 (2020).
gene-editing and induced pluripotential stem cell methodology142,143. This compilation provides recent data regarding cardiovascular risk factors in various
Beyond DNA and mRNA analyses, understanding the functions of regions of the world, and their import for cardiovascular diseases.

530 | Nature | Vol 592 | 22 April 2021

9. Després, J.-P. & Lemieux, I. Abdominal obesity and metabolic syndrome. Nature 444, This genome-wide association study pointed to lipoprotein(a) as causal for aortic
881–887 (2006). stenosis, which is a common concomitant of atherosclerosis.
10. Nordestgaard, B. G. & Varbo, A. Triglycerides and cardiovascular disease. Lancet 384, 44. Lee, S.-R. et al. LPA gene, ethnicity, and cardiovascular events. Circulation 135, 251–263
626–635 (2014). (2017).
11. Moore, J. X., Chaudhary, N. & Akinyemiju, T. Metabolic syndrome prevalence by race/ 45. Tsimikas, S. Potential causality and emerging medical therapies for lipoprotein(a)
ethnicity and sex in the United States, national health and nutrition examination survey, and its associated oxidized phospholipids in calcific aortic valve stenosis. Circ. Res. 124,
1988–2012. Prev. Chronic Dis. 14, E24 (2017). 405–415 (2019).
12. Ference, B. A. et al. Low-density lipoproteins cause atherosclerotic cardiovascular 46. Libby, P. & Hansson, G. K. From focal lipid storage to systemic inflammation: JACC review
disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus topic of the week. J. Am. Coll. Cardiol. 74, 1594–1607 (2019).
statement from the European Atherosclerosis Society Consensus Panel. Eur. Heart J. 38, This review provides an overview of various theories of atherogenesis, culminating in a
2459–2472 (2017). portrayal of the current view that posits a synthesis that combines elements of many of
13. Borén, J. et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease: the previous concepts.
pathophysiological, genetic, and therapeutic insights: a consensus statement from the 47. Xiao, L. & Harrison, D. G. Inflammation in hypertension. Can. J. Cardiol. 36, 635–647
European Atherosclerosis Society Consensus Panel. Eur. Heart J. 41, 2313–2330 (2020). (2020).
14. Goldstein, J. L. & Brown, M. S. A century of cholesterol and coronaries: from plaques to 48. Ridker, P. M., Koenig, W., Kastelein, J. J., Mach, F. & Lüscher, T. F. Has the time finally come
genes to statins. Cell 161, 161–172 (2015). to measure hsCRP universally in primary and secondary cardiovascular prevention? Eur.
A review of the involvement of LDL in atherosclerosis, which represents one of the Heart J. 39, 4109–4111 (2018).
major advances in cardiovascular science in the past century. 49. Ridker, P. M. A test in context: high-sensitivity C-reactive protein. J. Am. Coll. Cardiol. 67,
15. Domanski, M. J. et al. Time course of LDL cholesterol exposure and cardiovascular 712–723 (2016).
disease event risk. J. Am. Coll. Cardiol. 76, 1507–1516 (2020). 50. Ketelhuth, D. F. J. & Hansson, G. K. Adaptive response of T and B cells in atherosclerosis.
16. Ridker, P. M. How common is residual inflammatory risk? Circ. Res. 120, 617–619 (2017). Circ. Res. 118, 668–678 (2016).
17. Sabatine, M. S. et al. Evolocumab and clinical outcomes in patients with cardiovascular 51. Que, X. et al. Oxidized phospholipids are proinflammatory and proatherogenic in
disease. N. Engl. J. Med. 376, 1713–1722 (2017). hypercholesterolaemic mice. Nature 558, 301–306 (2018).
18. Schwartz, G. G. et al. Alirocumab and cardiovascular outcomes after acute coronary 52. Sage, A. P., Tsiantoulas, D., Binder, C. J. & Mallat, Z. The role of B cells in atherosclerosis.
syndrome. N. Engl. J. Med. 379, 2097–2107 (2018). Nat. Rev. Cardiol. 16, 180–196 (2019).
19. Kwok, C. S. et al. Unplanned hospital readmissions after acute myocardial infarction: a 53. Lorenzo, C. et al. ALDH4A1 is an atherosclerosis auto-antigen targeted by protective
nationwide analysis of rates, trends, predictors and causes in the United States between antibodies. Nature 589, 287–292 (2021).
2010 and 2014. Coron. Artery Dis. 31, 354–364 (2020). 54. Ketelhuth, D. F. J. et al. Immunometabolism and atherosclerosis: perspectives and
20. Brook, R. D., Newby, D. E. & Rajagopalan, S. Air pollution and cardiometabolic disease: an clinical significance: a position paper from the Working Group on Atherosclerosis
update and call for clinical trials. Am. J. Hypertens. 31, 1–10 (2018). and Vascular Biology of the European Society of Cardiology. Cardiovasc. Res. 115,
21. Münzel, T. Up in the air: links between the environment and cardiovascular disease. 1385–1392 (2019).
Cardiovasc. Res. 115, e144–e146 (2019). 55. Tabas, I. & Bornfeldt, K. E. Intracellular and intercellular aspects of macrophage
22. Drager, L. F., McEvoy, R. D., Barbe, F., Lorenzi-Filho, G. & Redline, S. Sleep apnea and immunometabolism in atherosclerosis. Circ. Res. 126, 1209–1227 (2020).
cardiovascular disease: lessons from recent trials and need for team science. Circulation 56. Rohlenova, K., Veys, K., Miranda-Santos, I., De Bock, K. & Carmeliet, P. Endothelial cell
136, 1840–1850 (2017). metabolism in health and disease. Trends Cell Biol. 28, 224–236 (2018).
23. Mozaffarian, D. Dietary and policy priorities for cardiovascular disease, diabetes, and 57. Sakash, J. B., Byrne, G. I., Lichtman, A. & Libby, P. Cytokines induce indoleamine
obesity: a comprehensive review. Circulation 133, 187–225 (2016). 2,3-dioxygenase expression in human atheroma-asociated cells: implications for
24. Malik, V. S. & Hu, F. B. Sugar-sweetened beverages and cardiometabolic health: an persistent Chlamydophila pneumoniae infection. Infect. Immun. 70, 3959–3961
update of the evidence. Nutrients 11, 1840 (2019). (2002).
25. Andersson, C., Johnson, A. D., Benjamin, E. J., Levy, D. & Vasan, R. S. 70-year legacy of the 58. Cuffy, M. C. et al. Induction of indoleamine 2,3-dioxygenase in vascular smooth muscle
Framingham Heart Study. Nat. Rev. Cardiol. 16, 687–698 (2019). cells by interferon-γ contributes to medial immunoprivilege. J. Immunol. 179, 5246–5254
26. Aragam, K. G. & Natarajan, P. Polygenic scores to assess atherosclerotic cardiovascular (2007).
disease risk. Circ. Res. 126, 1159–1177 (2020). 59. Baumgartner, R., Forteza, M. J. & Ketelhuth, D. F. J. The interplay between cytokines and
A recent review of the generation and use of polygenic risk scores for atherosclerosis. the kynurenine pathway in inflammation and atherosclerosis. Cytokine 122, 154148 (2019).
27. Elliott, J. et al. Predictive accuracy of a polygenic risk score-enhanced prediction model 60. Hansson, G. K. Inflammation and atherosclerosis: the end of a controversy. Circulation
vs a clinical risk score for coronary artery disease. J. Am. Med. Assoc. 323, 636–645 136, 1875–1877 (2017).
(2020). 61. Baylis, R. A., Gomez, D., Mallat, Z., Pasterkamp, G. & Owens, G. K. The CANTOS trial: one
28. Mosley, J. D. et al. Predictive accuracy of a polygenic risk score compared with a clinical important step for clinical cardiology but a giant leap for vascular biology. Arterioscler.
risk score for incident coronary heart disease. J. Am. Med. Assoc. 323, 627–635 (2020). Thromb. Vasc. Biol. 37, e174–e177 (2017).
29. Siddiqi, H. K., Kiss, D. & Rader, D. HDL-cholesterol and cardiovascular disease: rethinking 62. Ridker, P. M. et al. Antiinflammatory therapy with canakinumab for atherosclerotic
our approach. Curr. Opin. Cardiol. 30, 536–542 (2015). disease. N. Engl. J. Med. 377, 1119–1131 (2017).
30. Thomas, D. G., Wei, Y. & Tall, A. R. Lipid and metabolic syndrome traits in coronary artery Results of the clinical trial that first established the role of inflammation in human
disease: a Mendelian randomization study. J. Lipid Res., https://doi.org/10.1194/jlr. atherosclerosis by showing improved cardiovascular and other outcomes by targeted
P120001000 (2020). neutralization of IL-1β.
31. Nazir, S. et al. Interaction between high-density lipoproteins and inflammation: function 63. Ridker, P. M. et al. Effect of interleukin-1β inhibition with canakinumab on incident lung
matters more than concentration! Adv. Drug Deliv. Rev. 159, 94–119 (2020). cancer in patients with atherosclerosis: exploratory results from a randomised,
32. Shea, S. et al. Cholesterol mass efflux capacity, incident cardiovascular disease, and double-blind, placebo-controlled trial. Lancet 390, 1833–1842 (2017).
progression of carotid plaque. Arterioscler. Thromb. Vasc. Biol. 39, 89–96 (2019). 64. Tardif, J. C. et al. Efficacy and safety of low-dose colchicine after myocardial infarction.
33. Libby, P. Triglycerides on the rise: should we swap seats on the seesaw? Eur. Heart J. 36, N. Engl. J. Med. 381, 2497–2505 (2019).
774–776 (2015). 65. Nidorf, S. M. et al. Colchicine in patients with chronic coronary disease. N. Engl. J. Med.
34. Musunuru, K. & Kathiresan, S. Surprises from genetic analyses of lipid risk factors for 383, 1838–1847 (2020). Two studies64,65
atherosclerosis. Circ. Res. 118, 579–585 (2016). that report the results of large-scale clinical trials, showing that treatment with
35. Voight, B. F. et al. Plasma HDL cholesterol and risk of myocardial infarction: a mendelian colchicine can reduce recurrent events in patients with recent myocardial infarction or
randomisation study. Lancet 380, 572–580 (2012). stable coronary artery disease.
36. Do, R. et al. Common variants associated with plasma triglycerides and risk for coronary 66. Libby, P. & Everett, B. M. Novel antiatherosclerotic therapies. Arterioscler. Thromb. Vasc.
artery disease. Nat. Genet. 45, 1345–1352 (2013). Biol. 39, 538–545 (2019).
37. Khera, A. V. et al. Association of rare and common variation in the lipoprotein lipase gene 67. Ridker, P. M. et al. Low-dose methotrexate for the prevention of atherosclerotic events.
with coronary artery disease. J. Am. Med. Assoc. 317, 937–946 (2017). N. Engl. J. Med. 380, 752–762 (2019).
38. Lewis, G. F., Xiao, C. & Hegele, R. A. Hypertriglyceridemia in the genomic era: a new 68. Ross, R. et al. Waist circumference as a vital sign in clinical practice: a consensus
paradigm. Endocr. Rev. 36, 131–147 (2015). statement from the IAS and ICCR Working Group on Visceral Obesity. Nat. Rev.
39. Varbo, A., Benn, M., Tybjærg-Hansen, A. & Nordestgaard, B. G. Elevated remnant Endocrinol. 16, 177–189 (2020).
cholesterol causes both low-grade inflammation and ischemic heart disease, whereas 69. Jaiswal, S. et al. Age-related clonal hematopoiesis associated with adverse outcomes.
elevated low-density lipoprotein cholesterol causes ischemic heart disease without N. Engl. J. Med. 371, 2488–2498 (2014).
inflammation. Circulation 128, 1298–1309 (2013). 70. Genovese, G. et al. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA
This contribution from the Copenhagen group presents evidence that remnant TGRL sequence. N. Engl. J. Med. 371, 2477–2487 (2014).
produce a greater inflammatory response than does LDL. 71. Jaiswal, S. et al. Clonal hematopoiesis and risk of atherosclerotic cardiovascular disease.
40. Hansen, S. E. J., Madsen, C. M., Varbo, A. & Nordestgaard, B. G. Low-grade inflammation N. Engl. J. Med. 377, 111–121 (2017).
in the association between mild-to-moderate hypertriglyceridemia and risk of acute This paper describes a newly recognized, potent, age-related, independent and
pancreatitis: a study of more than 115000 individuals from the general population. Clin. common risk factor for atherosclerosis.
Chem. 65, 321–332 (2019). 72. Steensma, D. P. et al. Clonal hematopoiesis of indeterminate potential and its distinction
41. Tsimikas, S. A test in context: lipoprotein(a): diagnosis, prognosis, controversies, and from myelodysplastic syndromes. Blood 126, 9–16 (2015).
emerging therapies. J. Am. Coll. Cardiol. 69, 692–711 (2017). 73. Libby, P. et al. Clonal hematopoiesis: crossroads of aging, cardiovascular disease, and
42. Tsimikas, S. & Hall, J. L. Lipoprotein(a) as a potential causal genetic risk factor of cancer: JACC review topic of the week. J. Am. Coll. Cardiol. 74, 567–577 (2019).
cardiovascular disease: a rationale for increased efforts to understand its 74. Fuster, J. J. et al. Clonal hematopoiesis associated with TET2 deficiency accelerates
pathophysiology and develop targeted therapies. J. Am. Coll. Cardiol. 60, 716–721 (2012). atherosclerosis development in mice. Science 355, 842–847 (2017).
43. Thanassoulis, G. et al. Genetic associations with valvular calcification and aortic stenosis. 75. Fidler, T. P. et al. The AIM2 inflammasome exacerbates atherosclerosis in clonal
N. Engl. J. Med. 368, 503–512 (2013). haematopoiesis. Nature, https://doi.org/10.1038/s41586-021-03341-5 (2021).

Nature | Vol 592 | 22 April 2021 | 531

Review
76. Gisterå, A. et al. Low-density lipoprotein-reactive T cells regulate plasma cholesterol 113. Pasterkamp, G., den Ruijter, H. M. & Libby, P. Temporal shifts in clinical presentation and
levels and development of atherosclerosis in humanized hypercholesterolemic mice. underlying mechanisms of atherosclerotic disease. Nat. Rev. Cardiol. 14, 21–29 (2017).
Circulation 138, 2513–2526 (2018). 114. Franck, G. et al. Haemodynamic stress-induced breaches of the arterial intima trigger
77. Ramírez, C. M. et al. Caveolin-1 regulates atherogenesis by attenuating low-density inflammation and drive atherogenesis. Eur. Heart J. 40, 928–937 (2019).
lipoprotein transcytosis and vascular inflammation independently of endothelial nitric 115. Crea, F. & Libby, P. Acute coronary syndromes. Circulation 136, 1155–1166 (2017).
oxide synthase activation. Circulation 140, 225–239 (2019). 116. Libby, P., Pasterkamp, G., Crea, F. & Jang, I. K. Reassessing the mechanisms of acute
78. Kraehling, J. R. et al. Genome-wide RNAi screen reveals ALK1 mediates LDL uptake and coronary syndromes. Circ. Res. 124, 150–160 (2019).
transcytosis in endothelial cells. Nat. Commun. 7, 13516 (2016). 117. Kolte, D., Libby, P. & Jang, I-K. New insights into plaque erosion as a mechanism of
79. Huang, L. et al. SR-B1 drives endothelial cell LDL transcytosis via DOCK4 to promote acute coronary syndromes. J. Am. Med. Assoc., https://doi.org/10.1001/jama.2021.0069
atherosclerosis. Nature 569, 565–569 (2019). (2021).
80. Leibundgut, G. et al. Oxidized phospholipids are present on plasminogen, affect 118. Libby, P. Once more unto the breach: endothelial permeability and atherogenesis. Eur.
fibrinolysis, and increase following acute myocardial infarction. J. Am. Coll. Cardiol. 59, Heart J. 40, 938–940 (2019).
1426–1437 (2012). 119. Molinaro, R. et al. Targeted delivery of protein arginine deiminase-4 inhibitors to limit
81. Libby, P. Counterregulation rules in atherothrombosis. J. Am. Coll. Cardiol. 59, 1438–1440 arterial intimal NETosis and preserve endothelial integrity. Cardiovasc. Res., https://doi.
(2012). org/10.1093/cvr/cvab074 (2012).
82. Kruth, H. S. Sequestration of aggregated low-density lipoproteins by macrophages. 120. Khera, A. V. et al. Genetic risk, adherence to a healthy lifestyle, and coronary disease. N.
Curr. Opin. Lipidol. 13, 483–488 (2002). Engl. J. Med. 375, 2349–2358 (2016).
83. Llorente-Cortes, V., Martinez-Gonzalez, J. & Badimon, L. LDL receptor-related protein This paper presents data that show that healthy behaviours can modify coronary
mediates uptake of aggregated LDL in human vascular smooth muscle cells. Arterioscler. disease risk that is conferred by inherited factors.
Thromb. Vasc. Biol. 20, 1572–1579 (2000). 121. Ridker, P. M. et al. Rosuvastatin to prevent vascular events in men and women with
84. Robbins, C. S. et al. Local proliferation dominates lesional macrophage accumulation in elevated C-reactive protein. N. Engl. J. Med. 359, 2195–2207 (2008).
atherosclerosis. Nat. Med. 19, 1166–1172 (2013). 122. Collins, R. et al. Interpretation of the evidence for the efficacy and safety of statin therapy.
85. Lichtman, A. H., Binder, C. J., Tsimikas, S. & Witztum, J. L. Adaptive immunity in Lancet 388, 2532–2561 (2016).
atherogenesis: new insights and therapeutic approaches. J. Clin. Invest. 123, 27–36, (2013). 123. Cannon, C. P. et al. Ezetimibe added to statin therapy after acute coronary syndromes. N.
86. Gisterå, A. & Hansson, G. K. The immunology of atherosclerosis. Nat. Rev. Nephrol. 13, Engl. J. Med. 372, 2387–2397 (2015).
368–380 (2017). 124. Abifadel, M. et al. Mutations in PCSK9 cause autosomal dominant hypercholesterolemia.
87. Swirski, F. K. et al. Ly-6Chi monocytes dominate hypercholesterolemia-associated Nat. Genet. 34, 154–156 (2003).
monocytosis and give rise to macrophages in atheromata. J. Clin. Invest. 117, 195–205 This paper reports a molecular mechanism that regulates LDL concentrations, and that
(2007). led to the rapid development of a class of lipid-lowering drugs that lower
88. Tacke, F. et al. Monocyte subsets differentially employ CCR2, CCR5, and CX3CR1 to cardiovascular risk.
accumulate within atherosclerotic plaques. J. Clin. Invest. 117, 185–194 (2007). 125. Preiss, D., Tobert, J. A., Hovingh, G. K. & Reith, C. Lipid-modifying agents, from statins to
89. Bennett, M. R., Sinha, S. & Owens, G. K. Vascular smooth muscle cells in atherosclerosis. PCSK9 inhibitors: JACC focus seminar. J. Am. Coll. Cardiol. 75, 1945–1955 (2020).
Circ. Res. 118, 692–702 (2016). 126. Ray, K. K. et al. Safety and efficacy of bempedoic acid to reduce LDL cholesterol.
90. Kubo, T. et al. The dynamic nature of coronary artery lesion morphology assessed by N. Engl. J. Med. 380, 1022–1032 (2019).
serial virtual histology intravascular ultrasound tissue characterization. J. Am. Coll. 127. Ray, K. K. et al. Inclisiran in patients at high cardiovascular risk with elevated LDL
Cardiol. 55, 1590–1597 (2010). cholesterol. N. Engl. J. Med. 376, 1430–1440 (2017).
91. Deliargyris, E. N. Intravascular ultrasound virtual histology derived thin cap fibroatheroma 128. Tsimikas, S. et al. Lipoprotein(a) reduction in persons with cardiovascular disease. N.
now you see it, now you don’t. J. Am. Coll. Cardiol. 55, 1598–1599 (2010). Engl. J. Med. 382, 244–255 (2020).
92. Vergallo, R. & Crea, F. Atherosclerotic plaque healing. N. Engl. J. Med. 383, 846–857 (2020). 129. Libby, P. Lipoprotein (a): a frustrating final frontier in lipid management? JACC Basic
93. Netea, M. G. et al. Trained immunity: a program of innate immune memory in health and Transl. Sci. 1, 428–431 (2016).
disease. Science 352, aaf1098 (2016). 130. Pradhan, A. D. et al. Rationale and design of the pemafibrate to reduce cardiovascular
94. Christ, A. et al. Western diet triggers NLRP3-dependent innate immune reprogramming. outcomes by reducing triglycerides in patients with diabetes (PROMINENT) study. Am.
Cell 172, 162–175 (2018). Heart J. 206, 80–93 (2018).
95. Williams, J. W. et al. Single cell RNA sequencing in atherosclerosis research. Circ. Res. 131. Bhatt, D. L. et al. Cardiovascular risk reduction with icosapent ethyl for
126, 1112–1126 (2020). hypertriglyceridemia. N. Engl. J. Med. 380, 11–22 (2019).
96. Kalluri, A. S. et al. Single-cell analysis of the normal mouse aorta reveals functionally 132. Bhatt, D. L. et al. Effects of icosapent ethyl on total ischemic events: from REDUCE-IT.
distinct endothelial cell populations. Circulation 140, 147–163 (2019). J. Am. Coll. Cardiol. 73, 2791–2802 (2019).
97. Kalucka, J. et al. Single-cell transcriptome atlas of murine endothelial cells. Cell 180, 133. Mason, R. P., Libby, P. & Bhatt, D. L. Emerging mechanisms of cardiovascular protection
764–779 (2020). for the omega-3 fatty acid eicosapentaenoic acid. Arterioscler. Thromb. Vasc. Biol. 40,
98. Schloss, M. J., Swirski, F. K. & Nahrendorf, M. Modifiable cardiovascular risk, 1135–1147 (2020).
hematopoiesis, and innate immunity. Circ. Res. 126, 1242–1259 (2020). 134. Libby, P. & Plutzky, J. Diabetic macrovascular disease: the glucose paradox? Circulation
This paper summarizes work that links lifestyle and behavioural variables with 106, 2760–2763 (2002).
alterations in the bone marrow that modify cardiovascular diseases. 135. Zinman, B. et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2
99. Libby, P., Nahrendorf, M. & Swirski, F. K. Leukocytes link local and systemic inflammation diabetes. N. Engl. J. Med. 373, 2117–2128 (2015).
in ischemic cardiovascular disease: an expanded “cardiovascular continuum”. J. Am. Coll. 136. Perkovic, V. et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy.
Cardiol. 67, 1091–1103 (2016). N. Engl. J. Med. 380, 2295–2306 (2019).
This paper summarizes recent data that add the central nervous system and bone 137. Neuen, B. L. et al. Cardiovascular and renal outcomes with canagliflozin according to
marrow to traditional cardiovascular risk schemes. baseline kidney function. Circulation 138, 1537–1550 (2018).
100. Yurdagul, A., Jr, Doran, A. C., Cai, B., Fredman, G. & Tabas, I. A. Mechanisms and 138. Wiviott, S. D. et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes.
consequences of defective efferocytosis in atherosclerosis. Front. Cardiovasc. Med. 4, 86 N. Engl. J. Med. 380, 347–357 (2019).
(2018). 139. Marso, S. P. et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N. Engl. J.
101. Virmani, R. et al. Coronary artery atherosclerosis revisited in Korean war combat Med. 375, 311–322 (2016).
casualties. Arch. Pathol. Lab. Med. 111, 972–976 (1987). https://www.ncbi.nlm.nih.gov/ 140. Marso, S. P. et al. Semaglutide and cardiovascular outcomes in patients with type 2
entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=3307684&dopt=Abstract diabetes. N. Engl. J. Med. 375, 1834–1844 (2016).
102. Tuzcu, E. M. et al. High prevalence of coronary atherosclerosis in asymptomatic 141. Gerstein, H. C. et al. Dulaglutide and cardiovascular outcomes in type 2 diabetes
teenagers and young adults: evidence from intravascular ultrasound. Circulation 103, (REWIND): a double-blind, randomised placebo-controlled trial. Lancet 394, 121–130
2705–2710 (2001). (2019).
103. Fernández-Friera, L. et al. Vascular inflammation in subclinical atherosclerosis detected 142. Seeger, T., Porteus, M. & Wu, J. C. Genome editing in cardiovascular biology. Circ. Res.
by hybrid PET/MRI. J. Am. Coll. Cardiol. 73, 1371–1382 (2019). 120, 778–780 (2017).
104. Davies, M. J. Stability and instability: two faces of coronary atherosclerosis. The Paul 143. Karakikes, I., Ameen, M., Termglinchan, V. & Wu, J. C. Human induced pluripotent stem
Dudley White Lecture 1995. Circulation 94, 2013–2020 (1996). cell-derived cardiomyocytes. Circ. Res. 117, 80–88 (2015).
105. Waksman, R. et al. The lipid-rich plaque study of vulnerable plaques and vulnerable 144. Feinberg, M. W. & Moore, K. J. MicroRNA regulation of atherosclerosis. Circ. Res. 118,
patients: study design and rationale. Am. Heart J. 192, 98–104 (2017). 703–720 (2016).
106. Libby, P. Mechanisms of acute coronary syndromes and their implications for therapy. 145. Jaé, N. & Dimmeler, S. Noncoding RNAs in vascular diseases. Circ. Res. 126, 1127–1145
N. Engl. J. Med. 368, 2004–2013 (2013). (2020).
107. Libby, P. Collagenases and cracks in the plaque. J. Clin. Invest. 123, 3201–3203 (2013). 146. Owsiany, K. M., Alencar, G. F. & Owens, G. K. Revealing the origins of foam cells in
108. Stone, G. W. et al. A prospective natural-history study of coronary atherosclerosis. N. atherosclerotic lesions. Arterioscler. Thromb. Vasc. Biol. 39, 836–838 (2019).
Engl. J. Med. 364, 226–235 (2011).
109. The SCOT-HEART Investigators. Coronary CT angiography and 5-year risk of myocardial
infarction. N. Engl. J. Med. 379, 924–933 (2018). Acknowledgements P.L. receives funding support from the National Heart, Lung, and Blood
110. Douglas, P. S. et al. Outcomes of anatomical versus functional testing for coronary artery Institute (1R01HL134892), the American Heart Association (18CSA34080399), the RRM
disease. N. Engl. J. Med. 372, 1291–1300 (2015). Charitable Fund and the Simard Fund.
111. Libby, P. & Pasterkamp, G. Requiem for the ‘vulnerable plaque’. Eur. Heart J. 36,
2984–2987 (2015). Author contributions P.L. is entirely responsible for the conception and design of the work; the
112. Arbab-Zadeh, A. & Fuster, V. The myth of the “vulnerable plaque”: transitioning from a acquisition, analysis and interpretation of data; and the initial draft of the work and its revision.
focus on individual lesions to atherosclerotic disease burden for coronary artery disease P.L. approved the previously submitted version and has agreed both to be personally
risk assessment. J. Am. Coll. Cardiol. 65, 846–855 (2015). accountable for his own contributions and to ensure that questions related to the accuracy or

532 | Nature | Vol 592 | 22 April 2021

integrity of any part of the work are appropriately investigated, resolved and the resolution interests of P.L. were reviewed and are managed by Brigham and Women’s Hospital and
documented in the literature. Partners HealthCare, in accordance with their conflict of interest policies.

Additional information
Competing interests P.L. is an unpaid consultant to, or involved in clinical trials for, Amgen, Correspondence and requests for materials should be addressed to P.L.
AstraZeneca, Baim Institute, Beren Therapeutics, Esperion, Therapeutics, Genentech, Kancera, Peer review information Nature thanks Christie Ballantyne, Michael Holmes and Daniel Rader
Kowa Pharmaceuticals, Medimmune, Merck, Norvo Nordisk, Merck, Novartis, Pfizer and for their contribution to the peer review of this work.
Sanofi-Regeneron. P.L. is a member of the scientific advisory boards for Amgen, Corvidia Reprints and permissions information is available at http://www.nature.com/reprints.
Therapeutics, DalCor Pharmaceuticals, Kowa Pharmaceuticals, Olatec Therapeutics, Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
Medimmune, Novartis and XBiotech, Inc. The laboratory of P.L. has received research funding published maps and institutional affiliations.
in the past two years from Novartis. P.L. is on the Board of Directors of XBiotech, Inc. P.L. has a
financial interest in Xbiotech, a company developing therapeutic human antibodies. The © Springer Nature Limited 2021

Nature | Vol 592 | 22 April 2021 | 533