The aspirin heart disease prevention conundrum

Abstract

The use of daily aspirin in a primary prevention population to reduce cardiovascular disease event risk is a conundrum because the reduction in cardiovascular event risk is balanced by an equal increase in the risk of a serious gastrointestinal bleed. This conundrum can be partially clarified by genetic testing for LPA polymorphisms. A genetic polymorphism in the apolipoprotein (a) component of lipoprotein (a) has been identified that could help identify patients who may benefit the most from daily aspirin in regard to coronary heart disease event reduction. Recent clinical trials have demonstrated that carriers of the LPA variant (rs3798220) have a significantly greater risk for a coronary heart disease event compared to non-carriers. However, in carriers of the LPA variant randomized to aspirin, the incidence of major coronary heart disease events was reduced to the same level as subjects who did not have the LPA variant. This provides a clinically available tool that may allow identification of a subset of primary prevention patients who derive more cardiovascular risk reduction from daily aspirin than the risk of a serious gastrointestinal bleed. This provides an inexpensive treatment options for patients with the LPA variant.

Graphical abstract

The clinical conundrum to recommend daily low dose aspirin is illustrated. The Number Needed to Treat (NNT) in the group with the LPA polymorphism (LPA+) and those without the LPA polymorphism (LPA-), in the WHS and ASPREE trials, reveals a lower NNT in LPA+ subjects in response to daily low dose aspirin treatment.

1. Introduction

A conundrum can be defined as “A confusing and difficult problem or question”. The decision to recommend daily aspirin, in a primary prevention patient population, in order to reduce the risk of a cardiovascular disease (CVD) event, is one such medical conundrum. Recent studies have highlighted the medical conundrum of whether the daily use of aspirin (acetylsalicylic acid) is of cardiovascular benefit or does it increase the risk of a bleed that could have devastating consequences [1]. This question has been addressed by several advisory committees including the United States Preventive Services Task Force (USPSTF) [2]. The public has been alerted to this conundrum by news media and some confusion remains [3,4].

One aspect of this discussion, that we believe has achieved insufficient attention, is the role of genetic polymorphisms in the apolipoprotein (a) [apo(a)] component of lipoprotein (a) [Lp(a)] in identification of populations that may benefit from daily aspirin use. This topic is clinically important for two reasons. First, by the commercial availability of a specific polymorphism in the Lp(a) gene (LPA rs3798220), “genetic aspirin” test for approximately $25-$50 [5]. This polymorphism is expected to be available in whole exon/genome commercial tests as well. It is worth noting that the genetic test need be conducted only once in a patient’s life-time. Second, the relevance of this topic relates to the increased awareness of Lp(a) as an individual and causal cardiovascular disease risk factor and the new series of medicines designed to reduce the cardiovascular events associated with increased blood levels of Lp(a) such as HORIZON and OCEAN(a) [[6], [7], [8]]. However, there is a medicine that may offer benefits for the primary prevention of cardiovascular events in this high-risk group, despite not lowering Lp(a) levels as demonstrated in randomized primary prevention clinical trials. The medication is inexpensive aspirin. The clinical utility of using the LPA genotype test is to identify the primary prevention patient population most likely to benefit from daily aspirin, and help balance the risk of a significant aspirin associated bleed versus the reduction in a CVD event [9].

2. Methods

Three clinical trials have been published that genotyped subjects and tested the effectiveness of daily aspirin in reducing cardiovascular events. The investigations included the Women’s Health Study investigation, in 28,345 female health professionals over the age of 45 years, the Atherosclerosis Risk in Community study (ARIC) (a non-randomized epidemiological cohort) and the Aspirin in Reducing Events in the Elderly (ASPREE) which was conducted in12,815 subjects, greater than 70-year-old, with no CVD history randomized to either aspirin 100 mg/d or placebo for a mean of 4.7 years [10–12]. The results of these investigations in the presence or absence of the LPA genotype was assessed.

3. Aspirin background

Aspirin has been utilized since ancient times for a variety of medical ailments and derived from the willow bark [13]. The famous Greek ancient physician Hippocrates, has been reported to use a salicylic tea to reduce fever around 400 BCE. Aspirin was first synthesized in 1899 as a drug. The mechanism of aspirin has been extensively studied and involves many physiologic properties one of which is its effect on blood clotting as an antiplatelet agent [14]. This was discovered by Dr Lawrence Craven in 1950 and used to reduce the pain and inflammation after a tonsillectomy but he also noted an unusual increase in the number of patients hospitalized for severe bleeding [15]. In the 1960′s it was discovered that aspirin had an anti-adhesive effect on platelets. It was noted that fewer people on aspirin were admitted to hospital with a myocardial infarction and suggested the anti-platelet effects of aspirin may have been responsible [16].

4. Aspirin clinical trials

The evidence for a potential benefit of aspirin to reduce clinical events can be separated into two patient populations, those with no prior history of cardiovascular events (Primary Prevention) and those with a prior history of cardiovascular disease (Secondary Prevention). The Coronary Drug Project hinted that 324 mg aspirin three times a day may reduce coronary events in patients with coronary artery disease (CAD) [17]. In 1980 the Persantine-Aspirin Reinfarction Study (PARIS) hinted that aspirin treatment may reduce CAD risk [18].

A meta-analysis of four aspirin Primary prevention trials in 48,540 subjects randomized to aspirin or placebo illustrates the conundrum. These studies included the Physicians Health Study (163 mg/d aspirin), the UK Doctors study (500 mg/d), the Thrombosis Prevention study (75 mg/d aspirin) and the HOT study (75 mg/d aspirin) [1,[19], [20], [21], [22]]. While aspirin reduced all cardiovascular events 15 %, and myocardial infarction 30 %, it also increased bleeding complications 69 % and importantly, it was noted that GI bleeding risk increased with the age of the participants. It should be noted that bleeding complications were not classified or reported uniformly in the four trials. For the Physicians Health Study bleeds were defined as patients that required transfusion or operation or were fatal; for the UK doctors study it was all fatal and non-fatal bleeds; and for the thrombosis prevention trial (TPT) and the hypertension optimal treatment (HOT) trial. All the bleeds tabulated excluded minor bleeds.

The effect of aspirin on CHD event reduction, versus bleeds, is impacted by the estimated CHD event risk. The higher the estimated CHD risk, the lower the number needed to treat (NNT) to prevent myocardial infarction or a major bleed causing death, transfusion, or operation. For example, for a risk of 0.5 %/ year, 133 subjects would require aspirin treatment to prevent one CHD event while if the risk is higher (1.5 %/year) the NNT is substantially lower at 44. The NNT of 44 is in the same range as the NNT for statin therapy [23]. However, in the lower CHD risk group, the NNT to prevent a MI without cerebral hemorrhage or a major bleed is 256 and drops to 53 in the higher CHD risk group. This illustrates that the NNT with aspirin to prevent a CHD event, without causing a major bleed, is at a high of 256 in the lower CHD event risk group but drops to a NNT of 53 in the higher CHD event risk group and argues that the higher the CHD event risk, the greater the benefit of aspirin when balanced against the bleeding risk. Sanmuganathan and colleagues conclude that the coronary heart disease risk threshold of 15 % over 10 years, (equivalent to 1.5 %/year) is the point at which aspirin treatment is indicated [1].

The US Preventive Services Task Force (USPSTF) guidelines provides some assistance and defines patient risk categories base on 10-year coronary heart disease risk for age ranges of 45–59 years, 60–69 years, and 70–79 years for men and women [2]. The USPSTF guidelines recommend considering information about CVD risk and bleeding risk in assessing the risk/benefit ratio of aspirin use. The USPSTF provided some guidance in 2009 by recommending cut points based on the Framingham Risk Score (FRS) to identify patients who are more, or less likely to derive heart disease prevention benefit from daily ASA treatment. The risk / benefit cut point was defined by age and gender and was established as the FRS level at which the number of heart attacks prevented equaled the number of serious GI bleeds caused by daily ASA treatment. This conundrum continues to be discussed in the medical and lay communities [4,24].

More recently the potential benefit of aspirin has been investigated in three randomized clinical trials, ARRIVE, ASPREE, and ASCEND. The Aspirin to Reduce Risk of Initial Vascular Events (ARRIVE) study investigated the efficacy of 100 mg enteric-coated aspirin daily versus placebo in the reduction of incident myocardial infarction, stroke, and related cardiovascular conditions in people at moderate risk (defined as 10–20 % 10-year coronary heart disease, with the exclusion of patients with diabetes). The event rate was much lower than expected and the role of aspirin in primary prevention among patients at moderate risk could therefore not be addressed [25]. ASCEND is a study of 15,480 participants, with type II diabetes, randomized to 100 mg/d aspirin or placebo [26]. The ASCEND investigators concluded that aspirin use prevented serious vascular events in persons who had diabetes and no evident cardiovascular disease at trial entry, but it also caused major bleeding events. The absolute benefits were largely counterbalanced by the bleeding hazard. The ASPirin in Reducing Events in the Elderly (ASPREE) was a study of 19,114 persons who were 70 years of age or older randomized to 100 mg/d enteric coated aspirin or placebo for a median of 4.7 years [12]. The investigators concluded that the use of low-dose aspirin as a primary prevention strategy in older adults resulted in a significantly higher risk of major hemorrhage and did not result in a significantly lower risk of cardiovascular disease than placebo. Guidance has also recently been provided on the clinical use of aspirin by the 2019 ACC/AHA primary prevention guidelines and the 2024 AHA/ASA primary prevention of stroke guidelines [27,28].

5. LPA genetic polymorphism

Genetic polymorphisms, reflected by single nucleotide polymorphism (SNP) difference, have been linked to CAD risk independent of traditional risk factors. One of the first was a SNP on chromosome 9 with the reference sequence (rs) number rs4977574 [29]. The presence of this SNP conferred an approximate 2-fold risk for CAD, peripheral vascular disease (PVD), and abdominal aneurysm.

Genetic polymorphisms in the apo(a) component of Lp(a) are associated with an increased risk of CAD independent of traditional risk factors. A SNP in the LPA gene has been identified that helps to identify patients who benefit the most from daily aspirin in regard to CHD event reduction and is termed LPA (rs3798220) [30]. This polymorphism is in the protease-like domain of Lp(a) encoding an isoleucine to methionine substitution (Ile4399Met) at amino acid 4399. This polymorphism occurs in approximately 12 percent of Caucasians. In a combined analysis, the odds ratio for CHD is 1.47 per risk allele [31]. The prevalence varies significantly across different ethnic populations which could potentially impact the generalizability of findings related to this polymorphism [32]. The rs379220 polymorphism has a significant relationship with Lp(a) plasma levels. In the Women’s Health Study, it was reported that the mean Lp(a) in non-carriers of the risk allele was 10.0 mg/dl, in the heterozygotes 79.6 mg/dl, and in the homozygotes 153.9 mg/dl (p < 0.0001) [30].

6. LPA aspirin clinical trials

There are three investigations that support the relevance of LPA in identifying a subgroup of patients that benefit from routine aspirin use (Table 1). These studies varied in ethnicity, gender, and age.

Table 1.

Characteristics of three aspirin versus placebo studies reporting results by LPA genotype. LPA+ are subjects who carry the LPA risk allele.

	WHS	ARIC	ASPREE
N	25,131	6752	12,815
N LPA+	921	221	406
LPA+ %	3.70 %	3.30 %	3.20 %
Age (yr)	≥45	45–65	≥70
CHD baseline	No	No	No
Duration (yr)	9.9	7.2	4.7
LPA rs	3798,220	3798,220	3798,220
Aspirin dose	100 mg qod	Daily	100 mg qd

First, in the Women’s Health Study investigation, in 25,131 female health professionals over the age of 45 years, the primary endpoint was negative. However, in this cohort, 3.6 % were heterozygous and 0.06 % homozygous for the rs3798220 risk allele [10]. After 10 years of randomization to aspirin (100 mg every other day) or placebo, there was a significantly greater incidence of a major CVD risk in the carriers of the LPA variant compared to non-carriers (p interaction = 0.048). A post-hoc genetic analysis of 25,131 participants revealed that in the placebo group carriers of the 4399Met allele, compared with noncarriers, had over a two-fold increased risk for major cardiovascular events (hazard ratio [HR] 2.2) and that this increased risk was essentially eliminated by low-dose aspirin treatment. The effect of aspirin on subjects with Lp(a) blood levels greater or <30 mg/dl was also reported. The event rate was greater in the high Lp(a) group (3.6 events per 1000 person-years) compared to the low Lp(a) group (1.9 events per 1000 person-years). This difference was not statistically significant but was hampered by the small number of events in subjects with low Lp(a) levels. All of these risk estimates were essentially unchanged when further adjusted for traditional risk factors, including LDL-C.

Second, the genotype dependent results reported in WHS was initially replicated in the Atherosclerosis Risk in Community study (ARIC). ARIC was not a randomized study of aspirin versus placebo but data was analyzed based on LPA carriers (4399Met) and noncarriers who self-selected to take daily aspirin [11]. In subjects with the LPA variant who self-selected to take aspirin the Hazard ratio was reduced from 1.57 to 0.86. Among self-reported nonusers of aspirin increased risk for CHD among carriers of the 4399Met allele, compared with noncarriers was reported but did not reach statistical significance using a two-sided test. Carriers, compared with noncarriers, were not at increased risk for CHD among self-reported regular users of aspirin. The ARIC analysis results were in the same direction that was reported in the WHS, namely that aspirin treatment ameliorated the risk associated with 4399Met carrier status.

Third, the Aspirin in Reducing Events in the Elderly (ASPREE) was conducted in12,815 subjects, greater than 70-year-old, with no CVD history randomized to either aspirin 100 mg/d or placebo for a mean of 4.7 years [12,33]. Death from any cause occurred in 558 participants in the aspirin group (12.7 events per 1000 person-years) and in 494 participants in the placebo group (11.1 events per 1000 person-years), with a hazard ratio of 1.14 with 95 % confidence interval for the hazard ratio 1.01 to 1.29. In the placebo group, Major Adverse Cardiovascular Events (MACE) occurred, with an interaction observed between rs3798220-C and aspirin allocation (P = 0.049). The rs3798220-C carrier status was associated with increased MACE risk in the placebo group (HR: 1.90, p = 0.018) but not in the aspirin group (HR 0.54; 95 % CI; 0.17–1.70; p = 0.294). In all participants, aspirin reduced MACE by 1.7 events per 1000 person-years and increased clinically significant bleeding by 1.7 events per 1000 person-years.

ASPREE investigators additionally utilized a 43-variant lipoprotein(a) genetic risk score (LPA-GRS) which had been shown to provide modest improvement in risk prediction [34]. The ASPREE investigators reported results comparing the highest versus lowest quintile [12]. In the rs3798220-C and high LPA-GRS subgroups, aspirin reduced MACE by 11.4 and 3.3 events per 1000 person-years respectively, without significantly increased bleeding risk. In the carriers of the LPA variant who were randomized to aspirin, the incidence of major CVD events was reduced to the same level as subjects in the trial who did not have the LPA variant.

The decision to use daily aspirin to reduce the risk of a future cardiovascular event, in the primary prevention patient population, can be difficult and involve balancing the potential cardiovascular benefit against the risk of a serious gastrointestinal bleed associated with routine aspirin use. Three large clinical trials have now helped to clarify this conundrum. Two of the investigations were prospective randomized trials of routine aspirin treatment versus placebo. In a retrospective analysis, investigators utilized a specific LPA genotype (rs3798220) previously shown to be associated with the quantitative Lp(a) phenotype blood level, to explore the role of this LPA polymorphism [30].

Primary prevention subjects with elevated blood Lp(a), or LPA genotype, were shown to have increased risk of a future cardiovascular event, but, in subjects with elevated blood Lp(a) or LPA genotype, the use of routine aspirin treatment reduced the risk to the level of subjects without the LPA genotype [35]. One aspect of Lp(a) that contributes to CVD risk is its similarity to plasminogen and potential thrombotic associated issues [36,37]. Aspirin may exert its beneficial effects in the LPA subgroup through a thrombotic pathway and not related to Lp(a) blood levels [37].

A sense of the clinical relevance of a measurement or treatment can be derived from the concepts of Relative Risk Reduction (RRR), Absolute Risk Reduction (ARR), and Number Needed to Treat (NNT). RRR expresses the relative risk reduction in one group (such as a treatment group) relative to another (such as a placebo group). The ARR is similar but expresses the absolute value difference between the two groups. The NNT is the number of subjects that need to be treated to prevent one clinical event. In six large LDL-C reduction clinical trials, the RRR is −27.2 %, the ARR −3.4 %, and the NNT is 41 [23].

In the case of aspirin and the LPA genotype, both the WHS and ASPREE provide a reflection of the potential utility of using LPA genotype to identify a group of patients who benefit the most from daily aspirin therapy in regard to a reduction in cardiovascular risk. In the WHS the RRR was 5.3 % in the LPA-negative group with an RRR of 55.7 % in the LPA-positive group resulting in an NNT in the LPA-negative group of 833 compared to 38 in the LPA-positive group. Similar findings are reported in the ASPREE trial with an RRR of 21.1 % in the LPA-negative group with a RRR of 80 % in the LPA-positive group resulting in a NNT in the LPA-negative group of 250 compared to 33.8 in the LPA-positive group. (Table 2).

Table 2.

RRR, ARR and NNT in the randomized trials WHS and ASPREE. LPA-Negative are subjects who do not carry the LPA risk allele and LPA-Positive reflect subjects who carry the LPA risk allele.

	WHS	ASPREE
LPA-Negative
AR ASA %	2.13	1.5
AR Placebo %	2.25	1.9
RRR	5.30 %	21.10 %
ARR	0.12 %	0.40 %
NNT	833	250
LPA-Positive
AR ASA %	2.14	0.74
AR Placebo %	4.83	3.7
RRR	55.70 %	80 %
ARR	2.69 %	2.96 %
NNT	37.9	33.8

This genetic perspective merits three considerations. First, and most importantly, genetic associations in one ancestral heritage group sometimes do not hold up for the entire population. A report of the genetic association of SNPs in Lp(a) varying in diverse populations has appeared [38]. Second, increasingly, polygenic risk scores for CAD risk that aggregate the genetic association from multiple genetic variants have begun to appear. Elliott et al. report a polygenic risk score that may be pertinent to larger fractions of population but again this study has limited diversity [39]. However, the American Heart Association recently noted the opportunities and challenges of using polygenic risk scores [40]. Third, enteric coated aspirin is readily available that may decrease associated bleeding risk relative to cardiovascular benefit. Further, clinical trial results are anticipated that will address the issue of Lp(a) blood level reduction in secondary prevention and high-risk primary prevention trials are being designed (Fig. 1).

Fig. 1.

Percent of subjects in the Women’s Health Study and ASPPEE, in the placebo and aspirin (ASA) groups, who experience a cardiovascular event, who did not carry the LPA risk allele (T/T-Blue) compared to those that carried one risk allele (T/C-Red). Subjects randomized to placebo are on the left and subjects randomized to aspirin on the right. Subjects with the LPA risk allele randomized to placebo had greater CVD events than those without the LPA risk allele, and, subjects with the LPA risk allele randomized to aspirin had the same CVD risk as those without the LPA risk allele.

7. Conclusions and relevance

The decision to prescribe daily aspirin, to reduce cardiovascular risk in the primary prevention population, is a conundrum. Daily aspirin treatment may reduce cardiovascular risk but also increases the risk of a serious gastrointestinal bleed. The rs3798220 LPA polymorphism has a significant relationship with Lp(a) plasma levels, and cardiovascular risk. In primary prevention patients who do not carry the rs3798220 risk allele, daily aspirin treatment does not reduce cardiovascular events compared to placebo. However, in those that carry the LPA risk allele, daily aspirin treatment significantly reduces the risk of cardiovascular events down to the same level as subjects who do not carry the LPA risk allele. This has now been reported in three clinical studies in a total of 44,698 subjects. The use of the LPA genotype, provides the clinician with a relatively inexpensive tool with which to identify a subgroup of patients who have increase CVD risk but also significant CVD risk reduction with routine daily low-dose aspirin use. This information may help to clarify the “aspirin conundrum” and guide physicians and patients in the use of aspirin to prevent CV events in the primary prevention population.

Agreement

An ‘Author Agreement’ template could not be identified on the web submission site so I copied the Authors agreement statement from the instruction to authors.

Publishing agreement

Authors will be asked to complete a publishing agreement after acceptance. The corresponding author will receive a link to the online agreement by email. We advise you to read Elsevier's policies related to copyright to learn more about our copyright policies and your, and your employer’s/institution’s, additional rights for subscription and gold open access articles.

CRediT authorship contribution statement

H. Robert Superko: Writing – original draft, Project administration, Formal analysis, Data curation, Conceptualization. John Sninsky: Writing – review & editing, Resources, Investigation, Data curation. Brenda Garrett: Writing – review & editing, Formal analysis, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

H. Robert Superko reports article publishing charges was provided by Cholesterol, Genetics, and Heart Disease Institute 501c3. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.