Journal of the American College of Cardiology Vol. 62, No. 14, 2013� 2013 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2013.06.024

EDITORIAL COMMENT

Toward a Genomic Definitionof Aspirin Resistance*

Nanette H. Bishopric, MD

Miami, Florida

Drugs that target platelet function have a prominent role inthe prevention and treatment of cardiovascular occlusivedisease, and acetylsalicylic acid, or aspirin, is likely the mostwidely used. First synthesized by Felix Hoffman in 1897,a chemist at Farbenfabriken Friedrich Bayer & Co., thecompound was initially used for pain relief; its utility for theprevention of myocardial infarction emerged beginning inthe middle of the 20th century (1,2). The 1971 discovery ofits principal mechanism of action, irreversible acetylation ofcyclooxygenase and inhibition of thromboxane A2 produc-tion, won the Nobel Prize for Sir John Vane in 1982 (3).

See page 1267

Since that time, aspirin has proved its value and cost-effectiveness in the secondary prevention of major adversecardiovascular events in a remarkably broad range of patients,including smokers, the elderly, obese, patients with hyper-tension, and those with diabetes, as reviewed recently byHennekens and Baigent (4) and Hennekens and Dalen (5).The data for myocardial infarction are compelling, with a 23%reduction in total mortality when aspirin is given within thefirst 24 h of symptom onset (6). However, despite docu-mented efficacy, the use of aspirin in primary prevention islimited by a lack of data about optimal dosages and individualrisk versus benefit, information that is critical to the use of anydrug in low-risk or intermediate-risk populations.

Amajor concern surrounding the use of aspirin is that somepeople (reports claim anywhere from 2% to >50%) may beresistant to its antiplatelet action. These figures are based onresults with point-of-care clinical testing and laboratory-basedassays measuring platelet aggregation responses and serumthromboxane A2 levels (7,8). Looked at in another way, therelatively high rate of on-treatment myocardial infarctionsupports the existence of “clinical” aspirin resistance. Some of

*Editorials published in the Journal of the American College of Cardiology reflect the

views of the authors and do not necessarily represent the views of JACC or the

American College of Cardiology.

From the Cardiovascular Genetics Laboratory, University of Miami Miller School

of Medicine, Miami, Florida. This work was supported by grant NHLBI R01-

HL71094 from the National Institutes of Health and by funding from the Florida

Heart Research Institute. Dr. Bishopric has reported that she has no relationships

relevant to the contents of this paper to disclose.

this resistance may be due to noncompliance with therapy,malabsorption, up-regulation of alternative aggregation me-chanisms, or competition for the cyclo-oxygenase 1 active siteby coadministered nonsteroidal anti-inflammatory agents(9–13). Genetic variations, especially in cyclo-oxygenase 1and cyclo-oxygenase 2, have been linked to a poor on-therapyplatelet response (14), but the predictive value of these single-nucleotide polymorphisms for patients remains to be deter-mined, and their individual contributions to risk are likely tobe small. Whatever the cause, aspirin-resistant patients areless likely to obtain therapeutic benefit from aspirin, whilestill experiencing its side effects of gastrointestinal distressand bleeding. Badly needed are tools that could determine therisk/reward ratio before starting aspirin in an otherwisehealthy subject and thereby provide a guide to proper dosing.

Recent advances in genomic technology have allowedclinical investigators to look beyond standard functionaltests and protein assays for signs of disease at the level of thegenome. While the deoxyribonucleic acid sequence isessentially fixed and identical in all cell types, the activity ofeach gene is controlled differently in each cell, in a dynamicand reversible manner, through so-called epigenetic mech-anisms (see Lorenzen et al. [15] for an excellent review).Protein-coding messenger ribonucleic acid (mRNA) levelsare a direct readout of gene activity and as such can behighly responsive to environmental cues, disease states, anddrug effects. From a technical point of view, measurement ofmRNAs is rapid, specific, and sensitive; even very raremRNAs can be detected and quantified in specially preservedwhole blood using polymerase chain reaction assays (16).The entire population of mRNAs in a sample (the “tran-scriptome”) can be assayed using deoxyribonucleic acidmicroarray hybridization and increasingly by an emergingtechnology called ribonucleic acid sequencing (RNASeq)(17,18) that can provide both quantitation and ribonucleicacid sequence information. A number of groups haveexploited gene expression profiling to identify biomarkers ofdisease and to obtain clues as to how patients may respond todrugs or to other interventions (19–22).

In a study reported in this issue the Journal, Voora et al.(23) used whole-blood gene expression profiling to identifymRNA patterns that predict the therapeutic response toaspirin, finding that these same patterns are predictive ofmyocardial infarction risk and death in 2 cohorts of patientswith known coronary artery disease, independent of otherrisk factors. Demonstrating just how widely used andeffective aspirin is, it appears from this study that resistanceto aspirin creates a particular hazard for patients withcoronary artery disease.

An important factor in the success of this project is that itbegan with a well-defined biological response. Using a novelplatelet function score, a composite of 4 different functionalassays, the investigators were able to develop a robust readoutof therapeutic response to aspirin and link this to geneexpression patterns of biological relevance at the very begin-ning of the study. They then reproduced these expression

Bishopric JACC Vol. 62, No. 14, 2013Gene Signature of Aspirin Resistance October 1, 2013:1277–9

1278

patterns in 2 additional cohorts, including 1 with coronaryartery disease. This study design provides mechanistic in-sight specifically into the aspirin response, filtering out factorsrelated to baseline platelet function as well as atherosclerosisper se.

Microarrays generate vast amounts of data that can bechallenging to analyze. The challenge is to take that haystackof data and find the few needles that really track with thebiology of interest. Here, the subsequent bioinformaticanalysis involved several steps. First, the investigators iden-tified the entire pool of mRNAs that changed depending onthe presence of an aspirin response and then looked forgroups of mRNAs that changed in tandem (in the samedirection, by the same amount, at the same time) and thatwere therefore presumably regulated by the same upstreamfactors. Next, they looked for clues to those regulatoryfactors by searching for recurring functional motifs in thesesets of genes. This process generated multiple sets ofmRNAs (which the investigators call factors), each unifiedby function and regulation, that could then be tested forregulation in the 2 validation cohorts. The relative expres-sion of each mRNA within the set was totaled to arrive ata quantitative factor score that can be used in multivariateanalyses. Only 1 factor, a group of 60 mRNAs, emergedfrom this analysis to show a robust correlation with aspirinresponse in both validation studies and in the subsequentCATHGEN cohort.

Interestingly, although perhaps not surprisingly, thisgroup of genes, which the investigators term the aspirinresponse signature (ARS), appears to be highly reflective ofplatelet function. Twenty-four of 60 genes were found to bespecific to platelets or megakaryocytes. Six platelet proteinsencoded by mRNAs in the ARS also correlated with aspirinresponse. Hence, the ARS appears to be conveying an im-portant message about platelet biology in the presence ofaspirin. ITGA2B, the principal component, encodes theglycoprotein IIb half of the platelet glycoprotein IIb/IIIacomplex, an important drug target during interventions foracute coronary syndrome (24). However, this gene signatureis not merely a platelet microarray; two-thirds of the ex-pressed transcripts are not obviously of platelet origin. It willbe interesting to learn more about the biological roles ofadditional mRNAs within the ARS, which may give other,unexpected clues to the biology of aspirin resistance.

There are some unavoidable limitations of this well-devisedstudy. One is that it was based on a platelet function scoredevised by the investigators that may not be indicative ofclinically relevant aspirin resistance. Each of the individualmethods for testing platelet function that were combinedto produce this score has limited reproducibility (7). In addi-tion, no single test is sufficient to test the full range ofplatelet functions, as multiple cyclo-oxygenase-dependent andcyclo-oxygenase-independent pathways regulate platelet ad-hesion, aggregation, and granule release (25). Most important,no existing platelet function tests can reliably predict whichpatients will have myocardial infarctions while taking aspirin.

Any selection bias in favor of patients with recent athero-thrombotic activity (e.g., samples collected in the cardiaccatheterization laboratory) may lead to an overestimation ofbasal platelet reactivity and a spurious conclusion of aspirinfailure (8). The study design and multiple validation sets havelargely mitigated these concerns. Importantly, the ARS standsalone as a genome-based test for myocardial infarction risk,independent of the functional platelet tests with which itcorrelates.

Other limitations include the preponderance of Cauca-sians in the CATHGEN test population (26); eventually,the value of the ARS will need to be proven in larger, moreethnically and geographically diverse groups. It will beimportant to determine how newer antiplatelet and antico-agulant agents affect the ARS. Just as important, the utilityof the test in the presence of common comorbidities (e.g.,uncontrolled diabetes, acute inflammation, and plateletdyscrasias) will need to be verified. Overall, however, thisstudy has moved us closer to having a much-needed andhitherto elusive predictor of the clinical efficacy of aspirin.

Reprint requests and correspondence: Dr. Nanette H. Bishopric,University of Miami School of Medicine, RMSB 4054, P. O.Box 016189 (R-189), Miami, Florida 33101. E-mail: n.bishopric@miami.edu.

REFERENCES

1. Craven LL. Experiences with aspirin (acetylsalicylic acid) in thenonspecific prophylaxis of coronary thrombosis. Miss Valley Med J1953;75:38–44.

2. Mehta J, Mehta P, Burger C, Pepine CJ. Platelet aggregation studies incoronary artery disease. Past 4. Effect of aspirin. Atherosclerosis 1978;31:169–75.

3. Vane JR. Inhibition of prostaglandin synthesis as a mechanism ofaction for aspirin-like drugs. Nat New Biol 1971;231:232–5.

4. Hennekens CH, Baigent C. Prevention. Aspirin in primary pre-ventiondgood news and bad news. Nat Rev Cardiol 2012;9:262–3.

5. Hennekens CH, Dalen JE. Aspirin in the treatment and prevention ofcardiovascular disease: past and current perspectives and future direc-tions. Am J Med 2013;126:373–8.

6. ISIS-2 (Second International Study of Infarct Survival) CollaborativeGroup. Randomised trial of intravenous streptokinase, oral aspirin,both, or neither among 17,187 cases of suspected acute myocardialinfarction: ISIS-2. Lancet 1988;2:349–60.

7. Kasmeridis C, Apostolakis S, Lip GY. Aspirin and aspirin resistance incoronary artery disease. Curr Opin Pharmacol 2013;13:242–50.

8. Grinstein J, Cannon CP. Aspirin resistance: current status and role oftailored therapy. Clin Cardiol 2012;35:673–81.

9. Sweeny JM, Gorog DA, Fuster V. Antiplatelet drug “resistance.” Part 1:mechanisms and clinical measurements. Nat Rev Cardiol 2009;6:273–82.

10. Nielsen HL, Kristensen SD, Thygesen SS, et al. Aspirin responseevaluated by the VerifyNow Aspirin System and light transmissionaggregometry. Thromb Res 2008;123:267–73.

11. Crowe B, Abbas S, Meany B, de Haan J, Cahill MR. Detection ofaspirin resistance by PFA-100: prevalence and aspirin compliance inpatients with chronic stable angina. Semin Thromb Hemost 2005;31:420–5.

12. von Pape KW, Strupp G, Bonzel T, Bohner J. Effect of complianceand dosage adaptation of long term aspirin on platelet function withPFA-100 in patients after myocardial infarction. Thromb Haemost2005;94:889–91.

13. Frelinger AL III, Furman MI, Linden MD, et al. Residual arachidonicacid-induced platelet activation via an adenosine diphosphate-dependent

JACC Vol. 62, No. 14, 2013 BishopricOctober 1, 2013:1277–9 Gene Signature of Aspirin Resistance

1279

but cyclooxygenase-1- and cyclooxygenase-2-independent pathway:a 700-patient study of aspirin resistance. Circulation 2006;113:2888–96.

14. Wurtz M, Kristensen SD, Hvas AM, Grove EL. Pharmacogenetics ofthe antiplatelet effect of aspirin. Curr Pharm Des 2012;18:5294–308.

15. Lorenzen JM, Martino F, Thum T. Epigenetic modifications incardiovascular disease. Basic Res Cardiol 2012;107:245.

16. Miller TE, You L, Myerburg RJ, Benke PJ, Bishopric NH. Wholeblood RNA offers a rapid, comprehensive approach to genetic diagnosisof cardiovascular diseases. Genet Med 2007;9:23–33.

17. Perez-Rathke A, Li H, Lussier YA. In: Biocomputing 2013:Proceedings of the Pacific Symposium, ed. Altman RA et al., WorldScientific Publishing Company, Singapore, pp. 159–70.

18. Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool fortranscriptomics. Nat Rev Genet 2009;10:57–63.

19. Tu S, Jiang HW, Liu CX, Zhou SM, Tao SC. Protein microarrays forstudies of drug mechanisms and biomarker discovery in the era of systemsbiology. Curr Pharm Des 2013 Mar 19 [E-pub ahead of print].

20. Paik S. Is gene array testing to be considered routine now? Breast 2011;20 Suppl:S87–91.

21. Barth AS, Hare JM. The potential for the transcriptome to serve asa clinical biomarker for cardiovascular diseases. Circ Res 2006;98:1459–61.

22. PatinoWD,MianOY, Kang JG, et al. Circulating transcriptome revealsmarkers of atherosclerosis. Proc Natl Acad Sci U S A 2005;102:3423–8.

23. Voora D, Cyr D, Lucas J, et al. Aspirin exposure reveals novel genesassociated with platelet function and cardiovascular events. J Am CollCardiol 2013;62:1267–76.

24. Mukherjee D, Eagle KA. Pharmacotherapy of acute coronarysyndrome: the ACUITY trial. Expert Opin Pharmacother 2009;10:369–80.

25. Davi G, Patrono C. Platelet activation and atherothrombosis. N Engl JMed 2007;357:2482–94.

26. Crosslin DR, Shah SH, Nelson SC, et al. Genetic effects in theleukotriene biosynthesis pathway and association with atherosclerosis.Hum Genet 2009;125:217–29.

Key Words: aspirin - biomarkers - genes - myocardial infarction -

platelets.