Circulating Chemokines Accurately Identify Individuals with Clinically

Significant Atherosclerotic Heart Disease

Diego Ardigo * 1, 2

Themistocles Assimes *1

Stephen P. Fortmann 3

Alan S. Go 4

Mark Hlatky 1

Evangelos Hytopoulos 5

Carlos Iribarren 4

Philip S. Tsao 1

Raymond Tabibiazar 1, 5

Thomas Quertermous 1

for the ADVANCE Investigators

1Division of Cardiovascular Medicine, Stanford University, Stanford, CA 94305; 2Department of Internal

Medicine and Biomedical Sciences, Parma University, Parma, Italy; 3Stanford Prevention Research

Center, Stanford University, Stanford, CA 94305, 4Kaiser Division of Research, Kaiser of Northern

California, Oakland, CA 94611; 5Aviir, Inc., Palo Alto, CA 94303

This work was supported by a grant from the Donald W. Reynolds Foundation, Las Vegas, Nevada.

*These authors contributed equally to this work.

Correspondence should be addressed to R.T. or T.Q.

Stanford Medical School, Division of Cardiovascular Medicine

300 Pasteur Drive, Falk CVRC

Stanford, CA, 94305

Tel: 650-723-5012, Fax: 650-725-2178

Emails: rtabibiazar@cvmed.stanford.edu, tomq1@stanford.edu

1

Page 1 of 23Articles in PresS. Physiol Genomics (August 14, 2007). doi:10.1152/physiolgenomics.00104.2007

Copyright © 2007 by the American Physiological Society.

ABSTRACT

Background: Serum inflammatory markers correlate with outcome and response to therapy in subjects

with cardiovascular disease. However, current individual markers lack specificity for the diagnosis of

coronary artery disease. We hypothesize that a multi-marker proteomic approach measuring serum levels

of vascular derived inflammatory biomarkers could reveal a ‘signature of disease’ that can serve as a

highly sensitive and specific diagnostic tool for the presence of coronary atherosclerosis.

Methods: We simultaneously measured serum levels of seven chemokines (CXCL10 (IP10), CCL11

(Eotaxin), CCL3 (MIP1alpha), CCL2 (MCP1), CCL8 (MCP2), CCL7 (MCP3), and CCL13 (MCP4)) in

48 subjects with clinically significant CAD (‘cases’) and 44 healthy controls from the ADVANCE Study.

We applied three different classification algorithms to identify the combination of variables that would

best predict case-control status (logistic regression, linear discriminant analysis and recursive partitioning)

and assessed the diagnostic performance of these models with Receiver Operating Characteristic (ROC)

Curves.

Results: The serum levels of six out the seven chemokines were significantly higher in cases compared

with controls (p <0.05). All three classification algorithms entered 3 chemokines in their final model and

only logistic regression selected clinical variables. Logistic regression produced the highest ROC of the

three classification algorithms (AUC=0.95; SE=0.03), which was markedly better than the AUC for the

logistic regression model of traditional risk factors of CAD without (AUC=0.67; SE=0.06) or with CRP

(AUC=0.68; SE=0.06).

Conclusions: A combination of serum levels of multiple chemokines identifies subjects with clinically

significant atherosclerotic heart disease with a very high degree of accuracy. These results need to be

replicated in larger cross sectional studies and their prognostic value explored in prospective studies.

Multimarker approach utilizing informed biomarkers may ultimately lead to improved screening,

diagnosis, and monitoring of cardiovascular disease.

KEYWORDS: Coronary Artery Disease, Multi-marker, Biomarkers, Atherosclerosis, Protein microarray,

Proteomic, Inflammation, Chemokines.

2

Page 2 of 23

CONDENSED ABSTRACT

Chemokines have been correlated with cardiovascular disease. In a study of 48 subjects with confirmed

coronary heart disease and 44 subjects without known disease, we demonstrate that a multimarker

proteomic approach measuring serum levels of chemokines can distinguish case from control subjects

with a very high degree of accuracy. These results need to be replicated in larger cross sectional studies

and the prognostic value of the multimarker profile to be explored in prospective studies. Similar

approaches may ultimately lead to improved screening, diagnosis, and monitoring of cardiovascular

disease.

3

Page 3 of 23

INTRODUCTION

Atherosclerotic cardiovascular disease (ASCVD) is the primary cause of morbidity and mortality in the

developed world (3, 4). Despite the chronic nature of the disease, ASCVD is generally undiagnosed before

onset of symptoms or complications. The first clinical presentation of more than half of the subjects with

coronary artery disease (CAD) is either myocardial infarction or death (16). This grim reality is at least in

part due to the lack of markers that accurately identify active atherosclerotic disease before complications

occur.

Inflammation has been implicated in all stages of ASCVD and is considered to be one

pathophysiological basis of atherogenesis (10, 19, 24). Inflammation may therefore serve as a potential

marker of the disease process itself. In large epidemiological studies, various serum markers of systemic

inflammation such as C-reactive protein (CRP), fibrinogen, and Interleukin-6 (IL6) have been shown to

predict cardiovascular events and to correlate with response to therapy (22, 23). Although potentially

useful in risk stratification, the current systemic markers of inflammation lack sufficient disease

specificity to be used satisfactory as a screening tool in the diagnosis of CAD(21). The inaccuracy of

current markers may reflect the fact that they are neither derived primarily from the vascular wall nor

produced primarily by cells involved in the vascular inflammatory process. Furthermore, they may signal

inflammation in a number of different organs and tissues, which may or may not have direct implications

for the vasculature. Recently, a number of studies have examined several other biomarkers on an

individual basis as potential novel risk factors for ASCVD (12, 25). Although these studies demonstrate

that some of these inflammatory markers (those known to be expressed in the diseased blood vessel) can

predict the onset of clinically significant ASCVD, none of the markers provide clinically meaningful

incremental value over traditional risk factors in predicting CAD complications. However, it is highly

likely that, due to the heterogeneity of the disease phenotype in the population at risk, a single marker may

not provide sufficient biological information for an accurate assessment of vascular damage in the

coronary circulation.

Thus, there remains a critical need to develop non-invasive tests that more accurately detect the

presence and activity of ASCVD and improve our ability to predict and prevent clinical events. In this

proof-of-concept study, we hypothesized that a measure of multiple serum proteins of inflammation

derived from the blood vessel wall can be used to reveal a ‘signature of disease’ that reliably identifies

individuals with CAD. To test our hypothesis, we studied a subset of “cases” with confirmed CAD and

“controls” without history of clinically significant CAD from the ADVANCE (Atherosclerotic Disease,

VAscular FuNction, & GenetiC Epidemiology) study, a population-based case control study with a focus

of uncovering novel genetic determinants of coronary atherosclerosis. Using serum samples collected at

the time of first study clinic visit, we simultaneously measured seven chemokines with a commercially

4

Page 4 of 23

available protein microarray and compared the ability of these measures to identify subjects with clinically

significant CAD to that of traditional cardiovascular risk factors.

METHODS

Source population and Study subjects

The ADVANCE study was approved by the IRBs at both Stanford University and the Kaiser Permanente

of Northern California (KPNC) Division of Research. The source population included adults (age ≥18

years) living in or near the San Francisco Bay Area who were receiving medical care within KPNC, a

population generally representative of the local and statewide insured adult population. (13, 18) Between

October 28, 2001 and December 31, 2003, 3179 subjects from the source population living in or near the

San Francisco Bay Area were recruited into 5 separate groups: (1) subjects with clinically significant CAD

at an age < 45 years for males and < 55 years for females (n = 500), (2) subjects with incident stable

angina at an older age (n = 468), (3) subjects with incident acute myocardial infarction (AMI) at an older

age (n=924), (4) young subjects with no history of CAD (n = 264), and (5) subjects aged 60 to 72 at the

time of their study clinic visit with no history of CAD, ischemic stroke (CVA), or peripheral arterial

disease (PAD) (n =1,023). Extensive details of the eligibility criteria, sampling strategy and the

recruitment statistics for all cohorts are available elsewhere (http://med.stanford.edu/advance/) (11, 15,

29)

For this study, we selected a stratified random sample based on race, age, and gender of 50

subjects from group 3 with incident AMI to serve as our “cases” of and 48 subjects from group 5 to serve

as our healthy “controls”. All cases were in the age range of 60 to 69 years at the time of their qualifying

event and all controls were in the same age range at the time they were identified by the computerized

databases as eligible to participate in 2001(29). All subjects were also of white/European descent and half

in each group were selected to be female.

Clinical measures

Clinical information was derived from a comprehensive questionnaire completed by all participants which

was reviewed and collected at the first study clinic visit. For cases, this clinic visit occurred a median of

3.2 (range 1.9 to 18.8 months) after the qualifying AMI. Thereafter, a targeted physical exam was

performed to document the blood pressure, heart rate, height, weight, and the waist circumference of all

participants. Finally, blood was collected for measurements of various serum proteins including fasting

insulin, fasting glucose, and C-reactive protein (CRP). Plasma concentrations of glucose and insulin were

measured with standard methodologies. CRP was determined by standard high-sensitivity ELISA assay.

5

Page 5 of 23

Hypertension, high cholesterol, and diabetes were defined based on self-report. For cases, each of

these risk factors was considered present only if subjects reported that the risk factor was diagnosed by a

health care provider prior to the time of the qualifying AMI. Smoking was classified as “ever” vs. “never”

relative to the clinic date. “Ever” smokers were subjects who reported smoking cigarettes regularly for at

least 3 months at any time prior to the study visit as well as at least 100 cigarettes throughout their

lifetime.

Protein Microarray hybridization and Data processing

To assess the concentrations of 9 different chemokines CXCL10 (IP10), CCL11 (Eotaxin), CCL3

(MIP1alpha), CCL2 (MCP1), CCL8 (MCP2), CCL7 (MCP3), CCL13 (MCP4), CXCL8 (IL-8), and CCL5

(RANTES), we used a commercially available Schleicher and Schuell protein microspot array (FastQuant

Human Chemokine, S&S Biosciences Inc., Keene, NH, US). This array platform utilizes multiple

monoclonal highly-specific antibodies spotted onto standard microscope slides coated with a 3-D

nitrocellulose surface. The sensitivity and specificity of these markers and correlation with conventional

ELISA has been demonstrated previously (17). Lack of cross-reactivity among these markers has been

established previously (17, 28). Plasma samples were hybridized to protein arrays using manufacturer’s

instructions, followed by addition of a biotinylated secondary antibody and Cy5-streptavidine conjugate.

Resulting fluorescence intensity was measured using an Axon Genepix 4000B microarray scanner in

conjunction with feature extraction software (Array Vision Fast 8.0, S&S Biosciences) to convert the

scanned image into numeric intensities. Absolute concentrations were measured by interpolation of

intensity values with internal standard references run in parallel. Depending on the specific analyte,

FastQuant protein arrays present control variability ranging from 3 to about 15 % and sensitivity from 1 to

10 pg/ml. Accuracy of FastQuant protein arrays are comparable to the correspondent ELISA

determinations (1, 2) with a similar linear range. Detailed supplemental methods and quality control

results for the current study are provided online on publisher’s website, including array reproducibility and

standard curves (Supplemental Material). Numerical raw data were subsequently both analyzed in local

Windows workstations and migrated into an Oracle relational database specifically designed for

microarray data analysis.

RANTES and IL-8 were not considered for further analyses. The RANTES standard curve was

non-sigmoidal and, therefore, did not have a linear portion for calculating concentrations. In both case and

control subjects, most of the IL-8 values were outside the standard curve limits.

Statistical analysis

6

Page 6 of 23

Subjects missing only 1 chemokine value (6 cases and 1 control) were maintained in the analysis by

imputing this missing value using the K-nearest neighbor (KNN) method with the nearest 10 neighbors

(31). Subjects missing 2 or more of the 7 chemokine values (2 cases and 4 controls) were excluded from

further analyses since these methods are not robust with >15% missing values (31). Missing values for

glucose, insulin, and CRP for 2 subjects were also imputed using the KNN method.

We examined relationships between variables in our final dataset independent of case-control

status in several ways. First, we performed a two dimensional hierarchical clustering analysis (2D-HC)

and built heat maps using the open-source software TMev, ver. 3.0 (TM4 suite, The Institute for Genomic

Research, Rockville, MD) (26) with complete linkage and Pearson’s correlation as distance metrics.

Second, we used Multidimensional scaling (MDS) implemented as function “cmdscale” in R language.

Third, we performed a multivariate linear regression analysis to test the marginal effect of each traditional

risk factor of CAD and the use of specific medications on each of the 7 chemokines independent of case-

control status. Finally, we calculated Spearman correlation coefficients between chemokines levels and

the number of days elapsed from the time of the qualifying AMI to the time of the clinic visit in cases.

Detailed methods are provided in Online Supplemental Material.

Differences in clinical characteristics between cases and controls were examined using standard

parametric and non-parametric methods including a t-test for normally distributed continuous variables, a

Wilcoxon test for non-normally distributed variables, and a Chi-square test for all binary variables.

We explored the performance of three different classification algorithms on our data: logistic

regression (LR), linear discriminant analysis (LDA), and recursive partitioning (RP). Three different

algorithms were used to maximize the probability of identifying the combination of clinical and biomarker

variables that would most accurately predict case-control status: For logistic regression, we used a

forward sequential automated model selection technique setting the α (SLENTRY) to 0.10 to select

independent predictors of case-control status. A stepwise approach was used with LDA with the

performance metric being the “correctness rate” (9). For Recursive partitioning, we used the library

“rpart” of the R language (30). Further details on the LDA and RP algorithms we used are provided in the

supplementary appendix. We also used logistic regression to test whether a forward sequential automated

selection technique would select any of the chemokines after first forcing all traditional risk factors into

the model.

To quantify the diagnostic performance of the final model derived from each classification

algorithm, we built Receiver Operating Characteristic (ROC) curves and calculated the Area Under the

Curve (AUC) (33). Using LR, we also generated ROC curves to calculate the AUC for each biomarker of

inflammation in isolation and for a model that included all traditional factors without and with CRP. To

7

Page 7 of 23

test for differences in the AUCs of any 2 models derived by LR, we used a cross-validation approach (see

supplementary appendix for detailed methodology).

RESULTS

Unsupervised data analysis and independent predictors of chemokines levels

Two-dimensional hierarchical clustering indicated that case and control subjects tend to form large

homogeneous clusters, although occasionally individual cases and controls remain outside these large

clusters (Fig. 1) suggesting a common profile within each group. MDS demonstrates that chemokines

cluster together and away from traditional risk factors and parameters related to insulin resistance (Fig. 2).

They explain the most variance within the dataset. Of interest, CRP clusters more closely with metabolic

parameters rather than chemokines. After adjustment for case-control status, levels of all 7 chemokines

were significantly correlated with diabetes and BMI (Table 1). Chemokine levels varied less consistently

with age and hypertension. There were no significant associations between chemokine levels and other

clinical variables including the use of aspirin, statins, and ACE-inhibitors. We did not find an association

between chemokine levels and days elapsed from the time of the qualifying event in cases to the time of

the sample collection (Spearman’s correlations range: -0.08 to 0.06, p value range: 0.59 to 0.94).

Characteristics of the subjects stratified by case-control status

Most clinical characteristics of cases and controls differed as expected (Table 2). Despite sampling cases

and controls in the same age range, controls were slightly older than cases. This association, in the

direction opposite of what one would expect, is a consequence of the eligibility criterion for age set in the

two cohorts from which we randomly selected our cases and controls (5, 11, 29). Despite this slight

difference in age, cases as expected had a higher BMI, waist circumference, insulin, and glucose

compared with controls, as well as a higher prevalence of smoking, hypertension, high cholesterol and

diabetes. Cases also had a much higher prevalence of use of drugs that are routinely prescribed for

secondary prevention of CAD including aspirin, statins, and ACE-inhibitors. The majority of chemokines

measured were also significantly higher in cases compared to controls. None of the cases or the controls

reported suffering from a chronic inflammatory disorder such as rheumatoid arthritis, lupus, or

inflammatory bowel disease (details not shown).

Variables selected for prediction of case-control status using different classification algorithms

Each of the three classification algorithms entered 3 chemokines into its final model although each

selected a slightly different combination (Table 3). Only the logistic regression algorithm entered

traditional risk factors into the final model. The classification algorithm that produced the best AUC was

8

Page 8 of 23

logistic regression (AUC = 0.95). Of note, the variable “age” was forced into the LR and LDA models

given the spurious association between age and case-control status described above. For recursive

partitioning, we could not force “age” into the model. For LR and LDA, we also initially attempted to

force in the variables indicating aspirin, statin, and ACE-inhibitor use but doing so led to model converge

problems probably as a consequence of our small sample size and the high correlation between these

variables and case-control status (Table 1). Therefore, these medication use variables were omitted from

these analyses. In the absence of any significant association between these variables and each of the 7

chemokines (Table 1), it is unlikely that their omission had any meaningful impact on the model derived

by each classification algorithm. For the LR model where all traditional risk factors were forced in, we

had to reduce the SLENTRY from 0.10 to 0.01 to avoid convergence problems. At this SLENTRY, 2

chemokines entered the model (IP10, and MCP3).

ROC plots and AUC for individual biomarkers of inflammation, models using traditional risk factors, and

best models derived by classification algorithms

Figure 3a shows the ROC plots for each of the chemokines individually and for CRP while Figure 3b

shows the ROC plots for the combination of traditional risk factors of CAD (age, waist circumference,

diabetes, hypertension, high cholesterol and smoking) without and with CRP and for the model derived by

our LR classification algorithm. Several of the chemokines alone have a higher AUC than the models

using traditional risk factors but none had an AUC higher then the best model derived by LR. CRP did

not improve the AUC of the model using traditional risk factors. The best model derived by LR appears

far more accurate than the either model using traditional risk factors. The best model derived by LDA

(AUC=0.94; SE 0.03) and the best model derived by RP (AUC=0.89; SE 0.03) is also more accurate than

either model using traditional risk factors (data not shown).

Pair-wise comparisons of AUC for models derived by logistic regression

The AUC of the best LR model derived by a forward sequential automated model selection technique was

statistically significantly higher than the AUC of the model using all the traditional risk factors (Table 4).

The AUC for the best LR model was also higher than the AUC of the two chemokines that best predicted

case-control status in isolation. Although the difference in these AUCs did not reach statistical

significance at the 0.05 level, the results demonstrated a very strong trend towards significance (for IP10,

p = 0.06 and for MCP4, p = 0.1). IP10, Eotaxin, MCP1, and MCP4 each in isolation performed better

than the model using traditional risk factors without or with CRP (p < 0.05 for all, details not shown).

9

Page 9 of 23

DISCUSSION

There is a great need for improved tools to diagnose active ASCVD prior to clinical presentation.

Although insights into the mechanisms and circumstances of atherosclerosis are expanding, methods for

identifying subjects with active disease and predicting the efficacy of primary prevention strategies remain

suboptimal. We hypothesized that a multidimensional approach utilizing profiles of several biomarkers of

inflammation could reveal a ‘signature’ of atherosclerosis-related vascular inflammation. The present

study provides strong preliminary experimental support for this hypothesis and suggests that measurement

of multiple biomarkers may reliably identify subjects with confirmed coronary heart disease.

Since vascular inflammation is an underlying pathophysiological basis of atherosclerosis,

chemokines, which are produced in the atherosclerotic vessel, are prime candidates as markers of CAD.

Chemokines form a network of chemotactic proteins produced by activated leukocytes as well as vascular,

endothelial, and smooth muscle cells (7). Their main role is to promote accumulation and activation of

leukocytes in tissues, and their interaction with several cellular receptors contributes to the specificity of

the inflammatory infiltrate (20, 27). Chemokines are often present as groups with varying composition,

and the biological effect of such groups can be quite different from that of individual factors in isolation.

Thus, measuring global patterns of cytokine and chemokine expression plausibly may yield more relevant

biological information than individual protein assays.

Our data demonstrate that plasma concentrations of the chemokines are differentially regulated in

individuals with clinical CAD compared to subjects with no history of CAD. Although chemokines were

correlated with traditional risk factors, these correlations were weak for most risk factors. Furthermore,

our MDS analyses showed that chemokine levels did not cluster with traditional risk factors or CRP.

Models using multiple chemokines more accurately distinguished cases and controls compared to models

using traditional risk factors. These comparisons reached statistical significance despite a majority of

cases being on medical therapy which can acutely suppress markers of inflammation (22). Models using

multiple chemokines also demonstrated a very strong trend towards being more valuable than individual

chemokines despite the relatively small sample size. Of note, CRP was not selected by any of the

classification algorithms. Furthermore, CRP had relatively poor diagnostic performance alone or in

combination with traditional risk factors. Taken together, our findings suggest that the chemokine profile

represents a strong signal of vascular disease which appears to be much more specific than traditional risk

factors and CRP.

Our study has several limitations. First, the serum samples from the case subjects were generally

collected several months after the myocardial infarction. Prior studies indicate that inflammatory markers

such as CRP (6) and metalloproteinases (MMPs) (32) are labile after acute events. However, little is

known regarding circulating chemokines under these circumstances. As such, we cannot be certain that

10

Page 10 of 23

the levels we measured in cases after initial AMI are a stable reflection of the underlying biology.

However, given that measures of inflammation taken after acute events generally subside to pre-event

levels within days, it appears unlikely that the chemokine profile at 3-18 months is simply a reaction to

myocardial infarction. Moreover, a linear model to evaluate a possible relationship between chemokine

levels and time-from-event did not reveal a correlation supporting the notion that the chemokine levels are

not due to the acute event. Furthermore, the complete lack of correlation between chemokine levels and

time elapsed from the qualifying event to the clinic visit also suggests that the AMI had no enduring

impact on serum chemokine levels beyond perhaps a few days post event. Second, some of the clinical

variables not related to chemokine levels were not optimally measured which may have reduced the AUC

for the models using these variables. For example, replacing self report of hypertension and high

cholesterol with accurate measures of blood pressure and serum cholesterol would probably improve the

diagnostic ability of the models using traditional risk factors. Furthermore, the model including traditional

risk factors and CRP may have had a higher AUC if CRP was measured off medications known to

influence its levels such as statins and ACE inhibitors (22). Third, this cross-sectional study does not

establish prognostic value for the models derived by the classification algorithms. However, given their

high discriminatory performance and the participation of chemokines in atherosclerosis, it is likely that

some prognostic information can be gleaned with appropriate study design. In fact, two recent prospective

studies (8, 14) have demonstrated the prognostic value of MCP1. A third prospective study did not clearly

demonstrate incremental prognostic value of chemokines. The different results of the three studies could

be explained by the various follow up period of each study. Whereas the first two studies had a mean

follow up of <10 months and 5.3 years, the third study had a mean follow up time of 11 years. Based on

these trends in prospective studies and the results of several case control studies including ours, it appears

that plasma chemokine level may be most valuable in the prediction of near term CAD events. Lastly,

chemokine measurements used in this study have not been standardized, which precludes their routine use

in clinical practice at the present time. In our study, inter-assay coefficients of variation for these

biomarkers were relatively high (>5%) which might have reduced the strength of the associations

uncovered.

We do not consider the panel of biomarkers of inflammation studied here to be comprehensive.

Indeed, the use of a wider array of analytes may improve sensitivity and specificity for the diagnosis of

active ASCVD. However, this study demonstrates the feasibility of using protein microarrays to

simultaneously measure multiple biomarkers and the potential usefulness of these measurements for

identifying ‘signatures’ of active ASCVD based on the levels of these biomarkers.

In summary, serum levels of a combination of chemokines measured using a multi-marker

approach can accurately distinguish subjects with clinically significant CAD from those with no prior

11

Page 11 of 23

history of CAD. A larger scale study is needed to validate our findings as measures of model

discrimination are generally higher in derivative samples, and prospective studies are also necessary to

determine the incremental prognostic value of these measures over current risk stratification tools.

Algorithms based on panels of informed biomarkers may ultimately lead to improved screening,

diagnosis, and monitoring of cardiovascular disease.

ACKNOWLEDGEMENTS:

We wish to thank the many individuals who contributed to the ADVANCE study, including Joan Fair (co-

investigator); Phenius Lathon (patient recruitment); Malini Chandra, Solomon Henry, Gail Husson,

Mohammed Mahbouba, Balasubramanian Narasimhan, Richard Olshen, Ann Varady (data management);

and Mary Chen, Heideh Fattaey, Emiko Miyamoto (project administration).

REFERENCES:

1. http://www.fastquant.com/Information/AntibodyMenu.html.

2. http://www.fastquant.com/Information/Literature.html.

3. NHLBI fact book, fiscal year 2003. Bethesda, Md.: National Heart, Lung, and Blood

Institute, February 2004. 2003, p. 35-53.

4. NHLBI morbidity and mortality chartbook, 2002. Bethesda, Md.: National Heart, Lung,

and Blood Institute, May 2002. 2002.

5. Armstrong EJ, Morrow DA, and Sabatine MS. Inflammatory biomarkers in acute

coronary syndromes: part IV: matrix metalloproteinases and biomarkers of platelet

activation. Circulation 113: e382-385, 2006.

6. Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R, Joyal SV,

Hill KA, Pfeffer MA, and Skene AM. Intensive versus moderate lipid lowering with

statins after acute coronary syndromes. N Engl J Med 350: 1495-1504, 2004.

7. Charo IF, and Taubman MB. Chemokines in the pathogenesis of vascular disease. Circ

Res 95: 858-866, 2004.

8. de Lemos JA, Morrow DA, Sabatine MS, Murphy SA, Gibson CM, Antman EM,

McCabe CH, Cannon CP, and Braunwald E. Association between plasma levels of

monocyte chemoattractant protein-1 and long-term clinical outcomes in patients with

acute coronary syndromes. Circulation 107: 690-695, 2003.

12

Page 12 of 23

9. Garczarek UM. Classification rules in standardized partition spaces. Dissertation

http://deposit.ddb.de/cgi-bin/dokserv?idn=965050130.

10. Glass CK, and Witztum JL. Atherosclerosis. the road ahead. Cell 104: 503-516, 2001.

11. Go AS, Iribarren C, Chandra M, Lathon PV, Fortmann SP, Quertermous T, and

Hlatky MA. Statin and beta-blocker therapy and the initial presentation of coronary heart

disease. Ann Intern Med 144: 229-238, 2006.

12. Herder C, Baumert J, Thorand B, Martin S, Lowel H, Kolb H, and Koenig W.

Chemokines and incident coronary heart disease: results from the MONICA/KORA

Augsburg case-cohort study, 1984-2002. Arterioscler Thromb Vasc Biol 26: 2147-2152,

2006.

13. Hiatt RA, and Friedman GD. Characteristics of patients referred for treatment of end-

stage renal disease in a defined population. Am J Public Health 72: 829-833, 1982.

14. Hoogeveen RC, Morrison A, Boerwinkle E, Miles JS, Rhodes CE, Sharrett AR, and

Ballantyne CM. Plasma MCP-1 level and risk for peripheral arterial disease and incident

coronary heart disease: Atherosclerosis Risk in Communities study. Atherosclerosis 183:

301-307, 2005.

15. Iribarren C, Go AS, Husson G, Sidney S, Fair JM, Quertermous T, Hlatky MA, and

Fortmann SP. Metabolic syndrome and early-onset coronary artery disease: is the whole

greater than its parts? J Am Coll Cardiol 48: 1800-1807, 2006.

16. Kannel WB, and McGee DL. Epidemiology of sudden death: insights from the

Framingham Study. Cardiovasc Clin 15: 93-105, 1985.

17. Knight PR, Sreekumar A, Siddiqui J, Laxman B, Copeland S, Chinnaiyan A, and

Remick DG. Development of a sensitive microarray immunoassay and comparison with

standard enzyme-linked immunoassay for cytokine analysis. Shock 21: 26-30, 2004.

18. Krieger N. Overcoming the absence of socioeconomic data in medical records: validation

and application of a census-based methodology. Am J Public Health 82: 703-710, 1992.

19. Libby P. Inflammation in atherosclerosis. Nature 420: 868-874, 2002.

20. Luster AD. Chemokines--chemotactic cytokines that mediate inflammation. N Engl J

Med 338: 436-445, 1998.

21. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO, 3rd, Criqui M,

Fadl YY, Fortmann SP, Hong Y, Myers GL, Rifai N, Smith SC, Jr., Taubert K,

13

Page 13 of 23

Tracy RP, and Vinicor F. Markers of inflammation and cardiovascular disease:

application to clinical and public health practice: A statement for healthcare professionals

from the Centers for Disease Control and Prevention and the American Heart Association.

Circulation 107: 499-511, 2003.

22. Ridker PM, Cannon CP, Morrow D, Rifai N, Rose LM, McCabe CH, Pfeffer MA,

and Braunwald E. C-reactive protein levels and outcomes after statin therapy. N Engl J

Med 352: 20-28, 2005.

23. Rifai N, and Ridker PM. Inflammatory markers and coronary heart disease. Curr Opin

Lipidol 13: 383-389, 2002.

24. Ross R. Atherosclerosis--an inflammatory disease. N Engl J Med 340: 115-126, 1999.

25. Rothenbacher D, Muller-Scholze S, Herder C, Koenig W, and Kolb H. Differential

expression of chemokines, risk of stable coronary heart disease, and correlation with

established cardiovascular risk markers. Arterioscler Thromb Vasc Biol 26: 194-199,

2006.

26. Saeed AI, Sharov V, White J, Li J, Liang W, Bhagabati N, Braisted J, Klapa M,

Currier T, Thiagarajan M, Sturn A, Snuffin M, Rezantsev A, Popov D, Ryltsov A,

Kostukovich E, Borisovsky I, Liu Z, Vinsavich A, Trush V, and Quackenbush J.

TM4: a free, open-source system for microarray data management and analysis.

Biotechniques 34: 374-378, 2003.

27. Sallusto F, Mackay CR, and Lanzavecchia A. Selective expression of the eotaxin

receptor CCR3 by human T helper 2 cells. Science 277: 2005-2007, 1997.

28. Stillman B, Parker B, Tonkinson J, and Harvey M. Applying Multiplexed Microspot

Immunoassays. Genetic Engineering News 24: 6, 2004.

29. Taylor-Piliae RE, Norton LC, Haskell WL, Mahbouda MH, Fair JM, Iribarren C,

Hlatky MA, Go AS, and Fortmann SP. Validation of a New Brief Physical Activity

Survey among Men and Women Aged 60-69 Years. Am J Epidemiol 2006.

30. Team RDC. The R Project for Statistical Computing http://www.R-project.org/. [3/15/06,

2006].

31. Troyanskaya O, Cantor M, Sherlock G, Brown P, Hastie T, Tibshirani R, Botstein

D, and Altman RB. Missing value estimation methods for DNA microarrays.

Bioinformatics 17: 520-525, 2001.

14

Page 14 of 23

32. Webb CS, Bonnema DD, Ahmed SH, Leonardi AH, McClure CD, Clark LL, Stroud

RE, Corn WC, Finklea L, Zile MR, and Spinale FG. Specific temporal profile of

matrix metalloproteinase release occurs in patients after myocardial infarction: relation to

left ventricular remodeling. Circulation 114: 1020-1027, 2006.

33. Zweig MH, and Campbell G. Receiver-operating characteristic (ROC) plots: a

fundamental evaluation tool in clinical medicine. Clin Chem 39: 561-577, 1993.

15

Page 15 of 23

FIGURE LEGENDS:

Figure 1- Two dimensional hierarchical clustering of clinical variables and chemokines as well as cases

and controls. The heat map provides graphic representation of case-control status on the x-axis and row-

normalized expression levels of individual serum protein concentrations or clinical measurements on the

y-axis. (CCL11 (Eotaxin), CXCL10 (IP10), CCL2 (MCP1), CCL8 (MCP2), CCL7 (MCP3), CCL13

(MCP4), and CCL3 (MIP1alpha)).

Figure 2- Multidimensional scaling representation of the relative distance between the variables in two

dimensions. The distance matrix is based on the Euclidean distance of the scaled vectors. Variables with

high similarity cluster together creating distinct clusters as shown in the figure. (CCL11 (Eotaxin),

CXCL10 (IP10), CCL2 (MCP1), CCL8 (MCP2), CCL7 (MCP3), CCL13 (MCP4), and CCL3

(MIP1alpha)).

Figure 3- Receiver Operating Characteristics Curve plots for models derived by logistic regression;

Individual Biomarkers (right): Muti-variable models (left). Some individual chemokine measures had

higher AUC values than models that included clinical variables. However, the optimal combined

chemokine model had the best performance characteristics. (CCL11 (Eotaxin), CXCL10 (IP10), CCL2

(MCP1), CCL8 (MCP2), CCL7 (MCP3), CCL13 (MCP4), and CCL3 (MIP1alpha)).

16

Page 16 of 23

TABLES:

Clinical Variable CCL11 CXCL10 CCL2 CCL8 CCL7 CCL13 CCL3age 0.131 0.036 0.269 0.04 0.035 0.378 0.018Body Mass Index1 0.003 0.009 0.053 0.012 0.013 0.007 0.042Waist Circumference 0.076 0.095 0.291 0.222 0.288 0.082 0.116smoker (never vs ever) 0.927 0.761 0.803 0.898 0.409 0.987 0.826hypertension (self report) 0.002 0.129 0.013 0.019 0.194 0.012 0.231fasting glucose1,2 0.561 0.341 0.962 0.196 0.155 0.724 0.081fasting insulin1,2 0.49 0.848 0.477 0.205 0.527 0.707 0.943diabetes (self report) <0.001 0.004 0.032 0.005 0.025 0.002 0.016high cholesterol 0.216 0.823 0.219 0.207 0.452 0.339 0.45C-Reactive Protein1 0.194 0.562 0.994 0.506 0.27 0.779 0.97Use of Statins 0.848 0.699 0.581 0.897 0.767 0.677 0.874Use of Aspirin 0.883 0.34 0.764 0.491 0.348 0.892 0.631Use of ACE Inhibitors 0.53 0.966 0.131 0.959 0.815 0.198 0.6241 A natural log transformation of this variable was used in the regression2 Subjects with a history of diabetes or who were not fasting at the time of the blood draw were excluded

Biomarker1

Table 1. Summary of p-values for the marginal effects of each clinical variable on each biomarker adjusted for Case-Control Status (p values <0.05 in bold)

3 CCL11 (Eotaxin), CXCL10 (IP10), CCL2 (MCP1), CCL8 (MCP2), CCL7 (MCP3), CCL13 (MCP4), and CCL3 (MIP1alpha).

17

Page 17 of 23

Cases Controls(n = 48) (n = 44)

Risk Factors for Coronary Artery Disease Mean (SD) Mean (SD) T testAge (years)1 64.2 (2.6) 65.8 (3.3) 0.01

Median (Range) Median (Range) WilcoxonBody Mass Index 28.8 (21.8-41) 27.1 (20.6-39.5) 0.25Waist Circumference (cm) 96.9 (71.2-129.5) 87.3 (66.3-123) 0.02

Count (%) Count (%) Chi2

Male1 25 (52.1) 19 (43.2) 0.39 Current/Former smoker (self report) 38 (79.2) 25 (56.8) 0.02 Hypertension (self report) 22 (45.8) 17 (38.6) 0.49 Diabetes Mellitus (self report) 11 (22.9) 6 (13.6) 0.25 High Cholesterol (self report) 21 (43.8) 12 (27.3) 0.10

Medications at study visitAspirin 43 (89.6) 13 (29.5) <0.01Any antihypertensive 46 (95.8) 18 (40.9) <0.01

Beta Blockers 40 (83.3) 5 (11.4) <0.01Calcium Channel Blocker 6 (12.5) 4 (9.1) 0.60Diuretic 14 (29.2) 9 (20.5) 0.34ACE inhibitor 35 (72.9) 9 (20.5) <0.01Angiotensin Receptor Blocker 1 (2.1) 0 (0) 0.34

Any lipid altering medication 42 (87.5) 10 (22.7) <0.01Statin 41 (85.4) 7 (15.9) <0.01Fibrate 1 (2.1) 2 (4.5) 0.51

Any medication for diabetes 9 (18.8) 3 (6.8) 0.09

Median (Range) Median (Range) WilcoxonBioMarkersFasting Insulin 10 (1-39.9) 7.9 (1-30) 0.09Fasting Glucose 99 (80-286) 96 (80-141) 0.17C-Reactive Protein 0.6 (-3-4.3) 0.5 (-1.7-2.7) 0.36 CCL11 (Eotaxin) 763.7 (40-2500) 187.2 (20.6-662.2) <0.01CXCL10 (IP10) 1958 (200-12500) 215.9 (45.4-3100) <0.01CCL2 (MCP1) 516.5 (10-2500) 197.5 (89.8-624.7) <0.01 CCL8 (MCP2) 92.7 (10-2500) 24.6 (5-859.6) 0.04CCL7 (MCP3) 40 (13-3217) 10 (5-1588) 0.08 CCL13 (MCP4) 694.8 (10-2725) 164.1 (54.6-644.8) <0.01CCL3 (MIP1alpha) 123.4 (40-2500) 100.6 (17.8-1500) 0.04

P

Table 2. Characteristics of study sample according to case-control status

1Factor used in stratified sampling. See text for details.

18

Page 18 of 23

Classification Algorithm Variables selected (in order) AUC (SE)

Logistic Regression1 CXCL10, CCL3, waist circ, CCL7, smoking 0.95 (0.03)

Linear Discriminant Analysis1 CCL13, CCL3, CXCL10 0.94 (0.03)Recursive Partitioning CCL7, CCL11, CCL13 0.88 (0.04)

AUC = Area under the Receiver Operating Curve, SE = standard error1age was forced into model (see text for details)

Table 3. Summary of variables selected by each classification algorithm and the respective diagnostic performance in predicting case control status

Model1 AUC (SE) P

Best model derived by LR2 0.95 (0.03) referenceTraditional Risk Factors3 0.67 (0.06) <0.001Traditional Risk Factors3 + CRP 0.68 (0.06) <0.001CCL3 (MIP1alpha) 0.69 (0.06) <0.001CCL7 (MCP3) 0.74 (0.05) <0.001CCL8 (MCP2) 0.79 (0.05) 0.001CCL11 (Eotaxin) 0.86 (0.04) 0.037CCL2 (MCP1) 0.87 (0.04) 0.05CXCL10 (IP10) 0.90 (0.03) 0.06CCL13 (MCP4) 0.90 (0.04) 0.1AUC = Area under the Receiver Operating Curve, SE = standard error1variable 'age' was forced into all models (see text for details)2variables MCP3, IP10, MIP1alpha 3variables waist circumference, diabetes, hypertension, high cholesterol, and smoking

Table 4. Comparison of AUC values between the multimarker model and other models derived by logistic regression

19

Page 19 of 23

Figure 1

Page 20 of 23

Figure 2

Page 21 of 23

Figure 3A

Page 22 of 23

Figure 3B

Page 23 of 23