In the Providing Regional Observations to Study Predictors of Events in the Coronary Tree (PROSPECT) study, plaque burden, plaque composition, and minimal luminal area were associated with an increased risk of adverse cardiovascular events arising from untreated atherosclerotic lesions (vulnerable plaques) in patients with acute coronary syndromes (ACS). We sought to evaluate the utility of biomarker profiling and clinical risk factors to predict 3-year all-cause and nonculprit lesion-related major adverse cardiac events (MACEs). Of 697 patients who underwent successful percutaneous coronary intervention (PCI) for ACS, an array of 28 baseline biomarkers was analyzed. Median follow-up was 3.4 years. Beta2-microglobulin displayed the strongest predictive power of all variables assessed for all-cause and nonculprit lesion-related MACE. In a classification and regression tree analysis, patients with beta2-microglobulin >1.92 mg/L had an estimated 28.7% 3-year incidence of all-cause MACE; C-peptide <1.32 ng/ml was associated with a further increase in MACE to 51.2%. In a classification and regression tree analysis for untreated nonculprit lesion-related MACE, beta2-microglobulin >1.92 mg/L identified a cohort with a 3-year rate of 18.5%, and C-peptide <2.22 ng/ml was associated with a further increase to 25.5%. By multivariable analysis, beta2-microglobulin was the strongest predictor of all-cause and nonculprit MACE during follow-up. High-density lipoprotein (HDL), transferrin, and history of angina pectoris were also independent predictors of all-cause MACE, and HDL was an independent predictor of nonculprit MACE. In conclusion, in the PROSPECT study, beta2-microglobulin strongly predicted all-cause and nonculprit lesion-related MACE within 3 years after PCI in ACS. C-peptide and HDL provided further risk stratification to identify angiographically mild nonculprit lesions prone to future MACE.
Acute coronary syndromes (ACS) typically occur because of plaque rupture and thrombosis of a lipid-rich plaque. Unfortunately, atherosclerotic plaques that are prone to future ACS are typically mild to moderate in severity by angiography and, thus, are rarely prospectively identified before becoming unstable. The “Providing Regional Observations to Study Predictors of Events in the Coronary Tree” (PROSPECT) study included 697 patients with ACS who underwent coronary angiography and 3-vessel grayscale and radiofrequency intravascular ultrasound (IVUS) imaging. At a median 3.4-year follow-up, PROSPECT demonstrated that plaque burden and composition were predictive of future cardiovascular events arising from angiographically mild lesions. Three-vessel IVUS imaging, however, is impractical in most patients because of cost and complexity. Noninvasive and less costly means to identify potentially unstable patients and plaques are, thus, urgently needed. Beta2-microglobulin is a novel candidate biomarker that may detect progression of atherosclerosis. Beta2-microglobulin is an intermediate size (11.8 kDa) uremic toxin. It forms the beta chain of the human leukocyte antigen class I molecule and is present on the surface of most nucleated cells and in most biologic fluids. Recently, beta2-microglobulin has been identified by proteomic profiling as a biomarker of peripheral arterial disease. The role of beta2-microglobulin as a predictor of progressive coronary atherosclerosis has not been studied. We, therefore, examined the potential prognostic role of beta2-microglobulin and additional biomarkers in patients with stabilized ACS in the PROSPECT study.
Methods
The design, inclusion and exclusion criteria, end points, and definitions of the PROSPECT study have already been described in detail elsewhere. PROSPECT enrolled 697 patients admitted for ACS (recent ST-segment elevation myocardial infarction, non–ST-segment elevation myocardial infarction, or unstable angina pectoris) in whom successful percutaneous coronary intervention (PCI) had been performed of all angiographically significant lesions (culprit lesions). Study patients underwent coronary angiography and 3-vessel grayscale and radiofrequency IVUS (Volcano Corp., San Diego, California) imaging of the proximal 6 to 8 cm of each coronary artery. The PROSPECT study complies with the Declaration of Helsinki and was approved by the institutional review boards of the participating centers. All patients gave written informed consent to participate in the study. PROSPECT was registered at ClinicalTrials.gov ( http://clinicaltrials.gov/show/NCT00180466 ; NCT00180466 ).
The primary end point of the PROSPECT study was the composite incidence of major adverse cardiac events (MACEs), defined as cardiac death or arrest, myocardial infarction, or re-hospitalization for unstable or progressive angina, as adjudicated by an independent clinical events committee. MACE were classified as related to culprit lesions or nonculprit lesions according to whether at follow-up angiography they were shown to have originated from a lesion originally treated during the index event or from an untreated lesion, respectively. If follow-up angiography was not performed, the lesion causing the event was classified as indeterminate. The median follow-up time was 3.4 years.
Biomarkers for vulnerable plaques or progression of atherosclerosis testing were selected (a) according to published evidence or (b) if they were previously known and established risk markers or risk factors for the progression of disease (e.g., lipids). All markers were measured on the ARCHITECT platform as established or prototype assays (for details on prototype assays, see Supplementary Data ). Blood was drawn at admission for the index event in EDTA plasma tubes, centrifuged, and stored at −80°C within 2 hours. For biomarker testing, all samples were shipped to a central core laboratory (Charité-Universitätsmedizin, Berlin, Germany) and analyzed in batches. A complete list of the biomarkers analyzed in the present study is listed in Table 1 .
| Marker | Description | Role in atherosclerosis | Reference |
|---|---|---|---|
| LDL-cholesterol | Low-density lipoprotein cholesterol | Target marker for statin therapy | |
| HDL-cholesterol | High-density lipoprotein cholesterol | Inverse correlation to CV risk | |
| ApoA1 | Apolipoprotein A1 | Important cofactor for Lp(a) action | |
| ApoB | Apolipoprotein B | High levels correlate with CV risk | |
| Lp(a) | Lipoprotein(a) | High levels correlate with CV risk | |
| Cystatin C | Filtration marker for estimation of GFR ∗ | Associated with CV outcomes | |
| Creatinine | Filtration marker for estimation of GFR ∗ | Associated with CV outcomes | |
| NGAL | Neutrophil gelatinase-associated lipocalin | Sensitive early marker of kidney injury | |
| hsCRP | High sensitivity C-reactive protein | Associated with CV outcomes | |
| MPO | Myeloperoxidase | Associated with CV outcomes (additive to high sensitivity troponins) | |
| PLGF | Placental growth factor | Associated with CV outcomes | |
| sFLT-1 | Soluble fms-like tyrosine kinase 1 | Associated with endothelial dysfunction | |
| GDF-15 | Growth differentiation factor-15 | Relation to (subclinical) atherosclerosis | |
| Alpha-1-antitrypsin | – | Associated with progression of atherosclerosis | |
| Homocysteine | – | Associated with progression and outcome | |
| Hs-TnI | High sensitivity troponin I | Associated with CV outcomes in ACS | |
| BNP | B-type natriuretic peptide | Associated with CV outcomes | |
| C-peptide | Proinsulin C-peptide; corresponds to endogenous insulin production | Impact on development of vascular disease in diabetes mellitus | |
| Insulin | Endogenous insulin concentration | Insulin resistance (high insulin levels) associated with atherosclerosis | |
| AFP | Alpha 1 fetoprotein | Potential role in pregnancy induced hypertension | |
| CEA | Carcinoembryonic antigen | New candidate for risk in ACS | |
| Transferrin | Marker of iron metabolism | Relation to insulin resistance and progression of atherosclerosis | |
| TIMP-1 | Tissue Inhibitor Matrix metalloproteinase 1 | Associated with atherosclerotic plaque burden | |
| Beta2-microglobulin | Beta chain of HLA class I molecule; 11.8 kDa polypeptide | Associated with peripheral artery disease |
∗ Laboratory biomarkers analyzed in the current study and their role in atherosclerosis.
Categorical variables are reported as percentages and were compared using Fisher’s exact test. Continuous variables are reported as median (25% and 75% interquartile range) and were compared with nonparametric tests (Mann-Whitney and Kruskal-Wallis). Event rates were estimated using the Kaplan-Meier methodology, and log-rank tests were conducted to examine variables influencing all-cause MACE and nonculprit lesion-related MACE. Classification and regression tree (CART) analyses based on Kaplan-Meier estimates were performed including age, gender, risk factors, TIMI and Framingham scores and their components, and available biomarkers as potential predictors for both all-cause MACE and nonculprit lesion-related MACE. For the analyses of nonculprit lesion MACE, patients with other MACE (i.e., culprit lesion MACE and indeterminate MACE) were excluded. CART is a stepwise (basically bivariate) statistical procedure based on binary recursive partitioning. The 3 basic steps of CART are that the overall study group is split into 2 subgroups using the most powerful predictor of the outcome; for this procedure, all quantitative variables are split into deciles as the potential cutoffs; this splitting into 2 is repeated within the subgroups until no further significant splits are found or the subgroups become too small, and finally, the results are displayed in a binary tree structure which in a final step is pruned as necessary. Multivariable analysis was performed using Cox proportional hazards regression to support the explorative CART calculations. Variables were chosen based on bivariate findings and CART analyses.
Results
A total of 697 study patients were enrolled. The 3-year rate of all-cause MACE was 20.4%. Table 2 lists baseline patient characteristics according to the occurrence of all-cause MACE during follow-up. In the CART analysis, patients with beta2-microglobulin values >1.92 mg/L had a 28.7% risk for all-cause MACE within 3 years. A C-peptide value <1.32 ng/ml in this cohort was associated with a further increase in 3-year MACE to 51.2% ( Figure 1 ). The relation between beta2-microglobulin and 3-year all-cause MACE and its components is shown in Figure 2 and listed in Table 3 . By multivariable analysis, beta2-microglobulin was the strongest predictor of all-cause MACE during follow-up. Other significant predictors were high-density lipoprotein (HDL), C-peptide, transferrin, and history of angina pectoris ( Table 4 ).
| Variable | All patients (n=697) | No MACE (n=566) | All-cause MACE (n=131) | p-value | NCL-MACE (n=74) | p-value ∗ |
|---|---|---|---|---|---|---|
| Age (years) | 58 (51-67) | 58 (51-67) | 58 (51-68) | 0.57 | 58 (50-68) | 0.98 |
| Female | 24% (167/697) | 22.4% (127/566) | 30.5% (40/131) | 0.05 | 27.0% (20/74) | 0.38 |
| Body mass index (kg/m 2 ) | 28 (25-31) | 28 (25-31) | 28 (35-32) | 0.39 | 28 (26-32) | 0.17 |
| Systolic blood pressure (mmHg) | 130 (116-145) | 130 (115-145) | 132 (118-147) | 0.66 | 130 (117-149) | 0.73 |
| Diastolic blood pressure (mmHg) | 74 (67-84) | 75 (66-84) | 74 (69-82) | 0.90 | 74 (69-80) | 0.93 |
| History of coronary artery disease | 13.7% (94/685) | 11.6% (65/559) | 23.0% (29/126) | 0.001 | 21.4% (15/70) | 0.03 |
| Prior myocardial infarction | 10.5% (73/693) | 10.1% (57/562) | 12.2% (16/131) | 0.53 | 13.5% (10/74) | 0.42 |
| Prior percutaneous coronary intervention | 11.1% (77/696) | 9.5% (54/565) | 17.6% (23/131) | 0.01 | 17.6% (13/74) | 0.04 |
| Prior congestive heart failure | 1.9% (13/693) | 1.4% (8/563) | 3.8% (5/130) | 0.08 | 1.4% (1/73) | 1.00 |
| Family history of CAD | 44.8% (276/616) | 43.5% (220/506) | 50.9% (56/110) | 0.17 | 54.8% (34/62) | 0.10 |
| Hypertension | 44.1% (279/632) | 45.2% (227/517) | 43.9% (52/115) | 0.84 | 46.0% (29/63) | 0.79 |
| Hyperlipidemia | 46.3% (320/691) | 44.6% (251/563) | 53.9% (69/128) | 0.06 | 56.3% (40/71) | 0.08 |
| Diabetes mellitus | 17.1% (119/694) | 15.6% (88/563) | 23.7% (31/131) | 0.04 | 25.7% (19/74) | 0.05 |
| Insulin-treated | 3.0% (21/694) | 2.1% (12/563) | 6.9% (9/131) | 0.01 | 6.8% (5/74) | 0.04 |
| Smoker | 47.7% (328/687) | 47.9% (267/557) | 46.9% (61/130) | 0.85 | 48.6% (36/74) | 1.00 |
| Clinical presentation | ||||||
| STEMI | 30.3% (211/697) | 29.5% (167/566) | 33.6% (44/131) | 0.40 | 36.5% (27/74) | 0.23 |
| NSTEMI | 65.6% (457/697) | 66.3% (375/566) | 62.6% (82/131) | 0.48 | 60.8% (45/74) | 0.36 |
| Unstable angina pectoris | 4.2% (29/697) | 4.2% (24/566) | 3.8% (5/131) | 1.00 | 2.7% (2/74) | 0.76 |
| Medication at discharge | ||||||
| Statin | 85.5% (594/695) | 85.7% (484/565) | 84.6% (110/130) | 0.78 | 83.8% (63/74) | 0.73 |
| Aspirin | 96.8% (675/697) | 96.6% (547/566) | 97.7% (128/131) | 0.78 | 95.9% (71/74) | 0.73 |
| ACE inhibitor | 62.8% (437/696) | 62.8% (355/565) | 62.6% (82/131) | 1.00 | 62.2% (46/74) | 0.90 |
| ARB | 8.5% (59/694) | 8.5% (48/563) | 8.4% (11/131) | 1.00 | 9.5% (7/74) | 0.83 |
| Beta-blocker | 90.7% (632/697) | 90.6% (513/566) | 90.8% (119/131) | 1.00 | 89.2% (66/74) | 0.67 |
| Thienopyridine | 97.0% (676/697) | 96.5% (546/566) | 99.2% (130/131) | 0.15 | 98.6% (73/74) | 0.50 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
