Clinical Biomarkers of Cardiac Injury: Cardiac Troponins and Natriuretic Peptides

Abstract

Laboratory and clinical medicine groups are actively collaborating and optimizing their individual expertise to potentially prove standardized biomarker assays that will optimize patient care. It is critical as new biomarkers are discovered, that quality specifications be developed prior to approval by regulatory agencies that give supported endorsement for the worldwide marketplace. The laboratory medicine and clinical communities of scientists continues to challenge the biomarker field, working towards developing standard reference materials, and providing quality analytical specifications that, hopefully, will be endorsed by our clinical, in vitro diagnostic and pharmaceutical industry colleagues. This paper will specifically address cardiac troponins and natriuretic peptides.

In 2005, academics, industry, and regulatory agencies are challenged to provide quality assays for cardiac biomarkers; specifically cardiac troponin (cTn) and natriuretic peptides (NP). In part this includes developing primary reference materials to standardize these assays, specifically cardiac troponin I (cTnI) and B-type natriuretic peptide (BNP); both goals of the AACC and IFCC (Panteghini et al., 2001; Apple et al., 2005). The use of biomarkers in clinical practice, whether for diagnostics or risk assessment, requires that measures should be reproducible, and be highly sensitive and specific (Manolio, 2003). When a biomarker test is in use worldwide, or crossing from research to clinical practice, requires it to be replicated in multiple settings.

Standardization efforts have been underway for the past 8 years by the International Federation for Clinical Chemsitry (IFCC) Committee for Standardization of Markers of Cardiac Damage (C-SMCD) and the American Association for Clinical Chemistry (AACC) Committee of Standardization of Creatine Kinase MB (CKMB) Mass and Cardiac Troponin I (cTnI). A reference material that reduces inter-assay CKMB bias from 31% to 15% using a lyophilized recombinant CKMB2 material has been developed (Christenson et al., 1999). This material is commercially available through the AACC, but has not been certified internationally.

The AACC cTnI Standardization Committee in collaboration with the National Institute of Science and Technology (NIST), identified 10 candidate reference materials for cTnI, as a step towards standardization. Working with numerous in vitro diagnostic manufacturers of cTnI, a single reference material, Standard Material 2921, is now commercially available from NIST. Using this material as a traceable calibration material, assays have now been harmonized, not standardized; decreasing discordance between methods from 80 to 95% to 2 to 8%, for the majority of cTnI assays evaluated (Christenson et al., 2001).

Ideally, this material will be used by cTnI manufacturers as a traceable reference material to calibrate cTnI assays. However, it may be impossible to standardize cTnI assays due to the fact that cTnI exits in multiple forms in necrotic myocardial tissue and circulating blood following myocardial necrosis. These forms include the ternary TIC complex, binary IC complex, free I and various forms of all three in the oxidized, reduced, phosphorylated and N and C-terminal degraded forms (Figure 1) (Wu et al., 1998). Thus, because the manufacturers of the numerous cTnI assays use different antibody pairs in their assay formats, which often do not recognize the same epitope regions on the different cTnI molecules with equal molarity, standardization becomes almost an impossible task (Apple et al., 2002). As an example, cTnI concentration differences observed between 2 assays for 2 patients presenting with acute myocardial infarction (AMI) is demonstrated in Figure 2. The two different assays show different cTnI concentrations and different appearance and clearance profiles over time in the blood. Thus, while harmonization of cardiac troponin assays could be possible in the future using the NIST standard cTnI material, there are no regulatory agencies that mandate in vitro diagnostic companies to use the NIST material for this purpose in their assays.

The IFCC C-SMCD is actively undertaking an initiative in 2005 through 2006 to develop standard referent materials for BNP and NT-proBNP. As in the case for NT-proBNP, Roche has sublicensed their reagents (antibodies and antigen standard material) to other companies (Dade-Behring, recently FDA cleared for the NT-proBNP assay); thus, assays should be harmonized if using the same antigen and antibody pairs for their assays as provided by Roche. Two initiatives of the IFCC C-SMCD regarding BNP will be first to identify sources of BNP antigen that will be characterized for purity, working with NIST. Second, obtain antigen materials for BNP, proBNP and NT-proBNP and carryout a cross-reactivity study in parallel with the manufacturers that currently have assays in the marketplace (Table 1).

The impact of successfully providing standard reference materials for traceability of assays reaches well beyond Laboratory Medicine. Clinical shareholders that heavily rely upon both cTn and BNP in medical decision processes, include cardiology, emergency medicine, epidemiology and clinical trials and preclinical trials in the pharmaceutical industry. In addition, regulatory agencies both in the United States, such as the FDA, as well as diagnostic companies are substantially impacted by assay standardization and the quality or lack of quality of assays in the marketplace.

The initial objectives of the IFCC C-SMCD was to devise quality specifications of biomarker assays to have manufacturers endorse and consistently follow, allowing for uniform inclusion of information in their package inserts. Further, the C-SMCD encourages publications of data in the peer-reviewed literature to increase the worldwide evidence base medical practices, as well as encourage regulatory agencies to adopt a minimal and uniform set of criteria for manufacturers to provide when seeking assay clearance. To date, 2 sets of quality specifications have been published; 1 for cardiac troponin assays (Apple et al., 2005) and 1 for NP assays (Panteghini et al., 2001) (Table 2). Both address analytical factors such as antibody specificity, influence of anticoagulants, detection limits, total imprecision and interferents; as well as preanalytical factors, such as sample storage, gel separator effects, collection containers. Further, the impact of assay standardization greatly influences reference limit determinations, thus impacting clinical decision making.

In September, 2000, the joint ESC/ACC consensus group (Figure 3) (Joint European Society of Cardiology, 2000; Jaffe et al., 2000) endorsed cTn as the preferred biomarker for detection of myocardial injury, and in the setting of ischemia, any increase above the 99th percentile reference limit, was deemed an AMI (Joint European Society of Cardiology, 2000; Jaffe et al., 2000). This shift, from the traditional ROC curve derived cutoff optimized for sensitivity and specificity as previously used for CKMB, was endorsed by the IFCC. In addition, the ESC/ACC document endorsed that the total imprecisions of assays should be ≤ 10% (10% CV). While no assay could meet this 10% CV guideline, over the last several years, 2nd and 3rd generation cTn assays have improved their low end analytical performance and now, allow for improved diagnostic confidences of the many assays in the marketplace (Apple et al., 2002). Lowering the cutoff for diagnostics of AMI has been shown to increase the rate of MI inclusions 35 to 45%. In one study by Lin et al., (2004) in 1719 consecutive patients being admitted with symptoms suggestive of acute coronary syndrome, the rate of cTnI positive, but CKMB negative patients increased from N = 73 (4.2%) to N = 209 (12.2%); an absolute increase of 86%; potentially allowing for more optimized patient management, allowing for improved outcomes (Luepker et al., 2003; Lin et al., 2004). Currently both the National Academy of Clinical Biochemistry (NACB) and the Global Task Force for the redefinition of MI, are in the process of updating guidelines and will likely endorse the 99th percentile cutoff. Both groups will likely encourage improving imprecision at the 99th percentile cutoff, recognizing that differences in imprecision between two assays, for example one with a 10% CV vs. and one with a 25% CV, will not statistically impact misclassification of patients being ruled out for MI based on serial samples over a 24-hour period following presentation.

BNP presents the same clinical issues and potential for misclassification due to lack of standardization between commercial assays (Apple et al., 2005). Further, because BNP and NT-proBNP measure different antigens that circulate in the blood, these two assays should never be mixed in patient monitoring. Table 3 shows the current assays in the marketplace, demonstrating the differences found between median NP concentrations vs. New York Heart Association Stage Classification. This data shows that while BNP assays may be harmonized around an accepted ROC curve derived medical cutoff of 100 ng/L, assay correlations likely vary at lower and higher concentrations. Ongoing analytical and clinical studies will hopefully increase our understanding of what circulating biological NP antigens we should be measuring and standardizing, and whether they may differ by disease state. However, neither standardization or harmonization for the different NP assays will likely occur in the next 5 years.

In conclusion, laboratory and clinical medicine groups are actively collaborating and optimizing their individual expertise to potentially improve standardization of biomarker assays that will optimize patient care. Cardiac troponin I and T are assays that are key to the diagnosis of myocardial injury and acute myocardial infarction, as natriuretic peptides (BNP, NT-proBNP) are useful for the ruling out heart failure and assisting in the diagnosis and risk assessment of heart failure. Analaytical issues for both assays still need to be resolved with evidence based studies. It is critical that as new biomarkers are discovered, that quality specifications be developed prior to approval by regulatory agencies that give supported endorsement for the worldwide marketplace. The laboratory medicine and clinical communities of scientists continues to challenge the biomarker field, working towards developing standard reference materials, and providing quality analytical specifications that, hopefully, will be endorsed by our clinical, in vitro diagnostic and pharmaceutical industry colleagues. One initiative to support this goal will be to create an IFCC supported Web-site database of analytical characteristics, and reference limit cutoffs, linked to both industry package inserts and the peer-reviewed literature.