Characterising a novel biomarker of early myocardial injury

Student thesis: Doctoral ThesisDoctor of Philosophy


Background: We previously identified cardiac myosin-binding protein C (cMyC) in coronary venous effluent and developed a high-sensitivity assay by producing an array of monoclonal antibodies and choosing an ideal pair based on affinity and epitope maps. Compared to high-sensitivity cardiac Troponin (hs-cTn), we demonstrated that cMyC appears earlier and rises faster following myocardial necrosis and is also more abundant. Contemporarily, we investigated (i) analytic sensitivity, (ii) whether cMyC can aid in the diagnosis of Acute Myocardial Infarction (AMI) (a) amongst unselected patients presenting to the emergency department (ED) and (b) presenting in a pre-hospital setting, (iii) derived and validated optimal cut-offs for the use of cMyC in a 0/1h algorithm for the rule-out/rule-in of AMI. Methods: We compared abundance of cMyC to hs-cTn by spiking cardiomyocytes/cardiac tissue into aliquots of human serum. We evaluated the clinical utility of cMyC by calculating the area under the receiver-operating characteristics curve (AUC) in 1,954 patients (17% AMI), for presentation and 1h-change values. Cut-offs were derived using a derivation/validation split, determining optimal thresholds based on NPV/PPV/triage efficiency for >390,000 combinations. Net Reclassification Improvement (NRI) determined immediate triage effectiveness. In 776 patients (22% AMI) sensitivity & specificity were calculated from in-ambulance blood draws using a real and feasible Limit of Detection (LoD) on a point-of-care testing (POCT) device for cTnT and cMyC, respectively. Results: cTnT, cTnI, and cMyC increased by 3.9ng/L (3.6-4.3), 4.3ng/L (3.8-4.7), and 41.0ng/L (38.0-44.0) per µg of human myocardium. In the ED, the diagnostic accuracy for AMI was comparable between the three biomarkers in baseline blood samples. cMyC increased the diagnostic performance of hs-cTnI but not hs-cTnT 0&1h samples. NRI was up to 30% better when comparing cMyC to hs-cTnT/I, translating into a more effective triage into rule-out and rule-in of AMI using a single blood test at presentation. The best performing cMyC 0/1h rule-out/rule-in algorithm matched the ESC hs-cTnT/I algorithms in terms of safety, and specificity in comparison to hs-cTnI, but not hs-cTnT. cMyC increased triage-efficiency by 3.9-10.6%. Further, cMyC significantly increases the number of patients eligible for direct rule-out or rule-in based on a single blood test at presentation to the ED. In the pre-hospital setting, the diagnostic accuracy of cMyC was significantly higher than hs-cTnT (0.839 vs 0.813, p=0.005). The POCT threshold of cTnT (50 ng/L, 10-fold LoD of laboratory assay) achieved a sensitivity of 40.5% [33.6-47.6%]; cMyC (12 ng/L, 30-fold LoD) achieved a sensitivity of 94.8% [91.2-97.7%]. Risk prediction was superior for cMyC at the POCT-detection limit. Conclusions: hs-cTnT/I and cMyC are exquisitely sensitive biological signals – all assays are able to detect the equivalent of necrosis of a single cardiomyocyte in spiked human serum. cMyC is more abundant than cTnT/I and provides discriminatory power comparable to hs-cTnT/I for the diagnosis of AMI in all-comers, but identifies a greater proportion of patients with AMI in very early presenters. A standout feature is cMyC’s ability to more effectively triage patients into rule-out and rule-in categories, with comparable safety endpoints (as with hs-cTnT/I). This distinction is likely related to the documented greater abundance and more rapid release profile of cMyC. If used on a POCT platform, cMyC could significantly improve the early triage of patients with suspected AMI.
Date of Award2019
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorMichael Etter (Supervisor) & Manuel Mayr (Supervisor)

Cite this