Supplementary Material for: Body-wide expression profiles of commonly assessed cardiac biomarkers in a large cohort of human tissue donors

Background:

Blood biomarkers of myocardial damage or stress are routinely used to guide the diagnosis and management of both acute and chronic cardiac conditions. The circulating levels of these proteins are directly influenced by source tissue expression levels, and understanding where and to what degree they are expressed throughout the body can yield insights into their properties as biomarkers. Thus, in this descriptive study, we sought to comprehensively map the expression of twelve clinically established or emerging cardiac biomarkers across a broad spectrum of human tissues, and comparatively assess expression characteristics that could impact diagnostic performance.

Methods:

Existing genome-wide RNA sequencing data originating from 16,357 cardiac and non-cardiac tissue specimens harvested from 946 donors were used to quantify the expression levels of genes coding for the twelve proteins of interest (cTnT, cTnI, H-FABP, proANP, proBNP, Mb, CKtotal, CK-MB, LDHtotal, LDH-1, sST2, and cMyBP-C). Cardiac abundance, atrioventricular heterogeneity, and cardiac enrichment were subsequently assessed and compared between genes, both in the total pool of specimens, as well as subsets of specimens grouped by donor sex and age.

Results:

When considering the entirety of our analyses, the spatial expression characteristics of cMyBP-C, one of the emerging biomarkers we investigated, compared favorably to those of established biomarkers such as the troponins, suggesting that it may be a viable supplement to markers currently in clinical use. However, several other emerging biomarkers we assessed, including sST2 and H-FABP, displayed high expression in numerous non-cardiac tissues that could serve as diagnostic confounds and limit their clinical value. In addition, we also observed differences between the expression profiles of closely related established biomarkers that have often been used interchangeably, including cTnI and cTnT, and proANP and proBNP, that could explain recent reports of discordant blood measures. Finally, we observed notable age and sex-related differences in the expression of proANP and proBNP within cardiac tissue specifically which support calls for the use of tiered diagnostic cutoffs.

Conclusions:

Our findings provide insights into the potential utility of several notable emerging cardiac biomarkers, and new information that could mechanistically explain previously reported phenomena or highlight possible diagnostic advantages, disadvantages, or use-caveats regarding others currently measured in clinical care.