Integrating the myocardial matrix into heart failure recognition and management

Abstract

In contrast to public perception, the morbidity and mortality and the resultant healthcare costs associated with chronic heart failure (HF) are increasing and arguably reaching epidemic proportions. Although basic research efforts have provided unique insights into fundamental processes that govern myocardial extracellular matrix (ECM) growth and function, the translation of these findings to improved diagnostics and therapeutics for HF has not been as forthcoming. The factors that contribute to this relative paucity of new clinical tools for HF are multifactorial but likely include the need to recognize and differentiate HF phenotypes and to couple the use of biomarkers and multimodality imaging in early translational research studies. Recognizing the classification scheme of HF with a reduced ejection fraction (EF) to that of HF with a preserved EF and incorporating unique and differential measurements of ECM remodeling to these specific disease processes are warranted. For example, profiling pathways of ECM degradation such as the matrix metalloproteinases in patients with ischemic heart disease and HF with a reduced EF can provide prognostic information in terms of risk of progression to HF. In patients with chronic hypertensive disease and HF with a preserved EF, plasma profiling indexes of ECM synthesis and turnover, as well as advances in ECM imaging, have been shown to provide diagnostic and prognostic use. In terms of therapeutics, strategies to stabilize the ECM in HF with a reduced EF hold potential, whereas in contradistinction, selective antifibrotic agents may hold promise for HF with a preserved EF.

Keywords: cardiovascular diseases; diagnostics; myocardial remodeling; therapeutics; ventricular function.

Figure 1

(TOP) The structure of the ECM is highly complex, and this schematic is highly simplified, but is intended to identify the myocyte, fibroblast, and fibrillar collagen weave. There are key transmembrane proteins, such as integrins, that form cell-cell communications between myocyte-ECM, myocyte-fibroblasts, and fibroblast-ECM as well as a number of bioactive signaling molecules that traverse this interstitial space. A pathological stimulus, whether it be an ischemic event or increased pressure overload, causes specific changes within the ECM and can contribute to the development of either HFrEF or HFpEF, respectively. (MIDDLE) With HFrEF, and in this example, the MI region, transdifferentiation, and proliferation of fibroblasts occur, which is also accompanied by increased fibrillar collagen accumulation but also enhanced proteolytic activity, resulting in a poorly formed and unstable ECM. (BOTTOM) In HFpEF, such as with pressure overload (PO), fibroblast transdifferentiation/proliferation also occurs and is accompanied by a robust induction of pro-fibrotic signaling cascades ultimately leading to significantly increased and highly cross-linked fibrillar collagen within the ECM. Concepts and constituents presented in this illustration are adapted from references 17 and .

Figure 2

Changes in plasma miRs have been identified early post-MI, and in this example, an early and robust increase in plasma miR-29a was detected, which then fell rapidly at later post-MI time points. The significance of this observation in terms of pathophysiology of ECM remodeling post-MI remains to be established, but these early changes in miR-29a were associated with LV dilation at 90 days post-MI (r=0.77, p<0.05- CTL: Control values). Adapted from reference 57.

Figure 3

Molecular imaging approaches coupled with functional measurements will yield important diagnostic insights as well as identify potential therapeutic targets in HFrEF. In this example, dual-Isotope SPECT/CT imaging was performed by injection of thallium (²⁰¹Tl) and an MMP radiotracer (^99mTc-RP805) in a pig at 2 weeks post-MI. A clear perfusion defect at the site of the MI was evident (Top panel-arrow) and coregistered with significant MMP radiotracer uptake-indicative of heightened MMP activity. Adapted from reference 60.

Figure 4

Multivariate modeling has been used to identify potential ECM biomarkers to predict the development and progression of HFpEF. (A) In this example, a biomarker panel incorporating specific MMPs, TIMPs, collagen peptides, and NT-proBNP demonstrated predictive value in patients with HFpEF. Specifically, using a particular cassette of ECM related plasma biomarkers, which included MMP-2, MMP-8, TIMP-4, and the propeptide of collagen IIII (PIINP) as well as N-terminal BNP, an area under the curve of greater than 0.70 (p<0.01) was obtained indicating a high predictive value for identifying patients with HFpEF. (B) In order to more carefully examine the predictive value of this ECM biomarker panel in terms of HFpEF prediction, random subsets of patients from the entire cohort were sampled and the prediction algorithm tested (cross-validation), which was repeated 5 times (5 subsets). The 5 subsets from this cross-validation are plotted to demonstrate conformity to the original AUC curve, confirming the predictive modeling of this biomarker subset for HFpEF. Adapted from reference 66.

Figure 5

(TOP) Representative computations for ECM volume index based upon cMRI and gadolinium contrast taken from a referent normal subject. The signal intensities for the myocardium and blood pool (time sequence images shown as thumbnails-inset) were plotted against inversion time, and T1 curves of a region of interest are plotted whereby the shift in the T1 curves following motion and blood pool correction provided an ECM volume index. (BOTTOM) Using this approach, patients with increased ECM volume index could be identified, and in this example, identified an ECM volume index of approximately 23%. Using Cox regression models, it was demonstrated that increased ECM volume index was a significant risk factor for subsequent mortality as well as a major adverse event. Adapted from reference 78.