Imaging of infarct healing predicts left ventricular remodeling and evolution of heart failure: focus on protease activity

Today, state of the art care ensures that most patients with myocardial infarction survive the acute ischemic event. Often, reperfusion treatment is successful and timely, limiting the loss of myocardium. In that case, patients are discharged with a normal ejection fraction, and if atherosclerotic risk factors are controlled, the outcome can be favourable. Many patients are less fortunate; here, delayed or insufficient reperfusion results in prolonged ischemia and death of a significant portion of their heart muscle. Necrotic cell death initiates the wound healing process that parallels the response of the body to other sterile injuries. Mechanical stress imposed by the blood pressure in the left ventricular cavity, however; is unique to infarct healing as mechanic forces on the wound are considerable. Contrary to other injuries, for instance a bone fracture, no plaster cast can reduce tissue motion during repair. Hence, the efficacy of healing is important, and sub-optimal healing leads to weakened resistance to mechanical forces, infarct expansion, enhanced remodeling and finally heart failure. Because many infarct patients suffer from comorbidities such as diabetes, the quality of healing is frequently impaired and myocardial infarction is a dominant cause for heart failure.

The scope of the problem -- approximately every 25 seconds, an American will have a coronary event¹ -- has spawned vigorous research to better understand post MI remodeling. We use ACE inhibitors and beta blockers to reduce remodeling, and are still learning new aspects² of how these drugs work. Nevertheless, we could do better: with 280,000 deaths per year in the US, heart failure mortality is still unacceptably high¹. The desire to repair the heart has motivated work on stem cells, and recent discoveries on myocyte turn-over are fueling the hope that one day “regrown” myocytes can replace the lost contractile units. Since there will be no plaster cast for the heart, tweaking the body's inflammatory response to myocyte death and optimization of infarct healing could complement the efforts on regenerative strategies³.

There are two major aspects how imaging can empower above efforts. First, the ability to noninvasively study molecular and cellular biology provides an opportunity to understand and then therapeutically target key disease determinants. Why is that the case? We can avoid in vitro artifacts, follow the time course of imaging markers in their undisturbed environment, correlate molecular and cellular players to each other, to changes in heart function and anatomy (as done by Sahul et al.⁴), and to outcome⁵. Second, in parallel to driving the therapeutic discovery for more efficient means to attenuate left ventricular remodeling, we need to develop the tools to monitor therapeutic effects and identify patients at risk for post MI heart failure⁵. Such tools can accelerate research by using end points alternative to mortality, which could make clinical studies more efficient, faster and reduce R&D costs that are currently so high that pharmaceutical companies are shying away from cardiovascular drug development.

A variety of imaging approaches, spanning many healing biomarkers and all major imaging modalities, have been developed towards these goals (table). These include cell death⁶, upregulation of chemokines and adhesion molecules⁷, phagocytic^8–10, myeloperoxidase¹¹, protease^{4, 10} and transglutaminase activity¹², angiogenesis¹³, myofibroblasts¹⁴, collagen production¹⁵, and receptors that are targeted with current heart failure medication¹⁶.

Table.

process	target	time	probe	modality
cell death	phosphatidyl serine	hours	Annexin V-CLIO6	MRI (T2)
leukocyte recruitment	VCAM-1	days	18F-4V7	PET/CT
phagocytosis	neutrophils monocytes macrophages	days	microbubbles8 CLIO nanoparticles10 liposomes9	ultrasound, fluorescence molecular tomography, MRI (T2, F19 fluorine)
proteolysis	MMPs cathepsins	days to weeks	99mTc-Rp8054,18 MMPsense10 Prosense2,10	SPECT/CT, fluorescence molecular tomography
inflammation	myeloperoxidase	days	MPO-Gd11	MRI (T1)
matrix cross linking	transglutaminase Factor XIII	days	111Ind-FXIII12	SPECT
angiogenesis	integrin	2 weeks	18F-Galakto-RGD13	PET/CT
matrix	collagen	days	collagen-specific peptide EP-353315	MRI (T1)
myofibroblasts	integrin	3 & 8 weeks	RGD peptide RIP14	SPECT/CT

In their current study⁴, Dr. Sinusas' group takes its long-standing effort on imaging matrix metalloproteinases (MMPs) to the next level. Due to their central role in disease, proteases, and among them MMPs, are especially promising imaging targets. Some protease activity is likely needed during wound healing. Macrophage mobility in tissue depends on proteases, which are also crucial for the clearance of necrotic debris after ischemic tissue injury. However, if inflammation is exuberant and protease activity exceeds normal levels, the tissue is destabilized beyond integrity. Transgenic mice with increased MMP activity are prone to infarct rupture and post MI heart failure¹⁷. The Yale group has pioneered the use of nuclear probes that bind to the active site of MMPs, and hence report on the activity of the enzyme. The current work describes that previous data obtained in rodents¹⁸ translates into a clinically relevant large animal model, thus motivating a clinical trial. Importantly, the probe uptake correlated with left ventricular volume, which, in conjunction with the data on infarct rupture reviewed by Dr. Spinale¹⁷, suggests that higher MMP activity promotes infarct expansion and likely also side-to-side slippage of myocytes in the remote myocardium. Excessive protease activity may impair the integrity of the extracellular matrix, which then gives way to the intraventricular pressure leading to ventricular dilation^{17, 19}.

The elegant multimodality study correlated MMP activity, regional myocardial function and LV volumes in otherwise healthy pigs which had a comparable infarct size of about 22%. Despite this fairly homogeneous cohort, inter-individual differences in healing lead to data scatter and allowed significant correlation of the molecular signal and left ventricular size. Likely, there is even higher heterogeneity in patients, given their variability in age, infarct size, comorbidity and genetics. Importantly, patients -- unlike the pigs in the current study -- have pre-existing atherosclerosis, a chronic inflammatory disease associated with blood monocytosis. Clinical studies have shown that high numbers of protease-rich circulating monocytes correlate closely with outcome and the degree of heart failure²⁰. We have recently found that coronary ligation in apoE^−/− mice with atherosclerosis causes excessive monocyte recruitment, higher infarct protease activity, impaired resolution of inflammation and worse infarct healing¹⁰. In these mice, infarcts expanded and LV dilation increased. Taken together, these considerations support the hypothesis that the molecular imaging agent used by Sahul et al. in a porcine model will predict left ventricular remodeling and prognosis in patients after MI. Hopefully, we will soon read about a clinical protease imaging trial.