Quantitative cardiovascular magnetic resonance for molecular imaging

Abstract

Cardiovascular magnetic resonance (CMR) molecular imaging aims to identify and map the expression of important biomarkers on a cellular scale utilizing contrast agents that are specifically targeted to the biochemical signatures of disease and are capable of generating sufficient image contrast. In some cases, the contrast agents may be designed to carry a drug payload or to be sensitive to important physiological factors, such as pH, temperature or oxygenation. In this review, examples will be presented that utilize a number of different molecular imaging quantification techniques, including measuring signal changes, calculating the area of contrast enhancement, mapping relaxation time changes or direct detection of contrast agents through multi-nuclear imaging or spectroscopy. The clinical application of CMR molecular imaging could offer far reaching benefits to patient populations, including early detection of therapeutic response, localizing ruptured atherosclerotic plaques, stratifying patients based on biochemical disease markers, tissue-specific drug delivery, confirmation and quantification of end-organ drug uptake, and noninvasive monitoring of disease recurrence. Eventually, such agents may play a leading role in reducing the human burden of cardiovascular disease, by providing early diagnosis, noninvasive monitoring and effective therapy with reduced side effects.

Figure 1

In vivo CMR molecular imaging of angiogenesis in atherosclerosis. Serial imaging of the aortic wall (arrow) of an atherosclerotic rabbit up to 2.5 hours post injection of α_νβ₃-targeted nanoparticles allows quantification of contrast agent uptake. The temporal evolution of the nanoparticle concentration in the tissue and the blood was used to determine the pharmacokinetic profiles of targeted vs nontargeted particles. Reprinted with permission from Neubauer, et al [49].

Figure 2

Quantitative molecular imaging can predict response to anti-angiogenic therapy. CMR molecular imaging of atherosclerotic rabbits treated with α_νβ₃-targeted nanoparticles carrying an anti-angiogenic drug, fumagillin, demonstrated signal enhancement in the aortic wall. Follow-up imaging with α_νβ₃-targeted nanoparticles was performed 7 days later to assess residual angiogenic activity in the vessel. Quantitation of enhancement at the time of treatment was related to the amount of drug delivered to the growing atherosclerotic plaques and correlated to the change in signal 7 days after treatment. Sections of the abdominal aorta with the highest signal enhancement at the time of α_νβ₃-targeted fumagillin nanoparticle treatment showed the greatest reduction in α_νβ₃-integrin expression assessed 1 week later. Reprinted with permission from Winter, et al [56].

Figure 3

CMR image enhancement with α_νβ₃-targeted nanoparticles correlates to the density of angiogenic microvessels. CMR molecular imaging of angiogenesis with α_νβ₃-targeted paramagnetic nanoparticles was directly related to histological measurement of microvessel density in atherosclerotic rabbits. The number of microvessels expressing both α_νβ₃-integrin and platelet/endothelial cell adhesion molecule (PECAM), a general vascular marker, were counted in aortic sections. The microvessel density was correlated in a logarithmic fashion (R² = 0.84) to the CMR signal enhancement observed after injection of α_νβ₃-targeted particles. Reprinted with permission from Winter, et al [55].

Figure 4

Molecular imaging of tissue factor expression on cultured smooth muscle cells. Cell cultures were incubated with tissue factor targeted nanoparticles (T), nontargeted nanoparticles (NT) or no nanoparticles (UT). Left: Spin-echo images of smooth muscle cell monolayers acquired at 1.5T reveal image enhancement for the well treated with targeted particles, but not the wells receiving nontargeted or no nanoparticles. Right: A maximum intensity projection through the 3 D stack of T1-weighted images acquired parallel to the cell monolayers demonstrates the sensitivity of this targeting method for detecting labeled cells. Reprinted with permission from Morawski, et al [74].

Figure 5

Direct quantitation of contrast agent binding utilizing ¹⁹F CMR and fibrin-targeted PFC nanoparticles. (a) Optical image ex vivo of a 5-mm cross section of a human carotid endarterectomy sample. This section showed moderate luminal narrowing as well as several atherosclerotic lesions. (b) A ¹⁹F projection image acquired at 4.7 T through the entire carotid artery sample shows high signal along the lumen due to nanoparticles bound to fibrin. (c) Concentration map of bound nanoparticles in the carotid sample. Reprinted with permission from Morawski, et al [74].

Figure 6

Mapping tissue uptake of VCAM-1 targeted PFC nanoparticles with ¹⁹F CMR. Multinuclear imaging of kidneys from atherosclerotic ApoE^-/- (top) and wild-type control (bottom) mice imaged at 11.7T. (A) Proton MR of kidney anatomy. (B) ¹⁹F CMR for direct detection of VCAM-1-targeted PFC nanoparticles. (C) A composite ¹H/¹⁹F image allows precise overlay of molecular biomarker of inflammation and anatomical detail. Reprinted with permission from Southworth, et al [76].

Figure 7

Quantitative ¹⁹F spectroscopy of angiogenesis in aortic valve disease. The ¹⁹F signal was utilized to quantify binding of nanoparticles to the valve leaflets from (A) a rabbit treated with α_νβ₃-targeted nanoparticles and (B) a rabbit treated with untargeted nanoparticles. Both nanoparticle formulations consisted of a crown ether core, which generates a single peak. The PFOB peaks originated from a reference standard utilized for quantification. Nanoparticle binding in the rabbit treated with targeted particles was much higher (Crown Ether/PFOB = 4.6) than the rabbit treated with nontargeted particles (Crown Ether/PFOB = 2.2). Reprinted with permission from Waters, et al [84].

Figure 8

Quantitative comparison of nanoparticle binding in valve leaflets. The volume of nanoparticles (in nanoliters) bound to the valves was calculated from the ¹⁹F signal. The valves treated with α_νβ₃-targeted particles displayed three times higher signal compared to the nontargeted formulation and twice the signal of valves with competitive inhibition of α_νβ₃-integrin binding. Minimal nanoparticle deposition occurred in non-atherosclerotic animals treated with targeted nanoparticles due to the lack of angiogenesis in the valve. Reprinted with permission from Waters, et al [84].