Читать книгу Magnetic Nanoparticles in Human Health and Medicine - Группа авторов - Страница 30

The use of the MNPs as contrast agents in MRI was the first application for which Food and Drug Administration (FDA) and European Medicines Agency (EMA) approved their use in clinical practice (see Table 2.1). MRI is based on the Nuclear Magnetic Resonance (NMR) of the protons constituting the nucleus of the hydrogen atoms. In the human body, protons represent 63% of the body mass. NMR can occur when the proton's nuclear magnetic momentum, aligned parallel to a static external magnetic field, reverses its orientation to an antiparallel one, by resonantly absorbing a certain amount of external energy. More precisely, the energy of an external RF electromagnetic field, transverse to the static magnetic field and irradiating the proton‐containing sample, has to be equal to the energy difference between the two orientation states. Translating the space in terms of the values of the magnetic field, B, allows one to get information about the position of the protons, by using the measured resonance frequencies.

Table 2.1 Examples of the MNPs clinically approved or in the phase of clinical trials.

Commercial name	Formulation	Application	Status
Feridex (Ferumoxide)	Iron oxide coated with dextran	MRI	FDA approval discontinued in 2008
Combidex (Ferumoxtran)	Ultrasmall iron oxide coated with low Mw dextran	MRI	Approval in Europe, withdrawal in 2007
Resovist (Ferucarbotran)	Iron oxide NPs coated with carboxyldextran	MRI	Approval in Europe, production abandoned in 2009
Gastromark (Ferumoxsil)	Silicone‐coated iron oxide NPs	MRI	FDA approval, discontinued in 2012
Feraheme (Ferumoxytol)	Iron oxide coated Silicone‐coated iron oxide NPs with polyglucose sorbitol carboxymethylether	Iron replacement therapy in patients with chronic kidney failure, MRICNS imaging, macrophage imaging, blood pool agent, cellular labeling, lymph node imaging	FDA approval withdrawn from EU market
Abdoscan	Sulfonated poly(styrene‐divinylbenzene) copolymer	Oral gastrointestinal imaging	Clinical trial
Nanotherm	Iron oxide NPs with aminosilane coating	Hyperthermia in solid tumors	Clinical trial
Magtrace Sienna+	Superparamagnetic iron oxide particles	Sentinel lymph node mapping	Clinical trials

In MRI, the measurements are performed in pulses of RF electromagnetic field, which has the “role” to perturb the orientation states of the nuclear spins. The perturbed orientation will relax by the end of the pulse, a phenomenon that is known as spin relaxation. There are two mechanisms that are responsible for this phenomenon: the spin–lattice relaxation and the spin–spin relaxation.

The spin–lattice relaxation process is characterized by the time interval (time constant) in which the magnetization along the direction of the static magnetic field decreases to zero, when the static field is inactivated. This time constant is called the longitudinal relaxation time and usually is denoted as T₁. On the other hand, the spin–spin relaxation mechanism is characterized by the time interval (also called transverse relaxation time, T₂) in which the magnetization in the direction perpendicular to the external magnetic field B₀ decreases to zero. The spin–spin relaxation process is also strongly correlated with the existence of inhomogeneities in the magnetic field the protons feel that leads to a decorrelation in their rotation frequencies around the axis of B₀.

Human tissues have different values of both relaxation times due to their different proton concentrations and structure (environment). These differences are responsible for the “quality” of the images obtained in NMR. However, there are many situations in which these differences are small. In this case, a reduced contrast NMR image will be recorded. Usually, even for the same tissue, there can be differences in the values of T₁ and T₂ relaxation times; T₁ being usually longer than T₂. As such, it becomes possible to record either T₁ or T₂ weighted MRI images. The T₁‐weighted NMR images are obtained by reducing the repetition time (i.e. the time between two successive pulses), while the T₂‐weighted MRI images are recorded by reducing the echo time (the time after which the magnetic resonance signal is recorded).

Over time, it was demonstrated that the contrast of MRI images can be increased by using different contrast agents (CAs). The first CAs that have been employed are the paramagnetic ions possessing large magnetic momentum (Mn²⁺, Gd³⁺). Their effect on the effective relaxation time is measured in terms of relaxivity, according to the following equation:

where i is either 1 or 2, and R_i,obs is the observed relaxation rate which is defined as the reverse of the effective relaxation time T_i,eff. T_i is the relaxation time in the absence of any CA, c is the CA concentration, and r_i is the relaxivity, which is characteristic of CA. Higher the r_i values are, higher the effectiveness of the CA, meaning that a smaller concentration of CA leads to measurable effects. The effects of the CA on the two relaxivities are very often completely different. As such, CAs that reduce the T₁ values (high r₁) and increase the signal in T₁‐weighted images are called positive CAs, whereas CAs that reduce the T₂ values and subsequently reduce the intensity of the signal in T₂‐weighted images are called negative CAs.

The mechanism by which CAs influence the relaxation times is related to the value of their magnetic moment and the time spent by the protons (mostly belonging to water molecules) in their proximity (retention time). Thus, paramagnetic ions with large magnetic moments (Mn²⁺, Gd³⁺, Dy³⁺) allow a close interaction of the ions with water molecules present in the first hydration sphere, leading to an increase of r₁ relaxivity values. These paramagnetic ions are usually administrated in the form of ion complexes that have the role to decrease their mobility and to increase their relaxivities. The phenomenon is known as Proton Relaxation Enhancement (PRE). These ions are known as T₁‐positive CAs.

On the other hand, the advancement in the synthesis of different classes of MNPs with improved magnetic properties has represented a major discovery in the field of CAs. MNPs have huge values of their magnetic moments, with respect to paramagnetic ions, influencing the relaxation behavior of water molecules beyond the first hydration layer. Nevertheless, their coatings, used for improving their biocompatibility, impede a direct contact with water molecules. For these reasons, the MNPs are mostly used as negative T₂‐CAs. The first commercially available MNPs approved for clinical applications as CAs in MRI are Resovist and Feridex (Table 2.1). They consist of SPIONs of small magnetic cores (approximately ~10 nm) obtained by the coprecipitation method. Their r₂ values are lower than 100 s⁻¹ mM⁻¹. The relaxivity values are directly proportional to the square of MNPs magnetic moment, which in turn is proportional to their magnetization saturation (Ms) and volume. Therefore, MNPs with larger volumes and Ms values present several fold higher relaxivities. The use of thermal decomposition methods led to the synthesis of MNPs with improved crystallinity. This increases their magnetic moments as well as their relaxivities. However, monocrystalline MNPs with sizes higher than 20 nm are difficult to obtain because they involve laborious seed‐mediated growth synthesis methods. Moreover, the increase of MNP size can induce their transition into the so‐called “blocked state” in which they will have a nonzero remanent magnetization (Mr – induced magnetization remaining at zero magnetic field). A direct consequence of this nonzero Mr will be the occurrence of dipolar interaction and MNPs' self‐aggregation that make them less suitable for biomedical applications. For example, a 20 nm diameter appears to be an upper limit in the case of MNPs applications such as MRI CAs.

Another approach for improving the relaxivity properties of the MNPs is to include them in preformed aggregates or clusters of controlled structure, shape, and size. The existence of a cluster creates a stronger magnetic field gradient around it, thereby, changing the relaxation rate R₂ and the relaxivity r₂. However, the experimental results obtained for these types of structures show that the cluster size and MNPs composition influence the transverse relaxation rate and relaxivity in a nonmonotonous manner (Kostopoulou et al. 2014). With increasing the size and the number of clustered MNPs, the transverse relaxation rate increases, and this regime is called the “motional average regime.” Once the size limit is reached, the water molecules feel a constant magnetic field during their relaxation. This regime is called the “static dephasing regime” and determines the maximum relaxivity limit. If the MNPs' size is further increased, the relaxivity decreases and the measured relaxation rate depends on the echo time. This regime is called “echo limited regime.”

Interestingly, in the case of longitudinal relaxivity, it has been observed that its value tends to decrease with increasing the size of the cluster. This effect is explained by the reduction of the surface accessible for water molecules as the size of the cluster increases.

In fact, the accessibility of water molecules to the magnetic core is a key factor in determining the effect of MNPs on relaxivities. This means, on the one hand, that the optimum coating allows for sufficient diffusion of water molecules to affect the relaxivity of protons and, on the other hand, it assures a rapid exchange for a maximum number of water molecules near the CAs (Laurent et al. 2008). The diffusion of water molecules around the CA as a function of the coating might be assessed by measuring the NMR dispersion (NMRD) profile. In this type of measurement, the relaxivity r₁ is measured as a function of the magnetic resonance frequency (which is determined by the value of the static magnetic field B₀). It also facilitates the acquisition of the frequency or the value of B₀ at which the CA is the most effective. A typical NMRD profile of magnetite MNPs is presented in Figure 2.1.

Figure 2.1 Example of a NMRD profile for a colloidal suspension of SPIONs showing the evolution of the r₁ proton relaxivity with the applied external magnetic field/frequency.

Source: Reprinted with permission from Laurent et al. (2008). Copyright 2019 American Chemical Society.

By fitting the experimental data with adequate theories, various parameters can be extracted, such as R_max, the maximal relaxivity, M_s of the MNPs, t_D, water diffusion correlation time, and MNPs' anisotropy energy (Muller et al. 2005).

But the most common application of MNPs, like MRI CAs, is the early detection and diagnosis of cancer. The MNPs' effectiveness in cancer–tumor detection is mainly dictated by their fate after they are administrated in the systemic circulation.

There are three main mechanisms underlying the biodistribution of MNPs. First, the MNPs are rapidly captured by the Mononuclear Phagocyte System (MPS), also known as Reticuloendothelial System (RES). MPS is composed of the liver, spleen, lymph nodes, and bone marrow and leads to MNPs accumulation in these organs (Weissleder et al. 1990). Their accumulation in the liver's Kupffer cells will make the liver tissue to appear dark in the T₂‐weighted images. However, in the presence of a liver tumor, deficient phagocytic activity will be highlighted as a bright spot because MNPs do not accumulate in the tumor cells. Based on the same mechanism, normal lymph nodes will become dark in T₂‐weighted images, and the lymph nodes produced by the metastasis will remain white and will be easily identified in the MRI pictures (Figures 2.2 and 2.3) (Motomura et al. 2011).

Figure 2.2 (a) CT lymphography demonstrated a sentinel node (arrow). (b) The corresponding node was identified on T_2*‐weighted axial MR imaging (arrow). The node showed high signal intensity before the administration of superparamagnetic iron oxide (SPIO). (c) After the administration of SPIO, the node showed strong SPIO enhancement and was diagnosed as benign (arrow). (d) Histologic findings confirmed it as benign.

Source: Reproduced with permission from Motomura et al. (2011). Copyright © 2011, Springer Nature DOI‐https://doi.org/10.1245/s10434‐011‐1710‐7.

Figure 2.3 (a) CT lymphography demonstrated a sentinel node (arrow). (b) The corresponding node was identified on T_2*‐weighted axial MR imaging (arrow). The node showed high signal intensity before the administration of superparamagnetic iron oxide (SPIO). (c) After the administration of SPIO, the node showed no SPIO enhancement and was diagnosed as malignant (arrow). (d) Histologic findings confirmed it as malignant. This node was almost entirely replaced by metastatic tissue (arrowheads).

Source: Reproduced with permission from Motomura et al. (2011). Copyright © 2011, Springer Nature. DOI‐https://doi.org/10.1245/s10434‐011‐1710‐7.

The development of these applications led to the approval of the first MNPs that were used as contrast agents (Feridex and Resovist, see Table 2.1) in Europe and USA. In the organs not associated with MPS, or in the case of MNPs that are functionalized such as they escape the MPS, the Enhanced Permeability and Retention effect (EPR) in solid tumors plays an important role in their penetration into tumor tissue. Moreover, due to the lack of functional lymphatic vessels in malign tumors, the MNPs can accumulate in these tumors. This accumulation can be visualized in dark contrast images (Jain and Stylianopoulos 2010). Another approach for increasing the accumulation of MNPs in tumor tissues is to functionalize them with ligands targeting tumor markers like vascular or epithelial growth factors, α_vβ₃ integrins expressed by the endothelial cells in tumor vessels, folate receptors, and transferrin receptors. Apart from their use as CAs for tumor detection, MNPs in conjunction with MRI were used in MR angiography for the detection of cardiovascular diseases, vascular abnormalities, and inflammations (Nahrendorf et al. 2008).

Another important application of MNPs in MRI is the monitoring of cell therapies by cell tracking. In order to generate high‐contrast MRI images, the tracked cells are loaded with MNPs. Large numbers of MNPs can be loaded into the cells by electroporation, magnetofection, or cell‐penetrating peptides. In this manner, single cell detection can be achieved in vitro. Moreover, recent studies showed that the MNPs accumulate in lysosomes without affecting the cell functions (Ou et al. 2020). Several studies have shown that the delivery of antigen‐specific cytotoxic T‐lymphocyte, natural killer cells, and dendritic cells to the tumor or regional lymph nodes, could be monitored in vivo in real time (de Vries et al. 2005). Another important advantage of this type of approach is the possibility to monitor and to identify individuals who do not respond to the therapy. Other important applications of the combined use of MNPs and MRI are based on the observation that the MNPs could be taken up by monocytes and macrophages involved in the inflammation response; this provides information in different pathological processes such as atherosclerosis, pancreatic islet inflammation, or cardiac allograft rejection (Tong et al. 2019).

The development of a nanoplatform able to create dual T₁ − T₂ contrast agents would represent a great advancement for medical applications of MNPs. This implies to create a CA with a high r₁ but, in the same time, with a low ratio r₂/r₁ (close to 1) (Blanco‐Andujar et al. 2016). Several strategies have been tested so far: doping the MNPs with T₁ ions, attaching T₁ ions on the surface of MNPs, and elaboration of core‐shell nanostructures having a r₂/r₁ ratio close to 1 but with the cost of a low r₂ (Xiao et al. 2014). Core@shell structures led to MNPs with higher r₂ values, especially if the distance between the core and the shell is increased by adding a nonmagnetic layer (SiO₂). Recent studies show that, in a case of a dual core@shell nanostructure, by increasing the thickness of the nonmagnetic layer, the r₂/r₁ ratio decreased. Very interestingly, the r₂ was quite high, reaching a value of 312 m M⁻¹ s⁻¹ (Yang et al. 2015a). This finding demonstrates the huge potential of these nanostructures in MRI applications.