High-sensitivity in vivo contrast for ultra-low field magnetic resonance imaging using superparamagnetic iron oxide nanoparticles

Abstract

Magnetic resonance imaging (MRI) scanners operating at ultra-low magnetic fields (ULF; <10 mT) are uniquely positioned to reduce the cost and expand the clinical accessibility of MRI. A fundamental challenge for ULF MRI is obtaining high-contrast images without compromising acquisition sensitivity to the point that scan times become clinically unacceptable. Here, we demonstrate that the high magnetization of superparamagnetic iron oxide nanoparticles (SPIONs) at ULF makes possible relaxivity- and susceptibility-based effects unachievable with conventional contrast agents (CAs). We leverage these effects to acquire high-contrast images of SPIONs in a rat model with ULF MRI using short scan times. This work overcomes a key limitation of ULF MRI by enabling in vivo imaging of biocompatible CAs. These results open a new clinical translation pathway for ULF MRI and have broader implications for disease detection with low-field portable MRI scanners.

Fig. 1. The basis of high-sensitivity SPION imaging at ultra-low magnetic fields.

(A) Magnetization of 25-nm SPIONs (green), gadolinium CA (Gd-DTPA/Magnevist, blue), and water (red) as a function of magnetic field strength (B₀). (B) Magnetization as a function of magnetic field strength (B₀) in the ULF (<10 mT) regime for the materials shown in (A). Superparamagnetic materials, such as SPIONs, are highly magnetized even at ULF. Paramagnetic materials, such as CAs based on gadolinium, and body tissues (which typically have diamagnetic susceptibilities close to water) have absolute magnetizations that increase linearly with field strength. Curves in (A) and (B) were reproduced from data in (32, 53) and reflect the magnetic moment per kilogram of compound. (C) Highly magnetized SPIONs (brown) interact with nearby ¹H spins in water, shortening ¹H relaxation times, and causing susceptibility-based shifts in Larmor frequency.

Fig. 2. In vivo imaging at 6.5 mT of SPION biodistribution in a rat model.

(A) MRI scan of rat anatomy before CA injection. (B) MRI scan of rat 30 min after a tail vein injection of HS-PEG SPIONs at 5 mg/kg. Both three-dimensional (3D) bSSFP MRI datasets were acquired in 12.5-min acquisitions with tip angle α = 90° and a 2.0 mm × 1.6 mm × 5.9 mm voxel size. The 9 signal-containing slices of 11-slice datasets are shown in (A) and (B). Attention is drawn to the two central slices from the pre-injection dataset (orange outline) and post-injection dataset (blue outline). Outlines of the kidneys (pink), liver (yellow), and the bifurcation of the aorta and inferior vena cava (green) are provided for discussion in the main text. Field of view (FOV) in each slice is 155 mm × 73 mm. AU, arbitrary units.

Fig. 3. Susceptibility-based contrast at ultra-low magnetic fields.

(A) Simulated transverse magnetization of spins during bSSFP imaging. MRI signal magnitude is proportional to the transverse magnetization (M_xy) of ¹H nuclear spins, which is shown here as a function of frequency offset (the difference between RF pulse frequency and Larmor precession frequency). Spins not in the presence of SPIONs have zero Larmor frequency offset. As SPION concentration increases, nearby spins experience a susceptibility (χ) shift in Larmor frequency. Curves for RF pulse tip angles of α = 20° (red) and α = 90° (blue) are shown. The transverse magnetization is normalized by the fully relaxed longitudinal magnetization (M₀). T_R = 85 ms and T₁/T₂ = 1 in this model. (B) Phantom schematic. A small vial of HS-SPIONs at 300 μM (orange) is suspended in a larger vial of water (blue). The 6.5-mT static magnetic field of the scanner (B₀) is oriented perpendicular to the cylindrical vial axis. (C) Standard bSSFP MRI of the phantom shown in (B), acquired at 6.5 mT with a tip angle α = 90°. (D) Small tip angle bSSFP MRI of the same phantom, acquired with α = 20°. (E) Analytical calculation of the ¹H Larmor frequency shift expected in the water region of the phantom in (A) due to bulk magnetization of the adjacent SPION vial. (F) Signal magnitude predicted in a bSSFP MRI of ¹H spins shifted by frequencies depicted in (E) when α = 90°. (G) Signal magnitude predicted in a bSSFP MRI of ¹H spins shifted by frequencies depicted in (E) when α = 20°. Simulation data for the SPION vial are blacked out in (E) and (F) as well as (G) as a key assumption of the bSSFP trajectory model, that T_R ≪ T₂, does not hold in this region of the images. FOV in all images is 49 mm. The α = 90° and α = 20° images were acquired with 6.2- and 12.4-min scans, respectively.

Fig. 4. Sensitivity of CA imaging at ULF.

(A) Phantom schematic (left) shows vials of HS-COOH SPIONs at various concentrations (orange) in a larger water vial (blue). bSSFP MRI scans of the phantom with α = 90° (middle) and α = 20° (right) are also shown. (B) Phantom schematic (left) shows vials of Magnevist at various concentrations (purple) in a larger water vial (blue). bSSFP MRI scans of the phantom with α = 90° (middle) and α = 20° (right) are also shown. FOV in all images is 49 mm. The α = 20° SPION image was acquired with a 12.4-min scan. All other images were acquired with 6.2-min scans. AU, arbitrary units.

Fig. 5. Switchable susceptibility-based SPION contrast.

(A) Images of rat anatomy before CA injection acquired with bSSFP tip angles of α = 90° (top) and α = 20° (bottom). (B) MRI scans taken 30 min after a tail vein injection of HS-COOH SPIONs at 5 mg/kg with α = 90° (top) and α = 20° (bottom). The 5 central slices of 11-slice datasets are shown for each acquisition. Expanded images of individual slices from α = 90° (green outline) and α = 20° (purple outline) datasets are shown. FOV in each slice is 155 mm × 73 mm. The red dot below the liver indicates a region of “boosted” signal for further discussion in the text. The 3D datasets with α = 90° were acquired with 12.5-min scans. The 3D datasets with α = 20° were acquired with 25-min scans. AU, arbitrary units.