Duke | Duke Chemistry | Center for Molecular & Biomolecular Imaging
The Warren Research Group at Duke University

iMQC Imaging of Tumor Tissues Loaded with SPIONs


iMQC imaging can enhance regions of anisotropy on structured samples and help to visualize the local dipolar field created by superparamagentic iron oxide na-noparticles (SPIONs). Even when the SNR of each single iMQC image is poor, clean images that highlight regions of anisotropy can be obtained.

Magnetic resonance imaging is a highly sensitive technique, and was initially hailed as a means of making definitive noninvasive diagnoses. However, it has poor specificity, and relating image contrast to object contrast can be very difficult: many object parameters affect the image intensity, and their relative contributions are very dependent on the particular imaging technique used.

To increasing the contrast between normal and abnormal tissue, thereby enhancing clinical diagnosis of pathologic conditions, a large variety of contrast agents have been developed. Contrast agents are MR-detectable particles that are injected into the body, and can differentially be taken up by cancer cells. MRI contrast agents either act predominantly on T1 relaxation, which results in signal enhancement and ’positive’ contrast in a T1 weighted sequence, or on T2 relaxation, which results in signal reduction and ’negative’ contrast in a T2 weighted sequence.

The effect of the contrast agent on an iMQC signal is quite different. The effect of con-trast agents in iMQC sequence is not dominated by the T1 and T2 effects, but rather by induced changes in the local magnetic field, and susceptibility variations. With the stan-dard methods, we measure average susceptibility change over an entire voxel, but with iMQC methods, we measure susceptibility and magnetization density changes across the correlation distance, a specific, user-controlled subvoxel distance. The correlation dis-tance can be tuned by changing the strength of the gradient pulse, the correlation gra-dient, used in the standard CRAZED sequence. In a CRAZED experiment, the correlation gradient modulates the magnetization along a well defined direction. In the case of a homogeneous sample, this modulation will create a signal proportional to
fig1
This implies that when we combine the images obtained after selecting the correlation gradient along the three orthogonal directions, we will obtain a completely blank image.
fig1
However, the assumption that the magnetization is modulated along one direction does not hold for most imaging experiments. In the situation where local magnetic field gra-dients, as induced by susceptibility differences or paramagnetic particles, interfere with the correlation gradients, the assumption of a one dimensional modulation is invalid. Moreover, the combined image will not be null anymore, and can be used to detect anisotropy. The modulation of the magnetization, which creates the distant dipolar field between spins, is substantially perturbed by the presence of contrast agents such as SPI-ONs. The iron core of these nanoparticles creates long range field distortions, which directly affect the modulation of the magnetization introduced by the correlation gradients, and therefore the creation of the iMQC signal. One of our current project aims is to detect the loading of contrast agent on tumor tissues through iMQC imaging.

The kind of nanoparticle we use is a SPION coated with luteinizing hormone releasing hormone (SPION-LHRH). Compared with uncoated SPIONS, this new kind

of nanoparticle presents better specificity to the tumor region, since it binds directly to the LHRH receptor present in breast tumor cells. Once bounded to cancer cells, SPIONs are incorporated into the cancer cells through receptor mediated endocytosis. To detect the loading of nanoparticles in the tumor region, we collect three images, each with the correlation gradient pointing in orthogonal directions

We have demonstrated this technique on a breast tumor mouse model. Figure 2 shows that, though each single image has a poor signal to noise ratio (SNR), the combined im-age is a clean image, which cleanly highlights the loading of the contrast agent around the tumor tissue.
fig1
fig1



Website concerns - contact mike.conti@duke.edu