Powered by SearchMedica

JUNE 2003

Innovative agents boost molecular MR's sensitivity

Nanoparticles open doors to earlier cancer detection

By: Catherine Carrington

The gaze of researchers looking for clues to the future of molecular imaging increasingly falls on MR. Through a combination of sophisticated new contrast agents and changing technology, MR is making impressive advances in tracking gene expression, identifying the functional and metabolic changes that precede anatomic evidence of disease, gauging drug delivery and effectiveness, and monitoring basic cellular processes.

Several qualities make MR appealing for molecular imaging. Among the most important is its superb spatial resolution. Research magnets with field strengths exceeding 9T permit MR to achieve a spatial resolution in the range of 10 to 15 microns, the size of a single cell.

The chief weakness of MR is its low sensitivity for detecting metabolites and other molecules-at least 1000 times lower than nuclear imaging techniques. Researchers are overcoming MR's shortcomings with new contrast agents that amplify the power of the MR signal by several orders of magnitude or switch themselves on and off when targeted molecular events are encountered.

Monocrystalline iron oxide nanoparticles, or MIONs, were used in some of the earliest studies showing MR's potential for noninvasive imaging of transgene expression. Today, they and their offspring show promise in the identification of cancer in its earliest stages. MIONs are essentially iron oxide superparamagnetic nanoparticles that are wrapped in dextran.

In groundbreaking work led by Dr. Ralph Weissleder, director of the Center for Molecular Imaging Research (CIMR) at Massachusetts General Hospital, MIONs were hitched to the protein transferrin, enabling them to lock onto and shuttle inside tumor cells induced to display an overabundance of engineered transferrin receptors.

The next generation has improved on MIONs in important ways. Cross-linked iron oxide nanoparticles, or CLIOs, are much more stable than MIONs. Transferrin molecules can be attached to each iron oxide particle with chemical linkers, increasing the particle's affinity for the transferrin receptor and making MR about 16 times as effective in detecting changes in signal intensity. The addition of amine groups during dextran cross-linking makes it possible to attach a wide variety of homing molecules to the particle's exterior-not just transferrin, but antibodies, snippets of DNA, and other compounds. This advance has opened up entirely new avenues of research.

The CIMR team is exploring the use of CLIOs for detecting the merest hint of cancer-not the tumor itself, but the messenger RNA that relays its genetic instructions. In laboratory experiments, CLIOs have been attached to multiple sets of DNA building blocks, or oligonucleotides, each of which matches up with a complementary set of building blocks on the cancer cell's messenger RNA. Additional building blocks on that piece of messenger RNA in turn match up with other CLIOs, in a process that is repeated over and over. The final result is a large aggregate of nanoparticles easily detected by MR.

It's tempting to envision using messenger RNA-targeted CLIOs and a tabletop nuclear MR machine to detect cancer-and susceptibility to specific cancer drugs-using a single drop of blood, said James P. Basilion, Ph.D., an assistant professor of radiology at Harvard Medical School and codirector of the National Foundation for Cancer Research-Center for Molecular Analysis and Imaging at CIMR.

"That's one of the avenues that I think holds a lot of promise: the ability to use CLIOs to detect low levels-atomic levels-of mRNA," he said.

Teaming up with pathologists, molecular imagers have also found that in more than one third of women with breast cancer, tumor cells express about five times as many transferrin receptors as do normal cells. They are exploring whether transferrin-labeled CLIOs will enable MR to detect these tumors when they're still very small.

"Looking for targets that are overexpressed in diseases and then targeting with MR or other probes will prove to be a very useful diagnostic tool," Basilion said. "The more sensitive our imaging tools and the earlier we find these tumors, the more impact our therapeutic interventions can have."

LIPID-COATED PERFLUOROCARBONS

Researchers at Washington University in St. Louis are using a different MR contrast agent to target cell surface receptors in blood vessels. Dr. Gregory Lanza and Dr. Samuel Wickline have co-invented a nanoparticle composed of a perfluorocarbon emulsion coated with a layer of lipid. Into the lipid outer layer they can incorporate hundreds of homing molecules, such as antibodies, peptides, or even better, peptidomimetics. These organic molecules mimic short protein fragments but are so small that 200 to 300 can squeeze onto the surface of each nanoparticle, ensuring that the contrast agent adheres to several cell surface receptors at the same time and stays in place long enough to be imaged.

Also linked to the lipid layer of each nanoparticle are up to 90,000 molecules of gadolinium-DTPA, a payload large enough to overcome MR's partial volume effects and enable detection of the targeted cell. The perfluorocarbon nanoparticles open new doors in both cardiovascular and oncologic imaging. The particles may be able to identify dangerous atherosclerotic plaque, for example, by targeting the fibrin found in the blood clots that form on the plaque's ruptured surface.

The researchers are also developing ways to detect angiogenesis, a process that plays a key role in both atherosclerosis and cancer. By targeting a protein dubbed alpha v beta 3-integrin, it is possible to detect the immature blood vessels that characterize angiogenesis (see Molecular Imaging Outlook April 2003, p. 6). In atherosclerosis, these vessels grow from the wall of the blood vessel into the developing plaque, feeding its development with a toxic soup of cholesterol, fats, inflammatory cells, and nutrients.

At the annual scientific meeting of the American Heart Association held in November, a Washington University team reported on research showing that MRI was able to detect early aortic atherosclerosis in rabbits, using a 1.5T scanner and a perfluorocarbon nanoparticle labeled with a peptidomimetic targeted to alpha v beta 3-integrin.

A combination of fibrin- and angiogenesis-targeted MR imaging could be used to identify not only ruptured plaques that require immediate treatment, but also inflamed, developing plaques whose rupture might be prevented through delivery of drugs embedded in the nanoparticle's lipid surface.

"We're hoping that we can not only target where those plaques are, but also bring therapy to cool them down, while we wait for more long-term therapies like statins to take effect," said Lanza, an assistant professor of medicine and bioengineering at Washington.

The ability to detect angiogenic blood vessels could also play a key role in identifying tumors at very early stages, when they are both vulnerable and dangerous. As a direct measure of tumor vascularity, alpha v beta 3-integrin is an early indicator of oncogenesis and provides a measure of the newborn cancer's metastatic potential, according to Wickline, a professor of medicine, physics, and biomedical engineering at Washington.

One potential application is in the detection of tiny pockets of metastatic cells. After a primary tumor is removed, these once-dormant cells become active, prompting the development of new blood vessels to meet their metabolic needs. Lanza believes that identifying the blood vessels-and, thus, the tumors-at this early stage could improve the effectiveness of therapy.

Wickline and Lanza are attempting to determine if gadolinium-labeled nanoparticles can be also labeled with technetium-99m. They envision cancer and cardiovascular applications using MR and SPECT to precisely localize and evaluate cardiovascular plaques and early cancers. The studies would be combined using fusion software so both anatomic MR information and functional SPECT data can be viewed simultaneously on a single set of images. The researchers are also testing whether antifibrin agents labeled with gadolinium and Tc-99m can be used to image and treat vulnerable plaques.

'SMART' CONTRAST AGENTS

Although sophisticated new MR contrast agents often go after similar molecular imaging targets, "smart" agents do their work with special flair, remaining undetected until arriving at the job site. Smart agents cage gadolinium within the delivery molecule, shielding it from water protons and rendering it silent until the intended target-an enzyme or cellular ion, for example-opens the cage door. Then, the interaction between water and gadolinium results in changes in T1 relaxation that can be detected by MR.

The best known example of smart contrast agents is EgadMe. Thomas J. Meade, Ph.D., a professor of biochemistry and radiology at Northwestern University, spearheaded the development of EgadMe, which he described at the 2001 annual meeting of the International Society for Magnetic Resonance in Medicine. EgadMe consists of chelated gadolinium caged by a galactopyranose molecule. The cage door is removed only when EgadMe comes in contact with a β-galactosidase enzyme.

Meade and his colleagues have since moved far beyond EgadMe, refining the approach in ways that may permit imaging of the most basic processes of human disease. Like the Washington University group, they have turned attention to MRI of angiogenesis. Rather than targeting integrin, however, they are developing a smart contrast agent caged by a peptide that can be removed only by matrix metalloproteinases. These are the enzymes that digest proteins separating tumor cells from capillaries.

Meade's group is also focusing on work that may enable scientists to observe cellular processes as they turn off and on. To accomplish this, they are building contrast agents whose cage doors are not removed but instead opened and closed in a reversible process. In one example, the shape of Gd-DPTA changes in response to the concentration of calcium, an intracellular ion that plays an important role in cell signaling, regulation, and metabolism. Once the molecule changes shape, water can come in contact with gadolinium, with a detectable reduction in T1 relaxation.

In an extension of this model, Meade's team has begun to build a contrast agent that can track the activity of kinases, as they remove and add phosphate "doors" on the molecular cage. Kinase phosphorylation and dephosphorylation play key roles in cell signaling throughout the body. If development of such a contrast agent succeeds, it would represent a huge leap beyond what was possible with EgadMe, according to Meade.

"β-galactosidase is great for tracking gene expression and cell lineage. With the kinase family, we're talking about human disease," he said.

One challenge that has dogged smart contrast agents for years is their inability to cross cell membranes. The Northwestern researchers have found that linking a transport compound to DPTA-a backbone of smart contrast agents-enables it to cross the blood-brain barrier after injection into the tail veins of rats. The compound surprised researchers by binding to plaques that characterize Alzheimer's disease, raising the possibility of a new way to render an early diagnosis. The next step is to incorporate the transport molecule into a fully functional smart contrast agent.