Notes from the Weissleder Lab: Exploring the Frontier of Precision Imaging for Nanotherapeutics

Massachusetts General Hospital’s Center for Systems Biology is one of the world’s leading research sites employing nanoparticle imaging in the investigation of biology and diseases for the purpose of bringing new diagnostic tests and new approaches to therapy.

At its helm, is one of the center’s radiologists, Ralph Weissleder, MD, PhD, who gave the plenary talk at the 3rd annual Nanomedicine for Imaging and Treatment Conference at Cedars Sinai Medical Center in Los Angeles on March 13. His address, “The Ins and Outs of Nanoparticle Imaging,” provided an intriguing window on the work of more than three dozen scientists—including biologists, chemists, physicists, informaticists, biomedical engineers, cardiologists, radiologists and an oncologist—who have the potential to deliver on the promise of precision medicine and perhaps one day change the way that radiologists image, diagnose and treat disease—most immediately, cancer.

“When we think about imaging and using nanoparticles for imaging, it usually boils down to five things,” Weissleder said. Nanoparticles are used, first, to target and find tumors via positive contrast; second, to target normal host cells to provide negative contrast; third, as companion-imaging nanoparticles to understand how emerging nanotherapies work; fourth, for designing sensors to extract biophysical information from cancers; and fifth, for in-vitro diagnostic applications, some of which use imaging.

Weissleder focused on four topics: the history of magnetic nanoparticles as they relate to imaging (a story intimately related to his own history); an FDA-approved agent about which little is known about how it works at the cellular level; the role of imaging in helping collaborators understand how a new therapeutic nanoparticle works (much differently than anticipated); and the quest to design better immunotherapeutics. “Because at the end of the day,” Weissleder said, “innate immune cells, immunocytes, are the major targets for whatever nanotherapeutics we inject.”

In the beginning

How a radiologist ended up leading a band of basic scientists into the intricacies of nanoparticle imaging is a story that began more than 35 years ago, south of the border. Weissleder was a medical resident, and a manufacturer had placed an early MRI scanner in Mexico.

“They said, ‘Go do something useful,’” Weissleder recalls. “We had no idea what to do, but it became very clear after we put the first few patients in the prototype scanner that there was only limited contrast. So we started thinking about how we could increase the contrast and obtain truly molecular information.”

Weissleder and team didn’t have to look far to find a solution: They hit upon the idea of grinding up refrigerator door magnets, thinking that they could improve contrast by effectively putting nano-sized magnets inside of big magnets. “While that experiment didn’t really work, it ultimately led to the first poly-crystalline injectable iron oxides,” he said. Weissleder and co-workers conducted the initial clinical trial, first in Mexico and then at Massachusetts General Hospital.

Since then, improvements have been made in structure (monocrystalline versus polycrystalline) and polymer technology, resulting in more suitable, targetable magnetic nanomaterials for MR imaging, ultimately leading to hybrids and the single compound left in the clinic today, ferumoxytol, approved for iron replacement in anemia but used by many for MR imaging.

The monocrystalline materials were designed to provide negative contrast, but researchers always had it in mind to target macrophages, Weissleder said: Early experiments in the 1990s were undertaken with the thought that if the materials were small enough and long-circulating, they would extravasate and be taken up by macrophages in lymph nodes. Where there were micro-metastases in lymph nodes, less material would be taken up, resulting in negative contrast detectable by MRI imaging.

“If we did the same experiment today, I could tell you about 20 reasons why all of this would fail,” Weissleder said. “We were naive in our understanding and nevertheless did the experiments; first in mice then in patients, where it actually worked spectacularly well. Today, we know that lymph node micro-metastases cause alterations in lymphatic flow and transport, limiting the amount of nanoparticle uptake in individual nodes. All of this is detectable by MR imaging.”

Subsequently, these materials were used as physiological probes to create angiogenesis maps,

permeability maps and cellular uptake maps. Most of these maps were based on mathematical modeling of relaxation differences observed after systemic administration of nanoparticles.

“This is about as much as we know about the biology,” he said, acknowledging the limits of their understanding. “Although the maps looked great, we had no way of really proving that they were actually correct. That was a little bit frustrating.”

The next frontier

That frustration led Weissleder’s lab in a different direction about 10 years ago—from macroscopic to microscopic optical imaging technologies—to get a better understanding of what actually happens when the nanomaterials are injected systemically: where do they go and what effects do they have?

One of the first discoveries Weissleder’s lab made—by injecting red fluorescent nanoparticles into a green fluorescent mouse tumor with blue vasculature—was that there were a surprising number of macrophages present in the mouse cancers. “Even if one looks inside the cancers in the greenest portion there can be

extensive numbers of macrophages almost as many as the actual cancer cells,” Weissleder said. “They are fairly stationary in there, so not a lot of movement, unlike many lymphocytes”

Their next step was to undertake more detailed experiments in which mice were injected with differently colored nanoparticles and drug “payloads,” and then imaged over time to track where and to which cell types the materials distribute.

The lab developed numerous technological advances in order to image at the highest spatial resolution in live animals. For example they developed “moving z stages” and adhesive or compression devices that move with respiration so that the target does not come out of focus. “We can now image cancers in virtually all orthotopic locations,” he said.

The ferumoxytol mystery

One of the first clinical questions Weissleder and team asked was whether MRI could be used in the clinic to pre-select patients who would be most likely to respond positively to a nanotherapeutic formulation. Rather than taking a therapeutic nanoparticle and trying to convert it into an imaging agent—a task that faced many hurdles including cost and FDA approval—they began by investigating the in-vivo cellular distribution and the kinetics of ferumoxytol, and how well that correlated with the distribution of a “model” therapeutic pegylated nanoparticle.

Weissleder showed an accelerated two-hour loop of one of the first experiments involving a tumor in a live mouse to calculate how many nanoparticles would accumulate inside cells. After two minutes, the injected materials already were absorbed inside macrophages near tumor vessels. “These are unbelievably rapid effects,” Weissleder noted. “We have now seen this in many different orthotopic tumors.”

At 30, 60, 90 and 120 minutes, there appeared to be perfect correlation between the ferumoxyl and the therapeutic nanoparticle distribution, but at later time points and higher resolution, differences emerged. More ferumoxyl than the PLGA-pegged nanoparticle could be seen in the macrophages.

“It turns out that one-third of the material is actually in tumor cells,” Weissleder said. “Two thirds is in host cells, and in those, half are in the professional phagocytes, mainly macrophages and neutrophils, and the remaining one third accumulate in other cells including fibroblasts, endothelial cells and epithelial cells. That was quite surprising, because before, the prevailing paradigm was that this material ends up almost exclusively in macrophages, and it certainly does not.”

[[{"fid":"21273","view_mode":"media_original","type":"media","attributes":{"height":717,"width":1000,"style":"width: 610px; height: 437px; border-width: 1px; border-style: solid;","alt":" - Weissleder-fig1","class":"media-element file-media-original"}}]]
Figure 1. Schematic diagram illustrates how magnetic nanoparticles distribute to tumor microenvironments as a function of time. The relaxivity changes (from red to blue) upon cellular uptake. Based on temporal relaxivity changes, angiogenesis (bottom left) or EPR maps (bottom right) can be constructed. Image courtesy of Ralph Weissleder, MD.

Imaging and patient selection

To answer the clinical question—whether imaging can help in patient selection—Weissleder’s team performed MRI on mice, administered an IV dose of magnetic nanoparticle and then measured the accumulation of tumoral magnetic nanoparticles mediated by the enhanced permeability and retention (EPR) effect.

Difference maps were made for each mouse and the animals were treated with a paclitaxel therapeutic nanoparticle. Tremendous differences in response were seen in the mice in which the tumors took up either little or a lot of the magnetic nanoparticles.

“To distill this down, yes, MRI can be used to select patients,” Weissleder said. “Those patients that have higher ferumoxytol accumulation are expected to do much better when given a therapeutic nanoparticle.”

This now enables the team to go back into the clinic, give ferumoxytol to patients and analyze imaging to explore which cancer patients would do well with a nanotherapeutic drug. Weissleder’s and other labs also are formulating PET agents out of materials where the quantitation of nanoparticle uptake into tumor may be simpler in clinic than MRI analysis.

TAM controversy

Weissleder also shed some light on the most important questions perplexing the field of macrophage therapeutics: which sub-type of phagocytic cells are essential and which are harmful; if most new therapies have a dual influence on tumors and macrophages, how exactly does it all work; and, in directly drugging macrophages, should we eradicate them, activate them or repolarize them?

“These are all important questions and it will take some time to sort out the answers, but I will show you how we think about some of them,” Weissleder said.

He began with a clinical observation: tumor-associated macrophages are generally bad for you. “In several large studies (it was shown in breast, colorectal cancers, lymphoma), whenever there are inflammatory cancers, the more macrophages there are, the worse these patients do, it was this original observation that subsequently led other investigators to say, ‘Well, since macrophages are bad for you, why don’t we just eradicate them with taxol.’”

Not until more recently did it become clear that there is “a yin and a yang.” Weissleder said: “Sometimes macrophages are good for you and sometimes macrophages are bad for you.” He shared a table that listed instances in which macrophages limit therapy response and examples where they enhance it.

“Sometimes you see the same drug in both columns, so it is complicated,” he noted, “but for those of you who work in nanotherapeutics, macrophages are probably a good thing for the delivery and slow release [of therapeutics]. That is a sort of global view.”

The study of macrophages is a complex new field that researchers are just beginning to understand, including the fact that there are different origins of macrophages—or at least dual origins of tissue resident in macrophages, Weissleder said.

Weissleder also pointed out that there is a raging debate over the so-called M1–M2 phenotype paradigm, which holds that M1 macrophages are mostly responsible for killing bacteria and M2 macrophages promote cancers. “We hope that ultimately with additional research we will be able to decipher some of these subtypes through molecular markers and be able to target them directly,” he said.

His take-home message was total macrophage ablation is probably not desirable for several reasons: they serve as slow-release depots for nanoparticles, are likely needed for anti-tumor immunity, and removing them would increase infection risks and side effects in normal tissues.

The MGH lab

Outstanding questions from Weissleder’s perspective are related to both biology and therapeutics. “Everything that I have shown you up to now is in a mouse;

We don’t know how many subtypes of tumor associated macrophages (TAMs) there are in our patients and in different cancers,” he said. “We don’t know what their function is and we don’t know how they influence metastases formation.”

On the therapeutics side, researchers have yet to discover whether TAMs are relevant immunotherapeutic targets in humans. “We know that TAM targeting agents can control cancer progression in mice. Now it will be important to test whether, and if yes how, these findings can be translated to cancer patients. Some clinical trials are underway.” he said.

While he focused on its work in therapeutics, Weissleder’s lab also investigates next-generation diagnostics, employing imaging and “other gadgets” to understand how human biology works in diseases and to develop new diagnostic tests. “We have gone from macroscopic all the way to microscopic imaging to understand how nanomaterials actually work,” he said. “Now that we know better how they work, we are able to translate this knowledge back to clinical imaging.”