MPI tracers: an interview with Professor Kannan Krishnan, University of Washington

Thought leader series

Please can you give a brief introduction to your work?

I'm Kannan Krishnan, professor of Materials Science and Physics at the University of Washington. My main interest is magnetism and I work on a range of issues ranging from biomedical applications to information processing and storage.

What are you currently working on?

We are currently presenting the latest developments of the tracers used for MPI. The tracers are the key to MPI. They must be truly optimized not just for imaging performance, but also for their performance in the body; how they circulate in the body, how long for, where they go in the body and how they get cleared, all while simultaneously providing the best performance in terms of imaging quality.
We are therefore presenting the various aspects of developing these tracers for this performance. That includes developing the core, which is the central part of the tracer that gives you the best magnetic response.
It also includes putting some coatings on the outside of the tracers, which allow you to control their circulation time and where they get distributed, as well as allowing you to functionalize them so that they bind to the specific sites you want them to target.
I'm presenting some things on targeting and how we make the particles, as well as some aspects of the particle physics and how we optimize the particle response. Then, along with various other collaborators, we're presenting how we use the particles for imaging purposes.

Could you please tell us a little bit more about MPI?

MPI is a technique that purely images the tracer and not the body. Without the tracer, there is no MPI. As the tracer is so central to MPI, we want to develop it in such a way that provides the maximum signal and the best resolution possible.
The other important thing is that since MPI is tracer-based, all the contrast is coming from the tracer and nothing is coming from the body. You essentially get infinite contrast in the tracer, so the tracer really determines, to a large extent, how effective MPI will be.

What tracers are currently used in MPI?

Currently in MPI, the tracers have a central magnetic core and we need to optimize that core to provide the best response to the specific frequency of radiation being used. When you match those, you get the most optimal response.
We've done a lot of theoretical modelling and have identified a certain core size that gives the best magnetic response. According to our theory and models and now proven by our experiments, that core size is somewhere between 23 and 27 nanometers.
We can make these cores, ensuring that each is around that size, with a very narrow size distribution. Therefore, from our point of view, the first part of the tracer development is essentially done.

What is the next step?

The next step is to put the coatings on the outside, which largely determines where they go, how they move about in the body, how they get distributed and how they perform.
We are currently focusing on the coatings that can provide maximum circulation and we are even trying to tailor them to allow a circulation that may be maximal or minimal, depending on your interest.
We are also trying to put a material on the outside that will lead to a targeting ability for cancer, but we have to be able to do all of this in the body, without sacrificing the magnetic performance.

Currently, what are the main challenges with tracers?

The main issue with tracers is understanding the physics and the surface coatings, but ultimately, you need to put them into the body and for that, approval from the FDA or some other sort of regulatory body is required, which takes time.
This is challenging because you have to jump through various loopholes and hurdles to reach the point where you can actually put the tracer into the body. That’s the big challenge now – developing the tracer in conjunction with the people who make the hardware so that we are all ready at the same time to take it through the regulatory process.

What would an optimal tracer for MPI look like?

The fundamental idea here is to match your tracer to perform in the machines that are being developed by Philips and Bruker or other commercial partners.
Once the core has been sorted out, the next real question is what the biological or biomedical application will be?
If you want something for the first order low-hanging fruit, the application is cardiovascular imaging. You put the tracer in there and want to see how it flows through the blood, where it is getting blocked and so on.
For that, you need the tracer not to agglomerate and to circulate most efficiently and, depending on the application, you may want it to circulate for a single pass or multiple passes.
There is no single tracer that suits all needs; it depends on what people want to do with them in terms of the application they are using.
Finally, if you want to move towards molecular imaging, which is one of the areas we are working on, then you want a tracer that has a targeting moiety that binds to a specific target, so you can try to understand the local environment.
Another potential area is stem cell tracking, which would mean getting the tracer into cells or to bind to cells so it can be tracked. Therefore, the optimal tracer will evolve as the applications and the needs for the applications evolve.
However, somehow a consensus has to emerge among the community because a tracer has to be taken through FDA clearance. We can't just change the tracer on a daily basis and we have to get a consensus from the community on what the average required circulation time is. 
In my lab, we have developed the ability to tailor the tracer to any of these needs. Therefore, if you have a specific need and that's the general consensus in the community, we'll tailor it. In fact, we have already demonstrated a whole range that would be tailorable to specific applications.

What do you think the future holds for developing MPI?

In terms of developing MPI, the first order, low-hanging fruit is cardiovascular imaging. For that, we need a tracer that circulates in a reasonable amount of time (which would be determined by the radiologist) and then people can apply this to biomedical issues. You could see in the talks that we gave today that those problems are essentially solved.
The second question is whether you are going to apply this for molecular imaging, in which case we need to find out how we are going put something on the tracers that binds to a specific target. Again, we have started to move in that direction and we have some rather interesting results for targeting glioma cells and so on.
I think an optimal tracer will emerge in time, depending on the consensus of the larger medical community. In my lab, we are positioning ourselves in such a way that when that consensus emerges, we will have the optimal tracer at the same time.
Our vision is to have MPI hardware in every hospital and have the people who use it recognizing the value of the technology, which I believe is immense. It could do much better than technologies that use nuclear medicine and it should be seriously considered as replacement technology in that area because nuclear medicine is associated with all kinds of harmful effects.
We expect MPI to be used as a replacement for MRI applications and possibly for PET, so I think it has huge potential. That's what drives us and brings us to work every day and it is hopefully what we will see in the future.