Valentina Mazzoli on Muscle, Aging, and the ‘Magic’ of Imaging

Valentina Mazzoli, PhD, is an assistant professor of radiology at NYU Grossman School of Medicine and a scientist with the Center for Advanced Imaging Innovation and Research at NYU Langone Health. Dr. Mazzoli specializes in the development of MRI approaches to characterizing muscle tissue. She is currently leading an investigation into quantitative assessment of early structural and functional changes in aging skeletal muscle supported by the National Institute on Aging. Our conversation was edited for clarity and length.

How would you describe the main focus of your research?

I’m very interested in using MRI to study skeletal muscle—I was always fascinated by that tissue. One of the main focuses of my research right now is on trying to identify ways to study muscle aging: trying to understand who is going to age well and who will not. I like to work across the full range of imaging, from technology development all the way to translation, so I enjoy all the things in between.

At our research center, muscle imaging is a bit of a unique area of focus. Does this reflect the fact that it’s a relatively unique area of application in MRI research at large?

There aren’t a lot of people working in this field—it’s a small community and we pretty much all know each other—but it has got a lot of attention lately, especially in relation to aging. People have started to realize that, although aging of the mind is a serious problem, aging of the body is a serious problem, too. So, there are more and more research groups focusing now on noninvasive ways of assessing muscle quality, and not only with MRI.

What do you think makes MRI a promising technology for the study and measurement of musculoskeletal muscle?

The way I see it, the research we and other groups are doing in this area aims at coming up with a simple, ideally also cost-effective tool for screening the general population to identify the people at risk of bad deterioration of skeletal muscle down the line, so you can implement preventive strategies. There are screening tools for osteoporosis, for example, which is deterioration of bone tissue. Populations at risk, like women in their 60s and older, can have screening and implement preventive measures to slow the process down. It would be so beneficial to have a similar tool for muscle deterioration.

You’re leading an NIH-funded investigation on loss of muscle mass. The medical term is sarcopenia. How different is this condition from normal aging?

I think the concept and the definition of healthy aging is something a little bit philosophical. Aging is associated with deterioration of pretty much every tissue in our body. The mind deteriorates, the organs deteriorate, and of course muscles also deteriorate with age. It’s a normal process, and it’s to be expected that as you become older, you become a little weaker. This per se is not pathological—it’s just how nature works.

The problem is that for some people, up to 25 percent of older adults, the muscle deterioration becomes so bad that they become at risk of falling or they become unable to perform the activities of daily life. That’s what sarcopenia typically refers to. It’s the same process that happens in everyone, but in some people it’s so extreme that it really affects their quality of life. It can be very severe, and people can be bedridden. But for those who are able to walk, one of the main issues is falling.

If we had a screening tool for sarcopenia, what could we do about it? What is the state of potential downstream interventions?

There are several approaches being proposed lately and they seem to be pretty effective. Most rely on physical therapy combined with physical activity, usually some form of targeted strength training. If you’re able to implement the right strength-training strategy at the right moment, it can have a huge effect in slowing down the process and helping people remain active longer.

This is an area that you’ve been working in for about a decade. When you started, what was the state of MRI muscle imaging then and what would you say it is now?

In the last decade I’ve seen so many improvements and I’m so excited about all the things that are happening. Ten years ago I was a PhD student, and some of the techniques were very new. We were pretty much trying to come up with something and test it on a very small number of volunteers—and that was it. We would get numbers out of our measurements, but we were not able to fully connect these numbers to underlying tissue characteristics.

I think there have been two main driving reasons why things have improved so much. One is a lot of advances in acceleration techniques for imaging. Now we can do our measurements much faster, which means we can actually collect more information from the MRI exam. The second thing—and an area I’m actively involved in—is trying to establish a stronger connection between what we see and actual muscle function. That’s why I’m very interested in combining imaging with biomechanics to try to understand exactly how particular imaging findings correspond to functional behavior.

Can you give an example of something that was maybe not possible a decade ago with MRI but today is?

For instance, one of the techniques I’m very interested in is DTI, which stands for diffusion tensor imaging, originally developed for reconstructing white matter tracts in the brain. Over time, people have started translating DTI to other body areas, and because of the way skeletal muscle cells are arranged in the body skeletal muscle is a very good candidate for this technique.

When we started, it was very difficult to do these kinds of measurements. It was very complicated, and images had a lot of artifacts. It was almost unthinkable to scan a lot of patients with that approach. Now I think we’ve reached a stage where we’re able to run large studies.

Also, community efforts are coming together from all over the world to assemble data sets. Besides sarcopenia, which is a common problem, a lot of the diseases that affect skeletal muscle tend to be very rare. But we have improved so much that now we’re not only using DTI at our research center but we’re also able to pool data together from a lot of different places, compare them, and obtain meaningful statistics about rare conditions that we wouldn’t be able to get much information about at any one research facility.

Can you explain why the way muscle fibers are organized makes them a good candidate for a technique like DTI?

DTI is essentially exploiting the motion of water molecules inside biological tissue to try to say something about the tissue’s size and orientation. The most ideal condition is when you have a structure organized in cylindrical shapes, because it has a small cross-sectional area where water molecules can’t travel as much but a very long structure with a lot of motion in that direction. This is one of the typical features of white matter fibers in the brain and why DTI works so well there. Skeletal muscle to some extent is arranged very similarly. Muscle fibers are very long cylinders, up to several centimeters in length, packed together in a hierarchical arrangement, and surrounded by connective tissue. Those fascicles of muscle fibers are well aligned, which means that locally, in the same imaging voxel, you have all these long cylinders oriented the same way, so it’s very easy to use techniques like diffusion MRI to study the orientation of those muscle fibers in a very similar way to what you would do with white matter fibers in the brain.

You mentioned that about a decade ago it seemed borderline fantasy to imagine these techniques applied in a clinical setting, whereas today it’s within the realm of possibility. What were some of the challenges that applications of DTI to muscle tissue have had to overcome?

I think some of the main challenges are the specific magnetic properties of skeletal muscle—for instance, very short T2 relaxation, which makes it very challenging from an imaging standpoint. One of the main reasons we are now able to do things that were not possible ten years ago are great advances in hardware—not only the magnets but also the gradient systems. Also, there have been a lot of advances in terms of sequence development, so now we’re able to manipulate signal better. This means we’re both faster and able to get much more meaningful information.

In a review that you co-authored recently, you and colleagues write that “multiparametric muscle MRI holds great potential in the diagnosis and monitoring of muscle conditions in research and clinical applications.” Can you talk a little bit about this great potential?

One of the key strengths of MRI is that we can use it to measure different aspects of skeletal muscle quality and isolate signal contributions coming from different tissue characteristics. For instance, we can measure atrophy, which a process of muscle cells getting smaller due to disease or inactivity. We can monitor fibrosis, which is the deposition of connective tissue that also happens with aging and disease. We can study fat infiltration and inflammation processes. One of the unique capabilities of MRI compared to other modalities is that we can study all of these things at the same time, within the same exam.

Putting together all this information can allow us not only to say that there’s something wrong but also to identify exactly which component, which pathway is involved, and this can provide very specific targets for treatments. We are understanding more and more about the connections between what we measure and the underlying functional status of the tissue. Last, these techniques are becoming easier to implement, and that further increases their potential, because we’re now talking about an assessment that can be done virtually on any MRI scanner.

This is not happening tomorrow but I can easily foresee that fast multiparametric MRI will become a valuable screening tool for sarcopenia, in a similar way that DEXA scans are currently used in screening for osteoporosis.

When did your interest in muscle imaging start and what makes skeletal muscle MRI so exciting to you?

I’m a physicist by training. I did my bachelor’s in Italy and then moved to the Netherlands for my master’s, both in physics. I had no connection to imaging until the moment I had to do a master’s thesis—in the Netherlands it’s very common for people to work for a year in a lab and then write a thesis about their project. And it just happened by chance that I wound up with an MRI group, where I worked on ex-vivo samples of cardiac tissue, doing experiments on pieces of pig myocardium.

That was my first experience of being in a lab, doing something practical and seeing actual results. I always had a general interest in imaging because my father is a radiologist, so I grew up seeing medical images. But the moment I started, when I would prepare the sample, put it in the scanner, and then get images—it was almost like magic.

As a master student, I got to spend hours at the scanner, changing parameters and seeing what happens. It was very fun, and I enjoyed it so much. At the end of my master’s, the lab offered me a position as a PhD student, and that’s when I started working on technical development for musculoskeletal applications.

My PhD was very broad and I focused on a lot of different tissues, not just muscle. But in a way, I’ve always been fascinated by muscle because it’s so plastic and it’s so easy to modify. Think about doing strength training and getting stronger or bigger. That’s something everyone can experience. It’s not that easy to modify other parts of the body. That plasticity has always been very fascinating to me, and I think that was one of the main reasons why I started to be so interested in the topic. Another reason is that I’m very interested in the technological approaches and MRI sequences that can be used to study skeletal muscle.

I was very lucky in my PhD, but I didn’t get to appreciate the clinical utility of some of the things I was doing. So, for my postdoc, I changed labs, moved to the United States, and really wanted to work with a clinician to get a better idea of what the clinical issues are and where we could contribute through technical developments.

You did your postdoctoral training at Stanford University. How did that experience shift your investigations and orient them more toward clinical needs?

When you start framing your interest in terms of what is needed—either clinically or in research—you will also appreciate that advancing for the sake of advancing is not necessarily the best approach. Sometimes you have to stop and evaluate how to get the most out of the things you already have and make sure that everything works robustly: that you can reliably get consistent measurements over multiple days; that you can do the measurements in a way that’s patient friendly, because if your technique takes too long or requires the patient to be in an uncomfortable position, it’s never going to become adopted. My main interest is in technological developments, but my experience at Stanford was very useful because I got to appreciate all those other aspects of technical research that are sometimes thought a bit less about.

Muscle MRI studies often focus on the calf and the thigh. Can you talk about why certain parts of the anatomy may be more interesting or better suited to muscle imaging research?

There are two main reasons for that. One is very practical: you tend to have a lot of muscle tissue around your thighs and lower legs, so that makes it an easy area to image. When we’re developing a new approach, we always test it on the lower extremities, because if it doesn’t work there, there is usually no point in trying to apply it anywhere else.

The second reason, which has less to do with pragmatic considerations and more with clinical implications, is that impairment in lower extremities tends to have more severe consequences for quality of life than do impairments in upper extremities. For instance, having very weak musculature in your thighs, really puts you at risk of falling, which is something that significantly increases one’s mortality.

Is it the case that muscle tissue in our lower extremities is usually reflective of the state of skeletal muscle across the system?

This is something we’re in the process of figuring out. For some muscles we have established this connection, but we are still actively working on trying to answer this question. It’s a very interesting issue and the short answer is that we don’t know yet.