Susan J. Harkema, Ph.D.
By: Sam Maddox
Susan J. Harkema, Ph.D., is director of the Reeve Foundation NeuroRecovery Network®, professor of neurosurgery and rehabilitation at the University of Louisville, and research director of the University of Kentucky's Spinal Cord Research Center and Frazier Rehab Institute. Her career has been built around a basic concept in human biology: there are nerve bundles in the spinal cord that control major functions, such as stepping. These circuits are smart on their own; they don't require connection to the brain. Harkema's work with people with spinal cord injuries has shown that recovery of function is possible -- even in people thought to be completely paralyzed -- by activating spinal circuits. In 2011, along with her mentor Reggie Edgerton, Ph.D., from UCLA, Harkema's lab implanted an epidural stimulator next to the lumbar spinal cord of paraplegic Rob Summers. Surprisingly, he regained voluntary leg function when the stim was on. Two more subjects have since received an epidural stimulator and a fourth is set for January; Harkema said the results, to be published soon, are just as exciting as Summers' were. Harkema spoke with Reeve staffer Sam Maddox in her office at Frazier Rehab, overlooking downtown Louisville.
Q. Some scientists are inspired early on to their vocation. How were you drawn toward this career?
A. I tried a lot of other things. I was a computer science major, an electrical engineer major and then became an athletic trainer because, well, just because my dad was a college football coach, and he's like, 'I'm tired of you changing your major, so why don't you be an athletic trainer?' Being a trainer wasn't going to sustain me; medical school was a brief consideration. I took a gig as a research tech, this while I was at Michigan State. I loved it. I did multiple surgeries for many different projects for the physicians. I was very good technically. That was my strength. I mean, I transplanted rat hearts. I was really driven and loved the idea of doing something that hadn't been done before, figuring out how to do it. And one day, one of the scientists on one of the projects came in and said, 'You should get your Ph.D.'
Q. What did you get the Ph.D. in?
A. Physiology -- based on a really exciting question in the small-muscle energetics field, whether ATP was actually the substrate for oxidative phosphorylation in skeletal muscle. (It is not.) It was a great research experience, but it's not a question with any direct link to any disease state.
Q. You wound up on the West Coast for your post-doctorate....
A. I found Reggie Edgerton at UCLA. His lab fit the criteria I was looking for, including warm weather! His work was about muscle, so I proposed to him that I could look at metabolism in paralyzed muscle; I had done all my work in cats, and they had a cat model at that time. I did end up there but soon moved to a human project, and there started my career in spinal cord injury. Of course once I started doing it and learning about the lives of people with spinal cord injury my motivation and passion really changed. I love being a scientist. I love designing a question and figuring out how best to answer it and then actually getting the answer. It's still incredibly exciting to me. But now there's a true humanitarian mission; you can't get to know people living with spinal cord injury without them inspiring you. I always come back to the 'ASIA As,' those people with clinically complete injuries; my real core passion is to figure out how we can incrementally change the lives of these individuals.
Q. What were some of your early projects with the Edgerton group?
A. These were the beginning of the experiments we talk about now: exploring how the circuitry of the spinal cord is able to generate a stepping pattern, apart from any input from the brain. We looked at activity-based therapies, including locomotor training. Could people with clinically complete injury demonstrate that sensory processing occurs in the spinal cord? The first thing we did was attempt to generate a locomotor pattern; if central pattern generation doesn't exist in the spinal cord, then no matter what we might try, there should be no output whatsoever in a patient with a complete injury. Amazingly, we got a motor pattern. Then the next step is, can you modulate it? We said 'Okay, if we change load, does the pattern change?' And the answer was yes: change load, you get more EMG activity and get a better pattern. We kept showing that there was some sophistication in the neural networks of the spinal cord.
Q. When did you realize this work might lead to a therapy for people?
A. This is going to sound harsh, but it isn't meant to be. We weren't trying to get people better. We were trying to understand how this circuitry worked, with no expectation that people would regain function. But when we did these sensory experiments, we asked, 'can we train humans to be able to independently step?' And what happened to those with incomplete injury is that a significant number of them came into the clinic in wheelchairs and they walked out -- they didn't walk out like before their injury, but they could take steps.
We started noticing other things. People began to report to us that this really changed their daily lives. They felt better, and their circulation and healing were better. People who were C4-complete regained a lot of trunk and arm strength; some were able to move from powered wheelchairs into manual ones. And this was just from standing. People reported that their bowel movements were improving, as was their bladder function. We also started seeing changes in muscle and bone and all these other aspects -- respiration, cardiovascular function.
Q. It seems the notion of recovery, or cure, is defined on a sliding scale ...
A. I learned that we can't judge whether what we discover and are able to disseminate to the population is important or not -- that is up to the individual living with SCI and where he or she is in life, what that person's expectations are. I remember one individual who could stand whenever he wanted as long as he wanted, and there was another who could only stand on one leg, for a minute or two. And the person who could stand anytime he wanted was very disappointed; he wanted to walk. And the person who could stand for a minute or two was ecstatic. Obviously, we can't ever stop until we've completely solved the paralysis riddle. But maybe we need to translate our promising research earlier, and not wait to translate until we have the whole puzzle solved.
Q. That's how the NeuroRecovery Network (NRN) came about, right? Translating what we know, now.
A. Right. The only place people were able to get locomotor training was in the laboratory, and of course we couldn't keep training them. So Christopher and Dana Reeve saw this was happening and the Foundation got involved -- to develop research and a clinical component.
Q. And the NRN is of course a major Reeve Foundation program.
A. The NeuroRecovery Network has turned out to be something beyond our imagination. We've got all these amazing clinicians and scientists and administrators working together. We continue to study locomotor training. We're developing new outcome measures. We're also in a position now in a very cost-effective way to support clinical trials, because the infrastructure and capacity to standardize are in place. We have an amazing electronic Web-based system that can very inexpensively gather data from multiple sites. We're submitting a lot of different grants to advance evidence at different levels, branching out into the cardiovascular arena, the pain arena and bladder function as well as other health-related aspects of SCI.
There's also a significant pediatric component of the NRN; recent evidence tells us that the plasticity in the pediatric population may be even more accessible than in adults. Here at Frazier we're building a research-clinical pediatric program, led by Andrea Behrman, Ph.D. She was at the University of Florida for many years but she just moved her clinical studies here to Louisville.
Q. You spent 10 years with Edgerton at UCLA, got recruited to Louisville in 2005. You have a lot of duties here, between teaching, running a lab, writing grants, traveling to meetings. Plus, you are married, with kids. How are you balancing the Mom job?
A. You don't balance it. You just do the best you can every minute of the day. My husband is a comedian and an actor, and when we moved to Louisville he decided to stay home with the kids, so that helps a lot in having a parent available all the time. But it is a struggle to make sure that you're spending enough time with them. Graduate students will ask me 'When is the best time to have your family? When you're in graduate school, or is it after you get your first RO1 [major federal funding]?' I always say, 'When you have them, that's the best time.' Somehow, you figure it out.
Q. What's the take-home message for the series of NRN articles published in September in the Archives of Physical Medicine & Rehabilitation?
A. Those articles lay out what's possible with activity-dependent plasticity. It's not the end-all. It's the start. First of all, in relationship to spinal cord injury, or any neurologic disorder, recovery doesn't stop one year after injury. For me that's almost the most compelling message: We had people in the NRN who had been a decade, two decades in a wheelchair who recovered their ability to walk. Now, they aren't running a marathon, but they are walking around in their homes and in their communities. We were able to generate some predictive models for which person will recover what function based on our data of 400 people.
The other main point is that the outcome measures we have available are very limited -- they only are sensitive for a slice of the population that we're studying. This is really critical as interventions come forward; we may be missing potential options because we're measuring a change in one type of score, such as 'we're looking for a 10-point change in the ASIA' [a measure of motor and sensory function]. But you may have recovery that's unrelated to that change. So the NRN developed recovery curves that are applicable to any number of interventions. Now we can demonstrate that with this intervention (say, locomotor training), for this type of patient, this is the recovery curve that you can expect for this number of training sessions. We introduced the Neuromuscular Recovery Scale in the Archives; it has an advantage over current measures in that it is sensitive for the whole continuum of recovery; it can pick up changes at the earliest stage, things for people who can't stand or walk, but also through standing and walking.
Q. Would you say this is a hopeful time for the SCI world?
A. It's a very hopeful time because there's so much research going on, and there is a widespread commitment to translation. It's clear that activity-dependent plasticity can play a huge role in recovery. We have an obligation to try to make therapies accessible to the community. The NeuroRecovery Network is one way to do that, not only to introduce new activity-based therapies into clinical settings and figure out how to most cost-effectively provide them but then also to disseminate information about the benefits so individuals and rehabilitative centers can decide. 'Okay, look, I know that if I train this individual with this intervention for 80 sessions they'll be able to sit independently.' Well, then everybody can decide if that's something that's worth their investment, from the patient to the center to everybody else.
Q. What about for people with so-called complete injury?
A. I don't know of anyone who's walked who has been classified as a clinically complete injury but there are important effects in regaining trunk or arm control. But it's all the other kinds of side effects of locomotor training that are really critical for the clinically complete population: their cardiovascular function gets better; they report fewer autonomic dysreflexia episodes; they have better circulation; they feel better. When you're sitting all the time you're not activating the neuromuscular system because you are unloaded. And as a physiologist it's very clear to me that being unloaded throws off all the other aspects. The irony is that the patients most in need of locomotor training are probably those who aren't going to walk as a result of it. They need it because that's their only way to bear weight and get neuromuscular activation of their bodies, which in turn helps to improve and/or maintain critically important health outcomes.
Q. Epidural stimulation opens possibilities for complete injury, yes?
A. We selected motor-complete individuals but when the stimulator is turned on, they can move voluntarily; they intend to move a toe and they actually do move that toe. And so people say, 'Well, those patients must not be complete.' Well, okay, but that opens up a whole other set of possibilities. If we assume someone's complete, the interventions available to incomplete individuals are sort of closed off to them -- we know they are not going to respond because they're complete. Clearly epidural stimulation is helping us understand not only how we classify individuals but also that it is something that can be useful for people with the most devastating injuries. It is encouraging that individuals who, by conventional measure, are considered clinically complete actually have capacities for recovery.
Q. You have tested three patients with epidural stimulators now?
A. Right. We have Rob Summers and two more implanted with epidural stimulators; all three have had, in general, the same result. All have regained significant muscle mass. They all can stand with the stimulator, independent of any assistance. Rob is an ASIA B [no motor, some sensory function] and we have another B and an ASIA A; the B has regained more sensory control without the stimulator but each of them can move voluntarily with the stimulator on. Our second B has now officially converted to a C [some motor and sensory recovery]; he can do some voluntary movement of his toes and ankles without the stimulator on. We're three-for-three in the results. Clearly we have to expand the number of people we study to see how generalizable these results are.
Q. So complete may not be so complete.
A. Being complete may not be anatomically complete. Is there some tiny thread of a nerve tract in these individuals that just happened to get spared? Is that what's allowing this voluntary activity once we excite the spinal cord? We don't know. To me the most astonishing thing is that you turn the stimulator on and then you turn it off and there's a difference. We're assuming there must be some connection -- these patients intend to move their legs and they do, so there has to be a connection. Yet we have no way to measure that. So what is that tract that's still remaining? If we knew that, maybe that's where a regenerative therapy could focus.
Q. When does this become clinically relevant?
A. The question that needs to be posed to all the stakeholders is this: Look at the daily lives of people with spinal cord injury; if you can change one of the consequences, is that worth the translation; is that worth the investment? For me, having the privilege of interacting with people living with spinal cord injury every day, my answer is 'Yes.' If we could just maintain normal blood pressure in these individuals, that would represent an immense change in health and quality of life. Cardiovascular function in people with cervical injuries, it is a really life-limiting situation. Their pressures are dangerously low and they can become so hypotensive that they get dizzy, lose consciousness. Can we use an epidural stimulator just to maintain blood pressure at a normal level; is that incremental change worth the investment?
For our young men in the epidural stimulation program, one of the things that's most important to them is that their legs look 'normal' again. Is that clinically relevant? Well, think about it. People with spinal cord injury have a higher incidence of cardiovascular problems, a higher incidence of diabetes, a higher incidence of metabolic deficiencies, and why is that? It's related to muscle atrophy. If you can maintain muscle mass you lower chance of developing many other complications.
If we look at it from a cost-benefit ratio, what's the cost of cardiovascular disease to the healthcare system, and to an individual? What's the cost of a fracture? What's the cost of a pressure sore to the healthcare system? I'm a little biased. I think the primary stakeholder is the person with the spinal injury. In my perfect world they get to make the decision; but that's not reality. If you look at this from the point of view of caregivers, or the insurance companies -- I think there is value-added to all the stakeholders in starting to translate these therapies now.