By Nathan Collins
David Wilson doesn’t remember those moments all that well, but his wife will never forget. Wilson had come out of surgery to remove his thyroid — doctors found a malignant tumor — but three hours later he was still having trouble speaking and moving. Perhaps, Janet Wilson thought, it was just taking a long time for the anesthesia to wear off, but she was ill at ease. As the hours went by, David Wilson’s condition worsened, and unease turned to dread.
“I knew I had a problem,” David Wilson says. “I didn’t know what it was.”
Looking at him now, one wouldn’t necessarily know that Wilson suffered a massive stroke that day. At 63, he’s slim, energetic and quick with a joke. He drives a yellow Corvette, plays golf, reads Dickens and plays Magic, The Gathering, a hobby he picked up from his son. That he uses a small hiking pole when he walks could easily be the result of an old football injury, not because five years ago he found himself unable to walk, speak, read, write or take care of his most basic needs.
Even after six months of rehabilitation, “I was in a wheelchair and my speech was really bad,” Wilson says. “For 2½ years, she had to do everything for me,” he says, looking to his wife.
Every year, roughly 795,000 Americans have a stroke — a life-threatening event resulting from poor blood flow in the brain, often caused by a blood clot or other obstruction that blocks a blood vessel, as happened to Wilson. Despite advances in immediate treatments that greatly reduce long-term brain damage for many patients, 17 million Americans have stroke-related disabilities. Besides physical limitations, many have short-term memory loss, complicating their mental and physical recovery, and will lose long-term memories sooner than the norm.
Wilson’s recovery, he and his wife say, benefited greatly from his access to faculty, therapists and research at Stanford and elsewhere. And they believe that most, if not all, stroke survivors who have the same access to care Wilson had could see similar improvements. Still, many don’t — in part because doctors don’t have good data on what treatments work, let alone which treatments work best for which patients.
“The first thing that family members, or patients, if they can talk, ask the doctor is ‘Am I going to get better? How much better am I going to get?’” says Marion Buckwalter, MD, PhD, an associate professor of neurology and of neurosurgery and a member of Stanford Bio-X and the Stanford Neurosciences Institute. “The answers right now are, No. 1, I can make an educated guess about how much better you’re going to get, but I might not be that good at it, and No. 2, maybe rehab will help you get better, but there’s not a lot of data to show that it actually helps.
“And there’s really not a lot else,” Buckwalter says — no drugs and no proven therapies to make stroke survivors’ lives more livable.
Yet like the Wilsons, Buckwalter and her colleague (and Wilson’s doctor) Maarten Lansberg, MD, PhD, an associate professor of neurology and a member of Bio-X and the Neurosciences Institute, are hopeful — even a bit upbeat — about stroke recovery.
In 2015, the pair launched what is now the Stanford Stroke Recovery Program to remove barriers between engineering, medicine and basic science researchers to improve long-term stroke recovery. The researchers’ ideas are mostly in their infancy, and may sound a bit wacky — injectable, drug-delivering gels, magnetic pulses fired into the brain, even robotic ankles — but, Buckwalter and Lansberg say, the future could be bright.
It could use some brightening. Stroke remains the fifth-leading cause of death in the United States and a leading cause of disability, according to the American Stroke Association. Until the 1950s, researchers understood little about strokes. But doctors now know that high blood pressure and smoking increase the risk of strokes, and aspirin can help prevent them. In 1996, the Food and Drug Administration approved a drug that dissolves the blood clots that cause five-sixths of strokes, and there are now surgical techniques to break up or remove clots before too much damage is done.
Rehabilitation, however, has lagged and mostly involves relearning and practicing basic skills — how to walk, how to form sentences or just how to pick up a bowl with two hands. What’s more, most stroke patients don’t have the benefit of adaptable young brains. Efforts to do better are complicated because the brain is very complex, Lansberg says.
Preventing stroke and treating it as it’s happening are problems of plumbing — thinning blood and dissolving and otherwise breaking up clots. But stroke recovery is a neuroscience problem.That observation, Lansberg says, motivated him and Buckwalter to create their recovery program’s predecessor, the Stroke Collaborative Action Network, in 2015 with help from a Stanford Neurosciences Institute Big Ideas grant.
“We felt this was a big unmet need. Essentially, we do not have effective therapies,” Lansberg says.
Lansberg and Buckwalter are looking at how to prevent dementia, which affects about 40 percent of patients in the first 10 years after a stroke. Buckwalter’s lab recently showed in mice that stroke triggers something similar to an autoimmune disease in the brain, leading to dementia. What’s exciting, Buckwalter says, is that in mice, the disease can be treated with a version of an FDA-approved drug, suggesting doctors may already have a tool to fight dementia.
“Right now, we are doing studies in patients to determine if there are biomarkers in humans that show us that the same or a similar process is causing dementia” in stroke patients, Buckwalter says. (Wilson is a participant in Lansberg and Buckwalter’s study.) But that research is just the tip of the iceberg.
Sarah Heilshorn, PhD, an associate professor of materials science and engineering, is not who you might expect to study stroke recovery, but she was interested in finding ways to treat Alzheimer’s and other neurodegenerative diseases using an unusual approach: injectable gels that could deliver powerful, localized doses of drugs capable of rebuilding damaged brain tissues. She’s working with Paul George, MD, PhD, an assistant professor of neurology, whose research in rats uncovered a protein produced by immature nerve cells that helps restore function of the nerve cells in the area surrounding the stroke site.
If something similar happens in humans, a drug that is based on those proteins could form the basis of a stroke treatment. The challenge would be delivering the drug: Usually, proteins either dissipate throughout the bloodstream or are eaten up by enzymes. To counter this, doctors could inject a single high dose or use multiple smaller injections. Neither strategy is great — a single high dose leads the drug concentration to spike too high, then rapidly fall too low. Also, Heilshorn says, “no one wants multiple injections to the brain.”
Heilshorn’s injectable gels, which are being tested in rats, could solve the problem in an elegant way, she says. They are engineered to hold on to the proteins, preventing them from diffusing or being eaten up. The gels are also biodegradable, so, ideally all that’s left after treatment is a healthier brain.
Amit Etkin, MD, PhD, an associate professor of psychiatry and behavioral sciences, is hoping to heal the brain in an entirely different way: stimulating it with beams of magnetic energy. Transcranial magnetic stimulation has the potential to rehabilitate stroke-damaged brains, Etkin says,tailored to the way a stroke has impacted each patient’s brain.
The idea to use TMS in stroke patients came from frustration with conventional approaches to diagnosing, treating and simply understanding such psychiatric diseases as major depression. The problem, Etkin says, is that conventional methods uncover correlations, not causation. Researchers studying depression, for example, often look for correlations between specific behaviors associated with depression and brain activity as measured by an fMRI brain scanner. Similarly, stroke studies might look for correlations between aphasia, the language disorder that leaves David Wilson occasionally looking to his wife for the right words, and damage to a particular part of the brain.
But Etkin wants to understand causation. For instance: How does damage to one part of the brain change the circuits responsible for movement or speech? Using a combination of TMS and electrical recordings taken from a person’s scalp, the team injects energy into one part of the brain, then follows it through brain circuits, revealing “the causal influence of one brain region on other brain regions,” Etkin says. His team has tested its technique in about 20 stroke patients, creating a separate map of each person’s brain — a “neuroscience of the individual,” Etkin says.
TMS also may be able to reshape the connections between the neurons that make up those circuits — again, tailored to each patient. TMS temporarily improved aphasia in one of Etkin and Lansberg’s patients, and there are signs that continuing the treatment over several days could provide more lasting relief.
Beyond TMS, stimulation itself — maybe even just poking the brain with a needle — could be the key to rebuilding it, says Gary Steinberg, MD, PhD, the Bernard and Ronni Lacroute-William Randolph Hearst Professor of Neurosurgery and the Neurosciences. In rodents, Steinberg has shown that injecting stem cells into the area immediately around a stroke site can help resurrect damaged, although not destroyed, neural function. Recently, he has applied optogenetics, which uses light to activate specific genetically modified neurons, to rebuild neural circuits in mice that have had strokes. Others have used electrical stimulation to similar effect.
It’s not clear why stimulation seems to work, Steinberg says, but possibly it helps restore a balance between the left and right brain hemispheres upset by stroke. Or, it may simply help rebuild connections between individual neurons. “We need to figure out mechanisms,” he says, but “I’m an advocate of using every form of energy” to explore stimulating the brain.
The flip side of the promise of new techniques is that injectable gels and brain stimulation may not be widely available to patients within the next few years. Meanwhile, tens of millions of patients need help now just to walk around.
Enter Steven Collins, PhD, an associate professor of mechanical engineering, and his robotic ankles. The devices, which look like an exoskeleton around the ankle, help patients with gait abnormalities — for example, a limp resulting from damage to the muscle-control center on one side of the brain — by giving a small mechanical kick to the affected leg.
Collins does not view his ankles as a path toward rehabilitation, despite widespread interest. “People have been trying to develop robotic devices to replace clinicians or augment therapy, but they have not been successful, and there have been a couple notable failures.” In one test of a robotic exoskeleton intended to help stroke survivors move their legs in a more natural way during treadmill rehabilitation, patients mostly let the robots guide the walking, rather than engaging their own muscles and the brain circuits responsible for activating them. “In the near term, the more obvious application of robotic devices is for gait assistance” to help people walk better, Collins says.
But even robotic assistance has problems, Collins says. Often, engineers invest time and money into building lightweight, compact devices designed to help people, only to discover they don’t work. And, Collins says, most engineers don’t involve patients in the design process.
The solution is twofold. First, Collins and his team have built an “emulator” that helps engineers tailor a device’s specifications to patients before sinking resources into creating mobile, wearable designs. The emulator is a robotic device worn on a patient’s ankle, similar to what one would wear in the real world, except connected to powerful off-board computers, motors and other devices in the lab. The second idea, “human-in-the-loop optimization,” is to have patients participate in the design process.
Those ideas allow researchers, and eventually doctors, to try variations to see which design works best for a patient before trying to make the ankle lightweight and compact. “We measure your performance in real time while you’re using the device,” Collins says, and “we systematically vary the device characteristics for control so as to maximize your performance.”
Collins and his team reported last year in Science the results of an experiment that showed robotic ankles designed using the emulator cut the energy expended on walking by an average of 24 percent in 11 healthy adults. Working with Lansberg, Collins plans to begin studies in stroke patients this year.
Collins’ study and others at Stanford illustrate principles that Janet and David Wilson believe are essential to Wilson’s recovery: tailoring stroke therapies and treatments to individual patients, and trying something new when one approach isn’t working.
For Buckwalter, they also illustrate an important Stroke Recovery Program aim: connecting researchers directly to patients. “The barrier’s basically dropped for someone who’s in engineering or another field and might normally not have any contact with stroke patients or might not know how to do clinical research,” she says. “There’s an entrepreneurial and adventurous, innovative outlook that people have here. We’d like to tap into that spirit and try to move things forward with rehabilitation.”
Lansberg agrees. “I lived through this period where it seemed tough to get anything done for acute stroke. It seemed pretty insurmountable not too long ago,” yet dramatic progress has been made in the past 20 years. “So I’ve seen how if you bring the right technology together with the right clinical teams, you can make very rapid progress.”