Eduardo Martín, neuroengineer: ‘When we stimulate the spinal cord with electrodes, the stimulus goes up to the brain and something changes’

Eduardo Martín Moraud, 39, obtained his degree in Telecommunications Engineering from the Technical University of Madrid in 2007. In the years since, he has used his time to study humanoid robotics in Kyoto, artificial intelligence at the European Space Agency’s center in Leiden (Netherlands), or robots capable of smelling in Paris. He was also part of a team that designed a bionic hand at the University of Edinburgh, in Scotland.

Key to this whirlwind journey was meeting the neuroscientist Grègoire Courtine, while the latter was working on spinal cord stimulation with electrodes at the Swiss Federal Institute of Technology in Zurich. Since then – and after passing through the University of Oxford – Martín Moraud finally settled on what would be his vocation. He now leads a neural engineering research group at the University Hospital in Lausanne, Switzerland. The team develops neuroprostheses for Parkinson’s patients. Their latest achievement has been helping a man who has lived with Parkinson’s disease for 25 years be able to walk again.

Question. How does a telecommunications engineer end up researching Parkinson’s?

Answer. After graduating, I specialized in robotics. I did my final year project at a center in Japan, which used humanoid robotics to understand human locomotion. It was a well-known center in Kyoto. In 1998 or 1999, they put an implant into a monkey and, by reading its brain activity, managed to control the legs of a robot in Pittsburgh. It was one of the first brain-machine interfaces.

Upon returning [to Europe], I worked in artificial intelligence in France. With that initially more robotic orientation, I did my Master’s project in robotic prostheses. So, I had gone from telecom to robotics [and then] from robotics to artificial intelligence. From there, I ended up in neuroprosthetics. Around that time, I saw a position being offered to work with [Grègoire] Courtine. That was in 2009. [The project] was also a neuroprosthesis, but a different kind of neuroprosthesis – it was no longer robotic, it [involved] electrical stimulation of the human nervous system. It seemed like a good idea to go with him and Silvestro Micera – who co-supervised my thesis – to Zurich. I found myself working with rodents every other day. I had never worked with animals before. The entire doctorate was done with rodents. I then moved on to primates and, finally, to my first patient.

Q. What’s electrical spinal stimulation?

A. The idea of delivering electrical impulses to the nervous system has been around for 150 years. Basically, metal electrodes are placed [across nerves], electrical impulses are released at a certain frequency and the neurons [typically] respond.

During the time of the Soviet Union, the Russians conducted many experiments. Today, [electrical spinal stimulation] is used to treat chronic pain. If there’s no response to any medication, the most common therapy is to place electrodes in the spinal cord and, by stimulating it, you prevent that [pain] signal from reaching the brain. Now, the idea of stimulating the spinal cord for motor control hadn’t been carried out until today. That’s what Courtine and others were developing for 20 years.

Q. Why stimulate the spinal cord, particularly the lower spine?

A. The lumbar spinal cord is where all the neurons that control the legs are. When you want to walk, the brain thinks: ‘I’m going to walk.’ That signal then goes down along the spinal cord, activating it. The lumbar spinal cord coordinates the contraction of the muscles. If it splits, nothing reaches the area, which is still alive. The idea is to place electrodes in that lumbar region, stimulate it with electricity, reactivate it and – combined with physiotherapy and rehabilitation exercises – recover leg movements to a certain level after a spinal cord injury.

Q. But how is this stimulation coordinated with the intention to walk?

A. The first thing we understood may seem illogical. There are signals that go from the marrow to the muscle… and then there are nerves that go from the muscle to the marrow. This is the sensory part. That is, if I stretch my arm, the muscle is stretching and, as it does so, the signal it’s sending to the spinal cord changes: the muscle tells my spinal cord, ‘hey, I’m stretching.’ Our stimulation is acting on this sensory part – not on the motor part. In reality, they’re all reflex circuits. This means, if I stretch a leg, there’s a moment in which the spinal cord will automatically generate a reflex and contract the muscle, like when you get hit in the knee. When we manage to recover a certain locomotion in patients, it’s all based on reflexes: we’re stimulating the sensory part, so that when a leg is stretched, it will generate a reflex and the leg will move forward.

Q. And the second thing?

A. The second thing is that the spinal cord has a certain anatomy. If I stimulate a high lumbar area – for example, the L2 or L1 root – I’m mainly activating or modulating muscles that are leg flexors. However, if I stimulate at a lower lumbar level – L5 or sacral – [I’m mainly activating] extensor [muscles] in the leg or foot. Initially, we stimulated almost in any way possible. Now, we have electrodes positioned on the spinal cord in a very precise way. We know very well which electrode – at what precise moment – must be activated.

Q. And in this process, when does the brain come into play?

A. The third element is the voluntary part. All of the above is what the bone marrow does and how we reactivate it. We have videos of paralyzed rats or monkeys (with an electrode in their spinal cord) who walk apparently naturally… but they’re not aware of anything, because it’s completely involuntary. Depending on the level of severity of the injury, if it’s minor, there’s some residual [paralysis] that continues to decline. In that case, stimulation increases what decreases naturally. But when the injury is so brutal that there’s [no mobility at all], you can stimulate as much as you want – you’re not going to increase any signals. What we did in primates – and, a few months ago, in the first human patient – was to put an implant in the brain, a second implant, which decodes its activity. We then connected it with the other implant, the one for spinal cord stimulation.

Q. Is this how people with paraplegia are treated?

A. The first [case we worked with was] in 2018, a patient with a relatively minor [injury]. He was paralyzed, but he had some motor control that we were able to increase. The second research investigation was with three patients who had much more severe injuries [with significant brain damage]. We worked with them only on spinal stimulation.

[Patients who were] trained with spinal stimulation went out onto the street with their electronic walkers. They pressed a button and the stimulation to walk was activated. They were hitting the buttons: right and left, right and left.

Q. And at a certain point, you reoriented your work towards Parkinson’s disease.

A. I had a family member with the disease. From when I finished my doctorate in 2014, my idea was to apply these therapies to patients with Parkinson’s. Courtine was convinced his approach could be used on these individuals. I spoke with him and with Jocelyn Bloch (a key neurosurgeon when it comes to these implants) to [develop this]. Today, I’m the one leading this part of the research.

Q. When did you realize that what you had learned working with patients suffering from paraplegia could be transferred to those living with Parkinson’s?

A. Between 2017 and 2019, we were doing trials on primates. Monkeys don’t develop Parkinson’s disease, but you can induce it with a toxin that kills dopaminergic neurons, which is what happens with Parkinson’s. Unfortunately, you have to induce the disease in them… but this allows us to verify if the therapeutic idea works. Then, we placed the implants with three objectives. First, to know if [the prostheses] hurt: paralyzed individuals may not feel [the pain], but a patient with Parkinson’s will. The second thing was to see if the concept of stimulating reflexes was maintained, as there could be degeneration in the spinal cord that prevents this. And third – if [the implants] didn’t hurt and the biological basis was maintained – the response had to be modulated. It was time to test it on humans.

We implanted Marc – a patient who had been living with Parkinson’s for 25 years – in 2021. We did the rehabilitation and he went home. Now, we keep in touch to see how he’s progressing, or to change the configuration of the parameters, because Parkinson’s is always changing. He wrote to me a few days ago telling me: ‘I just ran 100 meters (330 feet) without falling.’

Q. What can you tell those who are living with spinal cord injuries or Parkinson’s who are reading this?

A. Not long ago, I gave talks at [two associations for patients living with paraplegia].

As for these patients, we’ve already calibrated the expectations and the pace of progress. With people with Parkinson’s, it’s different. I don’t think they had ever been taken into account when talking about these issues. It’s more complicated [for them], because we

know how to evaluate spinal injuries: you do an MRI and an evaluation and you know if the person has residual motor control. Furthermore, patients [with paraplegia who recover] will reach what is called a plateau and [the condition] won’t change much anymore.

The problems [faced by] patients with Parkinson’s are numerous. You have young and older patients – those with tremor profiles and those who don’t tremble, with more rigidity and slow movements. Then, there are those whose condition evolves for 10 years, while some [see it change] for 25 years. And then we have those who develop locomotion problems and those who don’t. The typical case is that of freezing up, when [a patient] can’t move. This is something we don’t know how to deal with today. The neurologist knows that this man is going to fall, that he could break a hip… but the most he can do is recommend that he use a walker and be careful.

Q. And the electrodes can prevent a patient’s body from freezing up?

A. What we’ve understood with this patient is that we don’t understand everything yet. He had asymmetry: that is, he was very affected on one side. He was very unstable, so he fell frequently. He also froze up a lot. We’ve [managed to improve] the first two problems by stimulating the spinal cord and playing with his reflexes. As for the [inability to move], we don’t know why it has improved… and I don’t think it’s solely due to spinal stimulation.

In my opinion, when we stimulate the spinal cord, the stimulus also goes up to the brain and this changes something up there that makes the blockage improve. Nowadays, Parkinson’s is treated quite well with dopamine and, when this stops working, there’s deep brain stimulation. And, when that doesn’t work, there’s the Duodopa pump treatment. But there were no tools for locomotion problems… that’s what we’re providing now. Still, like any tool, it may not work for everyone – it’s important to keep this in mind. That’s why, this past January, we expanded the research investigation to six other people. We look for [patient] profiles that are different from each other. In this second phase, we’ll have electrode systems specifically designed for Parkinson’s patients.

Q. What does the financing look like in this field?

A. I think maybe the most difficult thing is financing. These types of clinical trials are very expensive. Just one implant costs 50,000 euros ($55,000). We’re talking about hundreds of thousands of [dollars] per patient. And, when the grants we get are, say, $100,000, we’re not able to cover even a single patient. You have to pay for the technology, the surgery, the hospitalization, the physiotherapists...

In [some high-income countries], money is given out, there are grants… but it depends on the study. There are studies that may be done with as little as $20,000, [but sometimes] you can get a million or so for [a study with just] six patients.

Q. But if your system works for other patients, the economic and social return will be massive.

A. When a patient reaches these levels of illness and loses independence – when they begin to depend on family, on associations, on nursing homes – it’s very expensive for the patient, for the family, for society and for the health system. Being able to gain years of independence, to be able to go outside, to go for a walk, to not have falls... it helps them, it helps their families. It helps us all.

Sign up for our weekly newsletter to get more English-language news coverage from EL PAÍS USA Edition