21st Century Docs: MGH Neurotechnology Trials Unit Brings New Tools to Bear on Neurological Diseases

"In the last two decades, advances in the ability to image the living, working brain, and record from or stimulate neurons in vivo have revolutionized neuroscience research. But translating these engineering achievements or insights into practical tools for human neurological health is just beginning", says Mass General neurologist and neuroscientist Leigh R. Hochberg, MD, PhD.

Unlike research and development of new drugs, which follows a tried-and-true translational path, the testing of new neurotechnologies—including devices and imaging methods—has been less common, and almost always initiated by industry. That’s why, in 2008, Hochberg and colleagues founded the Neurotechnology Trials Unit (NTTU), a unique clinical research resource for Mass General researchers and clinicians who want to develop, test and apply new technologies and devices of all types for the treatment of neurological disease.

Restoring communication and mobility
Hochberg co-directs the NTTU and is the principal investigator on one of its flagship programs, a pilot clinical trial of BrainGate2, a neural prosthetic being developed to allow people with paralysis of the arms and legs to control external devices using only their thoughts.

A paralyzing event like a brainstem stroke or spinal cord injury disconnects the line of communication between the brain’s cortex and the rest of the body, but leaves both otherwise perfectly healthy, Hochberg explains. BrainGate2, an investigational medical device, is a combination of hardware and software that aims to reestablish the connection between thought and action. A pill-size array of miniaturized electrodes surgically implanted into the motor cortex reads neural activity that occurs when a person with paralysis imagines moving her arms or hands. An external computer then receives the information and translates the neuronal code to move a computer mouse, or a prosthetic hand or robotic arm.

Mass General is a clinical leader in neural interface technologies —Hochberg and his collaborators were the first to achieve completely thought-directed movement of a computer cursor by a person with tetraplegia (unable to move either arms or legs) 10 years ago. In a widely publicized 2012 study, a tetraplegic woman used the device, coupled to a robotic arm, to reach out and pick up a thermos of coffee. It was the first time she had taken a drink without assistance since suffering a brainstem stroke 14 years earlier.

The group is now carrying out a multi-site pilot trial of BrainGate2, which they plan to test on up to 15 people. The trial, funded by the NIH, the Department of Veterans Affairs, and by private foundations, is taking place at the Mass General NTTU and also at Stanford University, Providence Veterans Affairs Medical Center and Case Western Reserve University, together with Brown University. The trial is enrolling people with brainstem stroke, cervical spinal cord injury, amyotrophic lateral sclerosis, and muscular dystrophy.

Eavesdropping on epilepsy
At the heart of the BrainGate2 brain-machine interface is a sophisticated microelectrode array that can record the activity of single neurons or small groups of neurons in the brain. By allowing analysis of individual neurons, the devices promise to open up a new understanding of neurological disease. In the hands of Mass General neurologist Sydney S. Cash, MD, PhD, single neuron recordings are changing the view of epilepsy, a common and debilitating neurologic disease.

For 100 years, EEG has been the gold standard for clinical evaluation of epileptic seizures. But EEG records activity from large aggregates of thousands or millions of neurons, like using a microphone to record the roar of a crowd. “It’s useful, but crude,” says Cash. In contrast, implanted microelectrode arrays pick up the activity of individual neurons and small groups of neurons, like listening to a few dozen people out of the whole crowd, and clearly hearing each person talking.

Based on EEG data, epilepsy has long been viewed as a problem of abnormal, hypersynchronous brain activity, where during seizures all the neurons in an affected area fire together. But when Cash zoomed down to the level of individual neurons, he found that the neurons were doing lots of different things. His current work is digging deeper to try to understand which neurons are doing what, and how that can yield new information about how seizures start, how they stop, and how they might be controlled. “Knowing that the seizure isn’t a monolithic entity reveals nuances that might be exploited for better treatments, whether medications, devices, or other approaches entirely,” he says.  

That work also offers new hope for a thorny issue in epilepsy: how to predict seizures. Clinicians have been working for decades on EEG-based and other techniques to forecast seizures, but so far there is no system used in routine clinical practice.  In 2011, the Cash group published landmark data suggesting that the behavior of individual neurons may signal when a seizure is coming. “We think recording at this level of resolution is much more likely to be successful than anything that’s come before,” he said.

Right now this is purely research, but Cash says he hopes before long to begin a clinical trial of the technique. At the same time, he is working continuously to improve implantable recording devices using novel electrode designs and wireless technology, through collaborations between the NTTU, engineers and device manufacturers.

Traumatic brain injury: A prognostic puzzle
Every critical care neurologist has a trauma patient, or more than one, who comes into the hospital in a coma with an apparently devastating brain damage, and then proceeds to make an unexpected, and sometimes near total, recovery. These “miraculous” cases highlight how little physicians understand about forecasting outcomes from severe traumatic brain injury, says Mass General neurologist Brian L. Edlow, MD.

“One of the biggest challenges we face in the ICU is being able to provide our patients’ families and loved ones with accurate information about whether an individual with a severe traumatic brain injury is going to recover,” says Edlow, a critical care specialist and associate director of the NTTU. “The reality is that our current clinical predictors are not accurate enough for us to give anyone a clear sense of what to expect.”  

To remedy that, Edlow, who is also Director of Critical Care Research Neuroimaging, and his team are developing a suite of advanced structural and functional neuroimaging and neurophysiological tests that provide the fullest possible picture of brain injury and the prospects for recovery. In addition to the standard MRI and CT scans that reveal structural damage, the Mass General neurointensive care unit offers advanced functional MRI and diffusion tensor imaging that can determine whether or not brain networks are functionally intact. Research in Edlow’s lab suggests that patients who maintain connectivity in networks that play a role in arousal or wakefulness, for example, may have a better prognosis for emerging from coma.

In neurointensive care, all patients receive rapid, multimodal assessment, and state-of-the-art multidisciplinary care. The MRI scanner sits just down the hall, facilitating imaging of critically ill patients early and whenever needed. All beds are equipped for around-the-clock EEG recording monitored by epilepsy specialists, who sit just down the hall. The unit also performs high resolution quantitative EEG and analysis of evoked potentials to assess brainstem, somatosensory, visual, and auditory function.

Edlow believes this unique blend of high-tech, multidisciplinary and integrated care and research not only improves prognostication, but will also facilitate the development of new treatments, such as deep brain stimulation, that aim to correct network dysfunctions by modulating neural activity in the brain.

Beyond critical care, the new technologies should prove useful for diagnosing patients with long lasting conditions, too. As Edlow explains, patients in subacute or chronic stages of brain injury may have motor deficits out of proportion to their cognitive deficits. “The patient may be thinking or trying to express him or herself but not be able to make outward signs,” he says. These patients are difficult to diagnose with even the most expert bedside exam. By adding high resolution quantitative EEG, structural and functional MRI and a suite of evoked potentials, Mass General physicians are beginning to identify patients presumed to be in a vegetative or minimally conscious state, but who actually may have much higher levels of consciousness.

While research is ongoing, and many techniques require further validation, Edlow says they foresee in the near future a state-of-the-art clinical program that will provide informed diagnoses for patients who may otherwise be impossible to diagnose. That in turn will open new opportunities for rehabilitation or assistive communication tailored to each person’s precise situation.

Through the NTTU, Hochberg reflects, “We are focused on patients who have some of the most challenging neurologic disorders: coma, locked-in syndrome, epilepsy, traumatic brain injury, ALS, stroke, spinal cord injury. Our goals are not only to help provide world-class neurologic care – which is something the Mass General does every day – but to expand the understanding of these disorders so that we can develop ever-better therapies that restore lost neurologic function.”