It had been three years since Jim received his Parkinson’s Disease (PD) diagnosis. Initially, he was able to control his movement symptoms with a simple therapy regimen prescribed by his neurologist, but as time went on Jim had to keep increasing the dosage and the meds were starting to have some cognitive side effects. Other than his PD diagnosis, Jim was perfectly healthy and wanted to maintain an active lifestyle, traveling with his partner and continuing to do consulting work during his semi-retirement, but it was becoming difficult with his current medication program.
After listening to Jim’s concerns and hopes for his future, Jim’s neurologist recommended getting a brain implant which would provide relief for Jim’s motion symptoms by directly stimulating a small, deep part of the brain called the Subthalamic Nucleus. This implant would also record signals directly from Jim’s brain, which would be used to control and adjust his therapy and learn more about his specific pathology. The surgical procedure would last a few hours and would be guided by an MRI machine. After considering with his family, Jim agreed to undergo the procedure.
Six months later, Jim has received his state-of-the-art brain implant and is happy to be getting back to living a normal life. Initially, the implant used the signals recorded from his brain to automatically calibrate the stimulation therapy to Jim’s precise anatomy and neural circuitry. After a few months, enough data was collected from Jim’s brain that his neural waveforms could be compared to other patients’, and it was flagged that Jim had a phenotype of PD which had previously responded well in other patients to a non-traditional combination of pharmaceuticals and dietary changes in addition to the stimulation provided by his new implant. Using this information Jim’s neurologist was able to reduce Jim’s overall medication dose while substantially decreasing side effects. Excited by the success of his new therapy, Jim has opted in to a research program which shares his data with neuroscience researchers in their quest to better understand PD and search for a cure.
The above story might sound a little like science fiction, but at top movement disorders clinics around the world it is quickly approaching standard-of-care. The implant described is one of the latest generation of Deep Brain Stimulation (DBS) devices which have recording, computing, and wireless data transmission capabilities in addition to the ability to deliver precise electrical stimulation to deep brain areas. These DBS devices, of which more than 20k are implanted annually, are just one in an ever-expanding class of neuromodulation therapeutics for treating PD, Epilepsy, Depression, Chronic Pain, and even Alzheimer’s. There are also non-invasive neuromodulation systems that use magnetic coils, called Transcranial Magnetic Stimulation (TMS) which treat 70,000+ patients annually for severe depression with emerging indications in addiction and other psychiatric disorders. The last decade has seen rapid technical, scientific, and clinical progress in neuromodulation therapies for these brain diseases and in the 2020’s neuromodulation is poised to go into the clinical mainstream. These therapies will not only directly benefit millions of patients like the fictional Jim, but will create massive amounts of high signal, low noise, well-labeled brain data that can be used to test computational neuroscience models, build new theories of these pathologies, and gain insight into how our brains work at the circuit level.
The beginning of the 2010s saw the introduction of the first implantable neuromodulation devices with the ability to directly record brain signals — Medtronic’s “Brain Radio” research platform and Neuropace’s RNS system for treating Epilepsy. The previous generation of neuromodulation devices (approved for use since 1997 for movement disorders including PD) behaved more like pacemakers, providing “open loop” always-on stimulation — in contrast, these new devices could record highly localized electrical brain recordings (Local Field Potentials or LFPs) which were directly linked to the brain areas being stimulated. For the first time, researchers and clinicians could see how these brain areas were responding in real time to the neuromodulation therapies being applied and adjust the therapy parameters accordingly. They could also see how implanted patients responded at the neural circuit level to other important stimuli — pharma medications, diet, circadian cycles, etc. Unlike a traditional therapy clinical trial, these devices gave actionable information about how the therapy is actually working for each specific patient, enabling new N=1 and iterative trial designs.
The first product in the third generation of implantable neuromodulation devices with recording capability is being released by Medtronic this year, branded as “Percept”, with other companies including Newronika and BioInduction following close behind. On the non-invasive TMS side, Magstim released their first Navigated TMS system in 2019 and fully TMS-EEG integration is being actively explored by multiple academic and commercial teams. Future generations of DBS and TMS devices will be able to take advantage of the same miniaturization trends which have driven progress in smartphones: shrinking transistors, more efficient wireless communication, longer-lasting batteries, higher-performance sensors, and cheaper computing power. These trends will drive advances in neuromodulation hardware over the next decade approximately equivalent to the advances between an iPhone 4 and an iPhone 11 Pro — and that is before considering potential step-change technologies like Neuralink and Paradromics.
The Brain Initiative was announced by President Obama in 2013 as a new program “aimed at revolutionizing our understanding of the human brain.” Practically, this has meant very large annual expenditures by DARPA, the NIH, and the NSF promoting applied, basic, and theoretical neuroscience research. The federal dollars themselves are large — $424,000,000 in 2019 alone — but the initiative has also catalyzed philanthropic investment, as exemplified by the Weil Foundation’s recent $185,000,000 gift to UCSF’s Neuroscience Institute. Out of the 735 Brain Initiative awards so far, 141 have been specifically for neuromodulation, including some of the largest single awards for the SUBNETS (depression and PTSD), RAM (traumatic brain injury and memory), and ElectRx (peripheral neuromodulation) programs. These levels of expenditure are expected to continue through 2025. While ~25x short of the 0.4% of US GDP spent during the Apollo Program, the combined federal and philanthropic funding for the Brain Initiative will similarly leave behind a legacy of PhDs creating new applications for, technologies to serve, and therapies based on neuroscience research.
In its first five years, the Brain Initiative has already funded dozens of first-in-man studies for new neuromodulation therapies, many based on the recording-plus-stimulation devices described above. In PD, researchers have shown that recorded neural signals might be used to auto-tune therapy parameters, eliminating some of the costly and time-consuming therapy calibration steps that occur right now for each new patient. Perhaps even more exciting, researchers have found patterns of brain signals that are correlated with current and future mood states and memory recall, critical steps into building effective neuromodulation therapies for depression, addiction, and Alzheimer’s. Similar studies using TMS have revealed potential pathways for new treatments for schizophrenia. These types of studies are generating so much data that the NIH has started to fund purpose-built data-warehouses to encourage researchers to share data so others can replicate studies and learn from the results. Over the next 5 years of the Brain Initiative, researchers are expected to run studies validating more neural-signal based biomarkers as well as starting to use larger real-world patient brain data to decode the particular neural circuits underlying common brain diseases.
In parallel with the technological and scientific advances described above, there is an equally important shift in the cultures of neurology and psychiatry that is accelerating adoption and acceptance of neuromodulation therapies: the movement towards circuit-based theories for brain disease.
If you were in your residency for neurology or psychiatry in the 80s or early 90s, you more likely than not learned that brain diseases were primarily a result of “chemical imbalances”, and that the natural treatment for these chemical imbalances was to find the right cocktail of drugs. However, with the widespread availability of fMRI and other neuroimaging modalities in the 90s and into the 2000s, evidence has mounted that the “brain-as-a-soup-of-chemicals-theory” is at best an oversimplification, and at worst outright wrong. In its place are new theories for mental brain disease that look at pathologies in different “neural circuits” — interconnected networks of neurons that relate to specific behavioral outputs like movement, focus and attention, memory, and motivation/reward. Tom Insel took important steps towards formalizing this approach with the Research Domain Criteria while serving as the Director of the NIH, and researchers such as Amit Etkin at Stanford continue to push these ideas further — in 2030, patients may not be diagnosed with “depression” at all but with specific deficiencies in a neural circuit which might be corrected with targeted neuromodulation or precision pharmaceutical intervention.
Neuromodulation therapies, with their temporal and spatial precision, are a natural fit for this new world of circuit-based understanding of brain disease, perhaps more so than traditional pharma therapies. It’s not surprising, then, that prescription rates (and market sizes) for both noninvasive and invasive neuromodulation therapies are growing. With the larger scale commercial release of recording-capable DBS and TMS devices in the coming years, this is likely to accelerate further as the data from these therapies gain utility beyond the neuromodulation therapy itself and can be used as a map for each patient’s specific pathology, like Jim in the story that started this post.
The next decade promises to be the most exciting yet for neuromodulation therapies and the science that underlies them; however, many challenges remain for bringing neuromodulation therapies for disease into mainstream clinical practice. One significant challenge is that the software and data systems that supported older generations of “open loop” DBS and TMS devices are poorly suited to new needs for connected devices, aggregated data, and advanced analytics. To address this need, we founded Rune Labs to build a software and data platform to support the development and delivery of new neuromodulation therapies. Since starting the company in 2018, we’ve built out the initial platform and are piloting it with academic and commercial partners developing new closed-loop DBS therapies for PD, Obsessive-Compulsive Disorder, and Depression. We hope that our platform can join with advances in technology, science, and clinical practice around neuromodulation to help hundreds of thousands (even millions!) of patients find treatment and live fully, and help everyone to better understand what is going on in the space between our ears.
Our team of engineers and scientists is looking forward to meeting you.