as recent controversies It has been highlighted around new Alzheimer’s disease treatments that our limited insights into the brain lead to difficulties in describing disease pathology, flawed clinical trial designs, and underutilized diagnoses of treatments — which are critical for many neurological disorders and diseases. It is so. For better outcomes, the precision neurology movement will require a trajectory similar to patient- and disease-specific advances in oncology, where data from patients is used to develop and deliver highly precise therapies.
Of course, the challenges in the brain are different, but we have begun to carve out a parallel track by leveraging multimodal data from new and existing technologies, from medical imaging techniques to digital biomarkers to real-world data. To continue this process, we must eliminate data silos, creatively interconnect disparate amounts of new data, and train algorithms to parse all of that data to paint a more complete picture of precision medicine development and care.
Diagnose the brain
Currently, many brain disorders are diagnosed by symptoms rather than causes. For example, there is no blood test for depression and no single biomarker for Alzheimer’s.Functional diagnosis of Parkinson’s disease with drugs trial and error. These diagnostic challenges also have implications for disease progression and treatment development.All kinds of Parkinson’s syndrome overlapping symptoms But it’s caused by different proteins clumping together in different parts of the brain, leading to different rates and complexity of clinical trial designs.
The good news is that a wide variety of potential data sources are available or under development that could serve as biomarkers for different aspects of neurological disease. Wearable devices allow for real-time self-reporting and motion detection, while implanted devices can observe the brain from the inside.
Data scientists are training algorithms to detect signs of Alzheimer’s, autism spectrum disorder (ASD), Parkinson’s and depression using tools that analyze sound, smell, GPS or behavior. Our challenge now is to validate and integrate datasets from each source with the expectation that together they provide the context needed to impact diagnosis and treatment.
Data Dilemma
This new field requires access and processing of sensitive data. That means spending a lot of time scrutinizing the practical, ethical and legal implications of this work.
The utility of brain data is limited by how it is processed and shared. Providing a clear signal to the user (whether the clinician or the direct patient) depends on the reliability of the analytical process, which requires a certain level of transparency.
It is also important to consider legal restrictions on how health data can be shared and with whom, and we have an ethical obligation to do so in the appropriate context, especially when presenting patients with their own data. The mutual push and pull of transparency and privacy requires evolving regulatory guidance to ensure that we all do the best for our collective good.
bridging synapses
If data analysis can help predict which patients are likely to develop brain-related diseases, it will open the door to insights into disease causes, treatment goals and precise patient matching.
Diseases such as Alzheimer’s and Parkinson’s have long plagued drugmakers for a number of reasons, not the least of which is that neurodegeneration can begin years or even decades before symptoms appear. Therapeutics are developed to target the biology of the disease, but often appear too late to affect the course of the disease.The first amyloid-targeted therapy finally approved For example, last year, but it’s good may be restricted For patients diagnosed with mild cognitive impairment in the early stages of the disease.
Currently, beta-amyloid detection by PET scan is the gold standard for Alzheimer’s diagnosis—although cerebrospinal fluid is sometimes used, and commercial blood test Once the FDA approves it may be next.Deep learning algorithms have been shown to be able to Detect mild cognitive impairment From functional MRI brain scans.
Most likely, early detection will rely on a combination of methods, and data from consumer technology can play a role.Digital biomarkers based on algorithms gleaned from web browser, cell phone or GPS usage data are already available Identify symptoms from behavior. Clinicians, who have long hoped that wearable technology could provide useful data between clinic visits, are making progress. Consumer devices such as smartwatches can detect motion sickness symptoms, dangerous falls and typical movements of physiological indicators such as stress responses or sleep patterns that often lead to a surge in symptoms.
Patients can also use consumer devices to manually track their medication plans and eating patterns and log unique or worsening symptoms that cannot be detected automatically.
This level of patient feedback helps guide diagnosis and treatment optimization, but is only the tip of the iceberg. Implantable electrical stimulation devices are increasingly used to treat symptoms of a variety of neurological disorders, including Parkinson’s disease, obsessive-compulsive disorder (OCD), and depression.
Some of the latest deep brain stimulation (DBS) devices are adaptive, incorporating sensors that capture feedback from brain signals to modulate impulses. Feedback data is also sent out to improve the algorithms that indicate the timing and pattern of electrical pulses.
By linking directly recorded brain data with behavioral data, we can further optimize treatment, personalize medication regimens, and respond flexibly to symptom progression. A recent example is adaptive DBS in patients with severe OCD, a disorder for which no reliable biomarkers exist today.
in a Studying at Brown University, Adaptive DBS data was combined with computer-recorded facial expressions and body movements in the clinic as well as self-reported symptom intensity and biometrics from wearable devices. The researchers were able to develop machine learning algorithms to identify potential OCD biomarkers, which were confirmed in a larger study.
forward thinking
Distinguishing psychiatric, neurodegenerative, and other central nervous system disorders based on symptoms dates back to a time when cancers were identified only by the organ in which they were found.
Can aggregated brain data biomarkers differentiate what we classify as “Parkinson’s” today, but as Parkinson’s, multiple system atrophy, progressive supranuclear palsy, or corticobasal degeneration at autopsy? Can physiologically similar anxiety disorders, subtyped by symptoms such as social anxiety disorder and agoraphobia, be described in a way that guides treatment development? Everything we’ve done to bring precision medicine into neuroscience shows us that it’s possible.
The AI-driven movement for precision neuromedicine, bringing together data from direct brain measurements, wearable technology, and patient experience, will unlock a new generation of therapies across the spectrum of neurological disorders.
Photo: Jolygon, Getty Images
