Nicola Davies, PhD

About 2.8 million individuals worldwide, including 1 million in the United States, currently live with multiple sclerosis (MS), the leading cause of nontraumatic neurological disability.

 Investigators have been striving for decades to find more effective ways to diagnose and manage MS, but the “gold standard” of evidence-based care—use of randomized controlled trials—has not been as beneficial for this disease as it has for others. “MS is a condition that has a great degree of variability across people and across time,” said Aaron Boster, MD, president of the Boster Center for Multiple Sclerosis in Columbus, Ohio. As such, he explained to 

®, it can be difficult to draw evidence from the relatively small number of patients recruited for clinical trials and standard observational studies, as they do not accurately reflect the status of the disease in the real world.

Could big data provide a solution to this problem?

Big data refers, essentially, to a large volume of data that can either be structured or unstructured. In the medical context, these data are obtained from a variety of sources: medical records, hospital records, or data generated from wearable devices and health care apps. Over the past few years, efforts by players in the field have focused on creating repositories of information by establishing online registries for MS.

“Such volumes of information are not normally accessible to the average neurophysician, who, over the course of 20 to 30 years, usually ends up seeing only a couple of thousand patients with MS,” Boster said.

When analyzed statistically, big data can be used to spot patterns and generate evidence that is applicable to the entire MS population. “Big datasets essentially help us understand health. The greater the number of data points involved, the greater the learning,” explained Ava Battles, MPsychSc, chief executive officer of Multiple Sclerosis Ireland.

Multiple factors play a role in the severity of MS, a disease that is not completely understood. Its exact cause is poorly grasped as well, and boundaries remain extremely blurred between the 2 distinct disease forms: relapsing-remitting MS (RRMS) and secondary progressive MS (SPMS). These deficits, understandably, can hinder disease management.

MS registries can collect patient details and synthesize them into information that can help improve understanding of the disease process. For example, Boster cites the most accepted definition of SPMS, which was created by identifying 576 versions of the definition and comparing them with a cohort of patients from MSBase, an international database that contains information on patients with MS throughout the world.

 The definition that performed the best in terms of timely diagnosis, disease activity, and overall disease burden was selected from the data. Using the Expanded Disability Status Scale (EDSS), SPMS is defined by: a disability progression of 1 step in patients with EDSS 5.5 or less, and of 0.5 step in patients with EDSS 6 or higher in the absence of a relapse; a minimum EDSS score of 4; a pyramidal functional system (FS) score of 2; and confirmed progression over 3 months, including confirmation within the leading FS.

Big data may also help in the identification of various factors that can predict disease progression. “Currently, the standard for monitoring patients with MS is through clinical assessments and radiographic monitoring using MRI scans. This needs to be done yearly to look for the formation of new lesions,” Gauruv Bose, MD, clinical research fellow in MS at the Brigham Multiple Sclerosis Center in Boston, Massachusetts, explained to 

However, meeting this standard may not be feasible in locations where access to care is limited. MRI scans, in particular, can be expensive and are not always covered by insurance. Another use for these datasets is to identify cost-effective prognostic markers to help predict disease progression in different individuals. For instance, results of a longitudinal observational study published in 2020 explored risk factors for conversion of RRMS to SPMS by utilizing datasets from more than 15,000 patients.

 The investigators found that older age, a longer duration of illness, higher disability scores, and more relapses in the previous year could predict progression to SPMS. Conversely, other factors that were previously thought to be associated with progression, including oligoclonal bands in cerebrospinal fluid and evidence of spinal cord lesions, were found to not be associated with progression.

Another recent observational study attempted to identify early clinical markers for aggressive MS by collating data from MSBase datasets. After screening almost 10,000 patients whose data were recorded over 10 years, specific markers were pinpointed: older age at symptom onset and higher disability scores and occurrence of pyramidal signs during the first year of MS. Based on these data, clinicians encountering these features in a patient may choose to be more aggressive in their early intervention.

Patients with MS often suffer from other medical comorbidities, which may or may not influence MS progression, and big data helps comprehend the various effects of these comorbidities. For instance, an observational study identified that patients with psychiatric comorbidities such as mood or anxiety disorders were more likely to have increased neurological disability.

 Another recent cohort study retrieved data from MSBase and explored the association between pregnancy and clinically isolated syndrome (ie, the first episode of neurological symptoms in MS). Preliminary findings from this study suggest that pregnancy can forestall the onset of MS.

The detection increased risk of certain comorbidities in patients with MS may offer another role for these large datasets. A matched cohort study from 2 different databases showed that patients diagnosed with MS have a 2-fold increase in rates of venous thromboembolism and peripheral vascular disease when compared with patients without MS. Additionally, the incidence of myocardial infarction in female patients with MS was 2.5 times higher than in those without MS.

Increasingly, decision-making is a process shared between physician and patient. Patients cannot make informed treatment choices unless they are aware of the risks and prognosis of their condition. Particularly in young patients who are newly diagnosed with MS, the risks may not be immediately apparent. “There is a significant disease lag in MS,” Boster explained. “The brain damage that occurs in a 30-year-old does not cause clinical concern at that time because of neuroplasticity and a large neurological function reserve. Although there might be temporary loss of function, this is usually regained— but the brain damage is permanent. Over the years, increasing bouts of brain damage will eventually lead to accelerated volume loss and progression of disability.” Study results corroborate that MS risk knowledge can be relatively low among patients,

 confirming the importance of educating patients on self-care. Results gleaned from large datasets of patients with MS can be valuable in this endeavor.


There is a significant disease lag in MS. The brain damage that occurs in a 30-year-old does not cause clinical concern at that time because of neuroplasticity and a large neurological function reserve. Although there might be temporary loss of function, this is usually regained—but the brain damage is permanent.

One such tool that can be used to educate patients is the MS Severity Rank Calculator, developed by investigators at MSBase by plotting the EDSS scores of a consolidated large volume of patients over time. The calculator can be used to categorize individual patients into a specific decile. Patients can be shown where they rank on the graph, which is a powerful indicator of the severity of their disease status. “Showing patients their own natural history over the course of time is extremely valuable in terms of treatment adherence,” said Boster.

Randomized trial results may not always represent the real-world efficacy of a drug, but for MS agents, the differences may be even starker. Patients in randomized trials are usually treatment-naïve and therefore have less potential for interactions that could impact the efficacy of a drug.

 Assessment of the real-world efficacy of a given MS medication can be done on a much larger scale through big data than is possible with randomized trials. For example, research from registry datasets show that ocrelizumab is more associated with respiratory infections than are other drugs for MS— a finding that was not apparent in clinical trials.

“As opposed to almost no treatments 20 years ago, we now have 24 FDA-approved therapies to treat MS. But we still don’t know the best way of using them all,” said Boster. This situation is due in large part to the drugs’ diversity. “While some drugs have similar mechanisms of action, many have unique pharmacological properties, which in turn creates differences in efficacy, as well as different [adverse] effect risks, routes of administration, and monitoring requirements,” added Bose.

Consequently, patients aren’t always prescribed the medication that is best for them. “Not all countries allow patients to have the medication their neurologist recommends, without first ‘failing’ another choice, one that is often cheaper and relatively safer,” said Bose. “If clinicians could use evidence derived from big data to characterize and prognosticate their patients’ condition accurately, policymakers could be persuaded to [pay for] the right medication for higher-risk patients when it is first prescribed.”

All of these factors—correctly diagnosing and prognosticating the patient, educating patients about the importance of adhering to treatment, and choosing the right drug for treatment—contribute to making informed health care decisions. “With appropriate data, a clinician can take an individual’s physiology and lifestyle into account, and they can potentially provide a detailed, holistic treatment plan for a person living with MS,” said Battles.

Making informed treatment choices also includes timing treatments correctly. Although disease-modifying therapy is often only initiated after obvious neurological disability, such as walking issues, information from big data has shown that cognitive decline—a relatively less visible parameter—can often signal the beginning of MS.11 This is an emerging advance for care, in that it offers parameters to assess cognitive function earlier in the disease process.

While data is obtainable from various sources, the key sources of these datasets thus far for MS research have been disease registries. Unlike patient and hospital records, which capture data passively, disease registries actively encourage data entry and the sharing of information, and while relatively small, they have been growing. Several European MS registries have more than 500,000 total patients in their databases, potentially covering a large swath of the estimated 700,000 people with MS on the continent. As well, attempts have been made to harmonize the data between registries for additional insight, with the EUReMS and BMSD projects currently analyzing the results of this for datasets from over 60,000 and 100,000 patients, respectively.12 North America has been catching up slowly, and now several registries capture information from patients on this continent as well.

Four of the key big data sources for MS are as follows:

. An international registry, MSBase is open to neurophysicians worldwide. Patient data can be shared on this registry with appropriate consent, and these data can be used by investigators to track and evaluate outcomes. “The MSBase registry has capture visits from sites across many countries, and several excellent analyses have been done using this and similar datasets,” said Bose. MSBase currently has records from approximately 74,000 patients in 38 countries.

. Supported by the MS International Foundation, this database actively collects epidemiological information on MS worldwide. In addition to prevalence across various patient demographics, the database also offers information on health care resources’ availability and accessibility.

The North American Registry for Care and Research in MS

. This database links MS centers across the United States and Canada. A collaborative registry, it has repositories of clinical, genetic, and radiographic data of more than 800 patients with MS from 24 institutions.

. This new registry, whose name is an acronym for COVID-19 Infections in MS & Related Diseases, was formed in wake of the coronavirus disease 2019 (COVID-19) pandemic. It was established by the Consortium of Multiple Sclerosis Centers and the National Multiple Sclerosis Society to help investigate how COVID-19 affects the MS population, and it already contains data from about 900 patients with MS who have had COVID-19.

Challenges in Using Big Data for MS Research

The most common fear shared by both physicians and patients regarding these collections of information is breach of privacy. “Sharing your most private, most intimate information about your health has fear attached to it,” said Boster. “First is the fear that your data might fall into the wrong hands or be used inappropriately, and second, that datasets might actually limit your options for treatment.” Clinicians can mitigate patient fear by being open about how big data is used. Indeed, Battles stressed, “People consenting to their data being shared should be informed of who will have access to the data and what [they] will be used for. They should also know how to have their data removed, should they wish to do so at any time.”

Another key challenge is the inherent bias that these datasets may have. “There may be differences in the examination and rating tools used, although efforts are underway to standardize these,” said Bose. He added that incomplete or missing data can also create a bias. For instance, not all countries have the resources to allow frequent MRI examinations. Certain countries may not contribute to databases at all, creating a geographical bias. The Atlas of MS database records data from 138 countries, but it has excluded at least 80 countries where coordinators could not be identified. The world’s largest MS database, MSBase, has data from only 38 countries.

Despite these challenges, most patients are inclined to share their experiences and health data on global registries. “We see that patients are willing to fight back and do what it takes to contribute toward the understanding of the disease, whether this is participating in clinical trials or sharing their health information,” said Boster.

The Future of Big Data and Advances in MS

The wealth of insights that big data has the potential to offer is exciting, experts agree. According to Bose, one avenue of future research could be developing analytical tools that can be applied to big data to create and refine specific treatment algorithms. Another focus should be toward consolidating information across different databases, said Boster. “Increasing efforts at consolidating databases is essential so that we can have larger, robust, homogenous datasets to work from,” he noted.

Boster also expresses hope that consolidated big data can be applied to studying the prodromal stage of MS. “We know from the literature that for almost 5 years leading up to an MS diagnosis, patients have certain abnormal patterns, such as increased emergency [department] visits, pain, or psychiatric complaints. If we are ever to a priori identify the trappings of an MS prodrome, it is going to be through mining big data for clues and tips that lead us there,” he says. Identifying patients in the preclinical stages could certainly revolutionize the management of the disease, as the next logical step would be to identify treatment strategies that can stave off the clinical course altogether.

Multiple Sclerosis International Federation; 2020. Accessed January 9, 2021. https://www.msif.org/wp-content/uploads/2020/10/Atlas-3rd-Edition-Epidemiology-report-EN-updated-30-9-20.pdf
2. Lorscheider J, Buzzard K, Jokubaitis V, et al; MSBase Study Group. Defining secondary progressive multiple sclerosis. Brain. 2016;139(Pt 9):2395-2405. doi:10.1093/brain/aww173
3. Fambiatos A, Jokubaitis V, Horakova D, et al. Risk of secondary progressive multiple sclerosis: a longitudinal study. 

. 2020;26(1):79-90. doi:10.1177/1352458519868990
4. Malpas CB, Manouchehrinia A, Sharmin S, et al. Early clinical markers of aggressive multiple sclerosis. 

. 2020;143(5):1400-1413. doi:10.1093/brain/awaa081
5. McKay KA, Tremlett H, Fisk JD, et al; CIHR Team in the Epidemiology and Impact of Comorbidity on Multiple Sclerosis. Psychiatric comorbidity is associated with disability progression in multiple sclerosis. 

. 2018;90(15):e1316-e1323. doi:10.1212/WNL.0000000000005302
6. Nguyen A, Vodehnalova K, Kalincik T, et al. Association of pregnancy with the onset of clinically isolated syndrome. 

. Published online September 14, 2020. doi:10.1001/jamaneurol.2020.3324
7. Persson R, Lee S, Ulcickas Yood M, et al. Incident cardiovascular disease in patients diagnosed with multiple sclerosis: a multi-database study. 

. 2020;37:101423. doi:10.1016/j. msard.2019.101423
8. Giordano A, Liethmann K, Köpke S, et al; AutoMS Group. Risk knowledge of people with relapsing-remitting multiple sclerosis – results of an international survey. 

. 2018;13(11):e0208004. doi:10.1371/ journal.pone.0208004
9. Rojas JI, Pappolla A, Patrucco L, Cristiano E, Sánchez F. Do clinical trials for new disease modifying treatments include real world patients with multiple sclerosis? 

. Published online January 3, 2020. doi:10.1016/j.msard.2020.101931
10. Smoot KE, Stuchiner T, Lucas L, et al. Utilization, safety, and tolerability of ocrelizumab: year 2 data from the Providence Ocrelizumab Registry. Presented at: 35th Congress of the European Committee for Treatment and Research in Multiple Sclerosis; September 11-13, 2019; Stockholm, Sweden. Accessed January 14, 2021. https://onlinelibrary.ectrims-congress.eu/ectrims/2019/ stockholm/278213/kyle.smoot.utilization.safety.and.tolerability.of.ocrelizumab.year.2.data.from. html?f=listing%3D3%2Abrowseby%3D8%2Asortby%3D2%2Amedia%3D3%2Asearch%3Docrelizumab
11. Cerqueira JJ, Compston DAS, Geraldes R, et al. Time matters in multiple sclerosis: can early treatment and long-term follow-up ensure everyone benefits from the latest advances in multiple sclerosis? J 

. 2018;89(8):844-850. doi:10.1136/jnnp-2017-317509
12. Glaser A, Stahmann A, Meissner T, et al. Multiple sclerosis registries in Europe—an updated mapping survey. 

. 2019;27:171-178. doi:10.1016/j.msard.2018.09.032

Articles

Big data is on the precipice of revolutionizing multiple sclerosis knowledge and treatment.

How Big Data Can Drive Advancements in MS Research

Alzheimer disease (AD) is a leading cause of disability and the fifth leading cause of death in patients age 65 or older.

 As the aging population continues to grow, the incidence of AD also increases, spurring research into innovative ways of detecting, managing, and treating the condition. Digital biomarkers, in particular, have aroused interest from researchers and manufacturers, as they have the potential to provide precise, quantifiable physiological information that can be used as diagnostic and prognostic indicators.

Any medical data that are objectively collected through digital technology, such as wearable devices and smartphones, are referred to as “digital biomarkers.” Given that the wearable health care device market is predicted to grow by 20% in the next 5 years,

 digital biomarker are an attractive option within the medical community.

Digital biomarkers show definite promise in detecting AD in its early stages.

“Although the gold standard for diagnosis is an autopsy or positron emission tomography (PET) scan, clinically validated end points for digital biomarkers are on their way,” said Graham Jones, PhD, director of innovation at Novartis. Indeed, over the past few years, much research has focused on the utility of digitally acquired data in detecting AD and predicting cognitive decline.

Based on their method of acquisition, 2 major categories of digital biomarkers are currently being employed for neurologic conditions

Active biomarkers, as the name suggests, require users to be actively involved in recording the parameter being measured. These biomarkers generally assess cognitive function through digitally administered tests or quizzes.

Passively collected data, on the other hand, occurs without active user engagement and results in continuous, real-time data acquisition. For instance, data such as step count, gait symmetry, and clarity of speech can all be recorded during spontaneous activities, and without any engagement on the part of the user.

According to Mircea Balasa, MD, PhD, a neurologist at Hospital Clinic de Barcelona in Spain, innovative digital biomarkers based on cognitive testing are being devised to improve on existing neuropsychological symptom tests, making them more standardized and reliable. One such innovation is using gamified versions of standard memory tests to make it more attractive and increase patient engagement. For example, Balasa and his team have been working on validating the Altoida Neuro Motor Index device, which is based on augmented reality technology and consists of a 10-minute gamified assessment that tests a patient’s functional and cognitive aptitude.

 “At present, the Altoid device shows promise in distinguishing patients with mild cognitive impairment from controls,” Balasa told 

Research has shown that AD presents with a definitive decline in sensory and motor functioning, even before cognitive decline begins.

 These deficits include general visual parameters, such as visual acuity, color perception, contrast sensitivity, and visual integration, all of which are reduced in patients with mild AD.

 Investigators have suggested that passive tests can measure the amount of fixation time on certain graphics that are repeatedly viewed, to quantify visual parameters over time.

 Both peripheral and central auditory system dysfunction occur in the prodromal stages of AD,

 and passively monitoring conversations could be one way of screening for AD. For instance, a study showed that atypical repetitions in 2 conversations that occurred on different days was a strong indicator of AD.

 Reading comprehension can also be evaluated by tracking eye movements.

In addition, slow gait speed is a strong predictive factor for dementia and cognitive decline.

 Measurement of gait speed using an accelerometer or gyroscope sensors has the potential to improve the detection of AD. Also, touchscreen sensors can evaluate fine motor control capabilities such as typing skills, swiping, and other keystroke patterns.

Imaging studies have shown that neuropathological lesions can develop in the autonomic nervous system (ANS) in the earliest stages of AD. “The ANS is fertile ground for diagnosis,” Jones told 

. “The ANS impacts several factors that can be picked up by digital tools—heart rate, reflexes, and respiration, for instance.” Indeed, the earliest stages of AD have been associated with heart rate variability (HRV),

 and mobile devices with built-in electrocardiogram (ECG) capabilities, such as the Apple Watch, can be used to monitor HRV in at-risk patients.

 AD is also associated with disturbed circadian rhythms,

 so devices that have built-in ballistocardiography sensors can be used to monitor sleep patterns.

Determining the Most Accurate Digital Biomarkers

Although investigators have studied a variety of digital biomarkers, very few have dealt with real-world, home-based evaluations. In 2019, Evidation Health set the ball rolling when they partnered with Apple and Eli Lilly and Company for a real-world study that produced around 16 terabytes of digital biomarker data. The investigators were able to establish that specific motor skills, such as reading, typing, and keystroke tasks were performed more slowly in patients who had mild cognitive impairment and dementia.

In 2018, the United Kingdom (UK) Alzheimer’s Society and the University of Oxford’s Big Data Institute launched the GameChanger project, which employs a smartphone app called Mezurio.

 Volunteers across the UK are being recruited to complete activities each day using the app, from which the data collected will be used to study markers of healthy cognition and determine factors that would be affected by cognitive decline.

Another project in the European Union, RADAR-AD, is focused on exploring the potential of mobile technologies in measuring disability progression associated with AD,

 while the US-based Alzheimer’s Drug Discovery Foundation is supporting digital biomarker exploration through its Diagnostics Accelerator.

"There is a very real expectation that digital biomarkers can monitor progression from prodromal to diseased states...[and] some evidence that learned behaviors—such as performing specific functions on tablets—coupled with certain exercise regimens, can delay the onset and severity of diseases like AD."


Digital biomarkers have the potential to advance several stages of the disease process:

 The current standard for diagnosis of AD is through PET molecular imaging, which is expensive and often only available at regional facilities. In contrast, when patients use mobile or wearable devices, there is continuous monitoring of passive biomarkers and data can be compared to the patient’s previous data. This makes it easier to detect any decline in nervous system parameters. “Any deviations from normal biomarker levels can be used as an indication to recommend more specific investigations, such as PET scans,” Jones said.

 Currently, neuropsychological testing is used to monitor cognitive function in high-risk individuals. However, these tests are not always reliable and may fail to detect the earliest stages of cognitive impairment. According to Balasa, active digital biomarkers could be used instead to create prognostic models that can help monitor patients who are at higher risk of developing cognitive decline. “Such individuals do not normally seek medical advice, as they don’t usually have clinically relevant cognitive decline,” he said.

 Digital biomarkers are a cost-effective solution to collect objective, continuous, quantifiable data in a longitudinal manner. The precision of this data can increase the statistical power of clinical trials.

 In the advanced stages of AD, neurons are destroyed and the opportunity to study disease pathology is limited. By identifying patients in the early stages of the disease via digital biomarkers, researchers can gain a deeper understanding of the pathophysiology of AD, which can help develop novel treatment strategies.

Wearable devices still have a long way to go before they can be reliably used for the diagnosis and management of AD. “The issue for most of these devices is clinical validation with appropriate double-blinded trials,” Jones said. For effective real-world usage, clinical validation needs to be followed by regulatory approval. The Altoida device is currently the only device-specific for AD that has FDA approval and CE marking. While the Apple Watch has FDA approval for fall detection and ECG rhythm monitoring, its AD-specific features still need validation and approval.

Actively measured biomarkers require dedicated engagement on the part of the patient, who must manually enter inputs into the recording device. Lack of interest or even beginning memory loss are some challenges that can prevent the effective use of this kind of biomarker. While passive biomarkers do not face these challenges, the large volumes of data that are acquired create the need for extensive storage systems. Analysis of this data may require complex tools and trained personnel.

In addition to these disadvantages, both kinds of data face another challenge: wearable devices record vast amounts of medical data that must remain confidential. “We need to navigate data privacy and Health Insurance Portability and Accountability Act regulations,” said Jones. Some applications such as Apple Health allow users to access their health care data directly through their own devices, which may increase the need for secure access to such data.

In Jones’ opinion, apart from timely diagnosis, there are 2 future directions that hold promise. The first is monitoring disease progression. “There is a very real expectation that digital biomarkers can monitor progression from prodromal to diseased states,” he said, adding that cognitive performance could also be tracked digitally. Secondly, and more promising, is the possibility of actual treatment using digital tools. “There is some evidence that learned behaviors—such as performing specific functions on tablets—coupled with certain exercise regimens, can delay the onset and severity of diseases like AD,” he said. Of course, these digital forms of treatment need sound exploration and validation before they become a reality.

REFERENCES
1. Alzheimer’s Association. 2020 Alzheimer’s disease facts and figures. 

 2020;16(3):391-460. doi:10.1002/alz.12068
2. Wearable healthcare devices market by type (diagnostic (ECG, heart, pulse, BP, sleep), therapeutic (pain, insulin), application (fitness, RPM), product (smartwatch, patch), grade (consumer, clinical), channel (pharmacy, online) - global forecast to 2025. Markets and Markets. May 2020. Accessed November 19, 2020. 

https://www.marketsandmarkets.com/Market-Reports/wearable-medical-device-market-81753973.html


3. Kourtis LC, Regele OB, Wright JM, Jones GB. Digital biomarkers for Alzheimer’s disease: the mobile/wearable devices opportunity. 

. 2019;2:9. doi:10.1038/s41746-019-0084-2
4. Buegler M, Harms R, Balasa M, et al. Digital biomarker-based individualized prognosis for people at risk of dementia. 

. 2020;12(1):e12073. doi:10.1002/dad2.12073
5. Albers MW, Gilmore GC, Kaye J, et al. At the interface of sensory and motor dysfunctions and Alzheimer’s disease. 

. 2015;11(1):70-98. doi:10.1016/j.jalz.2014.04.514
6. Salobrar-García E, de Hoz R, Ramírez AI, et al. Changes in visual function and retinal structure in the progression of Alzheimer’s disease. 

. 2019;14(8):e0220535. doi:10.1371/journal.pone.0220535
7. Swords GM, Nguyen LT, Mudar RA, Llano DA. Auditory system dysfunction in Alzheimer disease and its prodromal states: a review. 

. 2018;44:49-59. doi:10.1016/j.arr.2018.04.001
8. Yamada Y, Shinkawa K, Shimmei K. Atypical repetition in daily conversation on different days for detecting Alzheimer disease: evaluation of phone-call data from regular monitoring service. 

. 2020;7(1):e16790. doi:10.2196/16790
9. Fernández G, Manes F, Politi LE, et al. Patients with mild Alzheimer’s disease fail when using their working memory: evidence from the eye tracking technique. 

. 2016;50(3):827-838. doi:10.3233/JAD-150265
10. Grande G, Triolo F, Nuara A, Welmer A-K, Fratiglioni L, Vetrano DL. Measuring gait speed to better identify prodromal dementia. 

. 2019;124:110625. doi:10.1016/j.exger.2019.05.014
11. Kim MS, Yoon JH, Hong JM. Early differentiation of dementia with Lewy bodies and Alzheimer’s disease: heart rate variability at mild cognitive impairment stage. 

. 2018;129(8):1570-1578. doi:10.1016/j.clinph.2018.05.004
12. Owens AP. The role of heart rate variability in the future of remote digital biomarkers. 

 2020;14:582145. doi:10.3389/fnins.2020.582145
13. Uddin MS, Tewari D, Mamun AA, et al. Circadian and sleep dysfunction in Alzheimer’s disease. 

. 2020;60:101046. doi:10.1016/j.arr.2020.101046
14. Chen R, Jankovic F, Marinsek N, et al. Developing measures of cognitive impairment in the real world from consumer-grade multimodal sensor streams. 

KDD ‘19: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining


15. Study information. Game Changer. Accessed November 19, 2020. https://joingamechanger.org/information/
16. Strategy. Radar-AD. Accessed November 19, 2020. https://radar-ad.org/about-us/strategy
17. Diagnostics accelerator: digital biomarkers. Alzheimer’s Drug Discovery Foundation. Accessed November 19, 2020. https://www.alzdiscovery.org/research-and-grants/funding-opportunities/digital-biomarkers

Digital biomarkers offer promise to revolutionize the diagnosis and treatment of Alzheimer disease via our increasing ability to gather active and passive data.

Data Evolution in Alzheimer Disease: Is the Future Digital?

Sleep-disordered breathing (SDB) refers to any disorder in which the patient has difficulty breathing while asleep, including obstructive sleep apnea (OSA) and central sleep apnea (CSA). OSA, which occurs because of upper airway obstruction during sleep, is far more common, with estimates of approximately 936 million people worldwide suffering from the condition.

“With OSA, snoring and breathing pauses and reduced airflow during the night leads to sleep fragmentation and unrefreshing sleep, which in turn impairs daytime functioning,” said Stuart MacKay, BSc(Med), MBBS(Hons), FRACS, associate professor at Illawarra ENT Head and Neck Clinic in Australia. 

“Patients can feel groggy, tired, [and] sleepy and have a higher rate of motor vehicle collisions and workplace accidents. In addition, via a variety of pathophysiological pathways, but especially due to repetitive oxygen desaturation, patients with more significant sleep apnea are at risk of hypertension, arrhythmias, ischemic events, and other cardiometabolic consequences.”

A small subset of patients may have CSA, when the brain fails to send signals to breathe. CSA is less common in the general population, but there are increased rates in people with specific conditions, such as heart failure and chronic opioid use.

Although OSA and CSA have different causes, both contribute to reduced quality of life and have been linked to decreased cognition, increased cardiovascular risk, and increased mortality.

Continuous positive airway pressure therapy: Not suitable for all

Most sleep experts recommend continuous positive airway pressure (CPAP) as the first-line treatment for SDB, regardless of the severity of apnea. The CPAP device continuously delivers pressurized air into the upper airway, which helps keep the air passages open. Results from studies have shown that using CPAP therapy can significantly reduce the severity and morbidity associated with sleep apnea.

The main limiting factor in the use of CPAP for OSA is the lack of compliance to therapy. The device delivers air under moderate to high pressure, which can cause a certain level of discomfort for patients. “Although it is minimally invasive, not everyone can tolerate CPAP,” said Peter M. Baptista, MD, PhD, a senior clinical consultant at the Otorhinolaryngology Unit at Clínica Universidad 

de Navarra in Spain. “In addition, CPAP can cause nasal obstruction and feelings of claustrophobia.”

One cohort study estimated that only 37% to 41% of patients adhered to CPAP therapy in the long term,

 which ultimately hinders the achievement of good treatment outcomes. A recent meta-analysis showed that the cardiovascular risk associated with OSA was reduced only when patients wore the device for longer than 4 hours per day.

For patients with CSA, it is more than just the discomfort of CPAP that can raise problems. “With central sleep apnea, there is an interruption in the neural drive to breathe,” Sanjaya K. Gupta, MD, assistant professor of medicine at the University of Missouri-Kansas City School of Medicine, told 

®. “There is no movement of the diaphragm, so using CPAP does not work unless the patient is making an effort to breathe.”

Clinicians have used a wide range of surgical procedures in OSA, including palatopharyngoplasty, tonsillectomy, and maxillomandibular advancement.

 However, each surgery targets only
1 anatomical area. Results from sonographic studies have shown that OSA presents with multilevel airway collapse, which means that a single surgery may fail to resolve the complete issue.

Surgery is not an option for patients with CSA. Apart from oxygen ventilation, the only other ways to manage CSA include treating the underlying heart condition that is present and cutting back
on opioid use.

Nerve stimulation for the management of sleep apnea

The search for newer, better alternatives to CPAP has resulted in 1 promising solution––nerve stimulation. By stimulating selected nerves, specific muscles that aid in airway relaxation or breathing can be made to contract. This addresses the problem of sleep apnea at a physiological level, which would reasonably be more effective and have fewer adverse effects.

For OSA, clinicians choose the hypoglossal nerve as a target for stimulation. “Hypoglossal nerve stimulation [HGNS] provides an opening of the airway at the level of both the tongue and palate,” said Baptista. Nerve stimulation causes the genioglossus muscle to contract, which moves the tongue forward. According to research by MacKay and colleagues, it is important to only stimulate the genioglossus. “Selective stimulation of the nerve fibers to the genioglossus, the main dilator of the airway, allows [the] avoidance of stimulating retrusor muscles that can compromise airway opening,” he said. “Results of such implantation are likely to improve patient and polysomnographic outcomes.”

Research results have shown that there is also palatoglossal coupling, which increases the airway space at the retropalatal level.

 This dilates and stabilizes the airway at multiple levels. In a recent systematic review, Baptista and colleagues reviewed 12 studies that used HGNS for OSA.

 They found the procedure reduced the number of apneic and hypopnea events per hour by 18 points, measured via the apnea-hypopnea index (AHI). There was also a reduction of 14.6 points in the number of oxygen desaturations measured via the oxygen desaturation index (ODI). According to MacKay, “The advantages of HGNS include the ability to avoid devices worn on the face and in the mouth, and avoidance of painful intra-oral/pharyngeal surgeries. Disadvantages include the requirement for commitment to titrations, follow-up, and ongoing device use.”

There are different types of HGNS, namely unilateral and bilateral. “Bilateral HGNS allows a higher likelihood of symmetrical tongue and, in cases of ‘coupling,’ palatal movement to improve airway patency in sleep,” MacKay told 

®. “Advantages over currently available unilateral stimulation include [fewer] components to implant, and external stimulation rather than an implanted ‘pacemaker.’ The incision is small and relatively painless.”

In CSA, stimulation of the phrenic nerve can be just as effective. This eliminates periodic respiration by causing consistent, regular diaphragmatic contractions, which generates negative intrathoracic pressure and restores a regular breathing pattern. Last year, investigators reported the results of using a phrenic nerve stimulator over a 3-year period.

 The result was sustained improvement across a wide range of sleep metrics, including the AHI, ODI, central apnea index, and arousal index. More recently, a pooled meta-analysis of 5 different studies showed that phrenic nerve stimulation could effectively reduce AHI scores by approximately 26.7 events per hour, as well as reduce the ODI scores by approximately 24.

The nerve stimulator needs to be implanted into the patient’s body to function. Several different devices are available to treat OSA through HGNS. In general, the device consists of a sensor electrode, a pulse generator, and a stimulation electrode.

 The sensor electrode is usually implanted into the intercostal muscles, or just below the ribs. During inspiration, it gives off an electric signal that is picked up by an implantable pulse generator (IPG). The IPG is usually placed in a pocket below the clavicle, medial to the delto-pectoral groove. After receiving an electric signal, it generates a pulse that is transmitted to the stimulation electrode, which is implanted around the medial branches of the hypoglossal nerve.

The device used in CSA has a slightly different design; a sensing lead is placed in the azygous vein, behind the heart, to the level of the diaphragm to detect breathing. A pacing electrode is placed in a vein adjacent to the phrenic nerve. “The device works 

automatically and in concert with patients’ normal respiratory cycle,” said Gupta. “The pacing and sensing leads are plugged
into a device that looks and functions similar to a cardiac pacemaker. Once connected, the device and leads are surgically placed under the skin in the right upper chest, just below the collarbone. The device is programmed to sense the patient’s breathing and the moment when the breathing stops, the device will deliver electricity through the pacing lead to the phrenic nerve, which will stimulate the diaphragm to contract.” All this happens automatically and seamlessly, in concert with the patient’s normal respiratory cycle. No intervention is necessary on behalf of the patient to initiate therapy, they don’t have to wear a CPAP mask, and the device turns off when they get up in the morning.

Can all patients with SDB receive nerve stimulation therapy? According to most experts, the answer is no. The type of patient selected can greatly influence treatment outcomes, and establishing which sleep apnea the patient has is a key priority. HGNS will not work if the patient has CSA. Conversely, phrenic nerve stimulation only benefits those with CSA; it is not indicated for those with primarily OSA. There is a third type of sleep apnea called “treatment emergent” CSA

 This is primarily OSA in which a central component emerges following the use of CPAP therapy. Uncertainty remains about whether nerve stimulation can benefit this subset of patients.

An obvious indication for HGNS is for patients who are unable to tolerate CPAP. However, before clinicians consider this procedure, the patient must be evaluated for other factors. “Patients should see credentialed sleep surgeons for a thorough assessment and discussion of treatment options if they cannot tolerate or adhere to CPAP or other devices long-term,” advised MacKay.

In a recent review study, Baptista highlighted 3 key factors that must be considered during patient selection.

 These include the severity of OSA, body mass index (BMI), and the severity of airway collapse at the palatal level. “If patients have [a] very high BMI (> 35) or concentric palatal collapse, HGNS is best avoided,” said Baptista. Obesity is a predictor for poor OSA treatment outcomes regardless of the type of treatment. In terms of severity of airway collapse at the palatal level, HGNS relaxes the palatal and retrolingual musculature, but if there is concomitant collapse in the posterior or lateral pharynx, this technique may not be effective. To ascertain the exact area of collapse, drug-induced sleep endoscopy can serve as a valuable diagnostic tool.

Nerve stimulation is a relatively expensive technique compared to CPAP and may not be financially feasible for patients with mild OSA. The FDA recommends using this technique for patients with moderate to severe apnea with an AHI score between 15 and 65.

For phrenic nerve stimulation, the severity of apnea and the proportion of central events are the main considerations. “We generally target patients with moderate to severe sleep apnea where at least 80% of the events are central in nature,” says Gupta.

Implantable nerve stimulation seems to be a reasonably safe method of treatment. In their meta-analysis, Baptista’s team analyzed the number of adverse events related to HGNS. Serious events were documented in 6% of all patients, for whom there was a requirement for surgical repositioning of the device or the replacement of electrodes.

 Minor adverse effects were reported by almost 60% of all patients, including discomfort from electrical stimulation and tongue abrasion due to constant movement. Although these effects were common during the first year of treatment, they reduced in subsequent years.

The meta-analysis for phrenic nerve stimulation showed serious adverse events in 3.9% of all patients. These included hematoma, migraine, and atypical chest pain, which necessitated the removal of the device. Minor adverse effects were reported in about 5.3% of all patients, including device discomfort, dislocation of leads, and lead failure.

 The more severe adverse effects may be avoided by refinements in surgical technique, whereas discomfort mostly resolves with time as the patient gets used to the device.

As the use of nerve stimulators continues to rise, the onus is now on making these devices more efficient and increasing their capabilities. Baptista believes that future research must focus on finding ways to circumvent lateral airway collapse at the palatal level. “We are just at the very beginning,” he said. “In a few years, there will be devices that are much smaller and perhaps capable of stimulating a specific group of fibers that allow airway opening.”

Research into nerve stimulation therapy still has a long way to go. Although prospective studies have illustrated the therapy’s standalone efficacy, there is no research comparing the use of this therapy to CPAP, which is still the gold standard for OSA. Longitudinal studies are also needed to assess secondary outcomes. “We still do not know if phrenic nerve stimulation can reduce morbidity and mortality rates,” Gupta pointed out.

According to Mackay, “Areas that represent the most exciting future investigation are patient preferences, combination therapies, further refinement of implantation and titration techniques, miniaturization of devices, and endophenotyping outcomes in low muscle hyporesponsive-type OSA patients.”

Most physicians are aiming for a more accurate diagnosis and evaluation, prior to wider acceptance of current treatment strategies. “For patients with an appropriate indication, the phrenic nerve pacing procedure can be life-changing,” stated Gupta. However, nerve stimulation has yet to gain wide acceptance. One main barrier, especially in the case of CSA, is the lack of a proper diagnosis. “CSA patients lack traditional symptoms such as snoring and daytime somnolence,” said Gupta, who feels that the only way to improve diagnostic accuracy is to maintain a high index of suspicion. “If you have a patient with heart failure, atrial fibrillation, or stroke who continues to complain of fatigue despite optimal medical therapy, consider the possibility of CSA,” he advised. A sleep study or polysomnogram is the best diagnostic test for sleep apnea and clinicians should recommend it to such patients.

Addressing sleep apnea at a physiological level may reasonably be more effective and have fewer adverse effects than more traditional treatment modalities that may only target one mechanism.

Stimulating the Future of Treating Sleep-Disordered Breathing

Although there have been relatively effective treatments for seizures since 1912, beginning with phenobarbital, the focus on the symptoms of epilepsy rather than the cause means that many patients with difficult to treat epilepsy still long for effective therapies. This is especially the case for autoimmune epilepsy, a under-recognized condition that is often unmanageable with conventional antiseizure medications.

 As a result, a truly “seizure-free” life has remained out of reach for many. This may be set to change as researchers continue to uncover the causes of different types of epilepsy and the mechanisms of action of different treatment modalities.

“Autoimmune epilepsy is essentially a seizure disorder where a neural-specific antibody is present, may be pathogenic, and may cause seizures,” said Lindsay Higdon, MD, a neurologist and epilepsy specialist at Jefferson University Hospital in Philadelphia, Pennsylvania. In one study, experts found that approximately 20% of unexplained epilepsy cases had evidence of circulating neural antibodies, suggesting an autoimmune cause.

 These antibodies can target antigens present on neural surfaces or within the neural cell.

Over the past decade, several neural surface antigens and intracellular antigens have been found and documented in patients presenting with epilepsy, and in 2017, autoimmune epilepsy was officially recognized as a distinct entity by the International League Against Epilepsy.

“There are several clues,” Sarosh Irani, BMBCh MA (Oxon), DPhil, head of the Oxford Autoimmune Neurology Group and associate professor at Oxford University in the United Kingdom, told 

. “These patients often present with an explosive onset, and a higher frequency of seizures than normal.” Most of these patients also present with other features of autoimmune encephalitis, such as memory disturbances, mood changes, and changes in personality. Some patients experience faciobrachial dystonic seizures, which are essentially diagnostic of an underlying antibody.

Patients with autoimmune epilepsy can also show characteristic MRI findings that are consistent with limbic encephalitis. A retrospective cohort study found that such patients usually demonstrated parenchymal atrophy that was disproportionate to age, or T2 hyperintensities in cortical or subcortical regions.

To help clinicians make an early and accurate clinical diagnosis, Divyanshu Dubey, MBBS, and a team of investigators at Mayo Clinic have developed and validated a predictive tool.

 The APE (Antibody Prevalence in Epilepsy of unknown etiology) score takes into consideration the onset and progression of the seizure disorder, clinical signs, and MRI and cerebrospinal fluid (CSF) findings. If the APE score is greater than 4, there is a high chance that the epilepsy is of autoimmune origin.

One of the main hallmarks of autoimmune epilepsy is that it is generally resistant to antiepileptic drugs. In a systematic review of 6 studies that included 169 patients with autoimmune epilepsy, antiepileptic drugs were effective in only 10.7% of cases.

 “Antiepileptic drugs may still be effective in a few cases, but they treat the symptom of the autoimmune condition, and not the cause,” Higdon said.

Once the cause is known, it makes sense to treat autoimmune epilepsy just like other autoimmune diseases—by modulating the immune system. “Along with antiepileptic drugs, we use either steroids, intravenous immunoglobulin (IVIG), or plasma exchange,” Higdon said. “All 3 treatments decrease immune-mediated inflammation in the central nervous system by suppressing or changing the immune system or removing anti- bodies.” Indeed, one study showed that immunomodulatory therapy led to a 19-fold increase in odds of a better seizure outcome, with plasmapheresis (odds ratio [OR], 32.5) proving to be more effective than intravenous steroids (OR, 8.25).

 In another study, about 75% of patients became seizure-free in the 2 months after starting immunotherapies.

Many researchers are exploring IVIG—a concentrate of human antibodies obtained from healthy donors&mdash;as an alternative to steroids and plasmapheresis. Its administration is a simpler procedure compared with plasmapheresis and can help avoid the adverse effects of steroid therapy.

This form of treatment—flooding the body with more antibodies&mdash;may seem counterintuitive. However, IVIG is believed to stabilize and strengthen the immune system. How does this happen? “Currently, no one can answer that accurately, but there are a few hypotheses,” said Irani. “The IVIG possibly saturates the immune system, preventing key causative antibodies like leucine-rich glioma-inactivated 1 (LGI1) and contactin-associated protein-like 2 immunoglobulin G (CASPR-2) from functioning. IVIG antibodies could also directly interfere with these antibodies, or with their targets.”

A recent randomized controlled trial conducted at Mayo Clinic provided further evidence that IVIG was effective in reducing seizures in a cohort who met criteria for autoimmune epilepsy.

 Patients who tested positive for LGI1 antibodies and CASPR-2 were enrolled in the study. Of 8 patients enrolled in the IVIG arm, 6 showed more than 50% improvement in seizure frequency compared with none in the placebo group. After 5 weeks, 6 patients in the placebo arm of the trial who had persistent seizures were switched to IVIG; 4 of these then reported more than 50% improvement in seizure frequency. IVIG is still a relatively new treatment strategy and needs more research before it is adopted as the main modality of treatment. “We don’t know how it accesses the central nervous system, and how it affects B cells that produce target antibodies,” Irani said. “Assessing these factors could help identify groups of patients who would respond better to treatment.” Indeed, not all studies have shown favorable results with IVIG. For instance, one study showed a poor response rate in patients who tested positive for antiglutamic acid decarboxylase 65 antibodies (GAD-65); only about 32% of these patients responded to treatment with IVIG.

Higdon clarified that the kind of antibody present could influence response to immunotherapy. “While antibodies to neural surface antigens (such as LGI1 and CASPR2) may respond well to immunotherapy, antibodies to intracellular antigens (like GAD-65) are less responsive,” she said. “This is because these antibodies may not be the cause of seizures, but more an epiphenomenon—a secondary symptom that occurs simultaneously with a condition but is not directly related to it.”

In a retrospective study of 50 patients, researchers showed that those who had antibodies to a neural surface antigen, the voltage gated potassium channel complex antigen, were more likely to become seizure-free compared with patients who had GAD-65 antibodies (

 Notably, those with antibodies to intracellular antigens may benefit from the addition of cytotoxic chemotherapy.

To predict which patients are more likely to respond to immunotherapy, Dubey and his team have also introduced and validated the RITE (Response to Immunotherapy in Epilepsy) score.

 This takes into account all the factors of the APE score, whether immunotherapy was started within 6 months of seizure onset, and, in addition, whether or not a neural antibody has been detected in plasma or CSF. Patients with a RITE score greater than 7 are more likely to respond to immunotherapy.

While IVIG may potentially be more convenient and safer compared with steroids and plasmapheresis, well-designed comparative studies are currently lacking. “Based on my current clinical experience, it is likely that steroids and plasma exchange will prove to be more effective,” Irani said. This opinion is partially shared by Higdon. Often, she said, “the response is transient and there needs to be repeated infusions or more aggressive immunotherapy with cyclophosphamide or rituximab.”

Many patients with autoimmune epilepsy can be successfully managed if a proper algorithm is followed, such as the one suggested by Higdon.

 “The first step, when you deal with a patient who has epilepsy of unknown origin, is to apply the APE score. If it is greater than 4, neural antibody testing should be carried out,” Higdon said. “Immunotherapy should ideally be initiated even before the antibody test results are received, either with steroids or IVIG.” She suggested that plasmapheresis be reserved for patients who fail to respond to previous treatments, or who have severe symptoms. If the antibody tests reveal the presence of an intracellular antigen, tumor screening must additionally be considered.

Of course, this fundamental algorithm does not cover all bases. “The ideal protocol for the duration of treatment and evaluation of response still needs to be developed,” Higdon said. It is also unclear if antiepileptic drugs can be tried again, if immunotherapy fails. In a retrospective study, a small proportion of patients did respond to antiepileptic drugs after immunotherapy had failed.

 The researchers stated that certain antiepileptics, like carbamazepine, also had immunomodulatory properties, and this dual effect could have been responsible for a seizure-free response in a small subset of patients.

Most experts feel that a better understanding of the pathophysiology of autoimmune epilepsy is the way forward. “We are currently focusing on how the B cells produce antibodies, and how these antibodies interfere with target antigens,” said Irani, whose most recent research has focused on exploring the characteristics and pathogenic potential of the LGI1 antibody. Understanding these key mechanisms could pave the way to establishing innovative methods of blocking the antibody effects.

1. Dubey D, Alqallaf A, Hays R, et al. Neurological autoantibody prevalence in epilepsy of unknown etiology. JAMA Neurol. 2017;74(4):397-402. doi:10.1001/jamaneurol.2016.5429

2. Scheffer IE, Berkovic S, Capovilla G, et al. ILAE classification of the epilepsies: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):512-521. doi:10.1111/epi.13709

3. Fredriksen JR, Carr CM, Koeller KK, et al. MRI findings in glutamic acid decarboxylase associated autoim- mune epilepsy. Neuroradiology. 2018;60(3):239-245. doi:10.1007/s00234-018-1976-6

4. Dubey D, Singh J, Britton JW, et al. Predictive models in the diagnosis and treatment of autoimmune epilepsy. Epilepsia. 2017;58(7):1181-1189. doi:10.1111/epi.13797

5. Cabezudo-Garci&#769;a P, Mena-Va&#769;zquez N, Villagra&#769;n-Garci&#769;a M, Serrano-Castro PJ. Efficacy of antiepi- leptic drugs in autoimmune epilepsy: a systematic review. Seizure. 2018;59:72-76. doi:10.1016/j. seizure.2018.05.004

6. Thompson J, Bi M, Murchison AG, et al; Faciobrachial Dystonic Seizures Study Group. The importance of early immunotherapy in patients with faciobrachial dystonic seizures. Brain. 2018;141(2):348-356. doi:10.1093/brain/awx323

7. Dubey D, Britton J, McKeon A, et al. Randomized placebo&#8208;controlled trial of intravenous immunoglob- ulin in autoimmune LGI1/CASPR2 epilepsy. Ann Neurol. 2020;87(2):313-323. doi:10.1002/ana.25655

8. Li TR, Zhang YD, Wang Q, et al. Intravenous methylprednisolone or immunoglobulin for anti-glutamic acid decarboxylase 65 antibody autoimmune encephalitis: which is better? BMC Neurosci. 2020;21(1):13. doi:10.1186/s12868-020-00561-9

9. Feyissa AM, Lo&#769;pez Chiriboga AS, Britton JW. Antiepileptic drug therapy in patients with autoimmune epilepsy. Neurol Neuroimmunol Neuroinflamm. 2017;4(4):e353. doi:10.1212/NXI.0000000000000353

10. Higdon LM. Autoimmune epilepsy. Practical Neurology. October 2018. https://practicalneurology. com/articles/2018-oct/autoimmune-epilepsy 

Could expanding insights into the pathophysiology of autoimmune epilepsy offer seizure-free hope for patients?

Nicola Davies, PhD

Articles

How Big Data Can Drive Advancements in MS Research

Data Evolution in Alzheimer Disease: Is the Future Digital?

Stimulating the Future of Treating Sleep-Disordered Breathing

Unraveling Autoimmune Epilepsy