Avia Logo
Avia Logo
Avia Logo

Products /

neuroQWERTY

neuroQWERTY

neuroQWERTY

By nQ Medical, Inc.

0 review

Contact

Overview


Select which hospital or health system you work at and see a personalized match score.
About

DESCRIPTION
Computational Biomarker for Neurodegenerative Diseases
Read more
CATEGORIES

None provided

KEYWORDS
neurology
CLIENTS OVERVIEW
Key Clients
Clients Your Size
USED BY
0 People
3 Organizations
EHR integration

Integrations:

None provided

EMR Integration & Relevant Hardware:

None provided

EMRs Supported:

None provided

Hardware Compatibility:

None provided

Description

nQ is a computational biotechnology company with experience in digital phenotyping through AI-aided analysis of personal device interactions. We have developed digital biomarkers for Parkinson's disease in early and newly diagnosed/untreated stages of disease. In addition to PD, we are currently conducting a number of separate clinical trials to develop digital biomarkers individually relevant to Multiple Sclerosis, ALS, Alzheimer's disease, and mTBI with industry partners and academic centers including Cleveland Clinic and Massachusetts General Hospital. OUr AD trial has advanced to yield early and promising results measuring cognitive decline and delineating PD and AD symptomatology. In prodromal stages of neurodegenerative diseases such as PD, digital biomarkers offer possibility of detecting actual subclinical symptoms known to predict phenoconversion better than biochemical markers which often predict disease risk but not timing. Some key advantages of our digital biomarkers include: -Continuous, longitudinal, remote, real world quantification of psychomotor symptoms; -Patients are not burdened with completing structured tasks or clinic visits for data collection. Patients use their personal devices as they normally would and data is collected passively and transparently, often 24/7; -Patient privacy and security is ensured at each step of algorithm. Biomarkers perform even though content of what patients type on their personal devices kept confidential and secure. Specific to Parkinson's Disease: -Five peer reviewed publications reflecting 4 clinical trials demonstrating technical validation and clinical validation in PD patients; -Performance in correlating to UPDRS-III and differentiating healthy controls from from Parkinson's disease with AUC>0.90 in early PD patients thus far published. Follow-up work from our group and academic research by others have demonstrated continued increase in accuracy with increased data and refinement; -Performance in longitudinal symptom monitoring, distinguishing medication responders from non-responders. Use cases in multiple phases of clinical trial and in the clinic post-approval: Trial: -Cohort enrichment - pre-screening of patients to undergo more expensive or detailed diagnostic testing; -Symptom Fluctuation/Disease Progression Monitoring - characterize subpopulations of patient cohorts: responders, early progressors etc; -Outcome measure - Smartphone interaction is inherently relevant to daily function. Precise quantification can detect treatment responses missed by cruder metrics. Post-approval: -Disease screening for diagnostic referral; -Symptom monitoring as an independent or companion software to aid treatment titration, monitor compliance; -Generate real world data and evidence (RWE); -Supporting telehealth visits, access to patients in under-served/remote areas, via telemetry.
Read more
Use cases

Description
  • nQ-Medical’s Immediate Impact on Efficiencies in Clinical Development of Neurologic Therapeutics

    Exorbitant Costs and Inefficiencies of Drug Development
    The pharmaceutical industry, on average, incurs a cost to bring each new compound through to approval at an estimated cost of $2.8B. For Alzheimer’s Disease (AD), as an example, drug
    development costs substantially exceed other therapeutic areas with total costs of a traditional AD drug development (including cost of capital, cost of failures) estimated at $5.6B often taking over 13 years from preclinical studies to approval by the FDA.1

    In comparison, estimated cost of a cancer treatment development is $794M (at a 9% cost of
    capital). Clinical trials represent the most expensive phase of the drug development process.1
    Drug developers face significant clinical trial challenges and cost burdens around patient
    recruitment, retention, and adherence. These increasing complexities and costs of on-site
    monitoring are further inflamed by increasing payor and regulatory pressure for proof of value.
    Forbes recently summarized that about 80% of pharmaceutical trials do not meet enrollment
    deadlines, resulting in an average loss up to $1.3M per day for a given drug candidate.2
    Additionally, about 37% of research sites fail to meet their enrollment targets, and 10% fail to
    even recruit a single patient for the study. Based on these industry estimates, a lack of patientcentric trial designs leads to 35% of patients dropping out of clinical trials. Another 35% do not adhere to study protocols, costing about $1M per trial in lost productivity alone.2
    Drug developers within Neurology space face even greater challenges as probability of success for CNS therapeutic Phase 3 Clinical trials is much lower than other disease areas at 33% (Cardiovascular 74%, Anti-Cancer 62%).6 Within Neurology, AD Trials have a specifically low rate of success evidenced by exorbitant development costs when failure rates are included (Table 1).1

    nQ Immediate Efficiency Opportunities - Patient Segmentation
    Patient segmentation is critical to the success of a drug development and clinical trial program. Taking AD as an example, multiple failures in clinical trials over the past decade are likely due to incorrect patient selection, eg, screening and testing on advanced dementia instead of Phase III trials are the costliest part of AD drug development. Table 1 from Cummings et al shows the average cost and duration of each phase of AD
    drug development. These figures include the cost of capital and the cost of failures that companies sustain (3rd column) working in the AD drug development arena. Even out-of-pocket costs for development of a single AD agent approach $500M (4th column). 1
    early/prodromal disease. It is quite possible that efficacious compounds have likely been
    shelved after being tested in inappropriate patient segments.

    The FDA’s 2019 Guidance to Industry on Enrichment Strategies for Clinical Trials encourages
    drug developers to identify patient segmentation strategies to 1) decrease variability; 2)
    increase prognostic enrichment by selection of patients at high risk of disease-related
    endpoints; 3) increase predictive enrichment by selection of patient sub-segments with disease stages more likely to respond to drug treatment. In the words of the FDA, these type of strategies can be expected to power an “increased number of events in a shorter time period, generally allowing for a smaller sample size…even an imperfectly characterized predictive marker can greatly increase the power and likelihood of study success.” 3

    nQ plays an important role in each of these patient segmentation and enrichment strategies:
    • Decreasing variability:
    - As the FDA suggests, “choosing patients with baseline measurements of a disease
    or a biomarker characterizing the disease in a narrow range” can decrease
    variability and increase study power.3 nQ’s ability to give granular quantitation of
    disease symptoms allows clinical trial designers to define a narrow range of
    scores for inclusion into the trial to decrease heterogeneity.
    - As the FDA further suggests “excluding patients whose disease or symptoms
    improve spontaneously or whose measurements are highly variable” can
    decreased intra-patient variability and increase study power.3

    nQ’s ability to provide repeated quantitative assessments continuously between clinic visits
    allows trial designs which include a pre-randomization baseline “run-in” period
    whereby patients whose symptoms resolve spontaneously or have highly
    variable baseline symptoms could be excluded. The decreased variability
    provided by these strategies would increase study power.
    - The FDA adds “identifying and selecting patients likely to adhere to treatment”
    would decrease variability.3

    nQ’s ability to detect known drug effect (eg, Parkinson’s Disease [PD] patients taking levodopa) allows for monitoring of drug compliance and, in the correct trial context, could allow for monitoring of drug compliance and be used to identify patients likely to adhere to treatment.

    Illustrative examples: The concept of increased clinical trial power can encompass both
    decreased enrollment numbers and shorter duration. Using nQ to decrease variability
    and increase clinical trial power would allow for designs with shorter duration. Again
    taking AD as an example, VitalTransformation predicts that reductions only in the
    patient identification times during recruitment phase by as little as 25%# for a net total
    decrease of duration by only 4.8 months across Phase 1-3 trials of an AD Drug
    development program would yield clinical R&D savings of $70M in cost of capital savings
    alone (at 11% cost of capital).6

    Increased study power can also allow for smaller trials with decreased number of
    participants. Not specific to AD, but CNS disorders in general, the cost per patient across
    for phases 1+2+3 is $34,000 + $39,500 + $40,500 respectively = $114,000 total/patient.7
    For a non-AD CNS clinical trial with 200 patients per arm = 400 total patients, reduction
    in number of patients by as little as 10%# could yield savings > $4.5M in a single
    development program.

    • Increasing prognostic enrichment (identifying high risk patients for endpoints):
    - Taking AD as an example, accuracy of diagnosis of AD by clinical assessment
    alone is only around 70-80% meaning 20-30% of patients in a whole generation
    of prior clinical trials likely did not even have the disease being attempted to
    treat.4 Some newer AD clinical trials (such as Biogen’s Aducanumab trial) have
    tried to confirm AD diagnosis by selecting patients with biomarker evidence of
    abnormal amyloid in the brain. Unfortunately, those biomarkers for amyloid
    currently consist of invasive lumbar punctures which many patients refuse or
    expensive PET scans which are difficult to schedule at a limited number of
    facilities and cost about $4,000 per scan.

    Illustrative examples: if the goal is to recruit 1,000 cognitively normal individuals who
    are PET or CSF amyloid positive into a Phase 3 clinical trial, and given that around 30% of
    cognitively normal individuals above age 65 are expected to be amyloid positive,
    without enrichment at least 3,334 individuals need to be screened. Using nQ metrics,
    the screening of patients for subtle symptoms of AD could be accomplished to enrich
    screening for amyloid using PET scans above the background rate of 30%. Even if
    performance of nQ metric in AD is imperfect (data pending) and the positive predictive
    value (PPV) of NQ screening was 60%#, the number of individuals required to be
    screened on PET/CSF amyloid to reach 1,000 participants would be cut in half. At the
    estimated cost of $4,000 per PET scan, the amount of money saved from a reduction of
    initial PET prescreening scans alone for 1,667# individuals is > $6.5M.

    The cost saving above reflects only the pure billing cost of PET testing. Not reflected are
    further savings due to faster recruitment time from having to schedule fewer scans,
    fewer patients recruited/screened, fewer clinic visits/clinic overhead, and more rapid
    study closure. The per-patient cost of recruitment/screening into an AD trial can exceed
    $100,000/patient.8 If validated, reducing the number of patients screened from 3,334 to
    1,667# patients yields savings of >$166M. 80% of clinical trials face delays of >1month and delays can cost up $1.3M per day;2 If the time savings from nQ deployment lead to reduction in delays by even 15# days this could yield savings of $19M.

    • Increasing predictive enrichment (Identifying more responsive patients for treatment):
    o As the FDA suggests “An initial screening for response — a biomarker
    measurement, eg, early clinical response, or full-fledged clinical response — in an
    open-label pre-randomization period can be used to identify a responder population that would then be randomized in the controlled study …identifying a responder population, eg, a subset of the overall population with a larger than average response to treatment and studying this population in a clinical trial can provide two major advantages: 1) increased study efficiency or feasibility; and 2) an enhanced benefit-risk relationship for patients in the subset compared to the overall population.” 3

    nQ’s ability to quantify symptoms and to track them longitudinally can be used
    to identify responders versus non-responders. The ability to execute this type of
    tracking in PD patients has clearly been demonstrated in our publication
    Matarazzo et al.5
    - Identifying responder and non-responder populations can provide other critical
    advantages to clinical trial design. Leveraging this type of data can produce
    unique efficiencies across multiple clinical trials. For example, the FDA suggests
    using patients who failed or were non-responders to one drug as control subjects
    in a trial for a different agent which works by a different mechanism. They state,
    “A population of non-responders to a different drug can be randomized to the
    new drug or to the drug they did not respond to. The comparison is enriched with
    respect to the active control comparison because the population is expected to
    have a poor response to the original drug compared to the test drug.” 3

    Illustrative examples: Identification of responder and non-responder groups can lead to
    increased study power and expedited screening between trials allowing for briefer trial
    durations and fewer participant numbers; the advantages of this is already discussed.
    In addition, and perhaps more importantly, clinical trials with known responder groups
    would increase the probability of success which is an extremely sensitive driver of total
    drug development cost. To see this effect, we can return to AD as an example and
    consider Table 1; increases in probability of success affect figures in 3rd column (total
    cost Phase 1+2+3 = 4.02 billion) which adjust for development failures and cost of
    capital. We can use a simple toy model which includes cost of capital and probability of
    success to calculate total cost for each Phase.

    Using data from Table 1 and 11% cost of capital yields probability of success for Phase 1-
    3 of 13.2%, 10.5%, 25% respectively. The current total cost of Phase 1 + 2 + 3 per Table 1
    is $4.02 billion. Very modest theoretical improvements in probability of success values
    by relative increases of 5%# (ie x1.05) to 13.8%, 11.0%, 26.2% predicts total cost of
    Phase 1 + 2 + 3 = $3.83 billion for a net saving of >$190M.

    nQ Immediate Efficiency Opportunities - Clinical Trial Outcome
    In addition to patient segmentation, nQ metrics can provide an early outcome measure in
    clinical trials indicating compound efficacy for informing critical go/no-go decisions. As
    mentioned above, Phase 3 clinical trials, especially in neurodegenerative conditions, are the
    most expensive phase of drug development. Correct decisions about terminating a program
    early after Phase 2 studies or well-informed early futility analysis during Phase 3 trials are
    critical decisions involving deployment of multiple hundreds of millions of dollars. Failure to
    collect or correctly interpret data to inform these decisions can have potentially disastrous
    consequences as in Biogen’s recent Aducanumab experience. Inexpensive testing which even
    partially informs these critical decisions even in one or two cases easily justifies its cost. Within PD, nQ has demonstrated it can clearly identify patients responding to dopamine therapy versus patients who are not responding5 providing early, continuous measurement of
    compound efficacy.

    In addition to providing useful proof-of-concept data for guidance of internal Go/No Go
    decisions, continued data collection and clinical experience with nQ metrics could lead to a
    superior FDA approval endpoint for neurodegenerative diseases. For an approval endpoint/
    outcome, FDA usually requires a new drug to show improvement in an established clinical
    endpoint with decades of experience and direct relevance to mortality, function, or other
    clinical meaningfulness. Typing represents an inherently meaningful task with direct relevance
    to patient function. Furthermore, unlike traditional endpoints which can be difficult to assess or accurately quantify due to subjective clinical assessment (such as UPDRS scale for PD), nQ
    metrics derived from typing data can be more easily measured and better quantified accurately to allow for increased power in detecting drug effects. Continued data collection with larger numbers of patients, demonstrating correlation to traditional outcomes of function will develop nQ metrics into powerful new approval endpoints for FDA submission in neurologic disease.

    Illustrative examples: In Biogen’s Aducanumab trial, CSF phospho-Tau (p-Tau) from patient
    lumbar punctures was reported as an additional endpoint. Although the FDA will not approve a
    medication based on improvement of a surrogate biomarker such as CSF p-Tau, improvement
    of p-Tau levels in patients after treatment with anti-amyloid antibody such as aducanumab
    lends powerful support to an argument for disease-modifying efficacy of the drug and can drive internal decision-making as well as support arguments made to FDA. FDA’s 2018 Guidance to Industry on Alzheimer’s Disease Drug Development promotes demonstrating improvement in multiple tests to make arguments for efficacy, “FDA will consider strongly justified arguments that a persuasive effect on sensitive measures of neuropsychological performance may provide adequate support for a marketing approval…..beneficial effects demonstrated across multiple individual tests would increase the persuasiveness of the finding; conversely, a finding on a single test unsupported by consistent findings on other tests would be less persuasive.”9

    Similar to p-Tau, nQ can play a role as additional endpoint to drive internal decision-making and lend support to FDA submissions. The out-of-pocket cost of a Phase 3 AD trial can be $287M per Table 1. While no single test will alone drive the decision to discontinue a development program from Phase 2 to Phase 3, a test such as nQ which drives even 15%# of the confidence in correctly making that decision provides value of >$43M.
    # For illustration purposes only If with enough experience, nQ could be developed into a surrogate outcome for direct FDA approval based on improvements in nQ scores, this would allow for increased power of clinical trials with smaller numbers and shorter durations, the advantages of which have already been discussed.

    Mediating TBI Trial Data Collection Barriers
    With continued development of nQ TBI digital biomarkers, efficiencies similar to the above
    discussion (AD and PD) can be expected in a TBI drug development program through improved data collection, patient segmentation, and outcome measurement. The ability to develop improved patient segmentation strategies and outcome measurement strategies for TBI using nQ technology is enabled by unique advantages in data collection and analysis:

    • Accurate baseline: Players eager to return to play are well known to falsely impair their
    baseline performance on standard concussion assessments. This makes detection of
    new impairments after a concussion in game harder to detect and allows players to
    remain or return to play. By using passive data collection during natural device use, nQ
    technology can capture an accurate baseline to enable accurate measurement of
    change.

    • Rapid assessment: Early assessment of TBI symptoms is important within 12 hours postinjury.
    An nQ score can be generated in as little as 15 seconds of smartphone typing.
    This could enable even in-game, sideline assessment of nQ score. While this could be
    useful in a clinical trial setting, if validated, an accurate sideline assessment of
    concussion symptoms would represent a valuable and marketable test independent of
    any drug development program. Alternatively, assessment of concussion symptoms
    could begin immediately post-game as the player resumes using his/her device.

    • Continuous measurement: The younger age group population at higher risk for TBI
    overlaps significantly with population of high smartphone usage. As already discussed,
    this provides opportunity for remote monitoring of symptoms and response to
    medication.

    • Rich data set, adherence: In addition to information derived from keystroke dynamics,
    nQ data also reflects smartphone usage patterns. This type of phone usage data is
    studied in correlation to mood (depression, anxiety, PTSD, etc.), circadian rhythms, and
    can also be used to monitor compliance with “cognitive rest” (abstinence from reading,
    TV, smartphone usage, etc.) often prescribed after concussion.

    Mediating TBI Patient Segmentation/Outcome Measurement Barriers
    A smart patient segmentation strategy is critical to success of a trial and when effectively done can increase power of trial design allowing for shorter clinical trial and smaller sample sizes. The ability to assess outcomes accurately and continuously informs critical go/no-go decisions involving hundreds of millions of dollars in the most expensive phases of a drug development program in the form of:
    • Pre-screening with nQ to enrich/increase yield of screening with more expensive tests.
    Similar to arguments provided above about pre-screening for amyloid PET scans, nQ can
    be used to increase the yield of more expensive testing such as MRI scans or blood
    based genetic and cellular markers.

    • By detecting subtle symptoms and quantifying them, nQ scores could be used to
    decrease variability in patient cohorts, confirm that patients included in a trial had
    concussions, and possibly quantify severity of TBI.

    • Longitudinal assessment of symptoms after concussion allows for identification of which
    patients are improving and can “return to play” versus patients with continued
    symptoms developing “post-concussion syndrome.” Assessment in even longer
    timeframes could be used to identify Parkinson-like symptoms and cognitive symptoms
    commonly found in CTE (chronic traumatic encephalopathy), an otherwise difficult to
    clinically diagnose condition which can develop after repetitive head trauma.

    • Longitudinal assessment of symptoms can identify responder versus non-responder
    groups which allow for efficient clinical trial design and efficiencies across clinical trials
    as non-responders from one trial may be effective controls in another trial.

    Future State Vision/Promise
    Ultimately, typing represents a complex reflection of integrated central and peripheral nervous
    system function encompassing behavioral, cognitive, language, sensory, psychomotor, and
    neuromuscular domains of function. Continuous, remote, unobtrusive/passive collection of
    this rich dataset and its appropriate analysis enables not only immediately visible efficiencies in clinical trial design but multiple other advantages in drug development including screening
    efficiency, generation of real-world-data and evidence, generation of superior endpoints for
    efficacy:
    • Widespread deployment of nQ to patient cohorts can allow for in-silico remote
    screening of patients for desired symptoms or probability of testing positive for other
    measures. This screening could be done prior to in-clinic visit allowing for efficient use of
    in-clinic time.

    • Social networking sites could be accessed for remote recruiting. Such remote electronic
    screening strategies can identify patient populations for clinical trials or drug treatment
    including patients with limited access to healthcare facilities such as rural populations.

    • Widespread deployment of nQ data collection would allow for generation of real-world
    evidence (RWE) in the context of potentially multiple neurologic diseases
    simultaneously. Analysis of this observational data in context of other patient data such
    as that obtained from the EHR can generate RWE for new indications of existing
    medications, access to new patient populations otherwise not included in clinical trials,
    and/or satisfaction of post-approval requirements.

    • Increasing experience with nQ data in multiple neurological diseases is inherently
    clinically meaningful and can lead to new clinical endpoints that can be used directly as
    surrogate endpoints for FDA approval in addition to acting as secondary/exploratory
    endpoints to help detect the early signals of compound efficacy that drive internal
    Go/No Go decision making.

    Conclusion
    Analysis of keystroke dynamics data can be viewed as a digital biopsy of complex central and
    peripheral nervous system function. Harnessing this data by efficient collection and analysis can allow for innovative enriched clinical trial designs with increased power (shorter durations and smaller size), efficiencies across different trials, and inform critical futility or Go/No Go
    decisions. Widespread deployment could further allow for remote screening, RWE, and novel
    approval endpoints. The total value provided by deployment of nQ will vary significantly
    depending upon details of its implementation, method of calculating valuation, and pending
    data about nQ performance within various disease areas but achievement in the millions of
    dollars can be reasonably anticipated.

    References:
    1 Cummings, J., Reiber, C. and Kumar, P. (2018) ‘The price of progress: Funding and financing
    Alzheimer’s disease drug development’, Alzheimer’s & Dementia: Translational Research &
    Clinical Interventions, 4, pp. 330–343.

    2 Das, Reenita. “Top Five Digital Health Technologies in 2019.” Forbes.
    https://www.forbes.com/sites/reenitadas/2019/02/04/the-top-five-digital-healthtechnologies-
    in-2019/ (November 9, 2019).

    3 U.S. Food and Drug Administration. 2019. “Enrichment Strategies for Clinical Trials to Support Approval of Human Drugs and Biological Products.” http://www.fda.gov/regulatoryinformation/search-fda-guidance-documents/enrichment-strategies-clinical-trials-supportapproval-human-drugs-and-biological-products (November 9, 2019).

    4 Archer, M. C., Hall, P. H. and Morgan, J. C. (2017) ‘Accuracy of Clinical Diagnosis of Alzheimer’s Disease in Alzheimer’s Disease Centers (ADCS)’, Alzheimer’s & Dementia. (2017 Abstract Supplement), 13(7, Supplement), pp. P800–P801.

    5 Matarazzo M, Arroyo-Gallego T, et al. Remote Monitoring of Treatment Response in
    Parkinson's Disease: The Habit of Typing on a Computer. Mov Disord. 2019 Jun 18.

    6 “Better Science, Better Health: Downloads.” Vital Transformation.
    https://vitaltransformation.com/better-science-better-health-downloads/
    https://vitaltransform.wpengine.com/wp-content/uploads/2014/10/DGS_17-10-Opt-in-Optout-
    Patient-Led-Databases-MAPPs-DG3.pdf (November 12, 2019).

    7 Corea, John. “Clinical Trials Impact State Economies.” https://catalyst.phrma.org/clinical-trialsimpact-state-economies (November 12, 2019).

    8 Kolata, Gina. 2018. “For Scientists Racing to Cure Alzheimer’s, the Math Is Getting Ugly.” The
    New York Times. https://www.nytimes.com/2018/07/23/health/alzheimers-treatmentstrials.
    html (November 12, 2019).

    9 U.S. Food and Drug Administration. 2019. “Alzheimer’s Disease: Developing Drugs for
    Treatment Guidance for Industry.” http://www.fda.gov/regulatory-information/search-fdaguidance-documents/alzheimers-disease-developing-drugs-treatment-guidance-industy
    (November 12, 2019).
Users:

Researchers ad clinicians in neurodegenerative diseases.

Differentiators

Differentiators vs Internal Options:

None provided

Differentiators vs Competitors:

None provided

Company information

Founded in 2016

4.5M total equity funding

Select which hospital or health system you work at and see a personalized match score.
ContactAVIA

About AVIA Connect

AVIA Connect is where innovative health systems and hospitals go to find the right digital health solutions. It’s built to efficiently search vendors and guide informed decisions, with tools like product comparisons, match scores, report generators, peer reviews, and market insights.

AVIA Connect is brought to you by AVIA, a trusted company and community of experts who guide health systems on their digital transformation journey.

Stay up to date

Keep up with the latest products and trends in digital health.