Our Research Projects & Strategy
Our Research Strategy
We aim to provide financial and logistical support for analyses of our data and use of our banked biosamples to create the fastest and most accessible routes to effective disease management. We are also interested in proposals dedicated to expanding existing diagnostic algorithms and measurement scales. Overall, our priority is to ensure efficient and effective research through collaboration.
Preferred projects for BIC support:
- Analyses of existing or anticipated Unhide data, with or without suggested platform enhancements
- Proof of concept clinical trials of therapies with suggested efficacy and tolerability
- Biological studies to run in conjunction with platform to support biomarker identification and/or repurposing of existing therapies
- Investigations of approaches to reduce burden of disease for patients and families
Our Current and Upcoming Projects
The Cunningham Lab has identified anti-neuronal antibodies that can be used to help diagnose autoimmune-mediated neuropsychiatric disorders associated with infectious triggers, such as sydenham chorea (SC) and PANDAS. Studies of other immune regulation disorders have indicated that saliva antibodies and soluble factors correlate with blood sera and plasma levels. In this study, we aim to determine if anti-neuronal (AAbs) from saliva can identify children with basal ganglia encephalitis (BGE) and stratify symptom severity. We will also determine if antibody isotype or secretory IgA deficiencies may correlate with symptom exacerbation or an increased susceptibility to infections. The Cunningham Lab, Microbiology & Immunology (University of Oklahoma), for the past 20 years, has focused on pathogenesis of infection.
Natural History
In this study, we will use cognitive metrics produced by the Lumosity app for the domains of overall cognitive performance, speed, memory, attention, flexibility, problem-solving, math, and language, to explore differences in cognitive metrics, and differences in cognitive training curves: (1) between patients with neuroinflammatory disease and normal controls; (2) across different neuroinflammatory disorders; (3) as a function of each of age and duration of illness; (4) as a function of treatments applied; and (5) as a function of inflammatory events.
Although it is now recognized that substantial overlap exists between POTS and post-infectious neuroinflammatory syndromes including PANDAS/PANS, ME/CFS, and Long Covid, the prevalence of dysautonomia in the PANDAS/PANS and Long Covid populations has not been well established. Part of the failure to recognize dysautonomia in these populations is that symptoms may not even be recognized by patients or caregivers, and likely often go unreported. In this study, data will be assembled from the wearable devices and symptom reports of people with known diagnoses of these conditions to establish whether and how these populations truly differ with respect to dysautonomic symptoms.
There is much overlap in symptomatology across disorders like PANDAS, PANS, POTS, Long Covid, ME/CFS, OCD, Anorexia Nervosa, Tourette’s, and Narcolepsy, that diagnosis applied to any given individual may well be contingent on knowledge/orientation of the healthcare practitioner, a reflection of the age/developmental/disease stage of the patient at the time of presentation, and/or the symptoms most emphasized by the patient at presentation. In this investigation, we will apply both traditional statistical approaches and AI to the entire database of ImPatient EmPowered Project patients with the above-listed diagnoses, to determine if there are indeed distinct symptom clusters that represent each of these diagnoses, and whether those are the same symptom clusters reflected in existing diagnostic criteria.
For any given patient, tracking and identifying the specific variables associated with symptom worsening and improvement over time can be difficult. Through the use of AI, it may be possible to identify associations between variables in one patients’ experience over time. Through applying Drumroll AI computer-assisted analysis, we will present case studies of three individuals who tracked symptoms, lifestyle variables, interventions, and medical and life events, in a systematic manner over XXX time.
Diagnoses & Descriptives
Biomarker Data
Abstract
The Cunningham lab has identified anti-neuronal antibodies that can be used to help diagnose autoimmune-mediated neuropsychiatric disorders associated with infectious triggers, such as sydenham chorea (SC) and PANDAS. Studies of other immune regulation disorders have indicated that saliva antibodies and soluble factors correlate with blood sera and plasma levels. In this study, we aim to determine if anti-neuronal (AAbs) from saliva can identify children with basal ganglia encephalitis (BGE) and stratify symptom severity. We will also determine if antibody isotype or secretory IgA deficiencies may correlate with symptom exacerbation or an increased susceptibility to infections. The Cunningham Lab, Microbiology & Immunology (University of Oklahoma), for the past 20 years, has focused on pathogenesis of infection.
Abstract
Although effective treatments for neuroinflammatory disorders exist, their application has been limited in part by failures to correctly diagnose. Neuroinflammatory disorders are phenotypically complex and can mimic those of related neuropsychiatric disorders, and readily accessible biomarkers have not been widely available for either this class of disorders in general, or specific diagnoses within the class. Ideal biomarkers would reflect pathogenic mechanisms and thus serve not only diagnostic purposes, but also to gauge the impact of treatments, and exacerbating elements, on those mechanisms. The Cunningham laboratory has identified a series of specific antineuronal antibodies – “The Cunningham Panel” – that are associated with SC and PANDAS, two clear-cut post-infectious neuroinflammatory disorders, and for which elevations are associated with symptom exacerbations. These autoantibodies have also been observed in attention deficit hyperactivity disorder (ADHD), Tourette’s syndrome, OCD, anxiety, autism spectrum disorder, and Long Covid, and may reasonably be expected to be present in neuroinflammatory syndromes in general. Although these AAbs may in fact be markers of this class of conditions, potentially distinguishing “mental health” symptoms of neuroinflammatory origin from those of psychogenic origin, this hypothesis has not been explored, and specific signatures associated with particular conditions or symptom clusters have not been identified. In this exploratory study, we will use clinical data collected from the ImPatient EmPowered Platform, together with Cunningham Panel data collected through Moleculera testing and donated to the platform by patients, at-risk siblings, and healthy controls, to investigate associations between neuroinflammatory symptomatology and Cunningham panel signatures.
Diagnoses & Descriptives
Clinical Markers
Abstract
Although diagnostic criteria have been established for each of PANDAS, PANS, POTS, Long Covid, ME/CFS, OCD, Anorexia Nervosa, Tourette’s, and Narcolepsy, there is so much comorbidity across these diagnoses, and so much overlap in symptomatology, that the diagnosis applied to any given individual may well be simply a matter of the knowledge or orientation of the healthcare practitioner, a reflection of the age or developmental or disease stage of the patient at the time of presentation, and/or the symptoms most emphasized by the patient at presentation. In the ImPatient EmPowered Project, comprehensive and validated symptom inventories are performed over time, allowing a more full view of the patient to develop than is normally captured by a diagnosing healthcare provider. In this investigation, we will apply both traditional statistical approaches and AI to the entire database of patients with the above-listed diagnoses, to determine if there are indeed distinct symptom clusters that represent each of these diagnoses, and whether those are the same symptom clusters reflected in existing diagnostic criteria. The importance of this work is that it will create an objective picture of key signs and symptoms that may be used for differential diagnosis, and that may therefore guide patients most efficiently to effective therapeutic and supportive strategies.
Abstract
The diagnosis of POTS is defined by the failure to properly regulate orthostatic homeostasis, among other issues impacting sleep, activity tolerance, and digestion. Although it is now recognized that substantial overlap exists between POTS and post-infectious neuroinflammatory syndromes including PANDAS/PANS, ME/CFS, and Long Covid, the prevalence of dysautonomia in the PANDAS/PANS and Long Covid populations has not been well established. Part of the failure to recognize dysautonomia in these populations is that symptoms may not even be recognized by patients or caregivers, and likely often go unreported. A result of this failure is unrecognized and hence untreated symptoms that are often not only distressing but disabling, even though if recognized, they could be improved by simple and accessible interventions such as consumption of electrolyte drinks and a high-salt diet. In this study, data will be assembled from the wearable devices and symptom reports of people with known diagnoses of PANDAS/PANS, Long Covid, and POTS, to establish whether and how these populations truly differ with respect to dysautonomic symptoms. Symptoms and signs of dysautonomia will also be compared between these populations and each of healthy controls and psychiatric controls.
Diagnoses & Descriptives
Therapies + Self-care
Abstract
Virtually without exception, patients who experience neuroinflammation report subjective alterations in multiple aspects of cognitive function including memory, focus, processing speed, mathematical ability and/or verbal fluency. Clinicians and others who interface with such patients confirm these perceptions. In this study, we will use cognitive metrics produced by a popular cognition app for the domains of overall cognitive performance, speed, memory, attention, flexibility, problem-solving, math, and language, to explore differences in cognitive metrics, and differences in cognitive training curves: (1) between patients with neuroinflammatory disease and normal controls (including from a popular cognition app normative dataset); (2) across different neuroinflammatory disorders; (3) as a function of each of age and duration of illness; (4) as a function of treatments applied; and (5) as a function of inflammatory events. Correlations between symptom profiles and cognitive scores, both cross-sectionally and over time, will be examined. All analyses will be exploratory. Since metrics derived from a popular cognition app are impacted by experience and training time in the app and not only the variables described, these variables will be accounted for in all analyses using a popular cognition app’s data.
Abstract
In clinical research it is currently not uncommon to apply AI tools to many-patient databases to reveal complex associations across variables associated with disease risk, diagnosis, treatment responsiveness, and/or prognosis. Less commonly practiced but also possible is the use of AI to identify associations between variables in one patient’s experience over time. Patients with autoimmune disease can experience highly variable symptoms over the period of their illness, which can be worsened or improved with life events such as infections, allergen exposure, stress, and/or changes in lifestyle. For any given patient, however, the specific variables associated with symptom worsening and improvement can be difficult to track and identify. Here we present case studies of three individuals who tracked symptoms, lifestyle variables, interventions, and medical and life events, in a systematic manner over XXX time, and for whom these longitudinal data were then subject to computer-assisted analysis using Drumroll AI. Potential benefits and risks of using this approach to help patients and physicians better understand influencers of symptom flares at the level of the individual patient are then discussed.
Prevention for at-risk individuals
Genetic risk factors
Abstract
Genetic studies, in collaboration with Galatea Bio, in Hialeah, FL, will evaluate the association between genotype and clinical phenotype, including responses to interventions, and may also be used to develop risk models, prognostics, and diagnostics, including ancestry-adjusted risk scores. Analyses of response to treatment and other phenotypic descriptions will utilize the endpoint parameters described in the protocol, derived from the PANS Rating Scale. The results of this research will identify genetic markers associated with a propensity to develop PANDAS, for use both in diagnosis and in prospectively identifying at-risk individuals for targeting of preventive measures. The results of this research will also begin to identify whether individuals with particular genetic markers are more or less likely to experience particular types of symptoms, aspects of disease course (e.g. earlier onset), or comorbidities, or to respond to particular types of interventions. Sample collection will be performed via a simple cheek swab kit sent directly to the patient’s home with instructions for self-collection and shipping. In lieu of sample collection, analyses can also be performed using existing data that participants may have obtained from 23andMe, Ancestry, or other genetic-testing programs.
