Research spanning several decades and published in more than 750 peer-reviewed papers has established that oculometric markers can be used to diagnose and predict the progression of a wide range of neurological disorders.
There are two major challenges with current neurological assessment:
Below, we consider how oculometrics may be used to address these challenges.
Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI)
Eye movement and pupil response abnormalities have been a topic of research in Alzheimer’s Disease (AD) patients for over 50 years, and have gained even more interest in recent decades. Mild Cognitive Impairment (MCI) with amnesia is considered the transitional state to AD dementia (ADD) and other types of dementia, and a growing body of research has been investigating the observed oculometric abnormalities, and the extent to which such deficits can be used to distinguish between different types of conditions, as well as the correlations between oculometrics and clinical scales.
Commonly studied oculomotor functions involve deficiencies in gaze fixation, smooth pursuit, reflexive and voluntary saccades and the inhibition of reflexive saccades (e.g., the anti-saccade task), as well as abnormalities of pupillary light reflex.
Studies have shown correlations of oculometric measurements with AD severity scales such as the Mini-Mental State Exam (MMSE)–repeatedly shown to correlate with saccadic latency–and the Activities of Daily Living (ADL) score–which is correlated with a preference toward horizontal microsaccades. Furthermore, saccadic measurements can distinguish between different dementias, such as Frontotemporal Dementia (FTD), Semantic Dementia, Progressive non-fluent Aphasia, Alzheimer’s Dementia, Corticobasal Syndrome, and Progressive Supranuclear Palsy, each having a unique oculometric “signature”[1].
The underlying premise of our work is that combining all relevant oculometric parameters will produce accurate, objective, and sensitive AD/MCI measures, which are superior to any outcome measure used today.
Oculometric measures have been shown to be sensitive enough to distinguish between patients with MCI, AD and other dementias (and between the phenotypes of FTD)[2], and sensitive enough to show statistically significant change in less than 12 months (both in MCI, mild AD and moderate AD patients)[3]. On top of that, oculometric proxies can separate preclinical AD patients into meaningful phenotypes[4].
We’re looking to partner with pharma on trials involving AD and MCI outcome measures such as:
Parkinson’s Disease (PD)
The effects of Parkinson’s Disease on eye movements, pupil responses, and blink rate have been the subject of extensive study in the last 30 years, and gained even more interest in the recent decade. The three main research directions are the effect of PD on oculometric markers (and thus their value for diagnosis, and for distinguishing PD from other Parkinsonisms, especially in early stages), the relationship between oculometric markers and both disease progression and symptoms (e.g. cognitive decline and freezing of gait), and explaining the effects of PD on oculometric markers using the pathophysiology of Parkinson’s Disease.
Clinical diagnosis of PD and distinguishing it from other Parkinsonisms is quite unreliable, and about 30% of patients with PD are misdiagnosed in early stages, which is true even for patients in clinical trials[5]. Part of this challenge is distinguishing between diseases that exhibit similar symptoms, e.g. Parkinson’s Disease, PSP, and MSA.
On top of that, as with other neurodegenerative diseases, there are three main challenges with PD evaluation: (1) no objective, accurate and sensitive measures for disease progression: the existing gold standard, MDS-UPDRS, suffers from high inter- and intra-rater variability[6], is non-linear, and is not very sensitive (e.g. it takes about 12 months to determine statistically significant progression, and much longer for early PD); (2) no granular phenotyping: there is currently no way to define and divide subtypes in PD; and (3) evaluation is episodic: MDS-UPDRS depends on the presentation of symptoms during the time of visit, however PD symptoms present high variability between days and even during the same day.
Eye-related symptoms and abnormalities are a significant part of the neurological examination for all neurological disorders, including PD. Several researchers have described the value of oculometrics for diagnosis of PD (distinguishing between PD patients and other Parkinsonisms, as well as healthy controls), and correlations between oculometrics and UPDRS or measures of specific symptoms. Note that most published studies focus on a single oculometric parameter, as they are mostly interested in the value of straightforward symptoms as biomarkers. Robust correlations between dozens of oculometric parameters and the main PD outcome measure are described in the literature. However, while demonstrating statistically significant correlations, their high variance precludes using a single oculometric measure as a predictor of clinical outcome.
As mentioned above, the underlying premise of our work is that combining all relevant oculometric parameters will produce an accurate diagnostic tool for PD, and also objective and sensitive PD measures, which are far superior to the methods used today. Several papers[7] have already shown the promise of this approach for diagnosing PD: one published paper shows that the simultaneous use of 5 oculometric parameters in a logistic regression model gives sensitivity of 80% and specificity of 95% for diagnosing PD, whereas another shows that a regression model with 6 (other) oculometric parameters achieves an AUC of 92% in diagnosing PD. Another study[8] shows that a decision tree using 3 oculometric parameters can predict UPDRS with an AUC of 91%.
Oculometric markers have been shown to distinguish between PD, Multiple System Atrophy (MSA) and Progressive Supranuclear Palsy (PSP)[9]. This can be critical for the success of a PD study. On top of that, oculometric markers can distinguish PD patients into meaningful phenotypes[10]. The markers were shown to be highly sensitive and detect statistically significant changes within as little as 3-6 months, even for early-stage patients[11].
We’re looking to partner with pharma on trials involving PD outcome measures such as :
Amyotrophic Lateral Sclerosis (ALS)
Historically, the oculomotor system was thought to be spared in ALS, so most research concerning the effect of ALS on oculometric markers dates to the last decade. A considerable amount of recent research shows that the effects of ALS include impaired saccadic and smooth pursuit, increase in error rate and latency in the anti-saccade task, and gaze fixation instability, as well as square wave jerks and saccadic intrusions.
The underlying premise of our work is that combining all relevant oculometric parameters will produce accurate, objective, and sensitive ALS measures, which are far superior to any outcome measure used today.
Multiple studies[12] have shown that oculometric markers can distinguish between healthy controls and ALS patients, between bulbar-onset and spinal-onset patients, between ALS and FTD patients, and between phenotypes such as the involvement or lack of cognitive impairment in ALS patients. Furthermore, correlations have been shown between oculometric measurements and scales such as ALSFRS, ALSFRS bulbar component, the STROOP test, and disease duration.
We’re looking to partner with pharma on trials involving ALS measures such as:
Multiple Sclerosis (MS)
There are three main challenges with MS evaluation: (1) no objective, accurate and sensitive measures for disease progression: the gold standard, EDSS, suffers from high inter- and intra-rater variability, is non-linear, is not very sensitive (e.g., it takes at least 2 years to determine statistically significant progression and even longer in early MS). Furthermore, EDSS is biased towards walking and does not capture critical aspects of disability such as upper-body function; (2) no granular phenotyping: the only phenotyping is MS type–relapsing/primary progressive/secondary progressive–which is determined only retroactively; and (3) no reliable measures for disease burden: about 60% of MS patients suffer from significant cognitive decline, and 80% of MS patients suffer from MS fatigue. While over 40% of MS patients report these as the worst symptoms, which constitute a large part of disease burden, they are difficult to diagnose and measure.
Eye-related symptoms and abnormalities are a significant part of the neurological examination for all neurological disorders, including MS. In fact, in the case of MS, bedside oculometric examination is correlated with disability, and predicts disability progression 2 years after the examination, see. A growing body of research describes correlations between oculometrics and EDSS as well as other outcome measures. Most papers focus on a single oculometric parameter, as they are interested in the value of straightforward symptoms as biomarkers. Robust correlations between dozens of oculometric parameters and the main MS outcome measure are described in the literature. However, while demonstrating statistically significant correlations, their large variability precludes using a single oculometric measure as a predictor of clinical outcome.
The underlying premise of our work is that combining all relevant oculometric parameters will produce accurate, objective, and sensitive MS measures, which are superior to any outcome measure used today. An indication of the promise of this approach comes from, where moving from 1 to 2 to 3 oculometric parameters dramatically increases the predictive value for EDSS (to around 80% AUC), as well as cognitive and fatigue measures, even though all 3 parameters are parameters derived from the same phenomenon (microsaccades). Adding Timed 25-Foot Walk (T25FW) increases AUC to 90%, and while T25FW is not an oculometric marker, it is highly correlated with smooth pursuit measures.
Oculometric proxies have been shown to be sensitive enough to diagnose MS patients before any clinical manifestation (with MRI findings, before McDonald criteria are satisfied)[13]. In addition, our measures are sensitive even in early-stage patients (EDSS median < 1), and these proxies show statistically significant changes as early as 6 months into the study[14]. Similar sensitivity of these features was also demonstrated in SPMS patients[15].
We’re looking to partner with pharma on trials involving MS outcome measures such as:
We are looking to partner with pharmaceutical and biotech companies, academic research institutions, clinical centers, patient advocacy groups, and other organizations promoting solutions for treatment of neurodegenerative diseases.
