Oculomics and oculometrics: Using the eye as a biomarker for neurodegenerative disease

Image Credit: AdobeStock/Ilya

A biological marker or biomarker is an objective, quantifiable, defined characteristic that is measured in the body or its products as an indicator of a normal or pathogenic biological process and/or biological response to an exposure or intervention.¹ In eye care, we have an extensive number of biomarkers, including molecular (tear osmolarity), histologic (corneal microscopy), radiographic (fundus photography, optical coherence tomography [OCT]), corneal topography), and physiologic (visual evoked potential, electroretinogram) biomarkers, which allow us to measure the biologic processes of the eye and visual function.

Biomarkers are critical to our understanding of what is going on inside the body and are needed to objectively measure if an intervention, like a pharmaceutical, is working. For example, when we are trying to understand the process of glaucoma, we use a combination of biomarkers including IOP measurement, visual fields, OCT, and optic nerve photos to objectively quantify the ocular structures and measure if the disease is progressing or stable. These biomarkers are instrumental in guiding our treatment decisions in whether we want to start a patient on a medication, observe, or modify a current treatment plan. The more sensitive the biomarker, the better able we are to modify our treatments sooner, treat conditions faster, and potentially delay progression of or prevent disease to improve patient function and reduce morbidity and mortality. Over the past few decades, eye care has had the luxury of plentiful biomarker technologies added to our portfolio that allow us to have an extensive understanding of what is going on inside the eye.

Our neurology colleagues have not been as lucky. The ability to study the central nervous system is limited due to the inaccessibility of the brain’s neural tissue within the skull. With the recent emergence of pharmaceuticals to potentially delay the progression of Alzheimer disease (AD) and rapidly changing treatment options for other neurodegenerative diseases such as Parkinson disease (PD) and multiple sclerosis (MS), demand for better biomarkers for neurology is at an all-time high. Neurologic conditions are often misdiagnosed in the early to moderate stages of the disease as presenting symptoms such as depression, anxiety, and fatigue are common, and have multiple etiologies. Thus, patients with early onset neurodegenerative disease are often misdiagnosed or unaware, laying in wait in primary care clinics, completely unaware of their silently progressing neurodegenerative condition. Current diagnostic biomarkers for neurologic function include behavioral symptoms surveys, neuropsychological tests, blood and cerebrospinal fluid analysis, neuroimaging data from PET and MRI technology, and electrophysiology tests like transcranial magnetic stimulation and electroencephalogram. These neurologic diagnostics are expensive, time consuming, invasive and/or not readily available in all clinics across the country, causing barriers to access for patients seeking diagnosis. Furthermore, many of them rely upon human interpretation, which has the potential for bias and error.

While current patient perception is that the aforementioned diagnostics are capable of easily detecting and diagnosing neurodegenerative diseases, in reality, these biomarkers have low specificity or diagnostic accuracy for individual disease states, which often leads to a costly examination of multiple diagnostics and a delayed diagnosis as it may take multiple visits and clinicians to accurately pinpoint the diagnosis causing a patient’s neurologic symptoms. The limitations in the ability to diagnose neurologic disease ultimately leads to misdiagnosis, missed diagnosis, and delayed diagnosis, further delaying drug development as the effectiveness of many therapeutics depends on the timely and early detection of neurodegenerative disease. The current state of neurodiagnostics has led to exhaustive costs to the patient and to society. Additionally, structural neuro-imaging techniques like MRI have a limited ability to correlate the imaging data with clinical measures of disease severity for neurologic disorders like MS, which can be frustrating for the patient and provider alike when the symptoms and/or possible progression of the disease aren’t detected in an objective granular scale nor reflected by the biomarker.² Thus, the development of novel, low-cost, easily accessible, noninvasive neurologic biomarkers that can aid in improving accuracy in differential diagnosis and early detection of neurodegenerative disease is a promising area of research.

Oculomics: The eye as a biomarker

The imaging of the eye has unlimited potential to be a noninvasive biomarker for systemic and neurologic disease as the eye is the only place in the body where you can easily visualize vasculature and neuronal tissue. Since the retina and optic nerve share the same embryological origin as the central nervous system, the evaluation of ocular biomarkers can allow for noninvasive visualization of neural integrity and potentially serve as surrogates for overall neurologic function.

The retina contains multiple cell types, including dopaminergic neurons, astrocytes, nonmyelinated ganglion cell axons, and neurons, that through the use of OCT can be noninvasively visualized with high resolution, in vivo, allowing for the study of neurodegeneration, neuroinflammation, neuroprotection, and neurorepair.² New advances in retinal layer segmentation and longitudinal studies on retinal structure in neurodegenerative disease have shown progressive retina nerve fiber layer (RNFL) thinning with time in conditions like MS,³ PD,⁴ AD,⁵ and traumatic brain injury.⁶ 

Cerebrovascular pathologies such as hypertension, diabetes, and abnormal blood flow dynamics are a notable contributor to developing neurologic disorders such as stroke as well as neurodegenerative diseases like vascular cognitive impairment and dementia. Evaluation of retinal vasculature using diagnostics like OCT and OCTA (OCT angiography) can give insight as a surrogate for intracranial vascular pathology.⁷

Although the obvious choices, the retina and optic nerve are not the only potential neurodegenerative ocular biomarkers. Anterior ocular structures including the cornea, lens, aqueous and vitreous humor also have potential roles.⁸ Amyloid-ß, a misfolded plaque that is the pathologic hallmark in the brains of patients with AD, has been found in the retina⁹ and in the lens,¹⁰ as have other neural proteins like phosphorylated tau in the vitreous humor of patients with chronic traumatic encephalopathy and AD.¹¹ With the addition of artificial intelligence (AI) and deep learning, the eye has the potential to be a biomarker for multiple organ systems.¹²

Oculometrics: Eye movements as a biomarker

Beyond ocular structures, the visual system provides another unique insight into neurologic function via evaluation of oculomotor movements. Eye movements require a vast neural network that requires every lobe of the brain, brainstem, cerebellum, thalamus, midbrain, basal ganglia, cranial nerves, and visual tracts.¹³ Every brain structure involved in eye movements has a specific role and lesions in these areas can cause specific deficits in either a specific type of eye movement like fixational eye motion, saccades, smooth pursuits, vergences, and/or accommodation, as well as in specific eye movement dynamics including abnormalities in amplitude, accuracy, direction, velocity, and latency. With emerging eye-tracking technologies that incorporate AI and machine learning, oculomotor metrics (oculometrics) has the potential to serve as a correlate to neurologic function; aid in early and differential diagnosis of neurologic disorders, clarification of neuropathophysiology, and patient stratification; and objectively quantify disease progression, as well as response to treatment.^14,15 Another promising aspect of observing the oculomotor system is that eye movements have been able to correlate to disease disability scores, giving the potential to quantify/correlate with clinical end points like patient symptoms.¹⁶

Current challenges

One current challenge with using retinal structures like the macular ganglion cell complex and RNFL in neurodegenerative disease is poor specificity, as retinal thinning can occur in not just neurodegenerative disease but in normal retinal aging and comorbid ocular pathology such as glaucoma and age-related macular degeneration.¹⁷ Additionally, the process of retinal degeneration and protein deposition in ocular structures is not yet understood in relation to the temporal onset of neurodegeneration—where does it happen first, the retina or the brain? Do they degenerate simultaneously? Is the degeneration dependent or independent? Further longitudinal studies in preclinical and clinical disease is needed to understand these relationships and gain better insight into the underlying pathophysiology of these conditions as well as determine the association and utility of the eye as a proxy for neurodegeneration. Further challenges relate to the need for increased oculomotor screening in primary optometric and ophthalmologic routine eye examinations, as many of the aforementioned technologies are currently used mostly in tertiary and subspecialty clinics, and only after a confirmed neurologic diagnosis has been made for the patient. Thus, there is an overall lack of screening as well as a lack of standardization for methodology and examination techniques within eye care.

Exciting times are ahead for eye care as technology, including AI and machine learning in combination with the emerging field of oculomics and oculometrics, is set to revolutionize the way we diagnose not only ophthalmologic but also neurologic disorders. Eye care professionals will play a pivotal role in the multidisciplinary management of neurologic disease in the future, including early diagnosis and monitoring of neurodegenerative disease progression. Are you ready?

References

1. Biomarkers, End pointS, and other Tools (BEST). National Center for Biotechnology Information. January 28, 2016. Updated November 29, 2021. https://www.ncbi.nlm.nih.gov/books/NBK338448/

2. Galetta KM, Calabresi PA, Frohman EM, Balcer LJ. Optical coherence tomography (OCT): Imaging the visual pathway as a model for neurodegeneration. Neurotherapeutics. 2011;8(1):117-132. doi:10.1007/s13311-010-0005-1

3. Paul F, Calabresi PA, Barkhof F, et al. Optical coherence tomography in multiple sclerosis: a 3-year prospective multicenter study. Ann Clin Transl Neurol. 2021;8(12):2235-2251. doi:10.1002/acn3.51473

4. Chang Z, Xie F, Li H, et al. Retinal nerve fiber layer thickness and associations with cognitive impairment in Parkinson’s disease. Front Aging Neurosci. 2022;14:832768. doi:10.3389/fnagi.2022.832768

5. Sheriff S, Shen T, Abdal S, et al. Retinal thickness and vascular parameters using optical coherence tomography in Alzheimer’s disease: a meta-analysis. Neural Regen Res. 2023;18(11):2504-2513. doi:10.4103/1673-5374.371380

6. Gilmore CS, Lim KO, Garvin MK, et al. Association of optical coherence tomography with longitudinal neurodegeneration in veterans with chronic mild traumatic brain injury. JAMA Netw Open. 2020;3(12):e2030824. doi:10.1001/jamanetworkopen.2020.30824

7. Abdelhak A, Solomon I, Condor Montes S, et al. Retinal arteriolar parameters as a surrogate marker of intracranial vascular pathology. Alzheimers Dement (Amst). 2022:14(1):e12338. doi:10.1002/dad2.12338

8. Dheghani C, Frost S, Jayasena R, Masters CL, Kanagasingam Y. Ocular biomarkers of Alzheimer’s disease: the role of anterior eye and potential future directions. Invest Ophthalmol Vis Sci. 2018;59(8):3554-3563. doi:10.1167/iovs.18-24694

9. Koronyo Y, Biggs D, Barron E, et al. Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer’s disease. JCI Insight. 2017;2(16):e93621. doi:10.1172/jci.insight.93621.

10. Moncaster JA, Moir RD, Burton MA, et al. Alzheimer’s disease amyloid-β pathology in the lens of the eye. Exp Eye Res. 2022;221:108974. doi:10.1016/j.exer.2022.108974

11. Vig V, Garg I, Tuz-Zahra F, et al. Vitreous humor biomarkers reflect pathological changes in the brain for Alzheimer’s disease and chronic traumatic encephalopathy. J Alzheimers Dis. 2023;93(3):1181-1193. doi:10.3233/JAD-230167

12. Babenko B, Mitani A, Traynis I, et al. Detection of signs of disease in external photographs of the eyes via deep learning. Nat Biomed Eng. 2022;6(12):1370-1383. doi:10.1038/s41551-022-00867-5

13. Coiner B, Pan H, Bennett ML. Functional neuroanatomy of the human eye movement network: a review and atlas. Brain Struct Funct. 2019:224(8):2603-2617. doi:10.1007/s00429-019-01932-7

14. Przybyszewski AW, Śledzianowski A, Chudzik A, Szlufik S, Koziorowski D. Machine learning and eye movements give insights into neurodegenerative disease mechanisms. Sensors (Basel). 2023;23(4):2145. doi:10.3390/s23042145

15. Mao Y, He Y, Liu L, Chen X. Disease classification based on eye movement features with decision tree and random forest. Front Neurosci. 2020;14:798. doi:10.3389/fnins.2020.00798

16. Sheehy CK, Bensinger ES, Romeo A, et al. Fixational microsaccades: A quantitative and objective measure of disability in multiple sclerosis. Mult Scler. 2020;26(3):343-353. doi:10.1177/1352458519894712.

17. Lim JK, Li QX, He Z, et al. The eye as a biomarker for Alzheimer’s disease. Front Neurosci. 2016;10:536. doi:10.3389/fnins.2016.00536