Minimally invasive detection of early Alzheimer*s disease pathology at the eye clinic using blood tests and eye-scans: the BeyeOMARKER study

Alzheimer*s disease (AD) is characterized by the presence of amyloid beta (Aβ)
plaques and tau neurofibrillary tangles. The disease course of AD is
increasingly recognized as a continuum in which pathophysiological changes
occur up to around 10-20 years before the expression of clinical dementia. The
window prior irreversible brain damage and cognitive decline is crucial to
implement strategies that could slow the disease or could improve quality of
life. However, intervention and diagnosis are intrinsically connected to each
other: an accurate biomarker-based diagnosis is imperative to provide accurate
prognostic information, appropriate intervention, access to possible clinical
trials or, in the future, available disease-modifying medication. Currently, a
biomarker-confirmed Alzheimer*s disease (AD) diagnosis requires expensive and
sometimes invasive procedures that are only accessible in highly specialized
clinics (i.e. a lumbar puncture or amyloid positron emission tomography (PET)).
Consequently, the time to diagnosis in current clinical practice is relatively
long and this contributes to the increasing societal and economic disease
burden of AD. Hence, alternative methods and an infrastructure is warranted.
Currently, plasma phospo-tau (p-tau) and eye-based hyperspectral (HS) retinal
scanning are emerging as highly promising tools to distinguish between subjects
with and without AD pathology. Thus far, studies on these biomarkers did not
resemble *real-world* clinical application as these were mostly retrospective,
included participants with high diagnostic certainty, few medical comorbidities
and did not use a priori defined cut-offs. The BeyeOMARKER study aims to
provide real-world proof-of-concept for early detection of AD pathology outside
the memory clinic through prospective evaluation of plasma p-tau in an eye
clinic with high patient-throughput. Eye clinics are an important setting with
a high likelihood of early detection of AD given the high patient throughput of
patients within the age-range where AD typically starts (i.e. 50-80 years old).
Furthermore, eye patients may be at increased risk for AD based on
epidemiological associations showing a higher risk on AD for certain eye
conditions (e.g. HR=1.26 for cataract) and visual impairment (e.g. HR=1.47),
and based on shared risk factors and pathological mechanisms between eye- en
brain diseases. Importantly, visual impairment may mask cognitive complaints
leading to significant diagnostic delays and underrepresentation of these
individuals in clinical studies. Altogether, the BeyeOMARKER study utilizes
this unique setting to assess whether these novel tools can facilitate early
detection of AD and an accelerated diagnostic process in this clinically
relevant group. Therefore, the BeyeOMARKER study will screen 700 healthy
volunteers from the eye clinic using a bloodtest en short cognitive and
visuoperceptive screening. Subsequently, 150 participants will be invited for
the BeyeOMARKER+ cohort which entails more extensive assessment
(neuropsychological testing, magnetic resonance imaging (MRI), amyloid- en tau
PET). Based on these data we aim to determine which biomarkers are, in
practice, most suitable for early detection of AD. The BeyeOMARKER study aims
to provide proof-of-concept for implementation of these promising tools for AD
in eye clinics, with the ultimate goal of providing recommendations for
minimally invasive detection of AD pathophysiology in alternative clinical
settings.

The BeyeOMARKER study*s primary aim is to (1) evaluate the real-world
predictive value of blood- and eye-based biomarkers for detection of AD-related
clinical progression and biological changes in eye clinics. Secondly, we aim
(2) to assess the individual and complementary clinical predictive value of
blood-based biomarkers and retinal scanning through multivariate modelling of
clinical outcomes, (3) to provide head-to-head comparison of the accuracy of
blood- and eye-based biomarkers for early detection of AD pathophysiology, (4)
to explore the pathophysiological mechanisms linking AD and eye disease by
comparing the clinical and neurobiological manifestation of AD in the
BeyeOMARKER cohort against a traditional memory clinic cohort, (5) to
investigate enrichment for AD in an eye clinic population based on a (relative)
prevalence estimate of p-tau positivity, and (6) to ultimately provide a
roadmap for future studies on minimally invasive early detection of AD in
alternative diagnostic settings.

A prospective observational longitudinal cohort study.

Screening requires a single visit (~45 minutes) at the recruitment site
(Bergman eye clinics) and includes collection of sociodemographic and medical
information, blood collection (1x 6 mL EDTA blood) for plasma p-tau and N4PE
analysis, optional blood collection for the BeyeOMARKER biobank (2x 6 mL EDTA
blood for plasma, 1x 6 mL EDTA whole blood for genetic analyses), and a short
cognitive and visuoperceptive screening battery. All screened participants
(n=~700) will undergo annual cognitive screening (remote T1 and T2, 9-12 month
interval) and are invited for annual questionnaires in collaboration with the
*A Personalized Medicine Approach for Alzheimer's Disease* (ABOARD) cohort
(METc 2022.0120). In addition, all plasma p-tau positive cases and matched
negative controls (n=150) will be included in the BeyeOMARKER+ cohort for more
extensive assessment at the on-site T0 (preferably 3, but max. 6 months after
screening) and the on-site T2 (21-24 months after T0). This includes standard
retinal imaging and HS retinal imaging using the Optina Mydriatic Hyperspectral
Retinal Camera (MHRC) (T0; 60 minutes), brain MRI (T0 and T2; 30 minutes),
amyloid- and tau-PET (T0; 70 and 100 minutes), and cognitive and cortical
vision assessment (T0 and T2; 90 minutes) at the Amsterdam University Medical
Center, location Vrije Universiteit medical center (VUmc). The study procedures
of T0 will be split across minimally two visit days, adding up to a total of
three visit days and one remote assessment across two years of participation in
the longitudinal subcohort. Potential burdens experienced by participants
include mild discomfort during the blood draw, temporary photophobia and
blurred view due to pupil mydriasis (tropicamide 0.5% drops) prior to retinal
imaging, discomfort during the MRI, and burdens related to the PET-scan (i.e.
radiation exposure, minimal risk on idiosyncratic reaction to the tracer and
placement of the intravenous catheter). Risks related to the Optina MHRC are
negligible according to the FDA 510(k) premarket notification of intent, and
light exposure during scanning is below the recommended limits. All other study
procedures are medical routine procedures with known and acceptable risks.
There are no expected medical benefits to participation. The BeyeOMARKER study
has a non-disclosure policy for personal study results and participants will be
informed of the BeyeOMARKER disclosure protocol prior to participation to
ensure that participants are supported in making informed decisions concerning
their own biomarker data. However, participants will receive consultation in
case of a clinically relevant incidental finding.