Opportunities to encourage adoption of a biomarker‐enabled care pathway for Alzheimer's in primary care

Abstract

Identification of early‐stage Alzheimer's disease (AD) remains a challenge due to limited specialist availability, diagnostic access, disease awareness, and cultural factors. Blood‐based biomarkers (BBBM) could play a critical role in the identification and referral of patients suspected of AD to specialty care. A multidisciplinary AD Biomarker Task Force was convened to evaluate current biomarker use cases, define an optimal biomarker‐enabled AD diagnostic care pathway, and understand factors impacting adoption. The Task Force identified opportunities to support biomarker‐enabled AD diagnostic care pathway adoption, including streamlining risk assessment and screening by leveraging digital tools, activating primary care providers through education, generating data to expand applicability to diverse populations, and advocating for aligned policies and quality measures. Adoption of BBBMs in the primary care setting will be critical to improve early AD detection. However, challenges to pathway adoption persist and will require action from clinicians, payers, policy makers, and patients to address.

Highlights

• Blood‐based biomarkers can streamline the identification of AD in primary care.

• Future biomarker‐enabled diagnostic care pathways will leverage digital assessments.

• Education, data generation, and policy advocacy are vital to encourage BBBM use.

• Implementation of AD care pathways requires the activation of diverse stakeholders.

1. INTRODUCTION

Alzheimer's disease (AD), the most common cause of dementia and the fifth leading cause of death among older adults in the United States, is projected to affect nearly 13 million people by 2050. ¹ , ² Confirming a diagnosis of AD as the cause of cognitive impairment has been challenging outside specialist settings, as clinical definitions might include several non‐AD etiologies. ³ However, the emergence and validation of biomarkers for AD enable both a clinical and biological definition of the disease, and evidence from clinical trials of disease‐modifying treatments specific to AD makes AD biomarkers a critical part of the pathway to diagnosis. ⁴ , ⁵ Traditional biomarker testing modalities to help with AD diagnosis and staging include invasive cerebrospinal fluid (CSF) analysis of various biomarkers like amyloid‐β (Aβ42/40), tau, and phosphorylated tau (p‐tau) as well as imaging including positron emission tomography (PET) scans that detect fibrillar amyloid‐beta plaques and tau tangles. Additional imaging modalities include magnetic resonance imaging (MRI) or computed tomography (CT), which can be used to assess structural features such as hippocampal atrophy. While in vivo methods are considered the gold standard for AD diagnosis, use of these traditional diagnostic modalities is limited by cost, availability to non‐specialists, risk to older patients, test burden, invasiveness, and inequitable access for many population groups and geographic regions. ⁶ , ⁷ Additionally, these tests cannot be conducted at point‐of‐care and require referral to specialists to order and interpret, often within large academic medical center environments that have substantial wait times. Such delays are not inconsequential, as patients may decline to the point that they are excluded from additional workup and access to promising resources or therapies. ⁸

Recent advances in measurement of digital and blood‐based biomarkers (BBBMs) such as phosphorylated tau (p‐tau), amyloid‐β (Aβ42/40), glial fibrillary acidic protein (GFAP), and neurofilament light (NfL) present opportunities for a potentially more cost‐effective, accessible, and less invasive testing option that could support diagnosis at earlier stages of AD within primary care settings as part of broader population health approaches. The emergence of these biomarkers has already begun to shift the clinical approach and framework of how the disease is detected, diagnosed, staged, and managed as evidenced by the Alzheimer's Association's 2024 “Revised Criteria for Diagnosis and Staging of Alzheimer's Disease.” ⁴

In conjunction with BBBMs, digital biomarkers and tools are additional emerging resources that can be embedded into clinical practice to improve the early diagnosis of AD. Digital biomarkers are objective, quantitative physiological, and behavioral data point metrics with the potential to enhance AD assessment. With the continued collection of electronic patient data, two classes of digital biomarkers have emerged as follows: risk stratification and natural language processing (NLP) tools. One electronic health record (EHR)‐based risk stratification tool of note includes the EHR Risk of Alzheimer's and Dementia Assessment Rule (eRADAR), which is a tool leveraging commonly collected structured clinical data that has exhibited external validity for the ability to identify adults with increased risk of having undiagnosed dementia across multiple health systems, with an area under the curve up to 0.84 (95% confidence interval: 0.84–0.85). ¹⁰ Additionally, artificial intelligence (AI) tools including NLP have shown initial promise in the ability to analyze unstructured data such as speech to help detect signs or risk of progression to AD, with one study exhibiting approximately 78% accuracy in using NLP and other machine learning techniques to predict a patient's progression to AD based on speech recorded during cognitive exams. ¹¹ While detailed knowledge of these tools is currently limited, moving forward, integration of these types of digital biomarkers and tech‐enabled tools in concert with other easily assessed metrics such as decline in grip strength or gait speed can provide a non‐invasive, more efficient method for screening, and identification of patients with AD or at risk of progressing to AD.

The use of blood‐based and digital biomarkers is also a critical component in achieving detection in the early stages of AD that can improve timely access to support or counseling resources, as well as novel, biomarker‐guided therapeutics when appropriate. As primary care practitioners (PCPs), including physicians, nurse practitioners, and physician assistants, have limited time to assess for AD and report having challenges with screening, these emerging biomarkers have the potential to help streamline clinical workflows for patients suspected of AD in the primary care setting. ¹² Despite the promising research demonstrating the analytical validity of various AD biomarkers, implementation within clinical practice has yet to be fully realized. As biomarker tests continue to develop, they hold significant promise for expanding access to early detection and treatment, particularly in primary care settings where most patients initially present with cognitive symptoms.

Recently, recommended AD care models considering the integration and use cases of BBBMs in clinical practice have begun to emerge. ¹³ , ¹⁴ However, it is essential to garner consensus among PCPs, neurologists, psychiatrists, and other stakeholders to both determine the optimal approaches for systematizing and operationalizing tools within existing clinical decision support infrastructure, as well as how to eliminate barriers for the successful adoption of AD care models. Rapid advancements in the AD care landscape such as recent United States Food and Drug Administration (FDA) approval of disease‐modifying therapies (DMTs), anticipated upcoming FDA approval of the first BBBM for AD this year, and launch of the Centers for Medicare and Medicaid Services (CMS) Innovation Center GUIDE model have illuminated a need for expert consensus on the role and optimal application of BBBMs within clinical workflows to impact patient care delivery and reduce inequities imposed by poor access to diagnostic testing. Leveraging findings from a literature review and their clinical expertise, the AD Biomarkers Task Force identified opportunities to spur systematized utilization of BBBMs and encourage adoption of a diagnostic care pathway to support streamlined identification of patients with AD in primary care.

2. METHODS

To understand the current usage of emerging biomarker tests and opportunities to integrate them within an AD diagnostic care pathway for widespread implementation, a multidisciplinary AD Biomarker Task Force was convened over the course of three virtual meetings to discuss the use of AD biomarkers within clinical practice, challenges experienced in implementation, opportunities to address key challenges (including gaps in data as well as PCP engagement and confidence), and a conceptual framework for a biomarker‐enabled AD diagnostic care pathway. The Task Force was comprised of 10 experts representing the clinical disciplines of neurology, neuropsychology, geriatrics, geriatric psychiatry, primary care, epidemiology, laboratory benefits management, and patient advocacy. Task Force members represented a diverse array of system archetypes, geographies, and patient populations.

3. RESULTS

Task Force members indicated that the use of BBBMs has largely been confined to the research setting to date, oftentimes to patients diagnosed in specialty care equipped with CSF and PET modalities. Task Force members attributed the lack of BBBM adoption to several key factors as follows: limited awareness of biomarkers outside the specialist community, process challenges (e.g., difficulty ordering, insurance coverage), and limited data to support the clinical use of specific cutoff points for BBBM measures across the spectrum of disease states, diverse patient populations, and comorbidities. To address these challenges, Task Force members created a call to action spanning four key categories as follows: streamlining risk assessment and screening, PCP education and activation, data generation, and policy reform.

3.1. Streamlining risk assessment and screening

Following the conclusion of the final virtual meeting, Task Force participants defined an optimal primary care biomarker‐enabled AD diagnostic care pathway conceptual framework (Figure 1).

FIGURE 1.

Proposed future biomarker‐enabled AD diagnostic care pathway for primary care clinical care settings. Description: A proposed AD diagnostic care pathway leveraging blood‐based biomarkers to improve early AD detection in primary care settings. AD, Alzheimer's disease.

As recommended by Task Force consensus, the proposed conceptual framework illustrates the role that AD BBBMs could play in future primary care settings to identify patients suspected of AD earlier in their disease progression.

While there is currently limited awareness surrounding the use of evidence‐based AI in systematizing the identification of patients with AD, the group expressed a need for an AI‐based and/or EHR‐integrated algorithm or digital risk assessment tools to help PCPs detect patients who may have undiagnosed AD. These tools will be a critical element in redesigning clinical pathways that will empower primary care to incorporate AD BBBMs into clinical practices for the diagnosis of AD, setting the stage for decisions about care and treatment. The Task Force agreed that moving forward, the optimal diagnostic care pathway should harness the analytic power of AI‐based algorithms leveraging unstructured clinical notes and/or existing structured data within the EHR to proactively assess an individual's brain health, identify patients at risk for cognitive impairment, and potentially flag those requiring HCP attention to further assess for brain health concerns. This method ultimately allows a practice to leverage existing data within the EHR and check for additional risks related to brain health.

The diagnostic care pathway for future clinical use of BBBMs proposed by the Task Force envisions a system that begins with AI risk stratification prior to the patient's visit to flag that the patient may be at higher risk of AD and dementia. At the visit, those patients identified as needing a cognitive assessment can be triaged to an extended care team member (i.e., non‐physician, nurse practitioner, or physician assistant provider) to deliver a digital brain health assessment or provide a self‐administered assessment if available. If these resources are unavailable, patients flagged as being at risk for cognitive impairment should be evaluated by their PCP to query brain health concerns, consider risk factors for cognitive impairment, and conduct a brain health assessment. While the use of digital tools and/or AI‐based algorithms can help streamline workflows, it is important to recognize that these tools are continuously evolving and may not always be feasible to integrate with all healthcare settings and geographies. Nevertheless, assessment for cognitive impairment should continue to be pursued by a healthcare provider during the patient encounter, ensuring that patients can be identified as being at risk for AD even without such technology‐enabled resources. Where mild cognitive impairment or mild dementia appears possible, team members should work to evaluate potentially reversible causes of cognitive impairment and conduct a diagnostic assessment leveraging AD BBBMs. In some instances, positive results from the biomarker assay can lead to a diagnosis of AD, while indeterminate findings will require additional workup through a confirmatory PET or CSF‐based biomarker test, and negative results can help rule out AD. In all these situations, evaluation—and when possible, remediation—of other patient‐specific factors that could contribute to or exacerbate cognitive impairment are important parts of ongoing patient care in the primary care setting. The diagnostic care pathway proposed streamlines the process within the primary care environment, permitting subspecialty resources to be reserved for further disease‐specific testing, treatment determinations, and links to pertinent health systems or community opportunities to enhance patient and care partner support.

3.2. Educating and activating PCPs

The Task Force recognized that PCPs play a critical role in the identification, diagnosis, and treatment of patients with AD, and that educational materials and approaches on AD biomarkers need to be tailored for primary care clinicians and their patients. Within primary care settings, Task Force members emphasized that while early detection of cognitive impairment, including AD, is possible in primary care, it is not currently a health system priority. Furthermore, while cognitive assessments play a critical role in diagnosing AD, uncertainty remains about how to accurately interpret results, especially in diverse populations. Adoption of BBBMs as part of the initial assessment could reduce uncertainty and encourage PCPs to adopt a more systematic approach to evaluating cognitive impairment in their patients. Task Force members recommended anchoring education to a standard of care algorithm or diagnostic care pathway; educational initiatives should ultimately include building biomarker‐focused learning modules for primary care clinicians that reflect their commitment to whole‐person care. Moreover, education should clarify why early attention to both subjective and objective cognitive impairment matters as well as explain the place of biomarker testing in establishing a causal diagnosis. This will ultimately empower clinicians to successfully interpret and explain test results to their patients while concurrently using the results to craft an actionable ongoing care plan. Training and learnings should also be adaptable to different practice environments and prepare clinicians to explain to practice managers and health system administrators how earlier, more complete diagnosis of cognitive impairment can be integrated into routine workflows. Outreach through PCP conferences, publications, Geriatric Workforce Enhancement and ECHO Programs, and hotlines attended by experienced experts can amplify the impact of well‐designed educational materials and encourage PCPs to maximize their contributions to the care and well‐being of patients and care partners. Ultimately, activating and empowering PCPs will be critical moving forward, especially given the increasing shortages of neurologists and geriatricians.

In addition to PCP education, Task Force members recognized the need for ongoing patient education to successfully integrate and use diagnostic BBBMs in clinical practice. Many symptomatic patients currently express discomfort with BBBM testing for AD assessment and diagnosis, while others without cognitive complaints or symptoms may express interest in such testing. These observations point to the need to continue working against fear and stigma associated with an AD diagnosis, and to craft new patient‐focused conversations that promote personal agency and shared decision‐making in adopting strategies to optimize their brain health.

3.3. Bridging data gaps

While the promise of BBBMs in facilitating AD diagnosis is clear, data remain insufficient to validate their use to inform clinical decision‐making in real‐world practice settings, particularly in marginalized and minoritized communities. For example, the APOE E4 allele, the strongest genetic risk factor for early onset AD, has less penetrance among individuals with greater African or Indigenous ancestry. Moreover, Black Americans with AD are less likely to have positive amyloid PET scans even in the face of clinical AD. This underscores the need to continue to study the distribution of AD BBBMs among populations and subpopulations and to better understand how to optimize interpretation of test results in complex clinical situations. ¹⁵ In addition, the prognostic value of AD BBBMs among cognitively normal individuals remains uncertain because many older adults die with significant AD pathology, never expressing clinical impairment. To promote clinician confidence in using BBBMs in practice, more robust evidence is needed demonstrating that the sensitivity, specificity, and longitudinal performance of these tests are comparable to that of PET or CSF tests. Furthermore, recognizing that multiple conditions can impact cognitive function, evidence, and data using BBBMs must consider comorbid conditions and how that may impact test results. Lastly, Task Force members identified the need for BBBM data to provide guidance pertaining to the interpretation of test results and how clinicians may act upon the results to inform the development of an optimal care plan.

In summary, while plasma biomarkers may not be ready for ``prime time’’ in primary care across a wide range of patient populations, they could serve as valuable tools that informed clinicians can leverage in the near future to assist with diagnosis and ultimately initiate appropriate referrals to specialty care for higher‐risk patient populations.

3.4. Implementing policy reform

Currently, there is no national quality measure that sets reportable standards for identifying, diagnosing, or managing cognitive impairment, including AD, in healthcare. With competing clinical priorities and challenging resource allocation decisions, population health leaders are unlikely to prioritize screening and diagnosis of patients with cognitive impairment, including AD, within their institutions unless clinical and financial incentives are aligned. To encourage successful future adoption of the biomarker‐enabled diagnostic care pathway, implementation of policy changes is necessary, along with the development of appropriate AD quality assessment and screening measures. This development should include engaging policy leaders and organizations that influence primary care practice, such as the United States Preventive Services Taskforce (USPSTF), CMS, the Agency for Healthcare Research and Quality (AHRQ), and the American Academy of Family Physicians (AAFP), which already has a well‐designed guide to clinical evaluation and care of patients with cognitive impairment. The three BOLD Public Health Centers of Excellence and 43 state and local programs, funded by the Centers for Disease Control (CDC) to establish a national public health infrastructure for ADRD risk reduction, detection, and caregiving, are natural allies in this process. A key step is the successful translation of advocacy for meaningful adoption of AD detection practices into primary care settings and the communities they care for, particularly in the rapidly changing realm of biomarker discovery and use. The Task Force recognizes the necessity of ensuring that early AD diagnosis policy and guideline development is mindful of protecting patients from discrimination by insurers, employers, and, in some states, Departments of Motor Vehicles (DMV).

4. CONCLUSION

The rapid evolution of diagnostic and treatment modalities for AD and the potential for similar discovery in other cognition‐impairing diseases highlight critical gaps in knowledge and practice. These include gaps in understanding how to use AD BBBMs across diverse populations and outside of specialized research settings, translating knowledge into clinical and community settings where most patients receive their care, and creating meaningful education that addresses the needs of primary care clinicians, their patients and care partners, and health systems. The growing national public health infrastructure for AD creates new opportunities to engage multi‐sector dialogue through its convening and partnership roles. The rapidly evolving therapeutic landscape makes the present moment one of great opportunity to advance understanding of how AD biomarkers can contribute to earlier detection and better care for patients. Future research should prioritize clear understanding of how to use non‐invasive BBBMs in a diagnostic care pathway that includes community‐representative populations. Mechanisms are needed to rapidly synthesize new findings into an updated knowledge base and use them to tailor education to clinicians, patients, communities, insurers, and health system leaders. Development and integration of AI and digital risk assessment tools into EHR systems, along with dedicated advocacy for aligned policies and quality measures, will be crucial for the successful adoption of a biomarker‐enabled AD diagnostic care pathway.

Promising diagnostic BBBM tests will accelerate the currently slow AD assessment process by facilitating quicker, more objective clinical evaluations and diagnosis of early AD while sustaining attention to other causes of cognitive impairment when AD is ruled out. While it is important not to let applications leap ahead of the science needed to justify their use, BBBM tests are no longer merely a nascent possibility; they are now tools poised to expand the reach of AD prevention and treatment. Advancing patient awareness of AD biomarkers will be critical to future pathway adoptability and is especially salient for older adults. The Task Force noted that a combination of research excellence, clinical experience, and engagement with digital medicine are now mandatory. Pathways alone will not counter skepticism among leaders or decision‐makers whose focus is on short‐term benefits. Systemic integration of the care pathway within healthcare delivery systems will require engaging and garnering buy‐in with multidisciplinary stakeholders, including primary care, specialty clinical champions, and administrative and IT/IS leaders in health systems, along with support from public health programs to build community awareness and public demand. Successful sponsorship for implementation will involve providing data that supports the clinical utility and cost‐effectiveness of biomarker testing and pathway implementation. With researchers, clinicians, and other stakeholders investing in efforts such as blood‐based AD biomarker discovery and AI in medicine, we are poised to support the move of BBBMs and digitally analyzed biomarkers from promise and potential to clinically relevant AD assessment and diagnosis approaches.

CONFLICT OF INTEREST STATEMENT

S.B. has served as a consultant to Novo Nordisk, Roche Genentech, Biogen, Eisai, Abbvie, Lilly, and Linus Health. S.B. has received compensation from Medscape and WebMD for non‐CME/CE services. R.A. has served as a scientific advisor to Signant Health and Novo Nordisk. A.C. has received funding from the National Institute of Neurological Disorders and Stroke, California Department of Health Care Services, California Department of Public Health, Gary and Mary West Health Institute, and Health Resources and Services Administration. A.C. has received honoraria from Emory University, Peerview, Medscape Education, and Advanced Health Media Push. Archstone Foundation has provided A.C. support for attending meetings. S.G. is a co‐founder of Recuerdo Pharmaceuticals. S.G. has served as a consultant in the past for J&J, Diagenic, and Pfizer, and currently consults for Cognito Therapeutics, GLG Group, SVB Securities, Guidepoint, Third Bridge, MEDACORP, Altpep, Vigil Neurosciences, and Eisai. S.G. has received research support in the past from Warner‐Lambert, Pfizer, Baxter, and Avid. S.G. currently receives research support from NIH grants U01AG046170, RF1AG058469, RF1AG059319, R01AG061894, P30 AG066514 to Mary Sano, and from the Cure Alzheimer's Fund. D.K. has been a paid consultant or served on the advisory board for Eli Lilly, Eisai, Roche/Genentech, Acumen, Wellin5/Therachat, and Med Learning Group. J.M. has served as a consultant for Genentech, Inc., AARP—Global Council on Brain Health (GCBH), ACADIA, Alliance for Aging Research, AiOmed, Axsome, Exciva, Lundbeck, Merry Life, Praxis Bioresearch, and Otsuka/Avanir. J.M. has received research support from National Institute on Aging, National Institutes for Health, Alzheimer's Association, Eisai Inc., GHR Foundation, Eli Lilly and Company, Cerevel Therapeutics, LLC, and GSK Research & Development Limited, National Endowment for the Arts. Additionally, J.M. has also received research support from American Association of Retired Persons (AARP), Karuna Tx, Alzheimer's Drug Discovery Foundation, Cognition Therapeutics, Inc., Suven Life Sciences Ltd., and Vivoryon Therapeutics. J.M. has received payments from ACADIA, Genentech, Inc., and Alliance for Aging Research for presentations. ACADIA, AARP—Global Council on Brain Health (GCBH), AiOmed, Alzheimer's Clinical Trials Consortium (ACTC), Alzheimer's Therapeutic Research Institute (ATRI), Axsome, Genentech, Inc., Lundbeck, and Merry Life have provided J.M. support to attend meetings. J.M. has served on the advisory board or data safety monitoring board for AARP—Global Council on Brain Health (GCBH), ACADIA, Alzheimer's Therapeutic Research Institute (ATRI), Alzheimer's Clinical Trials Consortium (ACTC), Alzheimer's Association, and Alliance for Aging Research. J.M. has also served on the served on the advisory board or data safety monitoring board for Axsome, Exciva, International Psychogeriatric Association (IPA), NAB‐IT: Nabilone for Agitation Blinded Intervention Trial, CALM‐IT: Cannabinoid Liquid Medication Intervention Trial, Technology Accelerator Company (TAC), and Praxis Bioresearch. J.M holds stocks from Biopharma Connex, NeuroQuest, and Recruitment Partners. S.M. served as a consultant to the University of Southern California, Alzheimer's Therapeutic Research Institute/Alzheimer's Clinical Trials Consortium (ACTC). M.M. received research funding from the NIH, DOD, Alzheimer's Association, and Davos Alzheimer's Collaborative. M.M. has consulted for or served on the advisory boards for Althira, Novo Nordisk, Biogen, Danahar, Eisai, Lilly, Roche, Siemens Healthineers, and Sunbird Bio. D.W. served as a paid consultant or advisory board member for SynapsDx and Novo Nordisk. Additionally, D.W. serves as the editor‐in‐chief of Alzheimer's & Dementia. All other authors report no conflicts of interest. Author disclosures are available in the supporting information.

CONSENT STATEMENT

All Task Force members provided informed consent to participating in the Task Force prior to the first meeting.

Supporting information

Supporting Information

Borson S, Au R, Chodos AH, et al. Opportunities to encourage adoption of a biomarker‐enabled care pathway for Alzheimer's in primary care. Alzheimer's Dement. 2025;17:e70095. 10.1002/dad2.70095