Abstract
After years of anticipation, non-invasive tests for detecting cerebral amyloidosis and Alzheimer’s disease (AD) are entering clinical care. The PrecivityADtrademark test from C2N is a plasma-based test yielding an Amyloid Probability score with high sensitivity and specificity for brain amyloid accumulation, but some samples may have inconclusive results. The AGREEDementia consortium raised concerns that the field needs study of how best to use and communicate results of PrecivityADtrademark. Continued attention and mindfulness should be applied to the whole class of dementia biomarker tests and directed in light of FDA biomarker context of use framework. Unintended uses of biomarkers tests may have unintended consequences, such as mislabeling patients. AD biomarker tests may efficiently stratify AD risk but will inevitably be included in electronic medical records and be subject to interpretation by medical personnel lacking proper knowledge or context to interpret results appropriately. Another way forward is mindful discussion and consensus among all stakeholders about the uses and limits of each specific test.
In this issue of the Journal of Alzheimer’s Disease, the symptomatic subcommittee of the Advisory Group on Risk Evidence Education for Dementia (AGREEDementia) reviews the current status of a recently developed, United States Food and Drug Administration (FDA) approved plasma biomarker and Apolipoprotein E (APOE) isoform test called PrecivityADtrademark from the C2N biotechnology company [1]. The PrecivityADtrademark test is a new blood test intended for use in patients with cognitive impairment. The test aims to help physicians better determine the presence or absence of amyloid plaques in the brain. The PrecivityADtrademark test relies on precise and robust quantitation of the amyloid-β 42/40 ratio (Aβ42/40) and detection of the Apolipoprotein E prototype (equivalent to APOE genotype) in blood samples, using C2N’s proprietary mass spectrometry platform. Galasko et al. [1] are concerned that the test may not be “ready for prime time”, as a standalone measure without careful attention to the context of use. That is to say, issues derived from information theory (e.g., sensitivity and specificity, etc.) and cost might limit the usefulness of this test in clinical practice, along with potential pitfalls in test interpretation without clinical context.
Mindfulness, the art of being actively engaged in the present moment, has achieved wide popularity in our culture [2]. It can be paraphrased as “What are we trying to accomplish in the here and now?” Mindfulness leading to deeper appreciation of the PrecivityADtrademark test can move us forward in understanding and improving the field of AD biomarkers in general and PrecivityADtrademark specifically. The test itself measures plasma Aβ species and APOE isoforms, to yield an amyloid probability score (APS) which performs well in studies to date in populations with a 55% AD prevalence, such as might be seen in AD clinical trials [3]. In memory clinics, the prevalence of AD may be lower than in clinical trials settings and confounded by mixed vascular-AD and Lewy body disease with AD pathology, and diverse populations. Further extension into non-clinical populations such as attendees at a health fair, would likely further erode the positive predictive value of the PrecivityADtrademark test. Looking forward, the PrecivityADtrademark website mentions that similar methodology could be extended to other unspecified dementing conditions, so careful consideration of the test’s utility and ethics in the current clinical testing environment is in order. Furthermore, lessons learned from PrecivityADtrademark are likely to apply in the development of other tests for AD and other dementing disorders.
The FDA has published and defined several contexts of use for biomarker development, explained in detail in the BEST (Biomarkers, EndpointS and Tools) publication ([4]; see Fig. 1). Application of the BEST criteria to an AD biomarker test such as PrecivityADtrademark indicates that at least four categories apply, and this is also likely to apply for future clinical biomarker tests.
The FDA BEST model for context of use of biomarkers, useful for considering how a given biomarker might be developed and used (https://www.fda.gov/drugs/biomarker-qualification-program/context-use).
These include: Diagnostic biomarker used to detect or confirm presence of a disease or condition of interest or to identify individuals with a subtype of the disease: selection of people with high likelihood of cerebral amyloidosis, possibly to be confirmed with more direct cerebrospinal fluid (CSF) or amyloid PET studies. Prognostic biomarker used to identify likelihood of a clinical event, disease recurrence, or progression in patients who have the disease or medical condition of interest: Stratify patients using the continuous APS variable and comparing other tests, whether cognitive, digital biomarkers, CSF assays, etc. Under this context, it might also result in enrichment for patients meeting inclusion/exclusion criteria of a clinical study using a plasma test, again with later confirmation with CSF or PET. Predictive biomarker used to identify individuals who are more likely than similar individuals without the biomarker to experience a favorable or unfavorable effect from exposure to a medical product or an environmental agent: Patients with high APS and APOE e4 genotypes may be at higher risk of amyloid related imaging abnormalities, for example, in a clinical trial, thus directing our attention to those at highest risk to maintain clinical trial safety. Susceptibility/risk biomarker that indicates the potential for developing a disease or medical condition in an individual who does not currently have clinically apparent disease or the medical condition: Consider use in different clinical environments such as academic memory clinics, community practice sites in neurology, geriatrics, and other specialties, and conceivably at health fairs. This may be a standalone usage, since community usage and certainly health fairs, are not going to be routinely performing confirmatory tests of cerebral amyloidosis, and the quality of test interpretation will also vary. Note is made that the current indication is limited to individuals with memory or thinking problems.
Since the PrecivityADtrademark appears to apply to several BEST contexts, it is unclear which is primary, and which secondary, and how the test is optimized for each one of these contexts, or how much our present knowledge informs these different uses. The danger of labeling has been highlighted in the work of Whitehouse et al. [5] and the testing physician needs to individualize who is tested, why they are tested, how to disclose this information, and who gets access to test results. A test result that means one thing to a patient and a health care provider, may mean something quite different to an insurance company, or to a physician treating the patient in another setting.
Another area of concern is that although PrecivityADtrademark is meant for use in symptomatic persons, there is the temptation/risk that it could be applied predictively in asymptomatic persons, especially in clinical trials, even though not optimized for this use, with greater danger of erroneous clinical labeling, especially as research data is added to clinical data in the Electronic Medical Record (EMR) systems. This latter scenario is hardly far-fetched. We have encountered several cases where clinical trials using the term “Alzheimer’s disease”, led to treating physicians entering a diagnosis “Alzheimer’s disease” in an EMR, whether that is correct or not. Expunging this “diagnosis” from an EMR can be difficult or impossible, and creates confusion, patient resentment of privacy infringement, and substantial work for clinical trials staff. Unfortunately, the proliferation of clinical trials data in EMR and billing systems is increasing and may be beyond simple palliative steps to control. Systems should be developed with this in mind (re: mindfulness), to ensure accurate diagnosis and maintain privacy that is routinely promised in the informed consent process.
The question is how to proceed in this rapidly changing field is central to how researchers think about these problems, as well as how the general medical community and the public will react and utilize this information. The recent FDA approval of Aducanumab provides a cautionary tale and a model, that may ultimately help how we develop this area and avoid confusion. The approach to the solution highlighted by AGREEDementia consists of several further steps. Additionally, I would recommend an outside panel representing all applicable stakeholders including patients and their families, working with corporate sponsors to develop a very robust user manual for the clinician and the patient. The issues raised by these new biomarkers and specifically the AGREEDementia group are technical and complicated to understand; a comprehensive resource explaining the use, deployment and interpretation of a given biomarker test in clinical practice is essential. This would be an iterative process and bring stakeholders together to make sure the test is utilized as designed and can be readily interpreted by the ordering clinician. In the broader commercial technology world, even the simplest electronic device comes with a manual describing proper use and care; the possibility of doing harm with biomarker testing is palpable and must be addressed in a systematic manner. This review focuses on AD, but the issues are generic enough to apply to other conditions, and virtually all health care providers treating AD also treat other AD related dementias and other neurological conditions.
While all diseases may be considered as social constructions, our enthusiasm for transforming AD from a clinical entity to a biological entity is incomplete, and clinical considerations are unlikely to go away as noted by Galasko et al. [1]. People are “messy” and clinical methods may lead to confusion, but it is unlikely that relying strictly on biomarkers is going to yield completely clean results. Our continued mindfulness of these issues is our best weapon to maintain the highest standards of care of our patients.
