Ethical Arguments Concerning the Use of Alzheimer’s Disease Biomarkers in Individuals with No or Mild Cognitive Impairment: A Systematic Review and Framework for Discussion

Abstract

Background:

The shift to defining Alzheimer’s disease (AD) as a biological continuum, which is characterized by the presence of biomarkers instead of clinical symptoms, has sparked a widespread debate. Insight into the given arguments and their underlying moral values is crucial to ensure well-considered and appropriate AD biomarker testing in the future.

Objective:

To critically review the arguments in favor of or against AD biomarker testing in people with no or mild cognitive impairment and to explicate their underlying moral values.

Methods:

Seven databases were systematically searched for publications mentioning arguments of interest. Arguments are identified using qualitative data-analysis and evaluated within an ethical framework.

Results:

Our search yielded 3,657 articles of which 34 met the inclusion criteria. We discuss the clusters of arguments separate from their evaluation and the assessment of the debate as a whole. The right to know, which derives from the moral value of respect for autonomy, is a central argument in favor of biomarker testing. On the other hand, fear of the disease and lack of a disease-modifying treatment may result in a negative balance of good over inflicted harms, which argues against its use.

Conclusion:

Critical evaluation and weighing of the given arguments in a specific context, within an ethical framework, demonstrates the necessity to differentiate between what we hope or expect from research and where we currently stand. While AD biomarkers may have an indispensable value for research, the current advantage for clinical practice appears limited.

Keywords

Alzheimer’s disease bioethical issues biomarkers disclosure early diagnosis ethics

INTRODUCTION

There is a strong movement in research toward a definition of Alzheimer’s disease (AD) as a pathophysiological process which may or may not lead to the symptoms of dementia that used to define AD [1 –6]. The recently proposed update of the National Institute on Aging and Alzheimer’s Association (NIA-AA) research criteria reflects the progression of this movement [7]. It is fueled by the hope to prevent or slow down cognitive decline by intervening early in the pathophysiological process of AD before the onset of dementia and by the practical possibility to assess the occurrence of this process, albeit indirectly, in vivo using cerebrospinal fluid (CSF) analysis and positron emission tomography (PET) imaging of the brain. The measures of these tests, such as amyloid-β levels in the brain, are often collectively referred to as ‘biomarkers’ of AD. Although the application of biomarkers in people with no or mild cognitive impairment is generally intended for research, such markers are also used in clinical practice [8]. Motivations that are described for clinical use of AD biomarker tests is the clinician’s aim to reduce uncertainty about the diagnosis [9] or to attain neurochemical confirmation of AD in case of psychiatric symptoms, such as depression [10].

It is, however, still unclear whether the movement toward biomarker testing in people with no or mild cognitive impairment is desirable from an ethical perspective. It may offer the advantage to anticipate future cognitive decline [11, 12] and people could be granted the right to know their biomarker status [13]. On the other hand, AD is among the most feared diseases of aging [14] and no disease-modifying treatment is currently available. For this reason, it is suggested that disclosing biomarker results to research participants or patients may have a negative psychological or social impact [15 –18], but empirical research is needed to verify this [19].

Numerous ethical arguments have been mentioned in favor of and against biomarker testing in people with no or mild memory impairment in the literature so far, but a critical review of these arguments is currently missing. Moreover, as the debate is centered within the biomedical discourse, the moral values that underlie these arguments are often overlooked. All arguments—even those that appear to be purely medical-technical or economic—imply that a certain value should be protected or pursued. Insight into these values is essential for clarifying and motivating one’s position on the ethical desirability of biomarker testing.

The movement toward a biomarker-based diagnosis of AD in research and routine biomarker use in clinical practice calls for a critical reflection on its ethical desirability. We provide a critical review of the arguments given in favor of and against biomarker testing and an insight into their underlying moral values. This will constitute the basis for a final weighing of the given arguments within any specific context.

We evaluate the arguments in two consecutive steps. First, we provide an overview of all arguments that were extracted from the literature through qualitative data-analysis. In the second step, the arguments are evaluated within the framework of the four basic principles of medical ethics: non-maleficence, beneficence, respect for autonomy, and justice [20] to elucidate their underlying moral values and provide guidance when weighing the given arguments within a specific context or future scenario. Lastly, we identified issues that in our view merit greater attention in the continuation of this debate based on the evaluation of given arguments. This overview and critical evaluation of the desirability of AD biomarker testing is in our view a crucial step to ensure a well-considered and appropriate practice of AD biomarker testing in the future.

METHODS

We performed a systematic review of argument-based scientific literature [21]. Different from a systematic review of scientific evidence, we aimed to review the scientific discourse: Which arguments have been given in favor of and against AD biomarker testing in people with no or mild cognitive impairment?

Search strategy

The systematic literature search was conducted on October 24, 2016 in seven electronic databases to identify all publications that included arguments of interest: Embase, Medline Ovid, PsycINFO, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Cochrane Central Register of Controlled Trials, Web of Science, and Google Scholar. The search strategy was developed in collaboration with an information specialist from the Erasmus Medical Center. Additional studies were identified from reference lists of included articles and by searching the authors’ personal collections of articles. For the specific search terms and strings per database, please consult Supplementary Table 1.

Inclusion and exclusion criteria

After identification of records from the search strategy, duplicates were removed. Papers were included when they were peer-reviewed, written in English, and on non-genetic biomarkers of AD. Publications on genetic testing for AD are outside of the scope of this article because, being constant over time, genetic results cannot serve as early markers of disease according to the definition of AD as a biological continuum. Since amyloid-β biomarkers have been an important drive toward this reconceptualization of AD, we only included publications that were published after 1985, as in that year amyloid-β was first identified as being the main component of the deposits that were first described by Alois Alzheimer [22]. Additionally, included articles needed to address arguments in favor of or against AD biomarker testing in people with no or mild cognitive impairment or arguments in favor of or against the disclosure of its results as a central aim of the article.

As a pilot, three reviewers (MSm, KT, ER) independently screened 50 titles and abstracts of identified records for relevance to the research question and adherence to inclusion criteria. Similarities and discrepancies were discussed to inform the screening strategy. In the screening phase, one reviewer (MSm) screened remaining titles and abstracts. In the eligibility phase, two reviewers (MSm and KT) independently assessed full-text articles for eligibility. In case of disagreement, a third and fourth reviewer (MS and ER) decided in consensus about inclusion. Reference lists of included articles were screened by MSm to include other relevant articles. Quality and risk of bias assessment was not part of the literature selection, as is done in a quantitative meta-analysis, because quality of the argumentation is discussed as part of the review itself [23]. The process of study selection is illustrated in Fig. 1.

Fig. 1.

PRISMA flow diagram illustrating process of study selection. ¹Publications were excluded when the use of Alzheimer’s disease (AD) biomarkers was not explicitly mentioned in publications on early AD diagnosis, when the publication concerned early diagnosis of dementia instead of AD or if the reflection on the desirability of AD biomarker testing was too limited.

Extraction and synthesis

We conducted a qualitative thematic analysis of included literature. For data-extraction and data-synthesis, we used Qualitative Solutions and Research (QSR) NVivo 11 qualitative data analysis software. The main outcome measures extracted were arguments in favor of and against AD biomarker testing in people with no or mild cognitive impairment. First, relevant text passages were extracted by MSm. Second, categories and subcategories of arguments (the coding tree) in favor of or against AD biomarker testing were generated based on the content of the extracted passages [21]. This initial coding tree was reviewed and discussed with all authors (KT, MSc, and ER). MSm coded the extracted passages with time intervals to check whether new meanings could be attributed to the transcripts and to minimize bias. To minimize further subjective interpretation, KT, MSc, and ER reviewed and discussed the coding scheme multiple times.

The arguments in favor of and against AD biomarker testing are presented separately from the ethical structuring and critical appraisal of the current debate to avoid bias in the presentation of the results.

RESULTS

Study selection and characteristics

The systematic search yielded 3,657 articles out of which we identified 34 publications that fulfilled our inclusion criteria and were included for systematic review [1 , 24–49]. The wide range of settings in which AD biomarker testing is discussed in the included literature, reflects the variable purposes of its (potential) use. Even though the content of the extracted arguments greatly overlaps, their weight is highly dependent on the specific context of AD biomarker testing. Three contextual differences, which are relevant for the weighing of the arguments, can be identified in the included literature. Firstly, the included literature covers different kinds of AD biomarkers. We included publications that solely discuss amyloid-β PET-imaging as a biomarker for AD [16–18 , 44–48], a selection of other biomarkers [34, 35], as well as publications concerning AD biomarkers in general [1 , 49]. Secondly, the setting in which AD biomarker testing is being considered will also impact the weight of the given arguments, though some of the included publications did not address one specific setting [1 , 47]. Other publications considered AD biomarker testing in the context of screening [27 , 49], providing care in clinical practice [13 , 45–48], or for research purposes [15 , 42]. Lastly, the population for which AD biomarkers tests are being considered should be taken into account. We extracted arguments that applied to cognitively healthy individuals, people with subjective cognitive decline or mild cognitive impairment. Arguments in favor of or against AD biomarker use that specifically apply to patients with a clinical diagnosis of AD dementia were not included because it is outside of the scope of our research question. Characteristics of the included studies can be found in Supplementary Table 2.

Qualitative synthesis of arguments

The arguments in favor of and against AD biomarker testing in people with no or mild cognitive impairment mentioned in the included publications are presented in the categories and subcategories of arguments that resulted from the qualitative data-synthesis. Tables 1 and 2 provide an overview of these results. Cited arguments are discussed in the referenced publications, but do not necessarily represent the position of the referenced author(s) in this debate. While cited arguments may refer to a specific context of biomarker testing, that is not addressed in the results.

Table 1

Arguments in favor of AD biomarker testing presented in the reviewed literature

Categories of reasons in favor of AD biomarker testing	Subcategories	Articles
1. AD biomarker testing and disclosure of its results offers benefits for research practices	a. Enrichment of clinical trials	[1 , 48]
	b. Cost reduction when using AD biomarker results as surrogate outcomes
	c. Disclosure of AD biomarker status in a trial accelerates recruitment
2. AD biomarker results offer an increased diagnostic certainty	a. AD biomarker testing increases diagnostic certainty	[1 , 48]
	b. Misdiagnosis could be avoided by using AD biomarkers
	c. AD biomarker results could contribute to the patient and caregivers acknowledging the diagnosis
3. AD biomarker testing offers future planning possibilities	a. Practical planning possibilities	[1 , 44]
	b. The possibility to reflect on the impact of an early AD diagnosis
4. AD biomarker testing offers possibilities to improve one’s health and well-being	a. Greater access to healthcare	[1 , 46–48
	b. Possibility to anticipate on future healthcare needs	]
	c. Encouragement to start a healthier lifestyle
	d. Intervention in AD pathology will have most chance of therapeutic success at an early stage
	e. Psychological benefits of an early diagnosis
5. People have a right to know their AD biomarker status	a. Right to access individual research data	[13 , 44]
	b. Right to receive a prodromal diagnosis of AD in clinical practice
6. The harmful consequences of AD biomarker testing are over-estimated	a. Disclosure of AD biomarker status will not result in severe psychological harm	[12 , 36]
	b. Potential diagnostic misunderstanding can be avoided
	c. There will be no increase in healthcare costs

Abbreviations: AD, Alzheimer’s disease.

Table 2

Arguments against AD biomarker testing presented in the reviewed literature

Categories of reasons against AD biomarker testing	Subcategories	Articles
1. AD biomarker testing offers no (additional) clinical utility	a. There is no disease-modifying treatment or prevention option available	[11 , 44]
	b. Biomarker results offer no (new) future planning possibilities
	c. Neuropsychological test results are more robust predictors for progression from MCI to dementia
2. AD biomarker information offers limited diagnostic or prognostic value	a. AD biomarkers are currently not sufficiently validated	[16 , 49]
	b. Standardization and reproducibility of biomarker tests are lacking
	c. AD biomarkers offer only partial insight into the multifactorial cause of the disease
3. AD biomarker testing will contribute to the medicalization of healthy people	a. AD biomarker testing in the context of early AD diagnosis raises fundamental questions on the concept of disease	[25 , 49]
	b. AD biomarker testing and early diagnosis will lead to overdiagnosis and medicalization of memory loss
4. The disclosure of AD biomarker status will cause psychological harm	a. Disclosure of biomarker status may cause stress and fear for the future	[1 , 46]
	b. Psychological harm may be worse in people with full insight in their situation
5. Biomarker positive individuals might experience stigma and discrimination	a. The stigma tied to AD may extrapolate to biomarker positive individuals	[1 , 49]
	b. Biomarker testing may result in discrimination or privacy or confidentiality violation
6. Lack of availability and affordability of AD biomarker testing	a. Applying AD biomarker tests in clinical practice would require an enormous financial investment	[26 , 49]
	b. AD biomarker testing will increase healthcare disparities when it becomes commercially available
7. The right not to know your AD biomarker status		[15]
8. AD biomarker testing itself is burdensome	a. Risk and burden of AD biomarker testing itself	[15 , 49]

Abbreviations: AD, Alzheimer’s disease; MCI, mild cognitive impairment.

Arguments in favor of AD biomarker testing

AD biomarker testing and disclosure of results offers benefits for research practices

In research efforts to arrest the pathophysiological process of AD before cognitive functioning starts to decline, AD biomarkers play a crucial role to identify only those individuals that are in the ‘preclinical’ or ‘prodromal’ stage of the disease [1 , 48]. Besides this so called ‘enrichment’ of clinical trials, AD biomarkers may also contribute to the efficiency of AD prevention research by serving as surrogate outcome measures before clinical outcome measures are available. This could reduce the length and cost of clinical prevention trials [31 –35]. To the extent that people are more likely to volunteer to participate in research if biomarker results are provided, disclosing results may also expedite the conduct of research [33].

AD biomarker testing offers an increase in diagnostic certainty

Insight into the presence of AD pathology is taken as improving the accuracy of AD diagnosis, as reflected in the incorporation of biomarkers in AD disease criteria [1 , 44]. AD biomarker testing may thereby improve a physician’s diagnostic confidence [48] and prevent misdiagnosis of AD [11 , 48], but it also has an intrinsic value, which may be especially appreciated by patients [16 , 48]. Informing people of their AD biomarker results may also motivate them, and possibly their caregivers, to accept the diagnosis and to act upon recommendations of the physician [40, 48].

AD biomarker testing offers future planning possibilities

The possibility to anticipate future cognitive decline when being (relatively) cognitively healthy is considered one of the major benefits of AD biomarker testing in an early or ‘preclinical’ stage of disease [12 , 44]. Planning possibilities mentioned are of practical nature (e.g. to lay down healthcare directives or to purchase long-term care insurance) [25 , 36] or to increase personal well-being, as to retire early or go traveling [11 , 24].

AD biomarker testing offers possibilities to improve health and well-being

Biomarker testing may offer opportunities to improve one’s health and well-being in roughly five different ways: First, positive test results are presumed to lead to greater access to help within the healthcare system and within communities [12 , 40]. Second, a physician could anticipate future healthcare needs based on biomarker results, for instance by starting better monitoring of patient adherence [26]. Third, informing people of their biomarker status may encourage them to undertake beneficial changes in lifestyle [18 , 44]. Fourth, future disease-modifying therapies are expected to show greater effect when applied in people with AD pathology before the onset of clinical decline [1 , 50]. AD biomarkers would facilitate the identification of these individuals, who may already benefit from participating in clinical trials where disease-modifying drugs are tested [12 , 34–37]. And finally, AD biomarker results may provide sufficient confidence in the diagnosis to reduce feelings of anxiety in people with memory impairment and their families, and to provide social exoneration [12 , 46–48].

People have a right to know their AD biomarker status

It has been argued that the decision to learn one’s biomarker status should be a matter of personal choice for individuals, not of doctors [13 , 44], out of respect for an individual as an autonomous subject [33, 43]. In the context of research, participants can be granted the right to decide themselves whether they want or do not want to receive their own research results [16]. In a clinical setting, as well, it is argued that people should have the possibility to decide themselves whether they want to receive a diagnosis of prodromal AD [13].

The suggested harmful consequences of AD biomarker testing will be modest

Some of the arguments against AD biomarker testing have been refuted in favor of AD biomarker testing. The concern that disclosing biomarker results would result is psychological harm is challenged by research showing that neither genetic AD risk disclosure nor disclosure of a genetic predisposition for other severe neurological diseases results in catastrophic reactions [15 , 33]. Potential misunderstanding of biomarker results may be prevented by proper guidance [16, 17]. And lastly, the implementation of AD biomarker testing in clinical care may, in the end, result in a reduction in healthcare costs, instead of a growth, because it will facilitate research that is expected to enable people to live independently for a longer time [12, 36].

Arguments against AD biomarker testing

AD biomarker testing currently offers no (additional) clinical utility

The lack of a disease-modifying treatment for AD is one of the key arguments mentioned in the literature against AD biomarker testing and disclosure of its results [11 , 43]. Claims on clinical utility which rest upon the possibility of future planning are disputed by the argument that future planning and lifestyle improvements should be part of everyone’s planning, regardless their biomarker status [32]. On top of that, neuropsychological test results may be more robust predictors for progression from mild cognitive impairment to AD dementia than most AD biomarkers [26, 32].

AD biomarker information offers limited diagnostic or prognostic value

Some recommend against the use of AD biomarkers in individuals with no or mild cognitive impairment because a significant part of the amyloid-β positive individuals will not develop dementia [30]. The literature shows disagreement on whether AD biomarkers— both CSF and amyloid-β PET— in people with no or mild cognitive impairment are sufficiently validated or even reliable outside of a narrow research setting [31, 43]. Some argue that more research, especially in a heterogeneous population, is necessary for further validation [26 , 47]. Others stress a need to improve standardization and reproducibility of AD biomarkers across centers [26 , 49]. A deeper concern, however, follows from the poorly understood relation between the presumed pathophysiological process that is made insightful by AD biomarkers and cognitive functioning [31 , 50], which undermines the relevance of measuring AD biomarker levels altogether [16 , 50].

AD biomarker testing will contribute to the medicalization of healthy people

The proposal to expand the AD label to cognitively healthy individuals based on their AD biomarker results has raised fundamental questions regarding the boundaries between ‘normal ageing’ and disease [35 , 46]. It has been suggested that it may lead to a medicalization of memory loss [25], to harming people by over-diagnosis and over-treatment [29] and a tendency to view aging negatively as a process of continuous decline [49].

Disclosure of AD biomarker information may result in psychological harm

The disclosure of AD biomarker results, possibly accompanied by an early AD diagnosis, is suggested to lead to psychological stress in patients and their families, ranging from a cloud of uncertainty [42], to anxiety, despair, or depression or even pre-emptive suicide [1 , 44–46]. Psychological harm due to an early diagnosis may be worse in people with full insight in their situation [38].

Stigma and discrimination due to AD label

The expansion of the AD label to people with no or mild cognitive impairment may cause an extrapolating of the stigma tied to AD dementia to biomarker positive individuals [30]. The negative image people may have of AD could have an impact on someone’s position in society, relationships with colleagues, partners and friends and on how people view themselves; a phenomenon known as self-stigma [12, 30]. The shift toward an earlier diagnosis of AD may also carry the risk of discrimination, especially when data is accessed by third parties such as insurance and pharmaceutical companies. In the latter case, people with an early AD diagnosis may be discriminated in the context of legal status, job positions, or access to health insurance [1 , 49].

Limited availability and affordability of AD biomarkers

Employing AD biomarkers in clinical practice would require an enormous financial investment of limited healthcare resources [26 , 49], given the size of the potential target population, the costs of testing and the limited availability of the technique and experts to perform them. Concerns about the availability and affordability of AD biomarker use are important globally, but particularly pertinent in developing countries [26]. Should AD biomarkers become commercially available, it may increase healthcare disparities [35].

The right not to know your AD biomarker status

Similar to the widely acknowledged ‘right to know’ one’s medical information, the ‘right not to know’ one’s medical information is widely defended, even when the data is medically actionable. This argument is brought forward in the context of neurodegenerative prevention trials against an enrollment strategy that imposes disclosure of biomarker status to research participants [15].

Burden and risk of AD biomarker testing itself

Lumbar punctures are uncomfortable, they carry the risks of an invasive procedure and its cultural acceptability varies between countries [26, 49]. Amyloid-β PET imaging involves being injected with a radioactive ligand and may therefore cause redness, itching, or pain from an allergic reaction to the injection site [17]. Even though the risks are small and the burden can be considered low, the cumulative burden of discomfort and risks that accompany a lumbar puncture and amyloid-β PET imaging [17 , 49] could be substantial [15].

DISCUSSION

This systematic review shows that there is a substantial and widespread debate on the desirability of AD biomarker testing in individuals with no or mild cognitive impairment. Although some of the arguments are presented as purely biological or technical issues, all of them carry a moral value, meaning that all arguments imply that a certain value should be protected or pursued. That is not to deny that scientific evidence has a crucial role in the moral weighing of the listed arguments. The suggestion that more research is needed for further validation of biomarkers, for example, is based on the current stance of scientific research but departs from the normative assumption that medical testing can only be justified when its outcomes are reliable.

Four basic principles of medical ethics

Four basic principles of medical ethics— non-maleficence, beneficence, respect for autonomy, and justice [20]— could be used to identify moral values that underlie the arguments that are brought forward in this debate and to structure and weigh them accordingly in deciding upon the ethical desirability of AD biomarker testing in people with no or mild cognitive impairment [20]. This normative framework elucidates which values are at stake and which aspects or consequences of AD biomarker testing warrant further attention in the future. In doing so, it facilitates a critical appraisal of the given arguments.

Non-maleficence

The principle of non-maleficence entails the physician’s moral obligation not to harm the patient. It is weighed against the principle of beneficence to strive for a positive balance of goods over inflicted harms. Arguments that appeal to the principle of non-maleficence in this debate relate to the risk of stigma, discrimination, over-diagnosis, and psychological harm in people with positive biomarker results, and the burden and risk of testing itself. These disadvantages of AD biomarker testing are especially worrying in those that will eventually never develop AD dementia.

Given the stigmatization of people with a diagnosis of mild cognitive impairment [51] or AD [52], it is likely that people with AD pathology, as indicated by positive AD biomarker results, may face stigmatization as well [30]. Presumably, this stigmatization is tied to the expected prognosis of cognitive decline rather than the AD label itself [53]. In countries where no laws to protect people with positive biomarker status exist, possible discrimination, violation of privacy, and exploitation may accompany biomarker testing [36].

Little is known about the psychological impact of disclosing AD biomarker results, other than genetic risk status, to cognitively healthy research participants [19, 54]. Disclosing an increased genetic risk to develop AD did not show an effect on levels of anxiety or depression in first-degree family members of AD patients [55], but did result in a worsening of cognitive functioning [56]. It could be argued, however, that being a close family member of someone with a neurodegenerative disease such as AD, one could already be wary of carrying a genetic risk factor but not of having brain abnormalities that indicate an increased risk for AD. Therefore, caution is warranted when assumptions about the impact of biomarker risk disclosure are solely based on research investigating the impact of genetic risk disclosure.

Being confronted with positive AD biomarker test results and no disease-modifying treatment available, people may consider treatments which are advertised without a solid evidence-base. Online marketing strategies for these treatments can take advantage of this attitude by fostering false hopes and spreading misinformation [57].

Research that has been done on the impact of disclosing amyloid-β PET imaging results showed that even though some amnestic mild cognitive impairment patients became more concerned about future cognitive decline and a potentially more patronizing attitude by relatives, others reported a positive impact on social relationships. Furthermore, they appreciated having more information about their health status and perceived increased planning possibilities [54]. Examples of future planning mostly address cases of positive biomarker results. Whether those receiving negative biomarker results abandon plans to improve their lifestyle or to set up advanced directives is currently unknown. Although questions remained on the exact meaning of a positive amyloid-β PET imaging result, the information received during its disclosure was accurately recalled by cognitively healthy research participants [58] and those with amnestic mild cognitive impairment [54]. Yet, it is currently insufficiently clear how the given arguments concerning potential psychological, social or behavioral benefits or harms of disclosing AD biomarker status should be weighed. Further research on the impact of AD biomarker disclosure would provide valuable input to advance the current debate.

Although partly dependent on individual and cultural experience [26], the burden and risks of AD biomarker testing appear to be relatively low. Still, in order for testing to be proportional, it should be outweighed by the potential benefits of testing.

Beneficence

The principle of beneficence refers to a moral obligation to act for the benefit of others, being subjects, patients, or society. These benefits should always be weighed against risks of harms, as described above. The benefits of testing AD biomarkers rely on an increased diagnostic certainty in people with cognitive impairment, better future planning possibilities, and improved health and well-being of those who receive the results. Arguments concerning benefits for research practices are indirectly related to the principle of beneficence. They concern potential benefits for future patients, rather than for the individual participant.

Biomarker testing may provide solace to those in search of an explanation for their cognitive decline or, similarly, as a predictor of future cognitive decline to those in search of more certainty about their future health. Whether AD biomarkers can indeed provide such diagnostic or prognostic assurance depends on the predictive value of the test. This, in turn, depends on the AD disease criteria being used, which biomarker is being tested and on the cognitive health and age of the tested individual [59, 60]. With increasing age, for example, the association between levels of amyloid-β and dementia weakens [61]. The added value of CSF markers of amyloid-β for the prediction of dementia, over a memory test, is therefore especially limited for patients older than 75 years [62]. Even if it lacks certainty, people appear to appreciate risk information [63]. However, neuropsychological testing, being a less invasive alternative, may provide comparable prognostic information when aiming to comply to this need [62]. According to the ethical principle of subsidiarity, according to which one should always choose the least invasive or burdensome alternative, this could therefore be preferable.

Whether AD biomarker information, as a predictor for the development of dementia, could inform future planning depends on the reliability and time-indication of such information. These factors depend, as described above, on the characteristics of the individual. People with no cognitive impairment and positive biomarker results may take up to fifteen years before they develop dementia— if they develop dementia at all [1]. Given this uncertainty and the lack of an AD prevention option, it can be questioned whether AD biomarkers can offer any new planning possibilities at all. Recommendations suggested in the literature, such as setting up healthcare directives and taking up a healthier lifestyle, do already apply to the general population [26, 29]. Besides, the personality changes that accompany AD dementia may create a great psychological distance between the person that decides on future plans and the person whom this decision ultimately affects [40]. People may, however, feel increasingly motivated to think about their future when confronted with positive biomarker results [63] and develop coping strategies to prepare for future cognitive decline.

Although there are no disease-modifying treatments or prevention options available, psycho-social interventions may improve quality of life and coping strategies of individuals, and possibly their surrounding family and friends, who feel insecure about their cognitive performance. Offering adequate care should, however, be driven by symptoms or a need of care, not by biological parameters. Therefore, increased access to healthcare is, contrary to what has been suggested in the literature [12 , 40], not a valid reason to pursue biomarker testing. In fact, large scale deployment of AD biomarker testing in clinical practice will only further exhaust existing healthcare facilities and thereby reduce its accessibility. The effectiveness of existing psycho-social or biomedical interventions is, after all, in no way dependent on AD biomarker results.

In clinical trials aimed at preventing cognitive decline by increasing clearance of amyloid-β in the brain (e.g., the A4 study [64]), AD biomarker testing can be valuable for screening purposes. It averts needlessly exposing individuals with normal levels of amyloid-β to the burdens and risks associated with trial participation. Regardless of cognitive benefit, individuals may perceive trial participation in itself as beneficial.

Respect for autonomy

The principle of respect for autonomy stresses the importance of individual freedom and choice. It involves fostering of autonomous decision-making and allaying fears or conditions that might disrupt this process [20]. Individual’s right to know, or not to know, their AD risk status and to plan their futures accordingly can be supported by the principle of respect for autonomy. Respect for autonomy does not, however, entail a positive claim on others, such as demanding an amyloid-β PET scan [17, 18].

The right to know one’s biomarker status is diminished, however, by the absence of treatment options and limited prognostic value of the test results. Nonetheless, in a research setting one can claim a right to disclosure of biomarker status because of the autonomy-based right to access one’s own data. For CSF results specifically, the National Academies of Sciences report on Return of Individual-Specific Research Results Generated in Research Laboratories applies. This consensus report recommends, in line with our view, to assess the risks and benefits of results disclosure by weighing relevant ethical principles on a study-by-study basis [65]. In a clinical setting, physicians should be confident that the tests they offer will benefit and not harm people; according to the principle of respect for autonomy, the individual involved ought to have the opportunity to make their own assessment to decide whether to accept an offer for AD biomarkers testing or not.

Justice

The principle of justice represents, at minimum, the moral aim to ‘treat like cases alike’ [20]. In practice, this means that discrimination should have no role in determining healthcare access. The listed arguments that appeal to the ethical principle of justice concern the high costs and limited availability of AD biomarker testing.

In the light of scarce healthcare resources, even in most high-income countries, it is debatable whether AD biomarker testing in people with no or mild cognitive impairment should be given priority when there is no proven effective treatment or prevention option available. Until then, clinical implementation of AD biomarker testing can only be expected to significantly raise healthcare costs considering the costs of AD biomarker testing and the size of the potential eligible population. The central role for assumptions about a hypothetical disease-modifying therapy for AD in models that have estimated the cost-effectiveness of AD biomarker testing explain why others came to a different conclusion [12]. Concerns on the issue of justice apply to commercial testing as well, since not everyone would be able to afford these expenses.

Strengths and limitations of the systematic review

This systematic review gives a comprehensive overview of arguments in favor of and against AD biomarker testing in people with no or mild cognitive impairment as debated in the scientific and clinical literature so far, but grey literature or publications on related policy and legal issues are not included. The included literature varies from a narrow research setting to screening for AD biomarkers on a wide population. This heterogeneity in studies might complicate the interpretation of the aggregated results—especially since the arguments are taken out of the context of the cited publication— but we feel it is essential to bring all these arguments together to provide an overview of the current debate and the considerations that may be relevant when determining a position upon the desirability of AD biomarker testing.

The framework of the four ethical principles does enable a systematic way to consider the needs of research subjects, clinical patients, and society when weighing the given arguments within a specific context of AD biomarker testing. Specific contextual values of moral relevance that should then be taken into account are, for example, characteristics of the individual involved, such as age and physical health, or the clinical or research setting in which AD biomarker testing is considered. Depending on the specific context, additional or other ethical principles may apply, especially in non-western settings.

Weighing the arguments and suggestions for future debate

Although the included literature reflects an overall tendency to presume a future move of AD biomarkers from research to clinical practice, with exception of some more cautious authors [31], it remains unclear whether biomarker testing in people with no or mild cognitive impairment is desirable from an ethical perspective. We believe that a focused debate on the ethical desirability of this movement, in which the moral values that underlie current reasoning are acknowledged, is timely and is a crucial step to ensure a well-considered and appropriate practice of AD biomarker testing in the future.

We identified a number of issues that in our view merit greater attention in the future continuing of this debate. First, the distinction between biomarker testing in a research or a clinical setting is highly relevant from an ethical perspective but not always distinguished in the literature. Contrary to physicians who have the moral duty to provide care to the best of their abilities, providing care is not the main purpose of researchers. This means that there is more room for uncertainty regarding the validity of biomarker results. Furthermore, test results may have an indispensable value for research while having only a limited impact on clinical management of patients. Though research and clinical care may be intertwined in practice, we emphasize the importance of making clearer distinctions between practices in the continuing of this debate because the advantages of biomarker testing are less clear in a clinical compared to a research setting.

Second, arguments should be weighed differently depending on the age of the individual involved and whether someone has objective cognitive impairment or is cognitively healthy. These factors affect the prognostic and diagnostic value of biomarker results [60], which in turn impacts the benefit of future planning possibilities and the right to know one’s biomarker status.

Third, the possibility that a disease-modifying treatment will not be found, will have little effect or will only elongate a stage of severe illness, has hardly been seriously considered in the literature so far. It is inevitable that the current debate is oriented toward and shaped by expectations for the future. Differentiating to a further extent between what we expect and hope from AD research and where we currently stand will, however, purify the debate and clarify in which scenario we would consider clinical biomarker testing to be appropriate. Furthermore, it would leave room to anticipate on less favorable future scenarios.

Fourth, speculations regarding the psychological, emotional, and social impact of biomarker disclosure, the possibility to change one’s lifestyle, or to plan ahead of cognitive decline need further empirical insight to properly weigh these arguments both in favor of and against biomarker testing.

Fifth, general measures to encourage people to settle healthcare directives, take up a healthy lifestyle, or to cope with the uncertainty about their future health may be less costly and invasive than biomarker testing, and equally or more effective. When these benefits of biomarker testing are brought forward in the literature, it should be clarified how biomarker testing itself is a prerequisite for this result.

And finally, it has been suggested that conflicts of interest and tensions between academia and the private sector may exist and could rise in the future [31 –35]. These should be shunned to avoid a biased perspective on study results, economic feasibility, and appropriateness of testing.

CONCLUSIONS

Several years after the introduction of biomarker-based disease criteria for AD, a broad range of arguments in favor of and against AD biomarker testing in people with no or mild cognitive impairment have been brought forward, as presented in this systematic review. We provide a critical review of these arguments, an insight into their underlying moral values and show how their weight depends on the specific context in which biomarker testing is considered. When reflecting upon the desirability of biomarker testing in a specific context, this systematic review offers input and guidance for weighing the given arguments. In the continuation of this debate, we think that differences between what we hope or expect from future research and where we currently stand and alternative, less invasive, ways to obtain the benefits of AD biomarker testing merit more attention.

Footnotes

ACKNOWLEDGMENTS

This research is supported by a grant from The Netherlands Organization for Health Research and Development (grant number: 731010012) and by support from the EU/EFPIA Innovative Medicines Initiative Joint Undertaking EPAD grant agreement n° 115736. We thank Wichor Bramer, information specialist of the library of the Erasmus Medical Center for his contribution to the search strategy. We are also very grateful to Richard Milne, from the Cambridge institute of Public Health, for his insight and expertise that greatly assisted the writings of this paper and for his critical revision of the manuscript.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Dubois

, Hampel

, Feldman

, Scheltens

, Aisen

, Andrieu

, Bakardjian

, Benali

, Bertram

, Blennow

, Broich

, Cavedo

, Crutch

, Dartigues

, Duyck-aerts

, Epelbaum

, Frisoni

, Gauthier

, Genthon

, Gouw

, Habert

, Holtzman

, Kivipelto

, Lista

, Molinuevo

, O’Bryant

, Rabinovici

, Rowe

, Sal-loway

, Schneider

, Sperling

, Teichmann

, Carrillo

, Cummings

, Jack

(2016) Preclinical Alzheimer’s disease: Definition, natural history, and diagnostic criteria. Alzheimers Dement 12, 292–323.

Dubois

, Feldman

, Jacova

, Hampel

, Molinuevo

, Blennow

, DeKosky

, Gauthier

, Selkoe

, Bate-man

, Cappa

, Crutch

, Engelborghs

, Frisoni

, Fox

, Galasko

, Habert

, Jicha

, Nordberg

, Pasquier

, Rabinovici

, Robert

, Rowe

, Salloway

, Sarazin

, Epelbaum

, de Souza

, Vellas

, Visser

, Schneider

, Stern

, Scheltens

, Cummings

(2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria. Lancet Neurol 13, 614–629.

Dubois

, Feldman

, Jacova

, Dekosky

, Barberger-Gateau

, Cummings

, Delacourte

, Galasko

, Gauthier

, Jicha

, Meguro

, O’Brien

, Pasquier

, Robert

, Rossor

, Salloway

, Stern

, Visser

, Scheltens

(2007) Research criteria for the diagnosis of Alzheimer’s disease: Revising the NINCDS-ADRDA criteria. Lancet Neurol 6, 734–746.

Sperling

, Aisen

, Beckett

, Bennett

, Craft

, Fagan

, Iwatsubo

, JackCR, Jr., Kaye

, Montine

, Park

, Reiman

, Rowe

, Siemers

, Stern

, Yaffe

, Carrillo

, Thies

, Morrison-Bogorad

, Wag-ster

, Phelps

(2011) Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 280–292.

Albert

, DeKosky

, Dickson

, Dubois

, Feldman

, Fox

, Gamst

, Holtzman

, Jagust

, Petersen

, Snyder

, Carrillo

, Thies

, Phelps

(2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270–279.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

, Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

Jack

, Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

, Elliott

, Masliah

, Ryan

, Silverberg

(2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

Kunneman

, Pel-Littel

, Bouwman

, Gillissen

, Schoonenboom

, Claus

, van der Flier

, Smets

(2017) Patients’ and caregivers’ views on conversations and shared decision making in diagnostic testing for Alzheimer’s disease: The ABIDE project. Alzheimers Dement(NY) 3,314–322.

Kunneman

, Smets

, Bouwman

, Schoonen-boom

, Zwan

, Pel-Littel

, van der Flier

(2017) Clinicians’ views on conversations and shared decision making in diagnostic testing for Alzheimer’s disease: The ABIDE project. Alzheimers Dement (N Y) 3, 305–313.

10.

Somers

, Struyfs

, Goossens

, Niemantsverdriet

, Luyckx

, de Roeck

, de Vil

, Cras

, Martin

J-J

(2016) A decade of cerebrospinal fluid biomarkers for Alzheimer’s disease in Belgium. J Alzheimers Dis 54, 383–395.

11.

Prvulovic

, Hampel

(2011) Ethical considerations of biomarker use in neurodegenerative diseases-A case study of Alzheimer’s disease. Prog Neurobiol 95, 517–519.

12.

Dubois

, Padovani

, Scheltens

, Rossi

, Dell’Agnello

(2016) Timely diagnosis for Alzheimer’s disease: A literature review on benefits and challenges. J Alzheimers Dis 49, 617–631.

13.

Molinuevo

, Rami

(2013) Applying the IWG research criteria in clinical practice. Feasibility and ethical issues. Med Clin North Am 97, 477–484.

14.

MetLife-Foundation (2011) What America thinks. The MetLife Foundation Alzheimer’s survey, New York. https://www.metlife.com/assets/cao/foundation/alzheimers-2011.pdf, Accessed 31 October 2017.

15.

Kim

SYH

, Karlawish

, Berkman

(2015) Ethics of genetic and biomarker test disclosures in neurodegenerative disease prevention trials. Neurology 84, 1488–1494.

16.

Lingler

, Klunk

(2013) Disclosure of amyloid imaging results to research participants: Has the time come? Alzheimers Dement 9, 741–744.

17.

Baum

(2016) Patient requests for off-label bioprediction of dementia. Camb Q Healthc Ethics 25, 686–690.

18.

Snider

, Buckles

(2013) “Will I get Alzheimer disease?” When cognitively normal patients ask to be tested for Alzheimer disease. Continuum (Minneap Minn) 19, 470–474.

19.

Bemelmans

, Tromp

, Bunnik

, Milne

, Badger

, Brayne

, Schermer

, Richard

(2016) Psychological, behavioral and social effects of disclosing Alzheimer’s disease biomarkers to research participants: A systematic review. Alzheimers Res Ther 8, 46.

20.

Beauchamp

, Childress

(2001) Principles of biomedical ethics, Oxford University Press, New York.

21.

Strech

, Sofaer

(2012) How to write a systematic review of reasons. J Med Ethics 38, 121–126.

22.

Blennow

, Hampel

, Weiner

, Zetterberg

(2010) Cerebrospinal fluid and plasma biomarkers in Alzheimer disease. Nat Rev Neurol 6, 131–144.

23.

McDougall

(2015) Reviewing literature in bioethics research: Increasing rigour in non-systematic reviews. Bioethics 29, 523–528.

24.

Arias

, Cummings

, Grant

, Ford

(2015) Stakeholders’ perspectives on preclinical testing for Alzheimer’s Disease. J Clin Ethics 26, 297–305.

25.

Boenink

, Cuijpers

, van der Laan

, Van Lente

, Moors

(2011) Assessing the sociocultural impacts of emerging molecular technologies for the early diagnosis of Alzheimer’s disease. Int J Alzheimers Dis 2011, 184298.

26.

O’Brien

, Dubois

, Gauthier

, Cummings

, Chiu, HF, Brodaty

(2013) For debate: Is very early diagnosis of AD using the new criteria beneficial for patients. Int Psychogeriatr 25, 177–181.

27.

Calza

, Beltrami

, Gagliardi

, Ghidoni

, Marcello

, Rossini-Favretti

, Tamburini

(2015) Should we screen for cognitive decline and dementia? Maturitas 82, 28–35.

28.

Comer

(2008) Early Alzheimer’s disease diagnostics: Wait! Wait! Don’t tell me! Alzheimers Dement 4, 368–370.

29.

Le Couteur

, Doust

, Creasey

, Brayne

(2013) Political drive to screen for pre-dementia: Not evidence based and ignores the harms of diagnosis. BMJ 347, f5125.

30.

Gauthier

, Leuzy

, Racine

, Rosa-Neto

(2013) Diagnosis and management of Alzheimer’s disease: Past, present and future ethical issues. Prog Neurobiol 110, 102–113.

31.

George

, Qualls

, Camp

, Whitehouse

(2013) Renovating Alzheimer’s: “constructive” reflections on the new clinical and research diagnostic guidelines. Gerontologist 53, 378–387.

32.

Gordon

(2013) Identification of potential or preclinical cognitive impairment and the implications of sophisticated screening with biomarkers and cognitive testing: Does it really matter? Biomed Res Int 2013, 1–5.

33.

Grill

, Johnson

, Burns

(2013) Should we disclose amyloid imaging results to cognitively normal individuals? Neurodegener Dis Manag 3, 43–51.

34.

Holt

(2011) Timely diagnosis and disclosure of alzheimer disease gives patients opportunities to make choices. South Med J 104, 779–780.

35.

Illes

, Rosen

, Greicius

, Racine

(2007) Prospects for prediction: Ethics analysis of neuroimaging in Alzheimer’s disease. Ann N Y Acad Sci 1097, 278–295.

36.

Karlawish

(2011) Addressing the ethical, policy, and social challenges of preclinical Alzheimer disease. Neurology 77, 1487–1493.

37.

Kutschenko

(2011) Diagnostic misconceptions? A closer look at clinical research on Alzheimer’s disease. J Med Ethics 38, 57–59.

38.

Leuzy

, Gauthier

(2012) Ethical issues in Alzheimer’s disease: An overview. Expert Rev Neurother 12, 555–567.

39.

Leuzy

, Zimmer

, Heurling

, Rosa-Neto

, Gauthier

(2014) Use of amyloid PET across the spectrum of Alzheimer’s disease: Clinical utility and associated ethical issues. Amyloid 21, 143–148.

40.

Mattsson

, Brax

, Zetterberg

(2010) To know or not to know: Ethical issues related to early diagnosis of Alzheimer’s disease. Int J Alzheimers Dis 2010, 841–941.

41.

Molin

, Rockwood

(2016) The new criteria for Alzheimer’s disease - implications for geriatricians. Can Geriatr J 19, 66–73.

42.

Molinuevo

, Cami

, Carne

, Carrillo

, Georges

, Isaac

, Khachaturian

, Kim

SYH

, Morris

, Pasquier

, Ritchie

, Sperling

, Karlawish

(2016) Ethical challenges in preclinical Alzheimer’s disease observational studies and trials: Results of the Barcelona summit. Alzheimers Dement 12, 614–622.

43.

Porteri

, Frisoni

(2014) Biomarker-based diagnosis of mild cognitive impairment due to Alzheimer’s disease: How and what to tell. A kickstart to an ethical discussion. Front Aging Neurosci 6, 41.

44.

Roberts

, Dunn

, Rabinovici

(2013) Amyloid imaging, risk disclosure and Alzheimer’s disease: Ethical and practical issues. Neurodegener Dis Manag 3, 219–229.

45.

Rosen

, Bokde

ALW

, Pearl

, Yesavage

(2002) Ethical, and practical issues in applying functional imaging to the clinical management of Alzheimer’s disease. Brain Cogn 50, 498–519.

46.

Schicktanz

, Schweda

, Ballenger

, Fox

, Halpern

, Kramer

, Micco

, Post

, Thompson

, Knight

, Jagust

(2014) Before it is too late: Professional responsibilities in late-onset Alzheimer’s research and pre-symptomatic prediction. Front Hum Neurosci 8, 921.

47.

Sperling

, Karlawish

, Johnson

(2013) Preclinical Alzheimer disease—the challenges ahead. Nat Rev Neurol 9, 54.

48.

Witte

, Foster

, Fleisher

, Williams

, Quaid

, Wasserman

, Hunt

, Roberts

, Rabinovici

, Levenson

, Hake

, Hunter

, Van Campen

, Pontecorvo

, Hochstetler

, Tabas

, Trzepacz

(2015) Clinical use of amyloid-positron emission tomography neuroimaging: Practical and bioethical considerations. Alzheimers Dement (Amst) 1, 358–367.

49.

Wright

, Hall

, Matthews

, Brayne

(2009) Biomarkers, dementia, and public health. Ann N Y Acad Sci 1180, 11–19.

50.

Albert

, McKhann

(2013) Neuroethical issues in early detection of Alzheimer’s disease. In Oxford Handbook of Neuroethics, Illes

Sahakian

, eds. Oxford University Press, Oxford, pp. 553–561.

51.

Beard

, Neary

(2013) Making sense of nonsense: Experiences of mild cognitive impairment. Sociol Health Illn 35, 130–146.

52.

Batsch

, Mittelman

, Alzheimer’s Disease International (2012) World Alzheimer Report 2012: Overcoming the Stigma of Dementia. Alzheimer’s Disease International, London.

53.

Johnson

, Harkins

, Cary

, Sankar

, Karlawish

(2015) The relative contributions of disease label and disease prognosis to Alzheimer’s stigma: A vignette-based experiment. Soc Sci Med 143, 117–127.

54.

Vanderschaeghe

, Schaeverbeke

, Bruffaerts

, Van-denberghe

, Dierickx

(2017) Amnestic MCI patients’ experiences after disclosure of their amyloid PET result in a research context. Alzheimers Res Ther 9, 92.

55.

Green

, Roberts

, Cupples

, Relkin

, Whitehouse

, Brown

, Eckert

, Butson

, Sadovnick

, Quaid

, Chen

, Cook-Deegan

, Farrer

(2009) Disclosure of APOE genotype for risk of Alzheimer’s disease. N Engl J Med 361, 245–254.

56.

Lineweaver

, Bondi

, Galasko

, Salmon

(2014) Effect of knowledge of APOE genotype on subjective and objective memory performance in healthy older adults. Am J Psychiatry 171, 201–208.

57.

Racine

, Forlini

, Aspler

, Chandler

(2016) Complementary and alternative medicine in the context of earlier diagnoses of Alzheimer’s disease: Opening the conversation to prepare ethical responses. J Alzheimers Dis 51, 1–9.

58.

Mozersky

, Sankar

, Harkins

, Hachey

, Karlawish

(2018) Comprehension of an elevated amyloid positron emission tomography biomarker result by cognitively normal older adults. JAMA Neurol 75, 44–50.

59.

Middleton

, Grinberg

, Miller

, Kawas

, Yaffe

(2011) Neuropathologic features associated with Alzheimer disease diagnosis: Age matters. Neurology 77, 1737–1744.

60.

Johnson

, Minoshima

, Bohnen

, Donohoe

, Foster

, Herscovitch

, Karlawish

, Rowe

, Carrillo

, Hartley

, Hedrick

, Pappas

, Thies

(2013) Appropriate use criteria for amyloid PET: A report of the Amyloid Imaging Task Force, the Society of Nuclear Medicine and Molecular Imaging, and the Alzheimer’s Association. Alzheimers Dement 9, e1–16.

61.

Richard

, Schmand

, Eikelenboom

, Westendorp

, Van Gool

(2012) The Alzheimer myth and biomarker research in dementia. J Alzheimers Dis 31(Suppl 3), S203–209.

62.

Richard

, Schmand

, Eikelenboom

, Van Gool

(2013) MRI and cerebrospinal fluid biomarkers for predicting progression to Alzheimer’s disease in patients with mild cognitive impairment: A diagnostic accuracy study. BMJ Open 3, e002541.

63.

Vanderschaeghe

, Schaeverbeke

, Vandenberghe

, Dier-ickx

(2017) Amnestic MCI patients’ perspectives toward disclosure of amyloid PET results in a research context. Neuroethics 10, 281–297.

64.

Sperling

, Rentz

, Johnson

, Karlawish

, Dono-hue

, Salmon

, Aisen

(2014) The A4 study: Stopping AD before symptoms begin? Sci Transl Med 6, 228–213.

65.

National Academies of Sciences, Engineering, and Medicine (2018) Returning Individual Research Results to Participants: Guidance for a New Research Paradigm. The National Academies Press, Washington, DC. 10.17226/25094.