A Decade of Protecting Progress: Ethics Review

A DECADE OF ETHICS REVIEW: OUR PARADIGM AND APPROACH

Ethics Review was launched a decade ago in 2014 1 with the goal of identifying points when ethical decision-making becomes paralyzed and inefficient, describing barriers, and moving toward solutions. A central hypothesis is that this paralysis in advancing innovation happens when ethical concerns are not clearly specified, and when solutions to those concerns are not apparent. We advocate that the search for solutions should be broad and draw inspiration from parallels with other fields or by leveraging domains such as implementation science and the neuroscience of decision-making. Furthermore, being an early adopter also takes the courage to face the uncertainty of being on the bleeding edge of innovation; however, the knowledge that others, particularly sites that were involved in the early studies who have real-world, personal experience with the innovation, can reassure institutions and disseminate guidance that avoids repeating mistakes. For this reason, the Ethics Review feature enables reviewers to write a Review Response to the original Ethics Review article to highlight other proponents of the innovation. The author of the original article also is given an opportunity to respond as well. The three papers are ultimately published together. What is most critical of these papers is that they are clearly focused and solutions-based to avoid contributing to vague worries that exacerbate paralysis.

To be most sensitive to barriers and solutions to those barriers, we encourage people in the dementia field to build personal connections across institutions with multiple and diverse people involved in the study and care of people living with and at risk for dementia. Formal organizations such as the Alzheimer’s Association, the Lewy Body Disease Association, the Association for Frontotemporal Dementia, and the Gerontological Society of America provide yearly meetings and increasingly more frequent online gatherings for people who cannot afford to attend. The structure and effective aspects of our online community, the Advisory Group on Risk Evidence Education for Dementia, are described elsewhere (AGREEDementia.org), 2 The group’s focus is on the communication and use of dementia risk information and an overview of ethical issues 3 and related publications from that special issue are publicly available and are listed on our blog. 4 AGREEDementia continues to support multiple projects including collaborative grants, foundation-supported expert guidelines, and small student projects. A central value of both AGREEDementia and Ethics Review is that all voices are important. While informal guidance from experts inside and outside the National Institutes of Aging and their funded centers keeps us busy and energized, some of the most creative and impactful projects come from supporting people who would otherwise be silent because their ideas may challenge institutional consensus or would otherwise lack the support to pursue them. For example, study participants led a particularly impactful proposal for a Bill of Rights enabling access to clinical results during research. 5 Contributions like these keep us innovative and responsive, and the Journal of Alzheimer’s Disease continues to support us in broadening our reach.

BALANCING ETHICAL PRINCIPLES

Central to the process of demystifying the paralysis that holds back innovations is to clarify and balance the core ethical principles: Autonomy, beneficence, nonmaleficence, and justice. Considering the potential good of an innovation (beneficence) while also working to minimize the potential to cause harm (nonmaleficence) depends on data that may be unavailable in the early stages of develop-ment. Because people’s value systems differ, it is important to consider the personal utility of an inno-vation that would shift one’s choices. For example, knowing one’s risk for Alzheimer’s disease (AD) or biomarker status may be more or less important to different individuals, even if it is imprecise or there is no treatment.6,7, 6,7 Engaging people living with and at risk for dementia through offering them access to their personal risk information and making choices based on their values supports their autonomy. Monitoring rationales for decisions is important because cognitive dysfunction challenges autonomy by undermining understanding and memory for information needed to make choices. Access to these innovations may not be equitable and hence studying care delivery to make them more available supports justice. There are, of course, other frameworks for describing ethical dilemmas and other contributors to lack of acceptance of innovation and we welcome the study of these factors. Application of these principles to innovation around AD and related dementias enriches the field of ethics.

DEFINING INNOVATION, SCOPE OF APPLICATION, AND THE NEED TO FOLLOW RELEVANT GUIDANCE

There are multiple forms of innovation, each with different ethical dilemmas, hence the importance of making explicit the ethical concerns and the context and referring to the appropriate guidance. For example, the ethics of research and clinical practice differ in multiple ways.8,9, 8,9 Research carries the burden of protecting the integrity of the scientific inquiry, whereas clinical practice must prioritize patient welfare over other benefits; thus, a researcher but not a clinician can stop working with a patient for failure to take a medication if this noncompliance violates a research protocol. The International Committee on Publication Ethics (ICMJE) 10 and Committee on Publication Ethics (COPE) 11 can support the ethical conduct of research, for example, by refusing to publish clinical trials that fail to report that informed consent for research was obtained from participants. 12

With respect to clinical ethics in individual patients, following and communicating with patients the current scientific evidence is critical. Even when there are novel therapies and diagnostic tests that are newly FDA-approved or that have sufficient evidence to receive acceptance by professional organizations as a new standard of care, some clinicians are reluctant to change practice and may not explain their rationale to their patients. The result is that some clinical patients may not even be aware of these innovations thus depriving them of choice. In contrast, patients with progressive illnesses sometimes do not wish to wait for years until clinical trials conclude because the therapies will be too late to give them any benefit. When there are unmet needs, clinicians may provide new and non-validated practice (NNVP) that lacks an evidence base. 13 This lack of evidence must be transparent. An example from AD of how this NNVP can be problematic was discussed regarding a metabolic enhancement protocol that required patients to purchase expensive supplements and that promised to reverse AD on the basis of weak, case-study research. 14 The NNVP was described as damaging on three levels. Individual clinical patients were hurt by the treatment expense, scientific communication provided overly optimistic predictions suggesting that these patients would likely receive clinical benefits, and there were unfounded claims on a population level that therapeutic benefits would lower the burden of care of people with dementia on society.

Furthermore, the ethics of applying innovations to populations have different ethical parameters that are controlled at several societal levels.15 –17 For example, the allocation of resources to make possible risk reduction may be most effective when applied across the lifespan rather than selectively to people at the highest risk of AD in order to address the longitudinal accumulation of risk. 16 Society can be structured to “nudge” healthy behaviors in subtle, barely intentional ways; 18 however, applying population-based benefits is constrained by structural-resource limitations such as the lack of autonomy of women in some societies. 17 There is much work to be done on the ethics of impoverished and diverse communities around the ethics of dementia prevention.

A REVOLUTION IN THERAPY, DIAGNOSIS, AND RESEARCH AS A CASE STUDY

With the increasing ease and precision of detecting evidence of brain pathology underlying AD with blood tests, e.g. 19 , and the development of drugs to reduce that pathology,20 –22 the field of neurodegeneration is poised for a transformation. There are an increasing number of papers discussing the ethical issues, pragmatic and legal concerns, and frameworks for sharing dementia risk-related information which we touch on briefly more as case examples rather than as a comprehensive description, e.g. 23,24–27 , 23,24–27. As these innovations evolve, their limitations introduce ethical dilemmas.

There have been ethical concerns about anti-amyloid therapies like lecanemab and donanemab in part because they were expensive and involved expensive, complex, care but also because patients risk amyloid-related imaging abnormalities (ARIA) and rare deaths; 28 however, when lecanemab demonstrated at least modest benefits and FDA gave approval, patients now can request the treatment even though the value of the improvement is debated.29 –31 Furthermore, access to the medication is limited to people without significant vascular risk and those not needing such treatments as anticoagulant therapy and people with milder disease. People from socially disadvantaged backgrounds also face significant barriers because they do not have easy access to sophisticated memory clinics and PET scans to identify amyloid in their brains. 32

Biomarker confirmation of brain pathology is crucial when interventions target the biological substrate it identifies. For example, administering drugs targeting amyloid would be ineffective in people with a different brain pathology such as limbic-predominant age-related TDP-43 encephalopathy (LATE), 33 a disorder with symptoms clinically similar to AD. The FDA has developed an approach to validating biomarkers for specific contexts of use described by the Biomarkers, EndpointS, and other Tools (BEST), 34 and is refining the levels of disease progression to make possible the study of degenerative illnesses earlier.

Whereas blood tests for amyloid pathology offer the opportunity to detect amyloid positivity in larger populations thus enabling early detection, their results are sometimes confounded. A rigorous evidence-based assessment of tests for different use cases and cut points is needed so that it is possible to know the level of certainty of conclusions. For a variety of reasons, marginalized cohorts and people with comorbid medical conditions have been minimally studied and there is evidence that their results have subtly different patterns.35,36, 35,36 Understanding the mechanisms underlying those differences could potentially move the field forward such as when genes identifying resistance to AD are found. 37 When measures either under or over-identify pathology (as compared to more established gold standard PET biomarkers) and are used to allocate therapy, the biomarker test can either be a barrier to care or result in over-treatment. 38

Various solutions are being discussed as evidence builds. Ultimately, finding measures that are unbiased and promoting honest dialogs with patients about levels of certainty will be critical to provide care as data accumulate. The Global CEO Initiative on Alzheimer’s (Global CEOi, https://www.alzbiomarkerhub.org/) organized an expert consensus on criteria for the precision of biomarker testing for blood tests in different contexts. 39 Biomarker tests marketed directly to consumers may someday empower individuals to discover their status confidentially and outside the healthcare system; however, without professionals to explain the results these tests could cause distress and misinformation if they are imprecise. There is also an initiative to pool datasets in order to understand how to interpret blood tests with comorbid conditions and diverse ethnicities and results will be posted on this website (https://www.adaccnetwork.net/).

Implementation science can optimize use of limited resources such as highly trained professionals. Whereas early detection of brain pathology can identify more people who can be treated with drugs like lecanemab that reduces amyloid, there are limited specialists able to manage this complex treatment that involves intravenous delivery and monitoring with imaging at this point. The Global CEOi convened a panel of experts to provide an implementation plan to increase access to treatments by moving diagnosis into primary care rather than waiting for specialized care.40,41, 40,41 Regardless, the opportunities to help at least some people receive this treatment that slows the course of disease progression necessitates that the field adapts to and prepares for potential solutions to these barriers. The first to receive these novel treatments may be people with financial means, unaffected by societal disadvantage; however, those people are also choosing to be exposed to the unknown risks of novel treatments. A critical feature of any novel therapy is making sure public and patient involved people understand that there may be as yet unknown risks. This choice requires clinicians to evaluate whether patients are making truly informed decisions in the face of a degenerative cognitive condition that ultimately robs them of decision-making capacity.

We have known that amyloid pathology develops over decades, long before standard clinical neuropsychological testing detects dysfunction 42 so that reducing amyloid to prevent the onset of cognitive symptoms in people with amyloid positivity on blood tests would be ideal. Unfortunately, uncertainty around amyloid positivity results prevents this use. Whereas this early warning offers an opportunity to intervene before cognition declines, the clinical standard of care is not to treat with medications until there are symptoms 43 because not all people with amyloid-positive biomarker test results develop AD and the treatment is expensive and has a risk ARIA. 44 A potential solution could be to improve prediction by integrating multiple sources of data, including genetics. Because there are many contributors and causes for dementia, there are many layers and nuances to ethical practice that should be considered both at an individual and societal level.45,46, 45,46

ADDRESSING THE EDUCATIONAL NEEDS OF PATIENTS

Decision-making becomes increasingly complex as multiple factors contribute to risk, and everyone, but especially people with memory and comprehension dysfunction, can benefit from written material to support their choices. The National Society for Genetic Counselors’ Dementia Working Group developed a brief handout for patients with mild cognitive impairment (MCI) or mild AD considering amyloid-reducing medication (e.g., lecanemab, donanemab) to explain why genetic testing is recommended or needed prior to treatment and the implications of that testing. The handout informs patients that ARIA is a risk of these medicines, the APOE ɛ4 variant is associated with an increased risk of ARIA, and having one or two copies of the APOE ɛ4 variant increases the risk for AD in their children. It is intended to augment, but not replace, a patient’s discussion with their healthcare provider. Wherever possible, asking permission prior to sharing risk information, such as APOE or amyloid status, is advisable, to respect people’s right not to know. The digital version of the handout was distributed as part of a Qualtrics survey delivered via QR code to 32 older adults who were Veterans (mean age = 75.5, mean years of education = 16, 91% male, 88% white), some with MCI. The feedback provided was informative. Many members of the community stated that they were unaware of the medicine and needed more information, a situation that will likely change as the VA healthcare system begins to deliver the therapy. Almost one-third of people felt omitting an explanation of what genes are would hinder their comprehension (“A gene is a segment of DNA that influences traits like eye color, height, or our chance of developing a disease.”). Whereas physicians often refer to the medication by its generic name, about one third of respondents found this confusing and preferred that only the trade name (e.g., Leqembi) be used, consistent with what they would be asking for at the pharmacy. Most participants responded accurately to questions assessing comprehension of the information on the form, although a few participants had some difficulty, highlighting the importance of the patient-provider relationship. Expanding beyond older adults outside of our Alzheimer’s Disease Research Center to include people with lower education and more diverse backgrounds will also be important moving forward.

DESCRIBING PARALYSIS, THE GULF OF DISAPPROVAL, AND THE STATE OF COMMON KNOWLEDGE

Novel therapies, techniques, devices, applications, decision tools, and other innovations can improve the care and study of people with and at risk for dementia; however, resistance to innovations has long been described by writers such as Rogers. 47 Rogers named the different populations of users in terms of their readiness to adopt innovation. First are a small number of Innovators, next are Early Adopters, then there are larger cohorts termed the Early Majority, and next the Late Majority, and the most reluctant are a small group of Laggards. In Fig. 1, Seth Godin graphically represented the timespan of acceptance of innovation during which there initially is a Gulf of Disapproval 48 during which time innovators face widespread resistance from a broad majority except from the early adopters (Believers). Ultimately, there is increased enthusiasm for helpful innovations and broad acceptance, until innovations become common knowledge. There may be several explanations for this early resistance. We believe that one core problem is that much is unknown and untested, thus undermining objective weighing of risks and benefits. This situation leads to a reliance on a more emotional process that is subject to known decision biases that were described by Kahneman 49 among others. Paralysis in enabling innovation is particularly damaging in people living with degenerative brain diseases because there is progression and disability that could potentially have been prevented by earlierintervention.

OPEN IN VIEWER

Fig. 1

The evolution of adopting innovation by Seth Godin in light of Rogers.

EDUCATING RESEARCHERS AND CLINICIANS ABOUT EMOTIONAL DECISION MAKING

Under-confident innovators such as students and new investigators must be wary of this Gulf of Disapproval. Some innovations were obviously damaging, such as the widespread delivery of nonspecific human lobotomies. Some Believers unjustifiably cling to their innovations due to a sunk cost bias, a failure to abandon a poor project because of extensive investments in it. But what is often missed are the opportunity costs of complete rejection when selective implementation in restricted circumstances could solve problems. Careful weighing of risks and benefits in the specific context can address concerns. For example, the risk of addiction to certain pain medications is unimportant in a terminally ill patient with limited life expectancy when the priority is quality of life. Societal biases against recreational drug use could delay the use of psychedelics as potential therapeutics in patients suffering from poorly managed mental health conditions. 50 When an author suggested engaging family members in cognitive assessments, 51 ethical concerns were raised about whether testing would be objective, and whether sharing assessments undermined test security; however, with an awareness of interpretive concerns, selective in-home assessment could enable a caregiver to identify delirium that needed urgenttreatment. 52

On the other end of the spectrum, at the stage of common knowledge, there may be a reluctance to study long-established practices in innovative contexts. For example, there is a growing appreciation for the value of repurposing established drugs for novel applications through the exploration of large, healthcare datasets. 53 These data are correlational and thus a clinical trial may be worthwhile if the indication is novel. Once a patent has expired, the pharmaceutical industry can make limited profits from it, and this may disincentivize funding rigorous trials. The allure of novelty has also faded, and thus, funding agencies may not find this work appealing. Thus, an opportunity is lost.

MOVING INTO THE FUTURE

In this transformative time in research in AD and related dementias, we hope to continue to embolden innovators. The central mission of Ethics Review is clearly identifying known or potential risks, balancing them against opportunities lost, and promoting solutions. Changing standards of care requires an evidence base and a critical mass of people willing to join the Believers, the brave souls and institutions willing to confront the bleeding edge of innovation. Introducing those Believers to one another in a Review Response and publishing rationales for adopting innovations emboldens more people to join a movement away from paralysis toward progress and the potential for a world without dementia.

AUTHOR CONTRIBUTIONS

Allyson C. Rosen (Conceptualization; Supervision; Writing – original draft; Writing – review & editing); James A. Lavacot (Conceptualization; Writing – original draft; Writing – review & editing); Victoria Klee (Conceptualization; Writing – review & editing); Yuval Luria (Conceptualization; Writing – review & editing); Malia C. Rumbaugh (Conceptualization; Writing – review & editing).