No evidence of cognitive or psychological impact after returning research Alzheimer disease biomarkers: A delayed-start, noninferiority, randomized clinical trial

The Ethics of Knowing: Can Alzheimer’s Biomarkers Harm the Healthy?

In the landscape of modern neurology, we are approaching a paradox of transparency. We are developing increasingly precise tools—blood tests and brain scans—that can detect the biological hallmarks of Alzheimer’s disease years before the first memory slips. Yet, a significant ethical tension remains: if we tell a cognitively healthy person that they have a high biological risk for a disease that currently has no cure, do we inadvertently cause them harm?

For years, researchers have hesitated to return these research results (RoRR) to participants. They fear the news might trigger clinical depression. They also worry it might induce anxiety that impairs cognitive performance during follow-up testing. This paper addresses that hesitation directly. By conducting the first randomized, delayed-start trial of its kind, the authors investigate whether knowing one's Alzheimer’s risk fundamentally alters a person's mental health or cognitive trajectory over time.

The Burden of Early Knowledge

The core question is whether "knowing" acts as a psychological or cognitive stressor. In clinical research, Return of Research Results (RoRR) refers to the practice of sharing specific biological data with the people who provided the samples.

While the medical community is increasingly advocating for this transparency, the fear is twofold. First, there is the psychological risk. A person might receive news of future decline and descend into chronic depression. Second, there is the procedural risk. The stress of receiving bad news might cause a participant to perform poorly on subsequent memory or thinking tests. This could "pollute" the very research data the scientists are trying to collect. The study asks if providing this information is a neutral act or a disruptive one.

The Biological Landscape of Risk

To understand the stakes, one must understand what is actually being disclosed. The researchers targeted cognitively unimpaired (CU) older adults. These are individuals aged 65 or older who show no clinical signs of dementia. They offered a suite of biomarkers:

1. APOE Genotype: A genetic test for the Apolipoprotein E gene. Carrying certain variants (specifically the $\epsilon4$ allele) is a known risk factor for late-onset Alzheimer's.
2. Amyloid Status: This measures the presence of amyloid-$\beta$ (a protein that clumps in the brains of Alzheimer's patients). This was assessed via amyloid PET scans (imaging that uses radioactive tracers to visualize plaques) or plasma $A\beta42/40$ ratios (a blood-based proxy for brain amyloid levels).
3. Hippocampal Volume: An MRI-based measurement of the size of the hippocampus. This is a brain region critical for forming new memories that often shrinks in early disease stages.

Previously, small studies suggested that while immediate emotional reactions might occur, these feelings tended to wane. However, no study had yet used a randomized controlled design to track whether these disclosures affected longitudinal outcomes. These are changes that occur over many years rather than a single moment.

Testing the Stability of the Mind

The researchers employed a "delayed-start, noninferiority" design to isolate the effect of the news. In a noninferiority trial, the goal is not to prove an intervention is "better" than a control. Instead, the goal is to prove it is "not worse" by a clinically meaningful margin.

As shown in, 147 participants were randomized into two groups.

The "RoRR arm" received their results almost immediately (2–4 weeks after consent). The "delayed-start arm" acted as the control. They received their results exactly one year later. This temporal gap allowed researchers to compare the trajectories of people who had processed the news against those who had not yet heard it.

The study tracked three primary outcomes annually: * Geriatric Depression Scale (GDS): A tool to measure depressive symptoms. * Clinical Dementia Rating® sum of boxes (CDR-SB): A scale used to assess the severity of cognitive and functional decline. * Global Cognitive Composite: An objective score measuring various aspects of thinking and memory.

The researchers stratified their analysis by amyloid status. They recognized that a person told they are "amyloid positive" ($A\beta+$) might react differently than someone told they are "amyloid negative" ($A\beta-$).

The results, summarized in [Table 2] and visualized in, showed no significant difference between the groups.

For the $A\beta+$ subgroup, the difference in GDS (depression) between the arms was only 0.7 points. This is far below the 4.0-point threshold required for a "clinically significant" decline. Similarly, for the CDR-SB (clinical cognition), the difference was 0.0. This is well below the 0.5-unit threshold for meaningful impairment. Even the objective cognitive composite showed no significant divergence between the groups.

Crucially, the researchers also looked at "proximity." They asked if it matters if a test happens one month after the news or ten months after. Through linear regression, they found no association between the timing of the disclosure and the stability of the participant's mental or cognitive state.

Shifting the Ethical Paradigm

These findings allow us to view the return of research results through a new lens. Until now, the debate has been framed as a choice between "transparency" and "protection from harm." This study suggests that, for cognitively healthy older adults, these two goals are not in conflict.

The data implies that the "harm" of knowing one's risk is not a driver of actual clinical decline. If receiving biomarker results does not accelerate depression or mask cognitive abilities, then the barrier to RoRR shifts. The question moves from a matter of safety to a matter of utility. We can now move past the fear of causing psychological trauma. We can begin focusing on how to deliver this information effectively and at scale.

The Limits of the Evidence

However, the study is not a universal mandate. The authors are transparent about several critical edges:

First, the demographic profile of the participants was narrow. The cohort consisted primarily of White and highly educated individuals. Because Alzheimer's disease affects Black and Latino populations at different rates, it remains unknown if these results would hold true in more diverse populations.

Second, the scope was limited to depression and cognition. While these are the most obvious risks, the study does not account for other potential impacts. These might include changes in social behavior or lifestyle modifications.

Finally, the study focuses on "cognitively unimpaired" individuals. The psychological impact of receiving a positive biomarker result in someone already experiencing mild cognitive impairment (MCI) remains an open question for future research.

Figures from the paper