Disclosure of amyloid positron emission tomography results to individuals without dementia: a systematic review

This systematic review is based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement [28]. We conducted a systemic literature search in four electronic databases on 28 March 2017: Embase, PubMed, the Cochrane library, and Web of Science. We developed the search strategy in collaboration with an information specialist from the VU Medical Center medical library. Search terms included controlled terms (MeSH in PubMed and Emtree in Embase) as well as free text terms. We used free text terms only in the Cochrane library. Search terms used were variations on the keywords ‘Alzheimer or dementia’, in combination with search terms comprising ‘disclosure’ and combined with search terms comprising ‘biomarker’ or ‘amyloid’. The full search strategies for all databases are presented in Additional file 1. The references of the included articles were further searched for relevant publications.

We included articles if they reported on disclosure of amyloid PET results in patients with MCI and CN individuals. Given the limited number of studies with empirical data (e.g., randomized controlled trials, surveys, multiple case studies) we also considered reviews, perspectives, and point-of-views addressing theoretical arguments in favor or against amyloid PET disclosure. Articles had to be published in peer-reviewed journals, and all languages were accepted.

Our database search resulted in 2664 unique articles after removing duplicates. Titles and abstracts of all the identified articles were screened by two independent reviewers (AdW and MMvB). Screening for the title and abstract resulted in 228 articles for full-text assessment for eligibility. The same two reviewers assessed all full-text articles. In case of discrepancy, consensus was reached after discussion and consultation of a third reviewer (WMvdF). We included 15 articles for data extraction and data synthesis [5, 10, 29‐41]. A flowchart that shows the results of the initial study selection is presented in Fig. 1. We additionally included two relevant articles that were published after we conducted our search [42, 43], resulting in a total of 17 articles.

Empirical data were extracted by two reviewers (AdW and MMvB), discussed, and checked by a third reviewer (WMvdF). We extracted data regarding main outcome measures and additional information on study design, population characteristics, methods, other relevant outcome measures, and the conclusions of the authors.

Arguments in favor or against disclosure in theoretical papers were identified and extracted by two reviewers (AdW and MMvB). In case of discrepancies between the reviewers, consensus was reached after discussion and consultation of a third reviewer (WMvdF). Due to a large variation in the content of the arguments, we composed a meta-summary. Finally, we grouped empirical data and arguments based on both the context (research versus clinic) and the population (CN versus MCI) for which the authors used them.

There was a large variation in the research questions and designs of the empirical studies, with two prospective cohort studies, three randomized controlled trials, three surveys, and a modified Delphi study. Table 1 shows an overview of the studies reporting empirical data.

Prior to the Food and Drug Administration (FDA) approval of ¹⁸F-florbetapir, Alzheimer’s Disease Neuroimaging Initiative (ADNI) investigators and research staff (n = 159) were interviewed about whether ADNI should change its policy of not returning amyloid imaging results [37]. Of interviewees with direct participant contact (n = 139), 45% reported that CN participants had never requested their PET results, and 40% reported this for participants with MCI. From the ADNI investigators, 94% reported that they “never” disclosed PET results to CN participants, while 90% did not return results to participants with MCI. A majority of respondents indicated that, upon FDA approval for ¹⁸F-florbetapir, they would support disclosing amyloid imaging results to MCI patients (73%) and participants with normal cognition (58%). Important reasons favoring disclosure (based on free-text answers) were based on the principles of ‘respect for autonomy’ and the ‘right to know’, and that it enables future planning and lifestyle changes, while concerns were expressed about potential psychological harms and impact on insurance. Respondents who endorsed disclosure stressed the need for developing standardized disclosure procedures, with the disclosure procedure needing to be rigorously studied with longitudinal outcomes to assess well-being.

Three studies assessed study participants’ interest in knowing their amyloid PET status, albeit using different research questions and study designs. One study surveyed the interest of CN participants (n = 164) in an Alzheimer’s prevention registry in knowing their amyloid status and their motivations [36], and 81% of participants expressed a wish to know their amyloid status, motivated by a desire to participate in AD research (73%) and to prepare their family for their illness (60%). Almost 12% indicated they would use the information to make plans for ending their life when memory loss becomes imminent. The main reason for not being interested in their amyloid status was because of the expectation of feeling depressed when amyloid was elevated (40%). The second study randomly assigned CN participants (n = 219) in a longitudinal aging study to an education intervention (n = 119) or placebo (n = 100) to measure pre- and postintervention interest in biomarker disclosure [32]. One of the items was interest in amyloid imaging results. The authors observed a high preintervention interest (mean 4.0 ± 1.1 on a five-point Likert scale, with 5 being extremely interested) in receiving amyloid PET results. After controlling for preintervention level of interest, an ordinal logistic regression showed that assignment (education versus placebo) significantly decreased interest (odds ratio (OR) 2.8, 95% confidence interval (CI) 1.6–5.1; p < 0.001), with the exemption of three subgroups: participants who 1) estimated their subjective AD risk greater than 50% (the mean), 2) reported having at least one parent with AD, and 3) attended few participant meetings. The third study assessed the likelihood of trial enrollment in older community volunteers with self-reported normal cognition (n = 132) after randomly assigning them to a hypothetical informed consent form where amyloid PET result disclosure was either present (transparent, n = 66) or absent (blinded, n = 66) [33]. There was no significant difference in the likeliness to enroll in clinical trials between the two groups (70% versus 61%).

Two studies assessed the impact of amyloid PET disclosure on depression, anxiety, and stress, while two studies performed semi-structured interviews to assess experiences after PET disclosure. The first study compared measures of depression, anxiety, and test-related distress at different time points between amyloid-positive (n = 27) and amyloid-negative (n = 70) trial participants [31]. Depressive symptoms were stable across visits and were not different between groups. For anxiety symptoms, there was a small increase immediately post-disclosure in the amyloid-positive group, which was not sustained. However, post-hoc analyses revealed no group differences at any time point. Amyloid-positive individuals had slightly higher levels of distress after PET result disclosure, and these were related to higher baseline levels of depression and anxiety. The second study assessed depression, anxiety, stress, impact of events, and subjective memory complaints in a small number of CN trial participants (n = 11) who explicitly requested their amyloid status and which they received after reading a psychoeducational brochure [35]. Despite an insufficient number to make formal comparisons, the impact of disclosure did not seem to be different between amyloid-positive (n = 3) and amyloid-negative (n = 8) participants. Participants considered the psychoeducational brochure to be very useful, and disclosure of amyloid positivity seemed to motivate lifestyle changes. The third study performed two semi-structured post-disclosure interviews with MCI patients (n = 38) as a substudy of a clinical trial in which participants could opt to know their PET result [42]. The results were assessed using qualitative content analysis. Most patients could recall the core message of their result disclosure in their own words. Two out of eight amyloid-positive patients experienced emotional difficulties (feeling worried, sadness) post-disclosure, whereas three of thirty amyloid-negative patients doubted whether they had received the correct result. Experienced advantages included the possibility of making practical arrangements. The fourth study performed semi-structured post-disclosure telephone interviews with 26 patient-caregiver dyads with whom a neurologist discussed the option of amyloid PET [43]. Clinical reasons for scanning varied but were generally considered consistent with the appropriate use criteria for amyloid imaging [26]. Most patients who chose to undergo amyloid imaging would opt for the scan again. Regardless of the PET result, patient and caregivers commonly expressed relief on learning the results. Some patients had expectations of the PET scan that are beyond its capabilities.

Two studies reported on the development of amyloid PET disclosure materials in the context of clinical trials. The first study developed a process to maximize safety and effectiveness of disclosing amyloid imaging results to CN older adults participating in AD secondary preventions studies. They used a modified Delphi Method, consulting experts in the field of genetic testing and amyloid PET, to develop a consensus on best practices. Consensus was reached on the text for a brochure and a disclosure process. Recommendations included pretest counseling, screening for anxiety and depression, separate days for consent procedure, imaging, and disclosure, and follow-up to monitor the impact of disclosure, anxiety, and depression. The developed process and documents are currently used in the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) study [29]. The second study developed an approach for disclosure of amyloid imaging research results in patients with MCI [30]. They used simulated sessions during which MCI patients and their care dyads received fictitious but realistic information regarding their amyloid status, satisfaction surveys, comprehension assessments, and focus group data to evaluate the disclosure material. Recommendations included pretest counseling, the use of participants’ own brain images during disclosure, take-home materials, and follow-up to address emerging questions. The materials are currently used for trials at the University of Pittsburgh Alzheimer Disease Research Center.