Visualizing brain amyloid-beta pathology: Toward implementation of amyloid imaging in daily memory clinic practice

SUMMARY The ABIDE study (chapter II) aimed to address the practical use of AD biomarkers by the clinician and how to take patients’ preferences toward testing and communication of test results into account. ABIDE was a 4-year project with a multifaceted design structured into interconnected substudies using both quantitative and qualitative research methods.1 In short, ABIDE assessed the diagnostic performance of CSF, the clinical utility of Aβ PET (chapter III and IV), biomarker-based personalized risk estimates for clinical progression in cognitively normal individuals and patients with MCI, application of the novel biomarker-based AD research framework (chapter VI), and how to optimize patient-clinician communication during the dementia diagnostic work-up.2–14 Our large and prospective diagnostic utility study (chapter III) assessed the feasibility and added value of Aβ PET in an unselected memory clinic cohort – reflecting real-world use of this diagnostic test. Over the course of almost two years (2015-2016), we implemented the test in the routine diagnostic process and offered Aβ PET to all patients visiting our memory clinic. We assessed the association of Aβ PET with changes in diagnosis, diagnostic confidence, treatment and patients’ experiences in 507 patients. We found that etiological diagnosis changed for 25% of patients post-PET, more often due to a negative than a positive PET result. Diagnostic confidence levels increased significantly, both as a result of positive and negative scan results. Aβ PET results led to a change in patient treatment for 24% of patients, more often after a positive than a negative result. The association of Aβ PET with changes in diagnosis, diagnostic confidence and treatment were observed across diagnostic groups. In addition, disclosure of PET results led to a decrease in uncertainty, whereas anxiety did not increase. In this study we attempted to bridge the gap between validating Aβ PET in a research setting and implementing this diagnostic tool in daily clinical practice. We observed that both Aβ-positive and Aβ-negative results had an important association with changes in diagnosis and treatment in patients with and without dementia. As a next step we were interested in the ability of the appropriate use criteria (AUC) to select patients most likely to benefit from Aβ imaging. In Chapter IV we retrospectively applied the AUC criteria and assessed post-PET differences in changes in diagnosis and treatment between AUC-consistent and AUC-inconsistent patients. In addition, we assessed the AUC’s test performance characteristics for detecting (combinations of) clinical utility outcomes. In our memory clinic cohort, almost half of patients were AUC-consistent. We found however that across the spectrum of cognitive impairment, ranging from SCD to dementia, proportions of changes in diagnosis, and patient management were substantial in both AUC-consistent and AUC-inconsistent patients. On the other hand, there were also many patients both AUC-consistent and AUC-inconsistent, for whom Aβ PET did not have an apparent clinical benefit. Further, we observed that the current operationalization of the AUC focuses on identifying patients where confidence will increase, and management will change after Aβ PET, therewith neglecting the clinical benefit of accurate alternative diagnosis - i.e. change in diagnosis – which sometimes comes with a decrease in confidence. These results reveal that the current AUC are not sufficiently able to discriminate between patients who will benefit from Aβ PET and those who will not. Alternative strategies are needed to identify patients most likely to benefit from ab-PET in clinical practice.