Amyloid positron emission tomography (PET) imaging in patients with suspected mild cognitive impairment (MCI) or dementia results in more accurate diagnoses and alters patient management, results of a large, longitudinal study show.
“We were surprised by the magnitude of the change. We had anticipated 30% based on the literature that was available at the time,” lead author Gil D. Rabinovici, MD, of the Memory and Aging Center in the Department of Neurology at the University of California, San Francisco, told Medscape Medical News.
“The thing that really stood out to me was the proportion of people suspected of having Alzheimer disease who were amyloid negative. Over 70% of them had a pre-PET diagnosis of Alzheimer disease as the most likely cause of their impairment,” Rabinovici said.
Amyloid PET detects amyloid plaques in the brain, a core neuropathologic feature of AD. Autopsy studies suggest amyloid PET has high sensitivity and specificity and may be a useful addition to clinical criteria in improving the diagnostic accuracy of AD and possibly cognitive decline.
However, in 2013 the Centers for Medicare & Medicaid Services (CMS) concluded there was insufficient evidence to justify routine coverage of amyloid PET scans in clinical practice. However, the agency did agree to provide amyloid PET scan coverage with evidence development in studies that would examine whether amyloid PET improved health outcomes in Medicare and Medicaid beneficiaries.
Rabinovici and his team heeded the call with a two-part Imaging Dementia—Evidence for Amyloid Scanning (IDEAS) study. Between February 2016 and September 2017, 946 dementia specialists at 595 US practices enrolled 16,008 Medicare beneficiaries. The final cohort included 11,409 participants with complete data.
Comprehensive diagnostic assessments included the Mini-Mental State Examination (MMSE) or Montreal Cognitive Assessment (MoCA), laboratory testing within the past 12 months, and head computed tomography or MRI within the previous 24 months.
All participants met “appropriate use criteria” for amyloid PET, which included diagnostic uncertainty, consideration of AD, and an expectation among clinicians that knowledge of amyloid PET status would alter diagnosis and management.
The percentage of clinicians who reported diagnostic confidence in the uncertain range, 4-7 on a Likert scale from 1 to 10, dropped from 72% before to 16% after amyloid PET scans.
PET results ‘contributed substantially’ to 85% of the instances in which physicians made post-PET patient management changes.
Many diagnoses changed. For example, 25% of those identified with AD had a different diagnosis after PET imaging. In addition, 11% of patients were newly identified with AD.
Changes in Management
Among patients with positive amyloid PET results, overall use of AD medications jumped significantly — from 40% to 82% in the MCI group and from 63% to 91% in the dementia group. Both of these changes were significant (P < .001). In the same group, the proportion of those diagnosed with AD increased from 80% to 96%.
Among patients with negative amyloid PET scans, AD diagnoses dropped from 72% to 10%.
“We as neurologists have to be humble. Our clinical diagnosis, while valuable, has a lot of limitations,” Rabinovici said. Supplementing clinical assessment with direct biological measures “is really where we need to go in the field.”
“While refinement of diagnosis and core elements of the management plan are valuable for the care of patients with cognitive impairment, an important future goal is to test whether amyloid PET is associated with changes in additional patient-oriented outcomes,” the researchers note.
Phase 2 of the ongoing IDEAS study aims to do just that, Rabinovici said. His team is assessing Medicare claims over 1 year to measure rates of hospitalization, emergency department visits, and overall resource utilization.
The investigators plan to compare these findings between Medicare beneficiaries with, and without, amyloid PET scans.
Another proposed study would enroll a more diverse population, assess the role of the APOE ε4 risk gene to stratify individuals who may benefit from amyloid PET imaging, and/or establish a repository of plasma and DNA samples for further research.
These include the large number of participants and the fact that the study outcomes were actual, rather than potential, changes in care in response to PET findings.
They note PET results influenced 85% of the instances when a change was made. “This supports (but without randomization does not prove) a relationship between the PET findings and post-PET changes in management,” Jack and Petersen write.
Jack and Petersen also point out that amyloid PET results were negative in 36% of patients in whom an AD diagnosis was suspected prior to imaging.
“Given that both amyloid-β and pathologic tau deposits are required for a neuropathologic diagnosis of AD (the accepted gold standard) this means that a presumed etiology of AD based on clinical assessment by dementia experts was incorrect for approximately one-third of patients,” they write. This result, they add, is consistent with prior evidence.
Based on the finding that amyloid PET imaging provides clinicians with more specific knowledge about the etiology of cognitive impairment and changes clinical management in this patient population, “a path forward seems apparent,” the editorialists say.
“More detailed etiologic characterization by deeper biomarker-based phenotyping will result in more precise, patient-specific management decision making,” Jack and Peterson conclude. “Ultimately, the hope is that management will include access to pathophysiologically appropriate, disease modifying interventions.”
The Alzheimer’s Association, the ACR, Avid Radiopharmaceuticals, GE Healthcare and Life Molecular Imaging funded the study. Rabinovici reports receiving grants from the American College of Radiology (ACR), the Alzheimer‘s Association, Avid Radiopharmaceuticals, GE Healthcare, Life Molecular Imaging, and Eli Lilly. He also receives personal fees from Eisai, Piramal Imaging, Merck, and Genentech.
Jack reports he consults for Lilly and serves on an independent data monitoring board for Roche, but receives no personal compensation from any commercial entity. He also receives research support from the NIH and the Alexander Family Alzheimer’s Disease Research Professorship of the Mayo Clinic. Petersen reports receiving personal fees from Hoffman-La Roche, Merck, Genentech, Biogen, GE Healthcare, and Eisai.