MedicalResearch: What is the background for this study?
Dr. Ossenkoppele: Since 2004, several PET tracers have been developed that measure fibrillar amyloid-β plaques, a neuropathological hallmark of Alzheimer’s disease (AD). Through visual assessment by a nuclear medicine physician or quantitative cut-points, the presence or absence of amyloid-β pathology can be determined in the living human brain. The FDA, in support of the clinical application of amyloid imaging, has recently approved three of these PET tracers. A proportion of patients with other types of dementia then Alzheimer’s disease that harbor cerebral amyloid-β pathology, however, potentially limits the clinical utility of amyloid imaging. When ordering clinical amyloid PET scans and correctly interpreting the significance of amyloid PET results, clinicians need to understand the prevalence of amyloid-positivity across different types of dementia. It is also important to be aware of the relationships of amyloid-positivity prevalence and demographic (e.g. age and sex), cognitive and genetic (e.g. presence of the AD-risk allele apolipoprotein E [APOE] ε4) factors. Most amyloid PET studies to date come from single centers with modest sample sizes. We therefore conducted a meta-analysis with individual participant data from 29 cohorts worldwide, including 1359 patients with clinically diagnosed Alzheimer’s disease and 538 patients with non-AD dementia. We also included 1849 healthy controls with amyloid PET data, and an independent sample of 1369 AD patients with autopsy data from the NACC database.
MedicalResearch: What are the main findings?
Dr. Ossenkoppele: In patients clinically diagnosed with Alzheimer’s disease, the prevalence of amyloid-positivity decreased from 93% at age 50 to 79% at age 90. The drop in amyloid-positivity was most prominent in older Alzheimer’s disease patients who did not carry an APOE ε4 allele (~1/3 of these patients had a negative amyloid PET scan). This most likely reflects a mix of
1) clinical misdiagnoses (i.e. non-AD pathology causing an AD phenotype),
2) false negative PET scans (i.e. abundance of cerebral amyloid pathology that is not detected by PET), and
3) possibly elder patients need less amyloid pathology (sub-threshold levels for PET) to reach the stage of dementia due to age-related reductions in cognitive resilience (“cognitive reserve theory”) or simultaneous presence of multiple pathologies (“double-hit theory”).
The relatively high rate of amyloid-negative Alzheimer’s disease patients highlights the necessity of biomarker-informed patient selection for Alzheimer’s disease clinical trials.
In most patients clinically diagnosed with non-AD, the prevalence of amyloid-positivity increased with aging and was ~18% higher in APOE ε4 carriers. Presence of amyloid pathology in non-AD dementia may reflect
1) clinical misdiagnosis (i.e. AD pathology is the causative pathology), or
2) comorbid pathologies, where amyloid may be secondary to other pathologies that are actually driving the clinical presentation. Interestingly, patients with a clinical diagnosis of non-AD dementia who harbored cerebral amyloid pathology showed lower Mini-Mental State Examination scores (measure of global cognition), suggesting that amyloid-β is not just an innocent bystander.
MedicalResearch: What should clinicians and patients take away from your report?
- Diagnosis, age and APOE ε4 are important factors to take into account when ordering and/or interpreting clinical amyloid PET scans.
- The convergence of amyloid-positivity between Alzheimer’s disease and non-AD dementias with increasing age suggests that amyloid imaging may be most useful for differential diagnosis in patients with early-onset (<65 years of age) dementia (i.e. behavioral variant frontotemporal dementia vs Alzheimer’s disease), since the risk of incidental amyloid-β pathology is relatively low in young patients.
- The correspondence of amyloid-positivity in Alzheimer’s disease patients with PET and at autopsy was reassuringly high, suggesting that amyloid imaging can be used to “rule-out” Alzheimer’s disease as the causative pathology regardless of age.
- Ordering an amyloid PET scan seems not justified in APOE ε4 carriers clinically diagnosed with Alzheimer’s disease, since the prevalence of amyloid-positivity remained >90% regardless of age.
- On the other hand, amyloid PET may be informative in APOE ε4 non-carriers over age 70 with a clinical diagnosis of Alzheimer’s disease, since the prevalence of amyloid-positivity declined to 78% and further decreased to 68% at age 90.
MedicalResearch What recommendations do you have for future research as a result of this study?
Dr. Ossenkoppele: In general, Alzheimer’s disease studies will gain reliability when patients are included based on thorough clinical assessment in conjunction with biomarker conformation. This and previous studies have shown that the clinical diagnosis alone, even in expert centers, has limited capacity to accurately predict the histopathological cause of dementia.
The clinical application of amyloid imaging would benefit from a better understanding of clinically diagnosed Alzheimer’s disease patients who appear amyloid-negative on PET. What proportion does show amyloid plaque burden at autopsy and which factors may contribute to false-negative PET scans? What other – non-amyloid-β – pathologies can mimic an “AD phenotype”? Studies assessing ante-mortem PET versus post-mortem neuropathological examination in the same patients are crucial.
Finally, the finding that amyloid-β pathology in patients with clinically non-Alzheimer’s disease dementia was associated with more severe cognitive deterioration deserves more attention because – of this appear to be true – these patients might benefit from anti amyloid-β therapies when they become available.
MedicalResearch.com Interview with:, Rik Ossenkoppele PhD. Postdoctoral researcher, & UCSF Memory and Aging Center (2015). PET Scan Can Detect Amyloid In Alzheimer’s and Other Dementias