Kenji Ishii of the Tokyo Metropolitan Institute of Gerontology updated the audience at the 5th annual Human Amyloid Imaging conference, held in Miami, Florida, on 14-15 January 2011, about where the Japanese ADNI stands at this point. His talk reinforced how age and genetics influence amyloid formation and AD. Responding to unexpectedly high interest from its participants, the Japanese ADNI ended up scanning 450 people with amyloid PET, Ishii said, three times as many as they had originally planned. The resulting data are compatible with those from the U.S. ADNI and the Australian AIBL study without much of an ethnic confound, Ishii said. This means J-ADNI adds a cohort for combined analyses of larger datasets.
There are differences, though. The normal controls in the J-ADNI cohort are about a decade younger than its U.S. counterpart, and have fewer ApoE4 carriers among them. Consequently, only 21 percent were amyloid positive, less than half the rate of the U.S. ADNI controls. Similar to U.S. cohorts, though, in the Japanese volunteers, tracer retention started rising in the late sixties in people without an E4 allele, earlier with one, and earlier still with two. The PIB positivity rate tracked that: Around age 60, it was only 5 percent in those without ApoE4, but 37 percent in those with one copy and 67 percent for people with two copies. “ApoE accounts for at least a 10-year age difference in amyloid deposition,” Ishii said.
A note on 18F amyloid tracers and histopathology correlations. While Avid and Lilly learned on 20 January at the Food and Drug Administration that their application for clinical use of Amyvid—the most advanced 18F compound in the running—will be approvable pending further data on reader training and consistency (see ARF related news story), the Miami conference featured new data on what the runners-up are doing and how well PET tracers in general detect fibrillar amyloid pathology in the brain.
Steve Arnold of the University of Pennsylvania, Philadelphia, reported that 18F-flutemetamol, the fluorine-18 version of PIB being developed by GE Healthcare, correlates well with histopathological amyloid as measured in biopsy samples. Brain biopsies are rare, but people with normal pressure hydrocephalus sometimes have a small piece of tissue taken out when surgeons place a shunt from a brain ventricle to the peritoneum to let excess fluid drain. (This therapeutic procedure came up at last Thursday’s FDA Advisory Committee meeting on Amyvid, as well, when scientists entreated the committee to recommend approval by citing a range of examples for how amyloid imaging would improve their clinical care. One such instance is hydrocephalus. Patients with this condition tend to respond well to shunting if they do not also have amyloid pathology, but poorly if they do, so amyloid PET would enable clinicians to pick the patients for whom this invasive treatment is right.)
In Miami, Arnold presented data on seven hydrocephalus patients who’d had a shunt placed, a biopsy taken, and a flutemetamol scan thereafter. Staining of the fixed specimen with antibodies to various fibrillogenic proteins and with thioflavin S showed that four of the seven people indeed had Aβ amyloid in their brains, and one had CAA as well. This matched entirely with the flutemetamol scans, Arnold said. Seven cases don’t make much of a quantification, but still, this is the first study to look for concordance of flutemetamol imaging with histopathological evidence of Aβ-related lesions, Arnold said, and it fits with the existing literature on PIB matching up against pathology. Juha Rinne of the PET Center in Turku, Finland, presented a similar, but larger biopsy study for hydrocephalus cases using PIB scans. This work reinforces again that this tracer correlates strongly with the number of Aβ aggregates in the respective brain areas. And on pathology-PET tracer correlation in general, outside of biopsies, Ira Driscoll, at the National Institute on Aging in Baltimore, Maryland, presented her results of six additional autopsy cases. In her hands, the tracer she tested, PIB, accurately reflected the amount of Aβ pathology in different brain regions. “PiB does a good job of capturing Aβ in vivo,” Driscoll later wrote to ARF.
On a poster, Milos Ikonomovic of the University of Pittsburgh Medical School showed biochemical and pathology data to suggest that flutemetamol retention in the brain reflects Aβ plaque load similar to PIB. This tracer is beginning to be more widely used in research settings as well. Ranjan Duara of Mount Sinai Medical Center in Miami Beach, Florida, showed data on using flutemetamol together with MRI for classifying different kinds of MCI patient. G.E. is currently performing a Phase 3 multicenter histopathology study.
Also on flutemetamol, Rik Vandenberghe of K.U. Leuven, Belgium, and colleagues presented research on automated software and on a separate supervised machine learning technique applied to the scans from a Phase 2 study (Vandenberghe et al., 2010). The ultimate goal is to use machines to interpret scans instead of relying on human raters. It was around consistency among raters in deciding whether a given amyloid scan is positive or not, especially in the early disease stage, that Amyvid’s bid on 20 January for FDA Advisory Committee approval ran into a temporary snag, even though the company had followed a statistical plan the FDA had previously okayed. Getting this call right in the clinically relevant population is a pressing concern now. In Miami, several speakers, including Wenzhu Bi of UPitt and Adam Fleisher of the Banner Alzheimer’s Institute in Phoenix, Arizona, presented methods to that end, namely to set thresholds for positivity of a scan with PIB and florbetapir, respectively.—Gabrielle Strobel.
- Committee Shoots Down Florbetapir, Raising Bar for Field at Large
- Miami: Women Rock at Human Amyloid Imaging Meeting
- Miami: Amyloid in the Aging Brain—What Does It Mean?
- Miami: Multimodal Imaging, New Way to Test Amyloid Hypothesis
- Miami: Astrocytes, Antidepressants, Microbleeds, and More
- Miami: Is Human Amyloid Imaging Ready for Clinical Trials?
- Miami: HAI Amyloid Imaging Conference Abstracts
- Vandenberghe R, Van Laere K, Ivanoiu A, Salmon E, Bastin C, Triau E, Hasselbalch S, Law I, Andersen A, Korner A, Minthon L, Garraux G, Nelissen N, Bormans G, Buckley C, Owenius R, Thurfjell L, Farrar G, Brooks DJ. 18F-flutemetamol amyloid imaging in Alzheimer disease and mild cognitive impairment: a phase 2 trial. Ann Neurol. 2010 Sep;68(3):319-29. PubMed.