3 December 2003. Raj Kalaria and Francine Gervais organized the Second International Conference on cerebral amyloid angiopathy (CAA), held December 4 to 6 in Newcastle upon Tyne, England. It was an outstanding event that focused on clinical, pathological, genetic, and animal studies related to CAA. This summary lists some of the highlights from the different speakers.
Blas Frangione gave a historical perspective on different forms of familial CAA and its pathogenesis. He and also Tamas Revesz highlighted some of the recent work on familial British and Danish dementia that develops due to the buildup of the ABri and ADan peptides in the brain in the form of CAA. In these disorders, as in several other forms of CAA, there is abundant neurofibrillary degeneration in the form of tangles as well as degenerating tau and neurites around blood vessels with CAA. Bernadino Ghetti described pathology in a number of cases of familial dementia associated with presenilin mutations, E4 homozygosity, and a rare PrP stop codon mutation. Ghetti noted that "cotton wool" plaques seen in some patients with PS-1 mutations are only weakly thio-S positive as compared with CAA. They are not associated with capillaries and have Aβ42 but not Aβ40 immunostaining. Orso Bugiani described a CAA syndrome in some Italian families with an APP E693K mutation. These patients present with a syndrome similar to the Dutch families that have an E22Q Aβ mutation (discussed by Remco Natte and Marjolijn Bornebroek) usually with brain hemorrhages, (see ARF related news story). The Dutch families frequently develop a large number of white matter infarcts due to CAA in addition to hemorrhage. Occasionally, some of the Dutch patients clinically present with dementia (not stroke) and are found to have severe CAA and white matter injury due to ischemia with small hemorrhages. Hannu Kalimo described the marked ischemic white matter lesions found in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL for short) with obliteration of long penetrating arteries. He contrasted this with the paucity of infarcts in the presence of CAA in a family with a PS-1 mutation. Peter St. George-Hyslop reviewed the genetics of AD.
Lenore Launer reported on the epidemiology of CAA based on findings from the Honolulu Asia Aging study of Japanese American men. Interestingly, diabetes and ApoE4 appeared to act synergistically to increase risk of CAA by eightfold. Paul Ince presented data from the MRC’s Cognitive Function in Ageing Neuropathology study. One suggestion from their community-based cohort is that CAA is incredibly common in the elderly and that it likely contributes to some of the deep white matter changes and possibly dementia in association with parenchymal AD pathology. Harry Vinters reviewed some of the pathological features and consequences of CAA, including hemorrhage, vasculitic changes, and leukoencephalopathy. He also mentioned a new approach to analyze brain tissue from subjects with CAA called tissue microarrays. This technique has the potential advantage of characterizing distinct reactive tissue patterns associated with CAA. Masahito Yamada examined CAA in elderly Japanese. Among other known risk factors such as ApoE4, a polymorphism in the neprilysin gene was associated with an increased risk for CAA in this population. Roy Weller described the fact that, in CAA, Aβ accumulates in the walls of capillaries and arterioles in interstitial fluid (ISF) drainage pathways, first in the vessel basement membrane. Based on beautiful morphological data from human cases, his group is assessing the hypothesis that arteriolar pulsations are the motive force for ISF (and Aβ) elimination and that reduced pulsations with cerebrovascular disease may decrease Aβ elimination and result in CAA. Alex Roher emphasized the potential role of hemodynamic alterations in AD. He noted significantly greater atherosclerosis in a large cohort of patients who died with AD vs. nondemented controls and noted that it was likely that these changes resulted in significant hemodynamic alterations.
Steve Greenberg reported that leukoariosis (a.k.a. subcortical white matter lesions) was associated with CAA as well as cognitive decline in sporadic CAA from a large series of consecutively studied patients. In addition, he reported a striking syndrome seen in a subset of CAA patients in which there is rapid cognitive decline and extensive white matter changes associated with a CAA-associated vascular inflammation. There appeared to be a clinical response to steroids in a few cases. Interestingly, James Nicoll reported the first autopsy of a patient who was involved in the Elan AN-1792 Aβ vaccine trial. There was a marked inflammatory T cell response surrounding CAA-laden leptomeningeal and cortical arterioles, suggesting a possible relationship between the inflammatory response and CAA.
Several talks focused on potential novel treatments of CAA and other amyloidoses. Francine Gervais and J. Laurin from Neurochem Inc. in Canada described GAG mimetic compounds that bind soluble Aβ. One compound, NC-758, significantly decreased both plaques and CAA in an APP-transgenic mouse model when given systemically. This orally administered compound, under the name CerebrilTM has completed phase I trials in humans and is going to enter phase II trials in humans for both CAA and AD. Einar Sigurrdson, who has worked on Aβ-immunization, presented new data suggesting that passive immunization with certain PrP antibodies appears to delay the onset of prion disease in some types of mice. Timo Erkinjuntti reviewed, and presented data suggesting that in vascular dementia, cholinesterase inhibitors appear to offer some cognitive benefit. Finally, Phillip Hawkins reviewed systemic forms of amyloidosis, the function of SAP, and the molecule CHPC, which crosslinks SAP in blood and is being used as a new approach to treat some systemic amyloid diseases (see ARF related news story).
The last talks focused on understanding the pathogenesis of CAA by using transgenic animal models. Dave Holtzman noted that CAA-associated microhemorrhage is found in different APP-transgenic mice, the critical role that ApoE plays in CAA and its consequences, and that both ApoE and ApoJ/clusterin appear to contribute to Aβ clearance and fibrillogenesis. Paul Fraser noted that AD- and CAA-like pathology started by 10 weeks of age in APP triple mutant (APPswe and APPV717F mutation) mice. When these mice were crossed with PS-1 transgenic mice, AD pathology started as early as 33 days of life. Mathias Jucker described in detail the CAA pathology and associated hemorrhage in APP23 mice as well as the results of his recent Science communication describing an increase in CAA-associated microhemorrhage in very old APP23 mice passively vaccinated with an anti-N-terminal Aβ antibody (see ARF related news story). He also reported that very old transgenic mice with the Dutch APP mutation (driven by the Thy-1 promoter) developed similar CAA-type pathology to what is seen in humans with this mutation (severe capillary involvement as well as an increase in the 40/42 ratio). Bill van Nostrand reported the development of APP-transgenic mice (with Thy-1 promoter) that contain a triple mutation including the Swedish, Dutch, and Iowa APP mutations. These mice develop extensive CAA involving capillaries (thio-S positive), as well as thio-S negative diffuse parenchymal plaques. He also discussed the potential role of uPA and MMP’s in the damage to vessels induced by Aβ.
Overall, this was an outstanding conference that illustrated some of the similarities and differences between "AD" and "CAA" pathologies, their consequences, mechanisms, and potential future treatments.-Dave Holtzman, Washington University, St. Louis, Missouri.