The multi-morbid old brain
The multi‑morbid old brain
Johannes Attems 0
0 Institute of Neuroscience, Newcastle University , Campus for Ageing and Vitality, Newcastle upon Tyne NE4 5PL , UK
-
While it is well known that the “old” brain is characterised
by the simultaneous presence of multiple
neurodegenerative pathologies, rather than by the hallmark pathologies
of one single age-associated neurodegenerative disease,
comprehensive articles putting this multi-morbidity in
context with clinical, neuropathological, genetic, and
experimental data are lacking. Therefore, this issue of Acta
Neuropathologica features two articles reviewing the
various aspects and implications of age-associated cerebral
multi-morbidity.
Spires-Jones and colleagues inform the reader about
the current evidence for interactions between different
pathological proteins and provide data from both human
brain tissue and experimental models [8]. The interaction
between amyloid-β (Aβ) and tau is clearly the best
documented one; while the “amyloid-hypothesis” in its strict
sense, i.e., Aβ “causes” tau pathology, has been disproven,
not least by the fact that tau pathology can develop in the
complete absence of any Aβ (i.e., primary age-associated
tauopathy (PART) [3]), we also know from human
studies that severe neocortical tau pathology as reflected by
Braak stages V/VI has never been described in the absence
of Aβ, which clearly indicates that Aβ has an aggravating
effect on tau pathology (in particular pyroglutamylated Aβ
[7]). However, evidence for similar interactions between
α-synuclein (α-syn) and both Aβ and tau is emerging and
TDP-43 likewise seems to aggravate AD pathology [2, 5].
It has often been suggested that the presence of
additional pathologies, such as cerebrovascular pathology
in Alzheimer’s disease (AD), lowers the threshold for
the amount of AD pathology necessary to cause clinical
dementia [1]. Kapasi and colleagues now try to clarify this
assumption and review data from large community-based
cohorts to evaluate the influence of multiple pathologies on
the clinical phenotype [6].
Despite recent advances in the development of
biomarkers for neuropathological lesions, such as Aβ and tau by
both imaging methods and cerebrospinal fluid assessment,
cerebral multi-morbidity is not accurately reflected in
clinical diagnoses. This has a detrimental impact on clinical
cohort studies as cohorts are assumed to be homogenous,
while they are in fact highly heterogenous; e.g., a clinical
AD cohort will have patients with (i) Aβ and tau
pathology only; (ii) Aβ, tau, and α-syn pathology (over 40%); and
(iii) Aβ, tau, and TDP-43 pathology (over 50%). In
addition, there will be a range of cerebrovascular pathologies
present and likely combinations of (ii) and (iii). Therefore,
the interpretation of clinical data will be highly biased and
there is an urgent need for tools that allow for a more
accurate cohort stratification based on the underlying
pathologies. To date, this can only be achieved by
neuropathological post-mortem examination as neuropathological data
can be used to retrospectively stratify clinical cohorts and
thereby facilitate the development of more accurate clinical
biomarkers. The latter will in turn lead to an improvement
in patients’ diagnostics and therapy.
Age-associated cerebral multi-morbidity challenges our
current concept of classifying age-associated
neurodegenerative diseases, which categorises cases according to the
main underlying pathology, e.g., AD equals Aβ and tau
pathology and dementia with Lewy bodies (DLB) equals
α-syn pathology. Such a concept is actually based on the
rather whimsical assumption that these diseases are
mutually exclusive. However, it should not come as a surprise
that age-associated neurodegenerative diseases are not
mutually exclusive as this would imply that the presence
of one disease protects the individual against the other.
Indeed, quite the opposite is likely to be the case as
different protein aggregates seem to aggravate each other [8].
Hence, it was not surprising when Irwin and colleagues
recently reported that 30% of clinical dementia with Lewy
bodies (DLB) cases showed full blown AD pathology in
addition to DLB at post-mortem examination, as AD is by
far the most common age-associated neurodegenerative
disease [4].
Instead of categorising cases into one main disease with
or without additional pathology, we should aim to establish
a concept that is rather based on the quality, quantity, and
topographical distribution of protein aggregates and takes
cerebrovascular disease into account. Such an approach
would not only be helpful in large clinico-pathological
correlative studies as it may unravel subtle
clinico-pathological phenotypes, but it would also provide a much clearer
and less convoluted picture of neurodegeneration in
general. Naturally, such a classification can currently only be
pursued in neuropathological diagnostics as intra-vitam
diagnosis relies on clinical symptoms, imaging, (...truncated)