Have not see this on these boards but wondering if this a significant step forward?  This appears to be a "new" theory or mechanisms to explain AD and a looks like it is a bit of a divergence from the standard Amyloid Cascade Hypothesis.  Seems like we need new theories and this sounds cogent enough so one wonders if it is a large step forward?

Sorting receptor SORLA – a trafficking path to avoid
Alzheimer disease 

Summary
Excessive proteolytic breakdown of the amyloid precursor protein (APP) to neurotoxic amyloid b peptides (Ab) by secretases in the
brain is a molecular cause of Alzheimer disease (AD). According to current concepts, the complex route whereby APP moves between
the secretory compartment, the cell surface and endosomes to encounter the various secretases determines its processing fate. However,
the molecular mechanisms that control the intracellular trafficking of APP in neurons and their contribution to AD remain poorly
understood. Here, we describe the functional elucidation of a new sorting receptor SORLA that emerges as a central regulator of
trafficking and processing of APP. SORLA interacts with distinct sets of cytosolic adaptors for anterograde and retrograde movement of
APP between the trans-Golgi network and early endosomes, thereby restricting delivery of the precursor to endocytic compartments that
favor amyloidogenic breakdown. Defects in SORLA and its interacting adaptors result in transport defects and enhanced amyloidogenic
processing of APP, and represent important risk factors for AD in patients. As discussed here, these findings uncovered a unique
regulatory pathway for the control of neuronal protein transport, and provide clues as to why defects in this pathway cause
neurodegenerative disease.

Loss of SORLA activity might underlie sporadic AD
Given all the evidence that implicates SORLA in the cellular
processing of APP, an important question concerns the genetic
and pathophysiological mechanisms that predispose individuals
to low receptor activity and, hence, to increased amyloidogenic
processing that is causative of sporadic AD.

 
Because gene expression profiling in lymphoblasts has
uncovered 60% less SORL1 transcripts in AD patients
compared with controls, it has been proposed that sequence
variations in SORL1 might affect transcript levels (Scherzer et al.,
2004). In accordance with this assumption, Grear and colleagues
found a correlation between lower levels of SORL1 mRNA and
the presence of a SNP in the 59 intronic region of the gene in
cases of AD (Grear et al., 2009). Recently, we identified the
presence of two closely linked SNPs in the 39 region of SORL1,
which correlate with low expression of the receptor in the
diseased brain. These gene variations alter the sequence of the
SORL1 transcript, and result in a change from a frequent to a rare
codon usage in the SNP genotype. Further studies in cultured
cells confirmed that these SNP-containing transcripts are less
efficiently translated into protein and result in a 30% reduction in
SORLA protein levels, which is similar to the reduction seen in
AD patients (Caglayan et al., 2012).

 
However, regression analyses suggested that sequence
variations only account for as little as 14% of variations in
SORL1 mRNA levels, and therefore indicate that additional
genetic or non-genetic factors modulate SORL1 expression in the
brain (Rogaeva et al., 2007). One such factor is docosahexaenoic
acid (DHA), an essential dietary v-3 polyunsaturated fatty acid
that has been implicated in AD, because low blood levels of DHA
have been associated with an increased risk of neurodegeneration
in a large epidemiological study (Johnson and Schaefer, 2006).
DHA was shown to markedly upregulate expression of SORLA
in primary neurons and in mouse models, arguing that the
increased disease risk when DHA levels are low might be
explained by defects in induction of SORL1 transcription (Ma
et al., 2007). Another protein that modulates SORLA levels in the
brain is brain-derived neurotrophic factor (BDNF), a growth
factor that provides trophic support to neurons (Chao, 2003).
BDNF potently activates transcription of the SORL1 gene by
more than tenfold through the extracellular-signal-regulated
kinase (ERK) pathway and yet unknown downstream
transcription factors (Rohe et al., 2009). It is possible that a
poor trophic support to neurons, as shown in chronic insults to the
nervous system, is deleterious through the associated loss of
SORL1 expression (Matrone et al., 2008; Rohe et al., 2009).
Because SORLA and its interacting adaptors have a pivotal
role in the trafficking of APP, it comes as no surprise that
additional components of this trafficking machinery emerge as
genetic risk factors in AD. Foremost, this applies to the
pentameric retromer complex, which is composed of VPS35,
VPS29, VPS26 and the sorting nexins 1 and 2 (reviewed in
Seaman, 2012). VPS35 and VPS26 are poorly expressed in the
brain of AD patients (Small et al., 2005), and disruption of Vps35
in mice results in memory deficits and synaptic dysfunction,
which are associated with elevated Ab levels (Muhammad et al.,
200. As retrograde transport of SORLA, which is mediated by
retromer, is important for processing of APP (Fjorback et al.,
2012), impaired retromer function, might be an underlying cause
of neurodegenerative processes (Nothwehr et al., 2000). In a
similar manner, decreased levels of GGA3 are observed in brain
autopsies of AD patients (Santosa et al., 2011). Moreover, the
loss of GGA3 expression enhances b-secretase activity (Tesco
et al., 2007), suggesting that impaired anterograde transport of
SORLA is an additional cause of neuronal dysfunction.

 
Conclusions
Recent years witnessed the detailed elucidation of cellular
trafficking pathways that control APP processing, and how they
contribute to neurodegenerative processes in patients. In
particular, the functional characterization of SORLA, a unique
sorting receptor for APP, sheds light on the targeted transport of
APP to distinct neuronal compartments that harbour the various
secretase activities – a process that was previously poorly
understood. Results from histopathological and epidemiological
studies further substantiate the crucial role of this receptor in
sporadic AD. Equally exciting is a recent report that mutations in
SORL1 might even be the cause of autosomal dominant forms of
early-onset AD (Pottier et al., 2012). Here, five missense
mutations, one of which maps to the APP binding domain in
SORLA, have been identified in familial cases of AD. Although
much still needs to be learned about this sorting pathway, it
certainly holds great promise as a genetic risk factor and new
biomarker for diagnosis, and perhaps it might even constitute a
druggable target for the treatment of AD.

http://jcs.biologists.org/content/126/13/2751.full.pdf+html

So it looks like one can get a hold of DHA quite easily and also appears to be supplements with BDNF available.  Do we add these to the regimine?