Before the business of 2018 consumes all your brain power, take some time to reflect on the major findings of the last year. We have compiled some of our favorite stories from 2017 into seven major themes, ranging from therapeutics in clinical trials to fundamental molecular biology. Be forewarned, there was lots of exciting news in 2017, so grab a seasonal brew, settle back in a comfy chair, and give yourself time to soak it all in. Alternatively, follow the tool bar above the story title to save and print a PDF that you can take on that plane or train journey home.   


Confronting failure in trying to stem symptomatic AD, the field’s main thrust has turned toward retooling its drug trials for ever-earlier disease stages (Dec conference news). To sustain enthusiasm among participants and sites, scientists were advised to focus on learning from failure rather than conveying a sense of nihilism, both in internal discussions and in speaking with reporters (Dec conference news). 

They got to practice a positive attitude when another setback hit home. Merck announced an end to the EPOCH mild to moderate AD trial of the BACE inhibitor verubecestat for lack of efficacy (Feb news), and data released later indicated that the drug had nudged down amyloid plaques without a hint of benefit, even in the more mildly symptomatic participants (Dec conference news). Even so, BACE inhibitors are very much alive and being evaluated to a collective tune of billions of dollars. Researchers believe EPOCH treated people too late, when they’d had brain amyloid for years and neuron loss was well underway. The hope now rests on Phase 3 trials of verubescestat, Eisai’s elenbecestat, and Lilly/AstraZeneca’s lanabecestat in people with mild AD (May conference news). Ongoing prevention trials, including Generation 1, Generation 2, and EARLY, are testing BACE inhibitors from Novartis and Janssen even earlier, that is, in asymptomatic people who are amyloid-positive or at genetic risk of AD.

In trials of anti-amyloid antibodies, the place to go in 2017 was up. The A4 trial joined the trend in Alzheimer’s immunotherapy when it quadrupled the solanezumab dose and extended treatment time to five years (Jun news). Solanezumab had shown hints of efficacy in its negative Phase 3 trial, suggesting a higher dose might work. Trialists had already quadrupled the crenezumab dose in the Phase 3 CREAD trial (Dec 2016 conference news), and Roche and Biogen are following suit with gantenerumab and aducanumab (Dec conference news; Dec conference news). 

The neuroAD therapy system from Neuronix combines transcranial magnetic stimulation with computerized cognitive training. [Courtesy of Neuronix.]

In a rare apparent success, a medical device for AD treatment cleared Phase 3. The neuroAD system, which combines repetitive transcranial magnetic stimulation (rTMS) with cognitive training, claimed to have maintained cognitive abilities in people with mild AD over three months, which is typically the time frame for Phase 2 trials in AD drug evaluation. The therapy is approved in Europe, and its sponsor, Neuronix, Israel, has applied for marketing clearance from the U.S. Food and Drug Administration (Apr conference news). Basic research supported the idea that mnemonic training may help with memories (Mar news). 

Antisense oligonucleotides (ASOs) built momentum after approval late last year of the first such drug to treat spinal muscular atrophy (Nov 2016 news). In a Phase 3 trial, Ionis-TTRRx knocked down expression of pathologic transthyretin and stalled familial amyloid neuropathy, albeit with side effects (May news). Ionis’ ASO for Huntington’s disease reduced mutant huntingtin in CSF, and was well tolerated in Phase 1 (Dec news). Approaching Alzheimer’s, Ionis published preclinical data on an ASO targeting tau (Jan news). 

Prominent Phase 3 flops this year included 5-HT6 serotonin receptor antagonists. Intepirdine’s failure followed on the heels of idalopirdine’s (Sep newsAug conference news). 

Phase 2 Promise, Woes 
While dangerous side effects have led physicians to cut back on antipsychotics in advanced AD (Gurwitz et al., 2017), the drug pimavanserin offered new hope. In a Phase 2 study, this serotonin 5-HT2A receptor antagonist safely reduced psychotic symptoms (Dec conference news). Pimavanserin was approved for psychosis in Parkinson’s in 2016.

Several tau immunotherapies cleared safety studies this year, including Genentech’s RO7105705, AbbVie’s 8E12, and BIIB092. Formerly known as BMS-986168, the latter was snapped up by Biogen (Aug conference news; Dec conference news). For target engagement, BIIB092 lowered CSF free tau by more than 90 percent, and is entering Phase 2 testing in both progressive supranuclear palsy and AD.

Two different α-synuclein antibodies advanced to Phase 2 (May conference news). Researchers desperately want biomarkers for the next round of α-synuclein trials, and the race is on for PET tracers that will detect it. This work currently plays out on the Parkinson’s front, but tracers and therapeutics for this protein will come in handy in Alzheimer’s and dementia with Lewy bodies, as well, as they will help scientists dissect the significant overlap of pathology and symptoms across the AD-PD spectrum.

Attempts to target amyloid via Aβ oligomers advanced when the glutaminyl cyclase inhibitor PQ912 appeared safe over three months, with EEG and biomarker data hinting at a synaptic benefit. More dose-finding is needed to find a sweet spot between target engagement and tolerability (Jun news). Another molecule, CT1812, may have signaled synapse protection in an elderly volunteer study and is entering Phase 2, where one trial will incorporate PET imaging with the emerging SV2A marker of synaptic integrity (Dec conference news; Dec conference news). 

Inspired by experiments in which plasma from young mice boosted memory in AD mice (Middeldorp et al., 2016), researchers showed it’s safe to try the same in people. A Phase 1 study giving plasma from young men to older folks with dementia had few adverse effects. It keyed up a proprietary fraction of 400 plasma proteins for Phase 2 testing to see if it improves cognition and function (Dec conference news).

December brought a reality check for the Biogen/Eisai Aβ antibody BAN2401. It missed the primary endpoint in a creative Phase 2 adaptive trial design, which uses Bayesian statistics in hopes of getting a result well before the full trial has run to its conclusion (Dec news; for more on Bayesian design see Mar 2011 news). 

Trial Logistics
The year saw the demand for volunteers to fill dementia trials explode, with an estimated 55,000 people needed to complete existing symptomatic and secondary prevention trials (Cummings et al., 2017). The pressure is on to recruit. Online registries such as the Brain Health Registry, the Alzheimer’s Prevention Registry with its GeneMatch ApoE genotyping program, EPAD, and GAP, are all refining ways to attract, retain, and refer cognitively healthy older adults in sufficient numbers to keep research moving (Dec conference news). A new trials consortium called ACTC intends to innovate around recruitment and outcomes in AD prevention trials (Dec news). 

Meanwhile, scientists are already at work on the first primary prevention trial for AD. The Dominantly Inherited Alzheimer Network trials unit (DIAN-TU) will enroll young adults who carry a familial AD mutation but have no detectable amyloid plaques on PET scans yet. Participants will take the JNJ-54861911 BACE inhibitor for several years to suppress Aβ production and prevent plaque accumulation (Aug conference news). The trial promises the most definitive test to date of the amyloid hypothesis.

In clinical news on related diseases, the free radical scavenger edaravone became the first treatment to be approved for ALS in the U.S. since 1995 (May news), while deutetrabenazine, a symptomatic treatment for Huntington’s chorea, got the nod for improving on its predecessor tetrabenazine. The deuterated version breaks down more slowly and requires lower, less-frequent dosing (Apr news). 



Biomarkers will continue to be a research priority until they are solidly in place as routine features of AD diagnoses and trials, and 2017 saw strides toward that end. Researchers improved the standardization of CSF Aβ and tau measurement with automated CSF assays that vary less between runs and can predict clinical progression in cognitively normal people (Dec conference news). Researchers validated Aβ and tau biomarker cutoffs that were determined in one AD cohort in a second, independent cohort—a key step in developing standardized cutoffs for diagnosing AD (Apr conference news). The year closed with the debut of certified reference solutions of CSF Aβ42 to calibrate assays worldwide (Dec news). 

Ultimately, clinicians prefer to use blood over CSF, and this year saw the first signs that this may be possible. A mass spectrometry method reliably detects a 15 percent drop in the Aβ42/Aβ40 plasma ratio in people who have brain amyloid, and an independent ELISA-based blood test generates similar results (Jul conference news; Dec conference news). The link between brain and blood Aβ drew further support from a genetics study. Carriers of the protective A673T APP mutation, which reduces Aβ production in the brain, also have 20 percent less of the peptide in blood than do noncarriers of the variant (Jun news). Trialists all over the world seek a blood-based indicator of brain amyloid deposition to help them cut down on the number of expensive amyloid PET scans currently needed to recruit for secondary prevention trials. Besides blood tests of Aβ itself, polygenic risk scores—calculated from the subset of all known GWAS AD variants a person has inherited—are beginning to look as if they might subserve this purpose, as well (Dec conference news)

While CSF Aβ and tau sharpen AD diagnosis, these markers are blunt on progression (Aug conference news). Enter neurofilament light. NfL rises in blood during preclinical and prodromal stages of AD, correlating with degeneration and cognitive decline (Mar news). Blood NfL flags severity in most other neurodegenerative disorders (Mar newsFeb news), and in people with multiple sclerosis, NfL falls during immunotherapy, suggesting it could work as a surrogate outcome marker (Nov news). 

PET amyloid imaging holds potential for both diagnosis and staging. A staging scheme based on regional amyloid deposition identified nascent buildup that fell below the radar of global PET measures, suggesting this approach could pick out people in the earliest phase of the disease (see image below, Oct news). The default mode network, which activates during daydreaming and memory retrieval, appears to be where amyloid pathology settles first (Nov news). ADNI participants with elevated brain Aβ, by PET or CSF, were far more likely to decline cognitively in the following decade than were people with normal baseline Aβ (Jun news). 

Amyloid Stages. Regional progression of Aβ deposition, as proposed by the four-stage model. Each stage includes affected regions from the previous stage (blue) plus newly affected regions (red). [Courtesy of Grothe et al., Neurology, 2017.]

Amyloid PET scanning is entering clinical practice and could revamp patient care. First data are in from the U.S. Imaging Dementia—Evidence for Amyloid Scanning (IDEAS) study to assess whether amyloid PET scans do the participating 4,000-plus Medicare beneficiaries any good. They show a bigger-than-expected effect, whereby scans prompted changes in treatment recommendation for two-thirds of participants and changes in diagnosis for one-third (Aug conference news). In the U.S., IDEAS data may support insurance coverage for the scans; in Europe, smaller studies are reporting much the same result. In Australia, a PET center provides sponsor-paid amyloid scans to local clinicians as a way to substantiate diagnosis and accelerate trial recruitment (Dec conference news). 

Tau PET tracers have yet to gain regulatory approval. While some early investigational tracers fell to off-target binding, a handful of next-generation ligands appear to have solved technical challenges, and the data across tracers and other AD markers are starting to match up well (Apr conference news). Tau pathology shows up in the entorhinal cortex in people with subjective memory complaints, an early sign of AD. Tau PET correlates well with cognition and predicts regional degeneration (Oct newsMar news; Aug conference news). Researchers are excited that the rate at which tau pathology builds up better predicts future memory problems than do baseline tau levels or amyloid imaging (Dec conference news). 

Because tau beats amyloid for measuring disease progression, tau imaging might also help monitor treatment response. Some researchers calculate that picking trial participants based on the amount of their tangle pathology would halve the number needed to see a drug effect (Mar news). 

Consequently, tau tracer development is off to the races. For their part, drug developers suddenly have options. In addition to flortaucipir, the known tracer from Lilly, and PI-2620, the new tracer by Piramal, they can now look to a consortium called Cerveau Technologies for a third, emerging tracer, setting up 2018 as a bumper year for tau PET inclusion into drug trials across the board (Dec conference news). 

Trial Ready? Piramal’s PI-2620 shows a typical tau distribution pattern with varying degrees of accumulation in three amyloid-positive people with probable AD enrolled in the Proclara NPT088 trial. [Courtesy of Andrew Stephens.]

Other targets in AD pathology are finally opening up to PET scanning, too. For example, UCB-J, a derivative of the anti-epileptic drug levetiracetam, joined the scene as a potential marker of synaptic density when, in a first study of AD patients, it showed less hippocampal binding than in controls (Dec conference news). 

Given this panoply of investigational biomarkers, each of which is most dynamic at a different stage of disease, researchers are starting to build models in which combining markers boosts their ability to predict an individual person’s risk of progressing to MCI or AD within the next one or three years (Aug conference newsOct news).

In ALS and FTD, CSF levels of polyglycine-proline protein identified carriers of the expanded C9ORF72 repeat that leads to these diseases (Mar news), and traumatic brain injury turned out to be detectable by plasma tau (Sep news). 



Pretty Pair.

Stacks of symmetrically paired C-shaped protofilaments generated the tau paired helical filaments found in an Alzheimer’s brain. [Courtesy of Fitzpatrick et al., Nature 2017.]

Behold the wonders of cell biology! The year 2017 saw extraordinary progress on how neurons manage the shapes and amounts of aggregation-prone proteins packed inside them. We group selected highlights into new structures of troublesome proteins, the mystery of neurodegenerative proteins coalescing into liquid droplets, and strange new ways by which neurons jettison protein ballast.

Stunning Structures
Pathological accumulation of aggregated proteins marks all neurodegenerative diseases. To find out how proteins aggregate in the first place, scientists like to look at them. Two thousand seventeen was a banner year for structural biology, which graced the field with dazzling new pictures of long-known culprits. Attribute the windfall in part to the technical maturation of cryo-electron microscopy, in which scientists rapidly freeze protein samples, blast them with electrons, and employ computational acrobatics to transform the diffraction patterns into atomic-resolution three-dimensional structures. The technique’s wizards garnered this year’s Nobel Prize in chemistry (Oct news). 

Cryo-EM yielded the year’s most striking exposé: a 3.4Å structure of tau fibrils from the brain of a person with AD (Apr conference news; Jul news). C-shaped structures, each formed by a molecule of tau, bound in pairs to form individual rungs along the filament. Tau’s third and fourth microtubule binding repeat domains (R3 and R4), plus an additional 10 amino acids C-terminal to R4, comprised the C-shaped core, while the rest of the protein dangled loosely away from the fibril axis in a nebulous “fuzzy coat.” 

Double Zipper.

By virtue of two steric zippers (boxed interfaces A, B), any one tau molecule (turquoise) can bind two others (green). The highly stable structure may then form fibrils (denoted here by stacking). [Courtesy of Seidler et al., Nature Chemistry, 2017.]

The findings suggested that efforts to dismantle tau aggregation might do well to target the R3/R4 core, though another cryo-EM structure blamed tau’s R2 domain for fibril formation (Dec news). In a 1.5Å structure of the VQIINK hexapeptide heading the R2 domain, researchers saw steric zipper interfaces they predicted would form fibrils. Inhibitors developed to unzip them indeed stymied aggregation of full-length tau in biosensor cell lines, but it remains to be seen how VQIINK’s structure and inhibitors relate to tau aggregation in the brain. 

Cryo-EM also offered an unprecedented view of Aβ fibrils (Sep news). A 4Å structure of synthetic Aβ42 fibrils was the most detailed to date, and the first to resolve the peptide’s wiggly N-terminus. This fibril sported a unique dimer interface, and a tilt to the subunits suggested a novel mode of fibril growth.

Plaque Patterns.

Fluorescence of LCO dyes at 502nm and 588nm suggests unique structures for each plaque, represented here by colored symbols. Signatures for each AD subgroup (see legend) were different, yet overlapped. [Courtesy of Rasmussen et al., PNAS 2017.]

Structural biology in 2017 opened up a broader view of fibril conformations. NMR revealed that a single Aβ40 fibril structure predominated in people with typical AD or its variant posterior cortical atrophy (PCA), while a mixture of Aβ40 fibril structures was present in people with a rapidly progressive form of AD (Jan news). A new class of fluorescent dyes called luminescent conjugated oligothiophenes (LCOs) enabled scientists to look directly at Aβ fibrils at the heart of plaques in AD brains. They saw “clouds” of similar variants among people with the same familial AD mutations or AD subtypes (Aug conference newsNov news). 

The findings link structural variations in Aβ fibrils to clinical manifestations of AD, and offer a humbling reminder that one structure does not speak for all fibrils. Details about relationships between different types of fibrils emerged, as well. Researchers reported that dystrophic neurites caused by Aβ neuritic plaques coax tau to form oligomers, which might seed neurofibrillary tangles (Dec news). Placing structural studies of fibrils in perspective, the work emphasized a pathogenic role of soluble oligomers, which evade structural analysis. Along those lines, one report claimed that for Aβ, the smallest oligomers were the most toxic (Jan news). Another used Fourier transform infrared microspectroscopy (μFTIR) to spot the initial appearance of small Aβ aggregates that have a β-sheet structure, and could be the forebears of fibrillar plaques (Mar news). Regarding pre-fibrillar aggregates of tau, one study drew notice for showing that the RNA-binding protein and stress granule component TIA1 stabilizes oligomers of tau and prevents their fibrillization; this intensifies toxicity in a transgenic mouse (Nov news). 

Drip Drop.

Mix tau and RNA together, and droplets form. They merge together (left panel, red circle) and contain tau (right, green). [Courtesy of Zhang et al., PLOS Biology, 2017.]

Phase Transitions
The idea that molecules condense into liquid droplets that control cellular processes took the neurodegenerative disease field by storm in 2015, and a flood of phase transition findings continued unabated in 2017. Several groups reported that tau undergoes liquid-liquid phase separation thanks to interactions with negatively charged RNA and/or tau’s own phosphate groups (May conference news; Jul news; Aug news). Others found that tau droplets might fertilize the growth of microtubules (Sep news). 

Shooting Stars? Added to tau droplets, tubulin polymerizes into microtubules, wrenching drops into liquid strings. [Courtesy of Hernandez-Vega et al., Cell Reports, 2017.]

ALS/FTD proteins joined the ranks. The RNA-binding protein FUS forms droplets, yet phosphorylation appears to have the opposite effect on FUS as it does on tau, as it broke up droplets rather than facilitating them (Aug news). Dipeptide repeats (DPRs) translated from hexanucleotide expansions in the C9ORF72 gene also condense into droplets, as do CAG-repeat laden RNA molecules (Mar news; Jun news).

A structural study addressed fundamental mechanisms of phase separation. Many proteins form droplets via interactions between their low-complexity (LC) domains. As the name suggests, LC domains were thought to lack structure, and they share little sequence similarity with proteins that form rock-solid amyloid fibrils. Even so, NMR of FUS’s LC domain revealed a highly structured core (Sep news). Like liquid droplets themselves, fibrils formed from this core easily fell apart in the face of phosphorylation or other electrostatic changes. 

Bad Hair Day? In FUS-LC fibril, disordered N- and C-terminal segments flap freely around a stable amyloid core. [Courtesy of Cell, Murray et al.]

The basic science findings have stirred up the field, but whether they translate to neurodegenerative disease remains to be seen. 

New Ways to Take Out the Trash
Once aggregates have formed, how do cells get rid of them? In 2017, researchers suggested some wild new ways. One, called MAGIC (mitochondria as guardian in cytosol), connects mitochondrial function and protein disposal, both of which falter in neurodegenerative disease. Aggregated proteins were seen lining the surface of mitochondria, where chaperones disentangle and transport them into the mitochondrial matrix to feed them to resident proteases (Mar news). 

Munching Mitos. Protein aggregates congregate on the mitochondrial surface, where Hsp104 unwads them and packs them off into the mitochondrial matrix for degradation. [Image courtesy of Chacinska, Nature N&V, 2017.]

If protein disposal within mitochondria isn’t peculiar enough, another group told of neurons spitting out toxic “exophers.” These are giant vesicles chock-full of aggregated proteins and entire mitochondria (Feb news). This dramatic purging, demonstrated thus far only in C. Elegans, appeared to make neurons feel  better.

Let It All Out. A neuron labeled with aggregating florescent mCherry (white) expels much of the protein in a bright exopher, leaving a relatively dim soma behind. [Melentijevic et al., 2017. Nature.]

Yet another baffling bit of cell biology in 2017 was the discovery of proteasomes on the neuronal cell surface. With the ends of this protein grinder reaching into both intra- and extracellular environs, these neuron-specific proteasomes reportedly process intracellular proteins and spit peptides into the extracellular space. These peptides, in turn, activate neurons by triggering NMDA receptors. The findings bestow a new function to proteasomes in neurons (Mar news). 

Plasma Proteasome. Three potential models for how neuronal membrane proteasomes function within or at the plasma membrane. [Image courtesy of Ramachandran and Margolis, Nature Structural and Molecular Biology, 2017.]



Whatever doubt might have lingered out there about microglia’s role in Alzheimer’s was put to rest when scientists fingered a protective polymorphism near the gene for a major microglial transcription factor. Called PU.1, it controls myriad responses, including expression of known AD genes TREM2, CD33, MS4A, and other AD GWAS hits. This protective variant reduces PU.1 expression, lowers amyloidosis, and delays onset of AD (Jun news). 

Researchers also associated late-onset AD with rare coding variants in TREM2 and variants in two other genes highly expressed in microglia—phospholipase Cγ2 gene (PLCG2) and the ABI family 3 gene (ABI3) (Aug conference news). 

DAM Plaque Eaters. In mice, microglia (red) surrounding Aβ plaques (gray) express CD11c (green), a marker of disease-associated microglia. [Cell, Keren-Shaul et al. 2017.]

An important insight of 2017 was that science needs to come to grips with the different populations and activation states, i.e. “multiple personalities,” that exist within the broad term microglia. For starters, single-cell transcriptomics identified a subset of disease-associated microglia around plaques (Jun news). These DAM microglia consume Aβ deposits (image above). TREM2, APOE, and RIPK1 kinase were shown to drive microglial transcriptional phenotypes that associate with neurodegeneration (Sep newsSep news). How these different phenotypes relate to each other, and what they do in neurodegeneration, needs to be defined.

A 2017 highlight came with the unveiling of the first comprehensive analysis of the human microglial transcriptome (Jun news). Alas, microglial transcriptional profiles were found to change profoundly when the cells were placed in culture (Jul news). While the human microglial transcriptome mostly matches that from mice, microglia from mice and men age differently. In mice, half of microglia live for their host’s entire life, but human microglia, with their average 4.2 year lifespan, are replaced many times over (Aug news). Also, glia evolve with their host’s age in that cells in different brain areas modulate the age-related expression of specific types of genes (Jan news). 

In a bizarre twist, 2017 ended on news that microglia not only help clear amyloid plaques, they may also help seed them (Dec news). Some activated microglia spew bundles containing a protein called “apoptosis-associated speck-like protein containing a caspase-recruitment domain.” These protein globs, aka ASC specks, power inflammatory cascades and also latch onto Aβ, driving plaque assembly. ASC specks were spotted in AD plaque cores 

Core Problem.

An ASC core (red) in an amyloid plaque (green) in hippocampus of AD brain. [Courtesy of Venegas et al, Nature Neuroscience, 2017.]

More evidence for the many ways in which microglia can bear malice came in a study where switching on ERK kinase in just 10 percent of microglia sufficed to cause severe neurodegeneration in wild-type mice (Sep news). In blood and immune cells, somatic mutations in ERK signaling cause leukemia. Some carriers also get a neurodegenerative disease, and scientists now think that might be wrought by microglia.

And activated microglia do not act alone. They appear to push resting astrocytes into a reactive A1 state. Astrocytes in this aggravated state weaken synapses and kill neurons as well as oligodendrocytes. Since scientists have newly managed to coax progenitor cells to mature into astrocytes in vitro, the field now has a better handle on these elusive glia, as well (Jan news; Aug news). Despite these advances, however, 2017 ended on a deeply sad note for glia aficionados. Ben Barres, the scientist who did more than anyone to alert the AD field to the importance of astrocytes in neurodegeneration, passed away on December 27, at age 63, after an intensely fought battle with cancer. 

In 2017, researchers doubled down on the TREM2 receptor on microglia, motivated by the nearly two dozen genetic variants currently known to increase a person’s risk of FTD, AD, and possibly ALS and PD. TREM2 responds to brain insults by jolting microglia out of a homeostatic state into clean-up mode; the receptor also maintains microglia's metabolic fitness (May news; Aug news). 

Still, there is much still to learn yet about TREM2 in microglia. That both receptor and cell assume different roles at different ages and different disease stages has complicated matters. For example, TREM2 signaling activates microglia, tempers tau aggregation in early stage tauopathy in mice, and accelerates neurodegeneration later on (Oct news). Confused yet?

Unsurprisingly, TREM2 is being eyed as a drug target. Three groups independently mapped the site where proteases shed the soluble extracellular portion of TREM2, raising the notion of blocking that process. One caveat is that soluble TREM2 promotes microglial survival (Aug newsFeb news). 

Systemic immunity/inflammation
Brain-resident microglia are not the only immune cells whose link to neurodegeneration tightened in 2017. Blood monocytes were found to ramp up expression of inflammatory genes in people with ALS, especially in rapidly progressing forms (May news). Three proteins implicated in blood inflammatory responses—ITGB2, LILRA2, and CEBPD—correlate with ALS progression, and certain types of peripheral immune cells proliferate in lockstep with worsening ALS motor symptoms (March newsSep news). What turns these cells on, and whether they hasten harmful neuroinflammation in the CNS, remains unanswered.

Evidence that the peripheral immune system has lasting effects on the brain surfaced in a 24-year prospective study that linked high levels of systemic inflammation markers in middle age to late-life losses in memory and brain volume, including in areas associated with AD (Nov news).



The gradual sickening and eventual death of neurons defines neurodegenerative disease, but how exactly do disease-related proteins do this to neurons? A single theme did not emerge from this line of research in 2017; rather, it seems toxic proteins have an arsenal of weapons at their disposal.

Neuritic tau.

Tau (green) accumulated within axons (red), adjacent to neuritic plaques (blue). [Courtesy of He et al., Nature Neuroscience.]

For tau, a clue to the protein’s toxicity came in where it is made. In healthy neurons, tau goes about its business of stabilizing axonal microtubules, and its translocation into dendrites—something commonly observed in degenerating cells—botches synaptic transmission. New research this year revealed that tau does not necessarily travel from axons to dendrites; rather, it gets made there. Strikingly, Aβ oligomers may instigate the process (Jul conference news). Aβ aggregates may provide a breeding ground for tau oligomers, as Aβ-laden dystrophic neurites were shown to house these soluble tau aggregates (Dec news)

Tau appears to mess with all manner of cellular functions. Researchers implicated toxic variants in mitochondrial dysfunction, bungling synaptic vesicle release, disrupting the nucleus, compromising the epigenome. No one mechanism rose to the fore, however (Apr conference news; Dec conference news). 

Vesicle Interloper?

Nanogold-labeled recombinant human tau clings to synaptic vesicles (SV) isolated from the rat brain. [Image courtesy of Joseph McInnes.]

Glial responses to tau pathology worked to the detriment of neurons. Researchers reported that ApoE4 jacked up the microglial response to tau pathology, which may turn astrocytes from good to bad (Apr news). 

On a broader level, research cemented the status of tau as a toxic factor in the brain. Tau pathology in the medial entorhinal cortex got tied to loss of excitatory neurons, disruption of the “grid cells” that control spatial navigation, and impairment of spatial memory in mice (Jan news). Tangles were tied to cognitive decline and blamed for excitotoxic cell death after stroke (Aug conference news; Oct news). 

Tau and Cognition. When AD patients (top row) voice concerns about their own cognitive abilities, they deposit tau in frontal areas of the brain. When family members notice a memory change, tangles appear more posteriorly (bottom row). [Courtesy of Shannon Risacher, AAIC2017.]

Two thousand seventeen also dished out more damning evidence against Aβ. Scientists reported that Aβ prevented dendritic spines from forming (Oct news). On the pre-synaptic side, a proposed way by which Aβ accumulation in axonal terminals leads to dystrophic neurites involved failure of a dynein motor transport pathway and build-up of BACE1 in axonal terminals, triggering Aβ accumulation there. Restoring BACE1 retrograde transport spared synapses and improved cognition in hAPP mice (Mar news). 

Nuclear Breakout.

In control motor neurons (left), the U2 splicing complex (red) occupies the nucleus, but in C9ORF72 motor neurons (right), much of it lurks in cytoplasm. [Courtesy of Cell Reports, Yin et al.]

On the perennial question of which species of Aβ causes most damage, the smallest of all possible oligomers, a soluble dimer, came in for scrutiny in 2017 (Jan news). And humans may be particularly sensitive to Aβ’s toxic influence. Human neurons transplanted into an AD mouse model withered and died, while their murine counterparts persisted despite Aβ pathology (Feb news). 

This year also saw new mechanisms for proteins implicated in other diseases, including FTD/ALS. One surprising paper showed hexanucleotide expansions in the C9ORF72 gene form DNA-RNA hybrids that broke DNA, and the dipeptide repeats (DPRs) translated from the expansions derailed the DNA repair machinery necessary to fix the damage (Jul news).

C9ORF72 DPRs also mangled RNA splicing by whisking U2, a key component of spliceosomes, out of the nucleus and away from its freshly transcribed, would-be clients (Jun news).

Synaptic Feast.

Microglia (green) lacking TDP-43 (cKO) are more phagocytic and contain higher levels of engulfed synaptic markers (PSD95, red) than intact microglia (WT). [Courtesy of Neuron, Paolicelli et al.]

Researchers exposed new pathogenic functions for TDP-43. This normally nuclear RNA-binding protein is known to wind up in cytoplasmic inclusions in ALS neurons, and earlier this year was found to associate with ribosomes there. The toxic relationship led to global inhibition of protein synthesis (Feb news).

TDP-43 was found to dampen microglia’s appetite for Aβ. Deleting or hobbling TDP-43 function ramped up microglial phagocytosis, to better clear Aβ plaques. The downside? The voracious microglia also devoured synapses. Perhaps dysfunctional microglia are responsible for neurodegeneration in ALS/FTD and other TDP-43 proteinopathies? The finding could also explain why people with TDP-43 pathology have a lower incidence of AD (Jul news). 



Vascular dementia research used to be a slow backwater relative to the flow of data every year on AD, but 2017 was different. Researchers made inroads into the physiology underlying this disease, for example by toppling a long-held dogma with their demonstration that the human brain does have a lymph system. Running through the dura mater, the vessels may provide a route for immune or other cells to exit the brain (Oct news). The finding comes two years after a dural lymph system was discovered in mice. Continuing this year, the rodent studies reported that the dural lymph vessels drain cerebrospinal fluid from the brain into the blood stream (Nov news). 

Look—Lymphatics! MRI reveals lymphatic vessels (green) in the dura of a healthy 47-year-old woman. [Courtesy of Absinta et al., eLife.]

The findings paint the meninges as a pivotal immune structure in the brain. Case in point: A high-resolution map of immune cells in healthy mouse brains revealed a menagerie of peripheral blood infiltrators, including myeloid and lymphoid cells, dendritic cells, monocytes, macrophages, T and B cells, and natural killer cells, all spotted in the meninges and choroid plexus (Jul news). 

Besides the excitement about lymphatics, there was buzz about the regulation of blood flow in the brain. Pericytes, the tiny cells that line capillaries, generate outsize controversy on their role in regulating vascular constriction and brain blood. This year, animals with fewer pericytes were found to have less blood flow and less oxygen in the brain, though others pinned smooth muscle cells lining larger vessels as the master controllers (Feb news).

Part of this controversy stems from disagreement over what constitutes a pericyte and how to detect them. A new dye may help settle the issue because it appears to selectively label a set of non-contractile cells on capillaries (May news). 

Perivascular macrophages that line the brain’s blood vessels also play a hand in regulating flow, and this year science showed them to do more harm than good where Aβ pathology is in play. The cells release a burst of reactive oxygen species in response to Aβ, which slows blood flow (May news). This could explain how Aβ diminishes neurovascular coupling, i.e., the changes in blood flow occurring in response to neuronal signaling, and may stymie Aβ clearance from the brain.

Adding insult to injury, researchers found that microinfarcts shut down local clearance of Aβ from the brain, at least in mice (Mar news). These tiny, “silent” strokes are known to occur in people with AD, and the findings suggest they could hasten Aβ buildup by blocking clearance.

Impaired Clearance. Inducing microinfarcts in the mouse brain weakens glymphatic function throughout the brain (green) compared to controls. [Reprinted with permission: Wang, et al. The Journal of Neuroscience 2017.]

While the neurovascular system receives much attention as an exit route for Aβ from brain, Aβ may also use the vessels to hitch a ride into the brain. In 2017, a parabiosis study showed how Aβ from one mouse made its way through the vasculature into the brain of another, where it reportedly formed parenchymal plaques surrounding vessels, and caused cerebral amyloid angiopathy and neuroinflammatory damage (Nov news). 

Human studies cinched the link between vascular problems and dementia. Compared to people with healthy vascular systems, 50-year-olds with two or more vascular risk factors had almost triple the risk of harboring Aβ deposits in their 70s (Apr news). Might this unholy alliance start even earlier in life? In women, but not men, one study linked hypertension in the 30s and 40s to increased dementia risk later on, though experts in the field attributed the sex effect to methodological bias (Oct news).

Holes Herald Trouble. Large perivascular spaces (white arrowheads) appear around blood vessels in the basal ganglia (left) and white matter (right). [© 2017, American Medical Association. All rights reserved.]

Biomarker data further raised the profile of vascular dementia in 2017. Enlarged spaces around blood vessels emerged as a promising MRI marker of this disease, while a new five-year project called MarkVCID began hunting for better markers more broadly (Jul newsMar news). 

Understanding how the neurovascular system both fends off and facilitates neurodegeneration is a tall order, given its complex architecture and the tiny size of many vessels. In a step toward modeling the system, researchers created an artificial blood vessel, complete with endothelial cells, smooth muscle cells, and astrocytes. They reported that Aβ injected into these cell-lined tubes aggregated there, and was ferried into the lumen by ApoE and high-density lipoproteins (Oct news). 



All things considered, it was a growth year for dementia research. The number of dementia researchers in the U.K. doubled, while worldwide dementia publications rose 40 percent (Mar news). In two major research initiatives, the National Institute on Aging awarded $70 million over five years to establish a new flagship clinical trials network, the Alzheimer’s Clinical Trials Consortium (ACTC). It also invested in a new research center to develop animal models of late-onset AD, dubbed the Model Organism Development and Evaluation for Late-Onset AD. The center will exploit common genetic risk variants for LOAD (Jan news). Late in the year, Microsoft co-founder Bill Gates pledged $100 million to fund new therapeutic approaches and start-ups (Nov news). 

On the downside, earlier in the year the White House proposed a drastic decrease in NIH funding, which could potentially derail recent efforts to back more research into AD and related dementias (Mar news). The future remains uncertain as Congress has yet to pass a budget and continues to operate the government on a temporary funding bill that expires on January 19.

On the policy front, the scientific evidence is still considered insufficient to warrant sweeping recommendations on AD prevention. In the U.K., a Lancet report concluded that up to 35 percent of dementia cases are preventable. But short of calling for a public information campaign, and recommending nothing more than treatment of midlife hypertension, scientists questioned how much public impact the report could have (Aug conference news). Likewise, in the U.S. a report from the National Academies of Science, Engineering, and Medicine concluded that cognitive training, blood pressure management, and regular exercise might prevent or delay cognitive decline, but held off on issuing specific recommendations (Jun news). 

Understanding of AD risk continued to evolve, with one study reporting that living near busy roads could increase incidence, and another debunking the idea that low body weight predisposes to the disease (Jan news; May news). Nevertheless, deaths due to Alzheimer’s climbed by 50 percent in the U.S. between 1999 and 2014, even as dementia incidence has been falling in developed countries. Scientist think a combination of increased longevity and greater acknowledgement of AD on death certificates drove the uptick (May news).—Alzforum Editorial Team


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. 10th CTAD: Finally, Alzheimer’s Field Is Serious About Prevention Trials
  2. No Negativity, Please. Clinical Trial Leader Urges Focus on Learning, Progress
  3. Merck Pulls Plug on Phase 2/3 BACE Inhibitor Trial
  4. Verubecestat Negative Trial Data: What Does it Mean for BACE Inhibition?
  5. Anti-Amyloid Drug Pipeline Shows No Sign of Drying Up
  6. A4 Researchers Raise Solanezumab Dosage, Lengthen the Trial
  7. Much ‘Adu’ About a Little: Phase 1 Data Feeds the Buzz at CTAD
  8. High-Dose Gantenerumab Lowers Plaque Load
  9. Transcranial Magnetic Stimulation for AD Boasts Success in Phase 3
  10. Mental Training Evokes Superior Memory, Transforms Brain Networks
  11. Positive Trials of Spinal Muscular Atrophy Bode Well for Antisense Approach
  12. What Price Success? Ionis Drug Worked in Phase 3 but Had Serious Side Effects
  13. Antisense Oligonucleotide Squelches Huntingtin Protein in Phase 1/2a Trial
  14. Antisense Oligos Tango with Tau Transcripts to Reverse Tauopathy
  15. Intepirdine Joins the Ranks of Failed Alzheimer’s Drugs
  16. At AAIC, Yet Another Phase 3 Flop While Phase 1 Trials Forge Ahead
  17. Pimavanserin Trial Raises Hope for Treating Dementia-Related Psychosis
  18. High-Dose Aβ and Tau Immunotherapies Complete Initial Safety Tests
  19. In the Running: Trial Results from CTAD Conference
  20. α-Synuclein Antibodies Enter Phase 2, Sans Biomarker
  21. New Alzheimer’s Drug Shows Safety, Hints of Efficacy in Phase 2
  22. Elusive or Not, Aβ Oligomers Are in BioPharma Crosshairs
  23. At CTAD, Tau PET Emerges as Favored Outcome Biomarker for Trials
  24. Blood, the Secret Sauce? Focus on Plasma Promises AD Treatment
  25. No Man’s Land: Neither Early Success nor Failure for BAN2401
  26. Can Adaptive Trials Ride to the Rescue?
  27. Don’t Be an Enrollment Loser: Throw Your Own Swab-a-Palooza!
  28. Clinical Trials Consortium Succeeds ADCS, Focuses on Prevention
  29. Planning the First Primary Prevention Trial for Alzheimer’s Disease
  30. FDA Approves Edaravone for Treatment of ALS
  31. In a First, FDA Approves Deuterated Drug for Huntington’s
  32. Automated CSF Tests: Check. Blood Tests: In the Works
  33. Are CSF Assays Finally Ready for Prime Time?
  34. It’s Official: CSF Aβ42 Reference Standard Will Unify Assays
  35. Finally, a Blood Test for Alzheimer’s?
  36. More Support for Amyloid Hypothesis: Protective APP Mutation Lowers Aβ in Blood
  37. Longitudinal Data Say: Nope, CSF Markers Do Not Track Progression
  38. Blood Neurofilament Light a Promising Biomarker for Alzheimer’s?
  39. Neurofilament Light Chain as Prognostic Biomarker in ALS
  40. Touchdown for NfL: Blood Test Tells Parkinson's from Related Disorders
  41. Serum NfL Detects Preclinical AD, Reflects Clinical Benefit
  42. PET Staging Charts Gradual Course of Amyloid Deposition in Alzheimer’s
  43. Daydreaming Network Serves as Ground Zero for Aβ Deposition
  44. At Risk, or Already Alzheimer’s? Elevated Aβ Predicts Cognitive Decline
  45. In Clinical Use, Amyloid Scans Change Two-Thirds of Treatment Plans
  46. Next-Generation Tau PET Tracers Strut Their Stuff
  47. Subjective Memory Complaints Tied to Tau
  48. Multimodal Imaging Ties Tau to Neurodegeneration, and Symptoms
  49. All Signs Point to Tau Tangles as the Culprit in Fading Memory
  50. Selecting Trial Participants Based on Tangle Pathology Might Improve Power
  51. Data from DIAN Revise Familiar Biomarker Trajectories
  52. Can Biomarker Data Predict Individual Risk for Alzheimer’s?
  53. Poly Dipeptide in Cerebrospinal Fluid Marks C9ORF72 Expansion Carriers
  54. Bloodborne Tau: Foggy Window into the Brain for TBI, Dementia
  55. Cryo-EM Developers Win Nobel Prize in Chemistry
  56. Location, Conformation, Decoration: Tau Biology Dazzles at AD/PD
  57. Tau Filaments from the Alzheimer’s Brain Revealed at Atomic Resolution
  58. From New Tau Structure, Bona Fide Aggregation Inhibitors?
  59. Amyloid-β Fibril Structure Bares All
  60. Do Palettes of Aβ Fibril Strains Differ Among Alzheimer’s Subtypes?
  61. Monomeric Seeds and Oligomeric Clouds—Proteopathy News from AAIC
  62. Paper Alert: Aβ Fibril Structures Vary by AD Subtype
  63. Sweat the Small Stuff: Teeniest Aβ Oligomers Wreak Most Havoc
  64. Infrared Microscopy Reveals Pre-Plaque Formation of β-Sheets by Aβ
  65. Stress Granule Protein Stabilizes Tau Oligomers, Hastens Neurodegeneration
  66. Protein Liquid-Liquid Phase Transitions: The Science Is About to Gel
  67. Tau Hooks Up with RNA to Form Droplets
  68. More Droplets of Tau
  69. Tau Droplets Sprout Microtubules
  70. Phosphorylation of FUS Does Away with Droplets
  71. ALS Dipeptides Drive Liquid-Liquid Phase Separation, Stress Granule Formation
  72. Not Just for Proteins—Expanded RNA Repeats Form Gels, Too
  73. Out of Chaos, Order: Reversible Amyloid Structure Seen in Phase Separation
  74. It’s MAGIC: Yeast Mitochondria Make Cytosolic Protein Aggregates Disappear
  75. Barfing Up Balls of Trouble Keeps Some Neurons Healthy
  76. From Garbage Disposal to Neuromodulation? Membrane Proteasomes Churn Out Stimulating Peptides
  77. Microglial Master Regulator Tunes AD Risk Gene Expression, Age of Onset
  78. Searching for New AD Risk Variants? Move Beyond GWAS
  79. Hot DAM: Specific Microglia Engulf Plaques
  80. Microglial Kinase Promotes DAM, Blocks Lysosomal Aβ Digestion
  81. ApoE and Trem2 Flip a Microglial Switch in Neurodegenerative Disease
  82. What Makes a Microglia? Tales from the Transcriptome
  83. Human and Mouse Microglia Look Alike, but Age Differently
  84. Long Live the Microglia! Studies Trace Their Lifespans in Mice and Humans
  85. Aging Causes “Identity Crisis” in Glia
  86. Do Microglia Spread Aβ Plaques?
  87. Somatic SNAFU—Can a Few Mutant Microglia Cause Neurodegenerative Disease?
  88. Microglia Give Astrocytes License to Kill
  89. Brain Spheroids Hatch Mature Astrocytes
  90. Paper Alert: TREM2 Crucial for Microglial Activation
  91. Without TREM2, Microglia Run Out of Gas
  92. Changing With the Times: Disease Stage Alters TREM2 Effect on Tau
  93. TREM2 Cleavage Site Pinpointed: A Gateway to New Therapies?
  94. Does Soluble TREM2 Rile Up Microglia?
  95. Inside Out, or Outside In? ALS Turns on Monocytes in Blood
  96. Can Immune Gene Expression Predict Pace of Motor Neuron Destruction?
  97. As ALS Worsens, Immune Cells Multiply in the Blood
  98. Inflammation in Midlife Portends Late-Life Brain Shrinkage
  99. A New Explanation for Dendritic Tau: It’s Made There
  100. Aβ Plaques: Breeding Ground for Toxic Tau?
  101. Is There No End to Tau’s Toxic Tricks?
  102. ApoE and Tau: Unholy Alliance Spawns Neurodegeneration
  103. Led Astray: Pathology Tied to “Grid Cell” Malfunction in Tauopathy Model
  104. Tau Toxicity Blamed for Widespread Cell Death After Stroke
  105. For Synapses, Are Aβ Oligomers a No-Go?
  106. Transport Breakdown Maroons BACE1 in Synapses, Boosts Aβ
  107. Chimeric AD Model Shows Human Neurons Are Uniquely Vulnerable to Aβ
  108. C9ORF72 Throws a Wrench into DNA Repair Machinery
  109. When C9ORF72 Silences U2, Spliceosomes Can’t Find What They’re Looking For
  110. In New Role for TDP-43, Scientists Say it Controls Protein Synthesis
  111. For Better and Worse, TDP-43 Controls Microglia’s Phagocytic Prowess
  112. Lymphatic Vessels Found in Human Brain
  113. In Mice, CSF Caught Draining Via Lymphatic Vessels, Not Veins
  114. New Technology Catalogues Immune Cells in Brain
  115. Pericytes Don’t Go With the Flow—They Change It
  116. Finally, a Dye to Visualize Pericyte Function
  117. Do Perivascular Macrophages Mediate Aβ Pathology?
  118. Mini Strokes Cause Mega Problems for Brain Cleansing
  119. Peripheral Aβ Can Accumulate in Brain, Trigger Degeneration
  120. Vascular Disease in 50s Begets Brain Amyloid in 70s
  121. High Blood Pressure in Younger Adults: Dementia Risk Only in Women?
  122. Ring Around the Vessel: Enlarged Spaces Signal Vascular Disease
  123. Consortium to Seek Biomarkers for Vascular Cognitive Impairment
  124. Artificial Human Blood Vessels: A Model for Cerebral Amyloid Angiopathy?
  125. Number of U.K. Dementia Researchers Doubles over Six Years
  126. Building Better Mouse Models for Late-Onset Alzheimer’s
  127. Bill Gates Throws His Weight Behind Alzheimer’s Research
  128. President Proposes Massive Cut to NIH; Potential Fallout for Alzheimer’s Unclear
  129. Lancet Commission Claims a Third of Dementia Cases Are Preventable
  130. Preventing Dementia: Getting Closer to Recommendations
  131. Dementia Risk Ticks Up Near Major Roadways
  132. No, Being Thin Does Not Lead to Alzheimer’s Disease
  133. Alzheimer’s Deaths on the Rise

Therapeutics Citations

  1. Verubecestat
  2. Elenbecestat
  3. Lanabecestat
  4. Solanezumab
  5. Crenezumab
  6. Gantenerumab
  7. Aduhelm
  8. Intepirdine
  9. Idalopirdine
  10. Pimavanserin
  11. Tilavonemab
  12. Gosuranemab
  13. Varoglutamstat
  14. Elayta
  15. Atabecestat
  16. Levetiracetam

Mutations Citations

  1. APP A673T (Icelandic)

Paper Citations

  1. . Reducing Excessive Use of Antipsychotic Agents in Nursing Homes. JAMA. 2017 Jul 11;318(2):118-119. PubMed.
  2. . Preclinical Assessment of Young Blood Plasma for Alzheimer Disease. JAMA Neurol. 2016 Nov 1;73(11):1325-1333. PubMed.
  3. . Alzheimer's disease drug development pipeline: 2017. Alzheimers Dement (N Y). 2017 Sep;3(3):367-384. Epub 2017 May 24 PubMed.

Other Citations

  1. Dec news

External Citations

  1. Generation 1
  2. Generation 2
  3. EARLY
  4. A4 trial
  5. CREAD
  6. Imaging Dementia—Evidence for Amyloid Scanning

Further Reading

No Available Further Reading