top of page
Business Meeting

Is Tau the New Amyloid?

By Selina Koch, Senior Editor

After finally getting the message that they need to look beyond amyloid to treat Alzheimer’s disease, pharmas are lining up behind tau as the next big target. The question is whether they can pick out the lessons from the amyloid saga to avoid following the same storied path.

At least four Phase II readouts of anti-tau antibodies are on the horizon. But there remain a number of questions, from the target’s biology to how best to translate that biology, plus the beginnings of a lemming effect, that serve as warning echoes (see “How the Amyloid Hypothesis Holds its Grip”).

The amyloid hypothesis was rooted in multiple lines of genetic data pointing to β amyloid as a causative agent in AD. In contrast, tau mutations are not found in AD patients, though they do occur in frontotemporal dementia (FTD), another neurodegenerative disease that leads to cognitive decline.

But tau is an obvious choice as the next-best chance for the disease because its aggregation in neurons — in neurofibrillary tangles — is one of the two major pathological hallmarks found in patient brains, the other being the extracellular deposits of β amyloid in plaques.

And there’s good reason to think tau could be a better target than β amyloid — the timing and location of tau aggregation correlate much better with the onset and nature of patients’ symptoms.

Tau builds up closer to clinical onset, and in the right places to explain early cognitive deficits, suggesting a role in progression to clinical disease. Amyloid accumulates decades before symptoms arise, and in a wide variety of brain regions, not specifically the areas that drive symptoms.

“Tau has really emerged as the leading target in Alzheimer’s,” Husseini Manji, global therapeutic area head for neuroscience at Johnson & Johnson’s Janssen Pharmaceuticals unit, told BioCentury.

Janssen is gearing up to run a Phase II study of its anti-tau mAb JNJ-63733657. Ahead of that is a small molecule from TauRx Pharmaceuticals Ltc. and the four anti-tau mAbs already in Phase II testing. A host of tau-targeted molecules trail it, including vaccine and antisense oligonucleotide therapies in clinical development, as well as small molecules and other therapies in preclinical testing (see Figure: “Tau in the Clinic”).

This piling of companies onto the same target in the absence of any clinical proof of concept is reminiscent of amyloid, as is the focus on a single aggregation-prone protein. That should set off some alarm bells, as the disease has many such proteins, plus dysfunctional immune cells, vascular abnormalities and other features (see “After Amyloid”).

The good news is that the Alzheimer’s field has more experience and better biomarkers than it had when it started testing anti-amyloid agents, suggesting it can avoid repeating some of its early mistakes. The risk is that the major players get tied up for another decade or two on a single target, iterating compounds and trial designs rather than casting a wide net (see “Amyloid: How Did We Get Here and What Can We Learn?”).

Over 50,000 Alzheimer’s patients have been or are being treated with amyloid-lowering compounds. So far, for tau, the tally is closer to 2,700.

Already one anti-tau agent, the small molecule LMTX from TauRx, has failed to meet its primary endpoints in two Phase II/III trials, and the company is trying again with a new dose. But other stakeholders dismiss the compound as a poor test case.

More specific tests of the hypothesis could come when results from the mAbs start rolling in, which should be by the middle of next year, according to

Not the same story

Despite parallels between tau and β amyloid, proponents of the tau hypothesis point to several advantages over the amyloid hypothesis (see Figure: “Tau vs. Amyloid”).

“In Alzheimer’s disease, tau is the one thing that correlates really beautifully with disease progression, much more so than Aβ [β amyloid] or any other markers that are out there,” Holly Kordasiewicz, executive director of neuroscience drug discovery at Ionis Pharmaceuticals Inc., told BioCentury.

Ionis has an antisense oligo therapy against tau, IONIS-MAPTRx, in a Phase I/IIa trial in AD.

Tau accumulates first in the hippocampus and other brain regions whose dysfunction is thought to drive early cognitive symptoms in patients, and then begins to aggregate in other regions, mirroring disease progression.

“That really strong correlation between tau pathology, tau spreading and disease progression suggest that this might be one of the underlying molecular causes of the disease,” said Kordasiewicz.

Despite the fact there have been many more studies on amyloid than tau, more is known about tau’s endogenous functions and how it might drive toxicity, she added.

For example, tau is an axonal protein that regulates the stability and flexibility of microtubules, as well as transport of cargo along them. Excessive phosphorylation of tau can cause it to detach from the microtubules, leading to loss of that function. Hyperphosphorylated tau also ends up in synapses, where it is thought to cause toxicity by a gain of function mechanism.

Tau is also prone to misfolding, and one concept — under debate — is that the misfolded protein passes from neuron to neuron, similar to prion proteins, accounting for the way the disease progresses from one brain region to another.

That provides at least two mechanistic footholds for companies to translate the biology into disease-modifying therapies — targeting different forms of the protein, and preventing its spread. Arguably, this puts tau a step ahead of the β amyloid field, which converged on the central hypothesis that reducing the amount of aggregated β amyloid would lead to cognitive improvement, despite the fact that no solid mechanistic connection has yet explained why that should be the case.

Spreading dispute

The fact that tau progressively spreads throughout the brain is not in dispute. At issue is how it spreads, which has implications for the best way to drug it, and how important tau spreading is in the disease process.

Antibodies directed against tau, which represent the largest class of clinical compounds

against the target, are based on the idea that tau spreads extracellularly.

“Our core therapeutic hypothesis is that we can intercept the cell-to-cell spreading of tau in the brain to reduce the progression of disease,” Casper Hoogenraad, senior director and staff scientist at Roche’s Genentech Inc. unit, told BioCentury. Genentech and partner AC Immune S/A have the anti-tau mAb RO7105705 in two Phase II trials.

There are also differences among those who think tau spreads this way. One theory is that extracellular tau gets trapped in exosomes or other protected compartments, meaning the target won’t be accessible to antibodies. Some researchers have suggested microglia — the brain’s phagocytic cells — ingest tau aggregates when they phagocytose dying synapses, then spread tau through the brain by releasing it in exosomes (see “Microglia Strike Again”).

Compounds that act intracellularly, such as small molecules and antisense oligos, would bypass these issues.

According to Manji, J&J is developing a preclinical small molecule inhibitor. AC Immune has a preclinical small molecule tau inhibitor partnered with Eli Lilly and Co.

“More people have focused on the spread of tau, and I think there’s good reason for that. But ultimately the toxicity is intracellular and can you target that with a small molecule,” said


Genentech’s Hoogenraad is not concerned about using mAbs, because he thinks release of

free tau is more common. “While there is cell culture data that suggests that some extracellular tau might be found in a membrane-bound component, those studies suggest that only a small proportion of the tau detected was found in such compartments,” he said. He noted that the compartment could be an exosome or ectosome, for example, and that the proportion found in them was less than 10%.

Hoogenraad declined to say when Genentech expects topline data from its two ongoing trials, but said the blinded portion of its first one, TAURIEL, runs for 18 months and dosed its first patient in 4Q17. Its primary completion date in is in June 2020.

Other anti-tau mAbs in Phase II trials are LY3303560 from Lilly, BIIB092 from Biogen Inc. and partner Bristol-Myers Squibb Co., and ABBV-8E12 from AbbVie Inc. and partner C2N

Diagnostics LLC. According to, the first Phase II trials from all these programs have primary completion dates in 2021.

Dennis Selkoe disputes the idea that tau spreads via any extracellular mechanism, and thinks that the pattern of progression is due to the fact that tau aggregation is more toxic to some neurons and parts of the brain than others.

Selkoe — one of the originators of the amyloid hypothesis, and one of the researchers who discovered in the 1980s that tangles were made of tau — told BioCentury the extracellular spreading hypothesis “is probably not correct.” If neurons are leaking or spitting out misfolded tau, then it is hard to explain why, “even after 20 years of clinical disease, some neurons have tangles, and neurons right next door are spared.”


“People who believe this say there must be some kind of receptor. But, usually, the more complicated you make your model, the less likely it is to be right,” said Selkoe, who is a professor of neurologic diseases at Harvard Medical School and co-director of the Center for Neurologic Diseases at Brigham and Women’s Hospital.

Form factor

The use of different modalities could tease apart which forms of tau drive toxicity, a goal that was not achieved for β amyloid.

Like amyloid, tau comes in different sizes, conformations and aggregation states, from small soluble oligomers to large insoluble aggregates. Tau is even more complicated, because its many phosphorylation sites increase the number of distinct states it can assume.

Hoogenraad said Genentech’s mAb “was selected to bind most forms of tau, since there is not yet conclusive evidence about which forms of tau are pathological.”

Janssen’s Manji said his company’s Phase I mAb JNJ-63733657 binds “a mid-region phosphoepitope, in contrast to most of our competitors who are targeting whole tau and terminal epitopes.”

Manji said the decision was based on extracting misfolded tau from postmortem brain tissue and finding “seeds” that are capable of inducing tau pathology in cell culture “fully retain the mid-region epitopes; whereas some of the N-terminal epitopes are cleaved, and so we thought that the mid-region phospho-tau is the more germane pathogenic form.”

Lilly declined to say which part of tau LY3303560 binds to, but told BioCentury the mAb has demonstrated about 1,000-fold higher selectivity for aggregated tau over tau monomers in preclinical studies.

Ionis’ Kordasiewicz noted that because her company’s antisense oligo targets tau’s mRNA, it does not depend on which form of the target drives toxicity. Nor does it depend on extracellular spread of the protein — a benefit it shares with small molecules.

“By stopping the production of tau we’re preventing the cells from ever having to deal with it in the first place, and we’re getting rid of all the toxic species,” she said.

Biogen has an option to develop and commercialize the Ionis compound, making Biogen one of several companies attacking tau via multiple modalities. In addition to its BMS-partnered mAb, Biogen has a Phase I anti-tau mAb from Neurimmune Holding AG. BIIB092 from BMS targets the N-teriminus of tau and truncated forms of tau containing the terminus, according to published studies. Neurimmune declined to disclose the tau epitope targeted by BIIB076, and Biogen did not respond to interview requests.

Manji said in addition to having a mAb in development and a preclinical small molecule inhibitor, J&J has partnered with AC Immune on a Phase I anti-tau vaccine.

“The vaccine is only a little behind the antibody,” he said, adding that the modality would be particularly useful in developing countries. “A vaccine could help more people in more parts of the world.”

AbbVie declined to be interviewed for this story.

Path to answers

At a minimum, the lessons from β amyloid mean that tau drug developers start off knowing that timing the treatment and finding the right patient population — rather than taking all comers — could dramatically increase their chance of success.

There are enough open questions around the biology of tau to suggest that the road ahead could still be long.

Because TauRx’s LMTX is a derivative of methylene blue, which has a wide array of effects in cells, many in the field don’t consider it a rigorous test of the tau hypothesis. TauRx argues that despite the two Phase II/III failures, a subgroup analysis supports continuing the program. The company is testing it in another Phase II/III trial as monotherapy and at a lower dose.

“If the first antibodies don’t read out positively, I would caution people not to throw the baby out with the bathwater,” said Manji. “A negative trial doesn’t mean every tau modality is doomed to failure. They’re not all the same.”

Lilly is trying to speed the process by homing in on patients at the right stage of disease. The pharma’s “Goldilocks tau” trial uses PET imaging of tau to enroll patients with low to medium levels of tau pathology, and are therefore likely to progress during the timeframe of the trial, but who don’t have so much tau aggregation that it’s too late to intervene.

Manji said J&J has set up a consortium through the Innovative Medicines Initiative (IMI) to “thoroughly characterize” the biomarkers and cognition performance of a group of patients prior to placing them in an interventional trial.

“Sometimes one of the challenges is you don’t have long enough baseline information on people. So if you already have this cohort that you’ve characterized well before intervention, then you’re more likely to pick up a correct signal after intervention,” he said.

J&J’s trial will likely enroll an “early, prodromal-type population” in its Phase II trial, said Manji.

By contrast, Hoogenraad said Genentech is not making assumptions about the best disease stage to target or how much baseline tau is optimal to achieve efficacy. Genentech is tracking tau PET signals in its patients but hasn’t set enrollment criteria based around the parameter. It is running two Phase II trials: one in early stage patients and the other in moderate stage.

Ionis may take a detour from AD after the safety studies, using patients with frontotemporal lobar degeneration (FTLD) who harbor tau mutations as a test case for proof of concept that the compound works. The company is still deciding if and when to run efficacy trials in AD and other tauopathies.

Kordasiewicz said the biotech believes sticking as close to the genetics as possible is the best bet for success. “That is a general principle we have.”

“We went into AD first because it is a larger patient population,” which makes recruitment faster, she said. “Going to the mutation carriers is just trying to give the drug the best shot for success, by getting it to the patients who are going to benefit the most.”

The approach is opposite to how pharmas approached β amyloid. Early β amyloid trials took virtually all comers, and over time became more restrictive. Only now, after almost two decades of failures, are companies testing amyloid agents in mutation carriers.

Biogen too is running a trial in tau mutation carriers. In addition to its AD studies, the biotech is conducting a Phase Ib trial of BIIB092 in four tauopathies that include FTD patients with tau mutations.

Hoogenraad said Genentech is “interested” in other tau-driven dementias but is sticking with its focus on AD. “Currently, cognitive/functional outcome measures are far better understood in AD than in other tauopathies, such as FTD,” he said.

August 5, 2019



AC Immune S/A (NASDAQ:ACIU), Lausanne, Switzerland

Biogen Inc. (NASDAQ:BIIB), Cambridge, Mass.

Brigham and Women’s Hospital, Boston, Mass.

Bristol-Myers Squibb Co. (NYSE:BMY), New York, N.Y.

C2N Diagnostics LLC, St. Louis, Mo.

Eli Lilly and Co. (NYSE:LLY), Indianapolis, Ind.

Harvard Medical School, Boston, Mass.

Ionis Pharmaceuticals Inc. (NASDAQ:IONS), Carlsbad, Calif.

Johnson & Johnson (NYSE:JNJ), New Brunswick, N.J.

Neurimmune Holding AG, Schlieren, Switzerland

Roche (SIX:ROG; OTCQX:RHHBY), Basel, Switzerland

TauRx Pharmaceuticals Ltd., Singapore



Tau (MAPT; FTDP-17) - Microtubule-associated protein tau



BioCentury’s mission is to provide value-added business intelligence & analysis for life science companies, investors, academia and government on the strategic issues essential to the formation, development and sustainability of life science ventures.



+1 650-595-5333; Fax: +1 650-595-5589


+1 312-755-0798; Fax: +1 650-595-5589


+1 202-462-9582; Fax: +1 202-667-2922


+44 (0)1865-512184; Fax: +1 650-595-5589

BioCentury®; Because Real Intelligence is Hard to Find™; BCIQ™; The BioCentury 100™; and The Clear Route to ROI™ are trademarks of BIOCENTURY INC. All contents Copyright © 2019, BIOCENTURY INC. ALL RIGHTS RESERVED. No part of BioCentury’s Publications or Website may be copied, reproduced, retransmitted, disseminated, sold, distributed, published, broadcast, circulated, commercially exploited in any form or used to create derivative works without the written consent of BioCentury. Information provided by BioCentury’s Publications and Website is gathered from sources that BioCentury believes are reliable; however, BioCentury does not guarantee the accuracy, completeness, or timeliness of the information, nor does BioCentury make any warranties of any kind regarding the information. The contents of BioCentury’s Publications and Website are not intended as investment, business, tax or legal advice, and BioCentury is not responsible for any investment, business, tax or legal opinions cited therein or for any decision made or action taken in reliance upon such information.

All use of BioCentury and its contents by current subscribers is governed by the BioCentury User Agreement and by all others is governed by the BioCentury Terms of Use, unless a written agreement to the contrary has been executed by BioCentury Inc.

USE OF IMAGES: Certain Images used in BioCentury Inc.’s Publications, Video Content, Websites, Services, Notices and/or Marketing Materials are licensed from Getty Images (US), Inc. Any such image of a person or object so displayed is being used for illustrative

purposes only and any such person or object depicted, if any, is merely a model. For more information see “Use of Images” found under the “Legal” section on the footer of the homepage at

PRIVACY & ADVERTISING: In accordance with its Privacy Policy, BioCentury does NOT sell its customer information or usage data to third parties. BioCentury does NOT sell advertising in the BioCentury, BioCentury Innovations or BioCentury Week in Review publications. BioCentury is pleased to acknowl edge its conference partners and sponsors through promotional announcements in its publi cations. BioCentury MAY accept paid promotional messages from sponsors, which are displayed only on BioCentury’s websites and in BioCentury Extra.


bottom of page