About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Oh Brave New World | Main | Watching A Palladium Reaction, In A Crystalline Flask »

May 8, 2014

Alzheimer's Antibodies: Can the Current Ones Even Work?

Email This Entry

Posted by Derek

As everyone in the field knows, there's a lot of work that has been poured into anti-beta-amyloid antibodies as a potential therapy for Alzheimer's. There are plenty of questions, starting with the amyloid hypothesis itself, but even if you stipulate that, the question that this paper asks is key: do these antibodies engage their target?

And the answer, according to this large team from Melbourne, is, well. . .

. . .All of the antibodies were able to bind Aβ in mouse tissue. How- ever, significant differences were observed in human brain tissue. While bapineuzumab was able to capture a variety of N-terminally truncated Aβ species, the Aβ detected using solanezumab was barely above detection limits while crenezumab did not detect any Aβ. None of the antibodies were able to detect any Aβ species in human blood. Immunoprecipitation experiments using plasma from AD subjects showed that both solanezumab and crenezumab have extensive cross-reactivity with non-Aβ related proteins.

Now that's interesting, isn't it? The paper uses SPR as one analytical technique, which showed that bapineuzumab and crenezumab bound to synthetic chip-immobilized Aβ at nanomolar levels, while solanezumab was probably picomolar. And mass spec showed that the antibodies recogized Aβ from transgenic mouse tissue. But human tissue, that was something else altogether. Bapineuzumab was able to pull out amyloid proteins in some experiments, but the other two were nearly inactive.

It appears that solanezumab and crenezumab are actually quite similar, and are both ineffective. The transgenic mouse studies may, according to this paper, have been a tremendous red herring:

Based on mouse studies, it has been suggested that solanezumab does not work directly on brain Aβ, but instead works as a peripheral sink targeting peripheral Aβ which in turn lowers CNS Aβ by mass action. Indeed when we examined plasma from tg2576 mice, solanezumab was able to detect Aβ1–40 in plasma. However, none of the therapeutic antibodies were able to detect Aβ in either the plasma or the cellular fraction from human AD subjects. . .

The authors suggest that the fact that these antibodies do recognize pure Aβ but still perform so poorly in vivo might be due to cross-reactivity. This is the sort of thing that would surely have been checked by the companies involved, but this paper does report substantial cross-reactivity in human plasma and against some other proteins. One in particular might be worth noting:

One of the proteins pulled down by both solanezumab and crenezumab was the Il12 receptor; this is interesting as a recent publication showed that modulation of the Il12 signaling pathway resulted in cognitive improvements in a transgenic mouse model of AD. It should be noted that a driving force for the selection of solanezumab as a drug candidate for the preventative trials was the post hoc analysis that showed a small, but significant improvement in cognition in the mild AD subjects in Phase 3 trials. . .

So has all this been a waste of time? And is Lilly's continued work on solanezumab, and Genentech's on crenezumab, likely doomed? Thoughts welcomed in the comments. What I'd like to know, even before we get to those big questions, is what these sorts of experiments showed internally, and why these (rather alarming) results haven't been seen by others so far. Or have they?

Comments (31) + TrackBacks (0) | Category: Alzheimer's Disease


1. Wavefunction on May 8, 2014 11:15 AM writes...

This reminds me again of the Whitesides quote on challenges in drug development: "Whatever else you may think of me, I am not a large, hairless mouse.”

Are the Aß sequences of mouse and human amyloid identical? Are there any covalent modifications in either? Could any of these changes result in differences in conformation that could affect antibody binding?

Permalink to Comment

2. entropyGain on May 8, 2014 11:17 AM writes...

Kinda sounds like the academic nature papers that make a splash, but can't be repeated. But in this case, instead of selling a bill of goods to editors/referee's, someone sold a bill of goods to some MBAs. Probably wasn't that hard: "everybody" knows antibodies are easy great drugs, and alzheimers is a huge market. Besides the other guys are doing it too.....

What's the emoticon for tongue in cheek?

Permalink to Comment

3. Anon on May 8, 2014 11:36 AM writes...

"One of the proteins pulled down by both solanezumab and crenezumab was the Il12 receptor; this is interesting as a recent publication showed that modulation of the Il12 signaling pathway resulted in cognitive improvements in a transgenic mouse model of AD."
Did they use a similar mouse model [and cell assays] during development?
This may be the case of "you get what you screen for"
If these guys paid a living wage, maybe they could attract the talent to avoid these mistakes?
And checking human blood or histology samples is summer project for an undergrad (or an MD). I'd even go as far as saying that it would be impossible for these experiments to not have been done. At the very least someone grabbed the wrong antibody or tissue specimen and had a result. That's VERY suspicious (if these published results are indeed correct).

Permalink to Comment

4. Erebus on May 8, 2014 12:18 PM writes...

@1 -
Reminds me a bit of the recent Irisin debates. That newly-discovered protein seemed like the greatest thing since sliced bread, until it was discovered that while Irisin is a very active metabolic regulator in mice and primates, it's effectively inactive in humans. (Our copy of the irisin precursor gene has a start-codon mutation.)

That said, mouse and human amyloid do not appear to be identical. The 42 amino fragment of human Aß is:


Whereas with mice and rats, it appears to be:


I've placed parenthesis around the two apparent differences.

Permalink to Comment

5. Ann O'Nymous on May 8, 2014 12:22 PM writes...

I can't quite tell from #3 if Anon thinks this was good science, or bad science?

To my mind, this is good science. Someone took a promising lead and found that it wasn't nearly as promising as they hoped. Isn't it better to openly share this than only publish "successes"? Otherwise the pressure leads researchers into murky stuff (like doctoring pictures, as Derek has shown at length in other posts...)

Anon - your thoughts?

Permalink to Comment

6. In Vivo Veritas on May 8, 2014 12:30 PM writes...

@ Erebus - that's not even the most egregious thing about irisin. Not exactly the most repeatable discovery I've ever looked into.

Permalink to Comment

7. Mac centric on May 8, 2014 12:47 PM writes...

Why do we think abs to ABeta should work in the first place? Most of the mouse AD models are contrived in that they are caused by transgenic over expression of Presenilin or Abeta (since we don't know what really causes AD). The increased Abeta in human brains may simply be a consequence of late stage disease. So mopping up excess Abeta may have little effect. Also, generally antibodies that bind their target at a site are taken up by local FcR-expressing cells such as microglial cells of the brain. If the microglial cells of the brain aren't working properly, they may ignore the antigen-bound antibodies. I suspect the microglial cells are simply pooping out in the elderly. These are the cells responsible for house cleaning in the brain and evidently, they aren't doing a good job in folks with AD.

Permalink to Comment

8. johnnyboy on May 8, 2014 1:08 PM writes...

To get to an IND for these, the companies would have to have done tissue cross-reactivity studies, which is part of the tox package for any biologic. However I'm not sure that plasma cross-reactivity is something that is part of the package.

In any case, my (limited) understanding of the approach is that anti-amyloid antibodies were pushed through to human trials based essentially on some evidence of plaque reduction in a mouse model, but no one had a good explanation or mechanism for this. The elephant in the room of anti-amyloid antibodies is that antibodies do not cross the blood-brain barrier, so how they would lead to plaque resorption in the brain is rather unclear. People usually mumble something about the BBB perhaps being disrupted in areas of plaque formation; then there's the hypothesis cited above, that clearing amyloid from the plasma might lead to reduction of amyloid in brain plaques by some sort of diffusion process. All this sounds more like wishful thinking than anything fact-based. And of course even if you did manage to reduce plaques in AD patients, no one knows if that would actually make a difference to the disease. All in all, the approach seems to me like one for the faith-based...

Permalink to Comment

9. anonymous on May 8, 2014 2:23 PM writes...

In a 12wk randomized Phase II trial in 52 patients with mild-to-moderate AD, who received 100 or 400mg 1x/wk or 100 or 400mg q 4wk iv infusion, and 16 healthy volunteers who received 100mg single dose, interim results showed that solanezumab bound to A-ß protein resulting in an increased amount of A-ß in the blood and cerebrospinal fluid (CSF). solanezumab weekly infusion up to 400mg was well tolerated, with no treatment-related side-effects. MRI and CSF assessment showed no evidence of brain inflammation or bleeding. There was no change in patients' cognitive scores, and no change in levels of A-ß plaque as measured by SPECT tracer. There was also an increase in 2 biomarker that lasted for several wk. In clinical studies, the most frequent side-effects was mild chills consistent with an infusion reaction (Press release, Lilly, 30 Jul 2008).

Permalink to Comment

10. luysii on May 8, 2014 3:07 PM writes...

#4 Erebus -- how about a link or a reference to the human Irisin mutation (if it can be called that). Has anyone assayed how much Irisin we actually have ?

Gene translation doesn't invariably begin at what we call a start codon -- the oncoprotein Myc is one example -- transcription initiates in 3 places -- the canonical 'AUG', an internal 'AUG' and a 'CUG' upstream of the canonical start codon. [ Cell vol. 149 pp. 22 - 25 '12 ]

Permalink to Comment

11. Anonymous on May 8, 2014 3:16 PM writes...

3/9 anon.
This is a Press Release. Press Release six year old about interim results. Press Release which is probably not worth the paper used to post it on the website.

Permalink to Comment

12. anonymous on May 8, 2014 3:22 PM writes...

11 anon
Study demonstrated binding to human abeta and dose dependent increases in abeta in blood and csf of treated patients. Worthless or otherwise.

Permalink to Comment

13. Anonymous on May 8, 2014 3:42 PM writes...

@10 -
Raschke S, Elsen M, Gassenhuber H, et al. Evidence against a beneficial effect of irisin in humans. PLoS ONE. 2013;8(9):e73680.

Also, hot off the presses, this review is also worth a look:

Elsen M, Raschke S, Eckel J. Browning of white fat: does irisin play a role in humans?. J Endocrinol. 2014;

Can't post links here for some reason. Sorry bout that.

Permalink to Comment

14. Anonymous on May 8, 2014 3:54 PM writes...

12 anon.
After some pubmed search, I think that study was published here:
Sorry, no access to the full length paper from home but from the abstract I deduce there was not a shred of proof of antibody binding to Abeta reported there. Actually, the abstract is perfectly fuzzy.

Permalink to Comment

15. Anonymous on May 8, 2014 4:01 PM writes...

#14 anon.
I have same problems and think corante blocks any links which do not start explicitly with http:

Permalink to Comment

16. Snickers on May 8, 2014 5:23 PM writes...

Bapineuzumab cleared amyloid from the brain. Something like a 25% reduction.

Permalink to Comment

17. QuinbusFlestrin on May 8, 2014 9:58 PM writes...

Isn't it telling that sola was so safe during the clinical trials? If it truly removed amyloid from the brain as Lilly claims it would have removed some amyloid from the brain capillary walls, leading to edema or microhemorrhage, effects clearly observed in the bapi trial, the only antibody that, from this new paper, really seems to engage the target.

Permalink to Comment

18. Lane Simonian on May 9, 2014 12:34 AM writes...

The question is likely a moot one since amyloid is at best a contributor to Alzheimer's disease.

There are two pathways to amyloid: one is beta secreatase and it is controlled by peroxynitrites and hydrogen peroxide.

These novel NO-mediated regulatory mechanisms likely protect BACE1 from being further oxidized by excessive oxidative stress, as from H2O2 and peroxynitrite which are known to upregulate BACE1 and activate the enzyme, resulting in excessive cleavage of APP and Aβ generation; they likely represent the crucial house-keeping mechanism for BACE1 expression/activation under physiological conditions.

The second is the release of intracellular calcium which results in the activation of the gamma secretase.

However, it is the link between γ-secretase cleavage and intracellular Ca2+ stores which is the most intriguing aspect. Cheung et al. [50] found that PS mutant-induced enhancement of Aβ secretion can be abolished by IP3 receptor knockout, indicating that γ-secretase activity is controlled downstream of PS by IP3. How this finding relates to the hypothesis that PS is the catalytic subunit of the γ-secretase needs to be explored further.

Presenilin gene mutations and peroxynitrites cut off the neuroprotective phosphatidylinositol 3-kinase/Akt pathway.

... it has previously been reported that PS1/gamma-secretase is involved in the activation of phosphatidylinositol-3 kinase/Akt (PI3K/Akt) pathway. PI3K is inhibited in Alzheimer's disease (AD) brain...

Together, our data indicate that the neuroprotective role of PS1 depends on its ability to activate the PI3K/Akt signaling pathway and that PS1 FAD mutations increase GSK-3 activity and promote neuronal apoptosis by inhibiting the function of PS1 in this pathway. These observations suggest that stimulation of PI3K/Akt signaling may be beneficial to FAD patients.

Peroxynitrite induces inactivation of the Akt pathway. Furthermore, overexpression of constitutively active Akt inhibits both peroxynitrite-induced Bax translocation and cell death.

The PI3/Akt pathway decreases the production of inducible nitric oxide and increases the brain's own antioxidant systems and in doing so inhibits the development of Alzheimer's disease.

Triggering of the macrophage cell line RAW 264.7 with LPS promotes a transient activation of phosphatidylinositol 3-kinase (PI3-kinase). Incubation of activated macrophages with wortmannin and LY294002, two inhibitors of PI3-kinase, increased the amount of inducible nitric oxide synthase (iNOS) and the synthesis of nitric oxide.

Inducible nitric oxide synthase is key to peroxynitrite-mediated, LPS-induced protein radical formation in murine microglial BV2 cells.

Brains from subjects who have Alzheimer's disease (AD) express inducible nitric oxide synthase (iNOS). We tested the hypothesis that iNOS contributes to AD pathogenesis. Immunoreactive iNOS was detected in brains of mice with AD-like disease resulting from transgenic expression of mutant human beta-amyloid precursor protein (hAPP) and presenilin-1 (hPS1). We bred hAPP-, hPS1-double transgenic mice to be iNOS(+/+) or iNOS(-/-), and compared them with a congenic WT strain. Deficiency of iNOS substantially protected the AD-like mice from premature mortality, cerebral plaque formation, increased beta-amyloid levels, protein tyrosine nitration, astrocytosis, and microgliosis. Thus, iNOS seems to be a major instigator of beta-amyloid deposition and disease progression. Inhibition of iNOS may be a therapeutic option in AD.

Amyloid oligomers by attracting copper and zinc can contribute to the reformation of peroxynitrites by increasing the amount of hydrogen peroxide that combines with nitrite anions (ONOO- + 2H+ + 2e- =H20 + NO2-; NO2- + H2O2= ONOO- + H20). But any approach that does not operate upstream of amyloid or beta secretase is bound to fail no matter how effective it is at removing in or all plaques.

For those who say this is pseudoscience, second-rate science, blah, blah, blah look first at the studies that link peroxynitrite formation to the death of neurons, then look at the studies that used peroxynitrite scavengers to treat Alzheimer's disease, then look at the studies that used gamma secretase inhibitors, beta secretase inhibitors, or amyloid antibodies to treat Alzheimer's disease, and then tell me what is the cause of the disease and how most effectively to treat it.

Permalink to Comment

19. Anonymous on May 9, 2014 7:17 AM writes...

Peroxynitrite blah blah pap. Suddenly a really interesting topic becomes really boring to read once again, sadly.

Permalink to Comment

20. Anonymous Pharmacologist on May 9, 2014 7:39 AM writes...

It is an obvious point to this jobbing antibody Clinical Pharmacologist that solanezumab clearly engages soluble aBeta in the periphery. How do I know this? Because total aBeta increases in a dose-dependent manner until saturation can be achieved. The accumulation ratio is just the ratio of Clearance of aBeta/Clearance of complex. The complex, being a larger molecule, is cleared slower and hence you see accumulation – proof of binding (as if you needed some). The preclinical tox studies in monkey will show exactly the same effects and will have achieved full saturation, perhaps around 30 mg/kg, but definitely at 100 mg/kg (they may have dosed even higher).

Subsequent clinical studies with labels and take up in the brain have measured reductions in amyloid load. I think there is a little more Pharmacology to be had with these agents, but not much. Sadly the link between Pharmacology and efficacy is not proven/absent and even worse, required a Phase 3 experiment to even ask the question. The results are in, there is no obvious link.

Permalink to Comment

21. Harrison on May 10, 2014 12:37 PM writes...

APP / PS1 = Familial Alzheimer's Disease
ApoE4 / ? = Sporadic Alzheimer's Disease

The big leap of faith taken by science decades ago was that Familial AD = Sporadic AD

We just might be on the verge of disproving this theory, especially if the Familial AD studies with crenezumab and solanezumab work and the sporadic AD studies don't.

Permalink to Comment

22. Snickers on May 11, 2014 3:41 PM writes...

"Peroxynitrite blah blah pap. Suddenly a really interesting topic becomes really boring to read once again, sadly."

Right on! @Lane Simonian is a serial waffler. Pap.

Permalink to Comment

23. ChangLao on May 11, 2014 6:28 PM writes...

@Anonymous Pharmacologis
"A total of 169 patients in EXPEDITION 1 (17%) and 97 patients in EXPEDITION 2 (9%) underwent
baseline and follow-up 18F-florbetapir–PET
scanning. The composite standardized uptake
value ratio for the anterior and posterior right
and left cingulate, plus right and left frontal,
lateral temporal, and parietal regions, combined
and normalized to the whole cerebellum, did not
change significantly in the solanezumab group
or the placebo group in either study."

Permalink to Comment

24. previous_alz_reseearcher on May 11, 2014 8:14 PM writes...

This is such a complex area.... it obviously can't be addressed in these short comments. I'm as always, however, impressed by the general quality of discussion and the issues raised. As an ex-Alzheimer's preclinical researcher, I'll just throw in an observation. First, until very recently, the assays that the field relied on to measure abeta levels in CSF and plasma (and which are still used by many researchers) had quantitation issues ( and no, I'm not an author ;). In a general way, this is symptomatic of the fact that the development efforts have outpaced both the basic science of the disease, and the clinical techniques required to perform meaningful trials.

Permalink to Comment

25. Lane Simonian on May 11, 2014 9:43 PM writes...

Yeah! This site is working again.

Just one more to try to put it in perspective.

Neurosci Bull. 2014 Apr;30(2):271-81. doi: 10.1007/s12264-013-1423-y. Epub 2014 Mar 24.
Oxidative stress in Alzheimer's disease.
Chen Z1, Zhong C.


Oxidative stress plays a significant role in the pathogenesis of Alzheimer's disease (AD), a devastating disease of the elderly. The brain is more vulnerable than other organs to oxidative stress, and most of the components of neurons (lipids, proteins, and nucleic acids) can be oxidized in AD due to mitochondrial dysfunction, increased metal levels, inflammation, and β-amyloid (Aβ) peptides. Oxidative stress participates in the development of AD by promoting Aβ deposition, tau hyperphosphorylation, and the subsequent loss of synapses and neurons. The relationship between oxidative stress and AD suggests that oxidative stress is an essential part of the pathological process, and antioxidants may be useful for AD treatment.

This old proverb was just passed along to me:

The dogs bark, but the caravan moves on.

Now all the barking dogs don't seem to bother me anymore.

Permalink to Comment

26. johnnyboy on May 12, 2014 7:19 AM writes...

@16: From the bapi phase3 clin trial result:

"...over the 71 weeks of the trial, bapineuzumab, given only once every 13 weeks at 0.5 mg/kg, reduced CSF p-tau. It also kept brain amyloid burden steady in ApoE4 carriers as it gradually increased in the placebo group. (...) Some researchers contacted by Alzforum cautioned that the effect seen in the ApoE4 carrier trial might be due to an uncharacteristic jump in amyloid burden in the last six months of the trial in the untreated group. Most data suggest that amyloid accumulates gradually and has neared its peak by the time patients are symptomatic"

This is NOT a reduction of brain amyloid - it is, at best, an interruption of brain amyloid deposition.

Permalink to Comment

27. Anonymous on May 12, 2014 12:26 PM writes...

@25 Lane Simonian: We're snoring, not barking at you.


Barking would imply that we bother to read what you write and get angry by it. I can assure you that we no longer do either.

Permalink to Comment

28. Snickers on May 12, 2014 7:15 PM writes...

It is indeed a reduction in brain-derived amyloid beta and a consequential reduction in downstream p-Tau in CSF. Even if it isn't removing existing plaques, isn't 'interuption' of amyloid deposition the goal of preventative anti-abeta immunotherapies? Solanezumab didn't even do that.

Permalink to Comment

29. anon on May 12, 2014 8:10 PM writes...

Seriously @25 Lane Simonian

There is nothing stopping you from writing and publishing a peer-reviewed paper on the subject.

Permalink to Comment

30. Lane Simonian on May 12, 2014 11:50 PM writes...

I understand. For some people this is just a passing intellectual debate. For those of us who have lost family members to Alzheimer's disease, understanding and treating the disease is an obsession.

Permalink to Comment

31. LauChang on May 15, 2014 7:00 PM writes...

@20. Anonymous Pharmacologist

I'm afraid you missed the point.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry