[ed note: Michael Jorrin, who I like call “Doc Gumshoe,” is a longtime medical writer who shares his thoughts with us a couple times a month — his articles are non-financial in nature, his words and opinions are his own, and you can see his bio and his past pieces here. Enjoy!]
Indulge me while I spin an anecdote that might have a tiny bit of relevance to Alzheimer’s disease.
In the dawn of the automobile age – and, no, I am not old enough to remember it! – the circuit that triggered the self-starter was called the “make-and-break.” In Cuba, where I was born and lived as a small child, this became the mequembreque. And when the car would not start, this was attributed to borronilla en el mequembreque. “Borronilla” literally means “eraser crumbs,” but the word is used loosely to describe crud in general. Then, when my father felt a mite under the weather, he would sometimes attribute it to borronilla en el mequembreque _i.e. nonspecific crud in the system.
When I think about AD, and in particular to the various root causes that are being investigated – amyloid beta (Aβ), tau protein, inflammation, toxins, fungi, life-style factors – I am irresistibly drawn to borronilla en el mequembreque. Any or all of those factors produce crud in the system, and this crud interferes with brain function. I do not say that continuing research into the pathophysiology of AD is in vain, and that scientists should just throw up their hands and accept the generalized crud hypothesis. No indeed! That research needs to continue. But the real test is whether reducing the buildup of crud results in getting the car going.
We have seen evidence that some individuals with abundant quantities of brain crud – e.g., amyloid plaque – do not experience significant cognitive decline. We cannot say whether this is because amyloid plaque is not really the root cause of AD or because these individuals have redundant cognitive resources. I suspect the latter. Yes, the buildup of Aβ in the space between neurons blocks some synapses, but these persons have other neuronal links that haven’t been clogged by crud, and their brains continue to function.
It is in the nature of scientists to search for single causes. The direct line from cause to effect is the prized goal of much research, and success is marked by the discovery of the means of interrupting that direct line: block the cause and the effect does not occur. This is the ultimate goal of a lot of AD research. If the buildup of Aβ plaque is the unitary cause of AD, and a means of preventing this buildup is found, and preventing that buildup then prevents the development of the characteristic AD symptoms, then the battle is won.
But so far, as we know, the battle has not been won – far from it! And not just relative to Aβ buildup, but relative to all the proposed causes of AD. In fact, up to now, evidence has been scanty that strategies that address any of the putative causes of AD result in significant clinical benefits.
That’s not a reason to quit trying. AD researchers are no doubt aware of the huge gap between the identification of cholesterol as the substance that forms arterial plaque and the first reliable data that lowering cholesterol improved survival in persons at risk for heart disease. Before 1910 there was extensive demonstration not only of the presence of cholesterol in arterial plaque, but that animals fed a diet rich in cholesterol developed atherosclerosis. But it was only about 90 years later that the famous 4S trial (Scandinavian Simvastatin Survival Study) for the first time demonstrated that interrupting that cause-effect link improved survival.
Let’s hope that similar success with AD does not take 90 years! For now, let’s check in on what’s going on currently with the development of prospective AD treatment options.
This class of agents offered early promise. They target an enzyme that cleaves amyloid precursor protein (APP), which includes the peptide sequence containing Aβ, thus resulting in Aβ accumulation in the synaptic space between neurons, and inhibiting the transmittal of neurotransmitters across this space. Since the transmittal of neurotransmitters from one neuron to another essentially constitutes brain activity, Aβ accumulation preventing this activity has been considered to be, if not the, at least an, essential cause of AD. The enzyme mostly responsible for creating that particular kind of crud is called beta (β) secretase, aka BACE, and several agents inhibiting the action of that enzyme have been in various stages of development for a number of years.
Most of these agents are in a class called monoclonal antibodies (mAbs), and they are derived from various sources. Researchers identify antibodies that have emerged in response to various challenges, some in mice (murine) or other animals, some in humans. These antibodies are then cloned and produced in the laboratory. The process of creating mAbs is basically the same whether the antibodies are in response to an immune process, as in rheumatoid arthritis, to a cancer, or, in the case of AD, to the enzyme that does most of the work of cleaving Aβ from the precursor protein.
Good or bad news about these agents? Well, it depends on how you look at it.
Solanezumab, from Eli Lilly (LLY), failed to meet its primary end-points in terms of cognition and activities of daily living in a Phase III trial. More recently, analysis of data from that trial indicated that in patients taking solanezumab, disease progression was slower than in patients taking placebo. Slowing disease progression is not as good as stopping it cold, but it’s better than nothing.
Bapineuzumab, from Johnson & Johnson/Pfizer (JNJ/PFE), was supposedly scrapped three years ago, for a similar Phase III trial failure. Although brain imaging showed positive changes in Aβ deposition, the patients taking bapineuzumab did not demonstrate clinical improvement in cognition and activities of daily living.
Ganterenumab, from Roche (RHHBY), has also been scrapped, or at least interrupted. The mechanism of action of this agent is similar to that of bapineuzumab in that it binds primarily fibrillar, deposited Aβ, not soluble monomeric Aβ as does solanezumab, or a mixture of Aβ species, as do crenezumab and Biogens’s aducanumab. Ganterenumab is hampered by the risk of an adverse effect known as ARIA (amyloid-related imaging abnormalities), which can lead to dangerous swelling of the brain, thus there is a reluctance to push the dose, and effective dosing levels may not have been reached in clinical trials. Roche is conducting another trial, called Marguerite Road, with ganterenumab, in patients with earlier AD, to see if the progress of dementia can be significantly forestalled over a two-year period.
Aducanumab, from Biogen (BIIB), formerly known as BIIB 037, was discussed at length in the previous Doc Gumshoe AD piece, which you can see here. As we reported in that piece, Biogen has a few other agents in development, including BAN2401 and E2609.
Crenezumab, from Genentech/Roche in partnership with AC Immune, also failed to meet endpoints in a Phase II trial, but reported some improvement in cognition in patients with mild AD. This constitutes a “glimmer of hope,” and in July, 2015, a large Phase III trial was announced, with results at least three years in the future.
Verubecestat (MK-8931), from Merck (MRK), has been touted as the leader in the BACE inhibitor race. A distinct advantage for this candidate is that it is a small molecule – not a monoclonal antibody – meaning that it has a better shot at passing the blood-brain barrier. A large Phase II/III trial in MK-8931, in 1,960 subjects with mild-to-moderate AD, two thirds of whom will receive the study drug, has recently been launched, with a 26 month duration, and another Phase III trial in 1500 subjects with early AD will announce results in about 5 years. Previous smaller trials in this drug have reported positive results, giving Merck incentive to push this candidate. Merck appears to be playing this close to the chest.
CNP520, an early stage BACE inhibitor, is a sort of joint venture by Amgen (AMGN) and Novartis (NVS). These two outfits have decided to pool their efforts in this field, sharing the costs now and presumably sharing the revenue later on. They reportedly have other candidates in the running, and they plan to work on them cooperatively.
On balance, I cannot say that the BACE inhibitor prospects are all that promising. This emphatically does not mean that the strategy of preventing or slowing Aβ accumulation is a failure. However, a problem with the mAb class is the difficulty these molecules have in passing the blood-brain barrier. The crenezumab molecule is huge – it consists of about 20,000 atomic units, with a weight of 144.88 kiloDaltons. (For comparison, TRx 0237, which we also discussed at length in the previous Doc Gumshoe piece about AD, consists of only about 40 atomic units.) As a result, the concentration of the antibodies within the brain is only about 0.1% of the concentration in the serum – or, to put it another way, to get enough concentration in the brain, the serum concentration has to be about 1000 times higher. This invariably entails a risk of adverse effects, so researchers are understandably highly reluctant to escalate the dose. No one at this point can be certain that beta-secretase does not play significant roles elsewhere in the body, so huge doses of a BACE inhibitor may be chancy.
The possibility of a vaccine that might protect against AD is obviously enticing. At this point, it’s no more than a possibility, but enticing nonetheless. Technically, the vaccine is active immunotherapy, meaning that it triggers the body’s immune system to create the antibodies to the formation of Aβ. Instead of trying to interrupt the formation of Aβ by dosing the patient with externally formed antibodies (passive immunotherapy), active immunotherapy stimulates the body to do the work of combating the β-secretase enzyme. This is not the kind of one-time vaccine that you only have to take once in life, or once a year – it needs to be taken on a long-term basis, but it would work the same way. This strategy would likely sidestep the blood-brain barrier obstacles that hamper the BACE inhibitors. However, the vaccines currently under development only address the Aβ hypothesis, so if the root cause of AD is something else, they’re not apt to do much good.
The first attempt at a vaccine was AN1792, originally from Elan Pharmaceuticals, which was acquired by the Irish outfit, Perrigo (PRGO). (Other Elan/Perrigo agents were discussed in the previous Doc Gumshoe AD piece.) AN1792 did provoke an antibody response, but only in about 20% of subjects. However, about 6% of subjects developed meningoencephalitis, so development was discontinued and AN1792 was scrapped.
CAD106, from Novartis, may be the most promising agent for active immunotherapy against Aβ. In two Phase IIa studies, over 90% of subjects developed antibody responses, and did not develop the Aβ-specific T-cell response that was the cause of the adverse events that sank AN1792. Previous studies have bolstered the safety profile of CAD106, and provided persuasive evidence that it is suitable for long-term therapy.
Affitope AD01 and AD02, from Affiris AG (Austria), completed Phase I trials, but a Phase II trial in AD02 has been terminated, for unstated reasons.
A number of other vaccines/active immunotherapy candidates are in early stages of research and development. At the risk of going over old ground, I should repeat that the process of bringing a drug of any kind, for any disease or condition, is necessarily slow, painstaking, and expensive. First, even if a potential mechanism for attacking the disease has been identified, and a specific agent that employs that mechanism has been found, that agent, whether a simple molecule or a hugely complex mAb, has to be shown to be safe for administration to humans. Then the researchers have to figure out a way to get the agent to where it’s needed. Most of the other vaccine candidates are still in those early stages, and early successes – e.g., we gave it to 20 subjects and it didn’t kill any of them – are no cause for shooting off fireworks.
Drugs aimed at the tangles formed by tau protein are nowhere near as far along nor as numerous as the agents directed at amyloid beta. The previous Doc Gumshoe piece discussed the tau hypothesis as well as the agent TRx0237, which is out in front in the tau immunotherapy sweepstakes. To brush up, when the normal chemical structure of tau protein is altered by the addition of extra phosphorus radicals, it forms fibrillary tangles inside neurons, and in particular, in the axons that connect neurons with the central nervous system. This is fundamentally different from what Aβ does, which is to accumulate in sheets in the space between synapses and prevent the passage of information from one neuron to the next. The damage wreaked by tau neurofibrillary tangles (these are now commonly abbreviated NFTs) consists in particular of starving neurons of nourishment, which is transported via the axons.
TRx0237, from TauRX, of Aberdeen, Scotland, has demonstrated promising results in early studies – in particular, a highly significant (P < 0.001) increase in cerebral blood flow in patients with mild AD. The NIH currently lists five clinical trials in TRx0237, one of which is a completed safety study. Three others are listed as “active, not recruiting,” and one is “by invitation only.” To me, this signals “go slow.” The drug, TRx0237, is a derivative of a fairly common compound, methylene blue, which has been used for the treatment of malaria and an uncommon blood disease. Their version is methylonimium chloride, and researchers are working on variations of the compound to improve absorption and distribution.
ACI-35, from AC Immune, is in the category of active immunotherapy (i.e., vaccines) against hyperphosphorylated tau. The rationale behind it is that the vaccine will treat tauopathy in AD by eliciting an immune response targeted to certain pathological forms tau without also mounting autoimmune B cell or T cell responses against normal forms of this ubiquitous intracellular protein. Results of preclinical studies in mice suggested that it might produce definitely favorable clinical responses in humans – among other outcomes, it increased mouse life-spans and did not result in inflammation in the brain. A Phase I clinical trial was started in 2013; results have not yet been made public. In 2015, AC Immune partnered with Johnson and Johnson to develop ACI-35.
AADvac1, from Axon Neuroscience, Bratislava, Slovac Republic, is also a vaccine directed against tau NFTs. In July 2013, a Phase I study evaluating AADvac1 immunotherapy against diseased tau protein was started. The primary goal of the Phase I study in 30 AD patients was to assess safety and tolerability of this compound.
The results, presented at the at the Alzheimer’s Association International Conference 2015 (AAIC) in Washington DC, reported the vaccine as safe and well tolerated. Most subjects experienced robust immune responses, and their average cognition remained stable over 6 months. These results should be viewed as promising, although the duration of the study was short in terms of evaluating changes in cognition.
Semagacestat (LY450139), from Eli Lilly, not only failed to demonstrate any benefit in a Phase III trial, but was associated with non-trivial adverse effects, including skin cancers and infections. Compared with placebo, semagacestat did not improve cognitive status, and patients receiving the higher dose had significant worsening of functional ability, so it’s curtains for that candidate.
I have no doubt that other tau-directed therapies are in the works. Doc Gumshoe will keep his ear to the ground.
Cholinesterase inhibitors and other agents meant to stimulate brain activity
I don’t want to spend much time on these. They do not pretend to alter the course of AD, only to slow the advance of dementia, which is, of course, a good thing in the present situation where no drug or therapy has actually demonstrated disease-modifying activity. Cholinesterase is an enzyme that degrades acetylcholine, which is a neurotransmitter essential for brain activity, so inhibiting that enzyme increases the availability of acetylcholine and should enhance cognition.
Donepezil (Aricept), from Eisai/Pfizer (ELSALY/PFE), is actually FDA approved for AD. It was originally marketed for Parkinsonism, and has demonstrated benefit in preventing persons with Parkinson’s from experiencing falls, which is one of the troublesome symptoms of that disease. The effect on cognition, both in Parkinson’s and AD, is acknowledged to be small; however, it has become part of the standard treatment.
MK-7622, from Merck, is a cholesterinase inhibitor being studied as adjunctive treatment with donepezil. No studies have been published on this compound in the peer-reviewed literature or other publicly available documents. In October 2013, Merck began a large Phase 2 trial (830 subjects) in the US to evaluate the compound’s tolerability and efficacy as an add-on to donepezil in patients with probable AD.
RO53313534, from Roche, was discussed in the previous Doc Gumshoe AD piece. It is not a cholesterinase inhibitor, but an acetylcholine receptor agonist. (Apologies for using the word “agonist,” which has nothing to do with agony, but is merely the opposite of “antagonist” – i.e., it activates the acetylcholine receptor). It was evaluated, also in combination with donepezil, and an early trial reported raid improvement in cognition as early as 8 hours post-dosing. Results of a more recent clinical trial have not been announced.
Memantine HCl, marketed in the US as Namenda, by Forest, addresses glutamanergic receptors in the brain and presumably stimulates cognitive activity. It is also marketed under a number of names in other countries, by Actavis and Merz Pharma. It is approved for AD in the US and also used in the treatment of Parkinson’s disease. As with donepezil, its benefit in terms of cognition is small, but under the circumstances of limited treatment options, it continues to be prescribed.
Agents and supplements with other mechanisms
Azeliragon (TTP488), from vTv Therapeutics (VTVT), is a small-molecule inhibitor of a receptor that has been reported to be involved in the toxic effects of Aβ in neurons and also in inflammation and oxidative damage in the brain. The rationale for blocking this receptor is that it would result in a multiple treatment effect. The agent was initially developed by TransTech Pharma as TTP488 and licensed to Pfizer as PF-04494700. A Phase II trial, funded by Pfizer and the National Institute on Aging, initially reported no benefit, but subsequent analysis reported significant clinical benefit for a low-dose regimen. Pfizer discontinued its work with this agent in 2011, and licensed it to vTv Therapeutics, which has initiated an 18 month Phase III trial in 800 subjects with mild probable AD. The outcome measures are the cognitive instruments described in the previous Doc Gumshoe piece. Results are expected in 2018.
Rosiglitazone (Avandia), from Glaxo SmithKline (GSK), is a drug whose primary use is in type 2 diabetes. I hesitate to mention it at all, since it has demonstrated no benefit whatever in AD, but a brief discussion may serve to puncture the notion that AD is nothing other than diabetes of the brain. The theory is that the AD-associated cognitive decline is due to the inability of the brain to derive energy from glucose due to problems with insulin sensitivity. That does not appear to be the case, although researchers continue to look for ways to nourish neurons other than glucose, e.g., Ketasyn and Axona, discussed in the previous Doc Gumshoe.
Anavex 2-73, from Anavex (AVXL), has been discussed by Travis on a couple of occasions in the past. The company keeps proclaiming that their drug does wonders in enhancing cognition. In the opinion of the ever-skeptical Doc Gumshoe, their claims are hooey. For example, this past month they announced – to great fanfare – the results of a small uncontrolled clinical trial in which they gave about 50 patients their drug, following which they gave them the usual tests of cognition. The patients did a little better, but there is no way of knowing whether this was because of the drug or a placebo effect. All of the patients in the trial got Anavex 2-73, and it would be entirely to be expected if these patients, pleased and excited to be in a clinical trial of an exciting new drug, actually did do a little better on those instruments. The absence of a placebo group ramps up my skepticism quotient by a factor of ten at least, and the ballyhoo of their announcement by another ten. Anavex has another candidate, Anavex 3-71, which targets the same receptors. So far it has been tested in mice only. This is a Wait and See proposition.
NSAIDs have been thought possibly to slow the onset of AD, or, possibly, to delay disease progression. The basis for this thinking has been the intuitively reasonable supposition that inflammation plays a role in AD, and that NSAIDs, active against inflammation, could counter AD. However, a large controlled trial – the Alzheimer’s Disease Anti-inflammatory Prevention Trial (ADAPT) in more than 2,000 subjects, reported contradictory results. In persons with symptomatic AD, who represented a small fraction of the 2,000 subjects, NSAIDs appeared to increase the risk of AD progression more than placebo, while in asymptomatic persons, one of the trial NSAIDs (naproxen) appeared to delay the risk of onset, after about 3 years of treatment. The bottom line is blurry; it can’t be said for sure whether NSAIDs help or hurt, but they are for sure not the magic bullet.
Aspirin has a number of mechanisms that the other NSAIDs don’t share. In addition to being an anti-inflammatory and analgesic, aspirin acts to prevent the formation of blood clots and also lowers fevers. Aspirin has been reported to improve cognitive function in some subjects, but there’s no evidence that it slows or prevents the onset of AD. Salicylic acid (aspirin is acetylsalicylic acid) has been reported to bind with an enzyme called GAPDH which is thought to contribute to cell death; when salicylic acid binds with GAPDH the resulting particle can no longer pass the blood-brain barrier. Thus, aspirin may possibly prevent cell death in the brain. The objective of further research would likely be finding a way of delivering the supposedly beneficial salicylic acid effect without incurring the bleeding risk that would accompany high dose aspirin.
Several Gumshoe Denizens have requested a whatever I could dig up on therapies that are in early stages. Here’s a quick take on some of them:
- Cerefolin NAC is a food supplement meant to nourish brain activity; there is no data on its efficacy.
- Huperzine A is a supplement with cholesterinase-inhibiting properties, meaning that its mechanism is similar to that of donepezil. At present, there is insufficient safety information, and until there is more data, most doctors don’t recommend taking huperzine A instead of the FDA-approved cholesterinase inhibitors.
- Bryostatin-1, from Neurotrope (NTRP), has previously aroused considerable interest as a cancer agent. It inhibits the enzyme protein kinase C, which results in the inhibition of tumor cell proliferation and the induction of tumor cell death. In AD, bryostatin-1 has been used in the treatment of just one patient, with positive results. The Blanchette Rockefeller Neurosciences Institute has received permission from the FDA to conduct Phase II clinical trials with bryostatin in patients with AD, initially to access safety. This one is a bit closer to having results.
- Anatabine citrate, from Rock Creek Pharmaceuticals (RCPI), is an alkaloid of the tobacco family, found also in tomatoes, eggplant, and other vegetables. It has anti-inflammatory properties, and an anatabine supplement called Anatabloc is marketed for relief of joint pain. A paper in PLOS reports that anatabine reduces Aβ in mice and also improves the cognitive performance of mice. A trial of Anatabloc in patients with AD was terminated for unstated reasons. The plan to conduct Phase I clinical trials with anatabine in AD was announced in May 2015.
And, finally …
The BRCA1 gene, which we have all heard about exhaustively in connection with breast cancer, may also have a connection with AD. A group of NIH-funded scientists showed that lower levels of BRCA1 cause neurological problems by experimentally reducing BRCA1 in the neurons of mice, which developed learning and memory deficits, perhaps as a consequence of lower BRCA1 expression. BRCA1 has so far been studied primarily in dividing (multiplying) cells, such as in cancer, which is characterized by abnormal increases in cell numbers. But now it also appears to play important roles in neurons, which do not divide. The causal link between BRCA1 and cognition may be due to the gene’s ability to repair damaged DNA. BRCA1 was found to be reduced by 65% to 75% in the brains of human Alzheimer’s patients. It is possible that manipulating BRCA1 may ultimately be used to prevent neuronal damage and cognitive decline in patients with Alzheimer’s disease.
What should we make of all this?
I haven’t counted up how many treatment options have been discussed here, however briefly, but many more than in similar Doc Gumshoe pieces about subjects such as cancer treatment. Why is this? The quick answer is that finding an effective intervention for Alzheimer’s disease is equivalent to coming upon the pot of gold at the end of the rainbow. The disease can be seen as threatening to just about everybody who lives long enough, and treatment costs are expected to soar in coming years, perhaps to the trillion dollar range by midcentury. And if some form of intervention were found that could reliably prevent the onset of AD, you can bet that just about everybody who could afford it would start taking it by about age 60, or even younger. Only people with elevated cholesterol take statins; only people with high blood pressure take antihypertensives; only people who actually have cancer take cancer drugs. But a drug that would prevent AD? The lines to get it would go around the planet.
Therefore, there’s an immense incentive to try anything at all that might work. The strategy being employed by many of the small biotechs – not naming names here! – is to pick a mechanism, do some studies that they hope will come out with encouraging results, and then release glowing reports about these results, particularly directed to the financial media, in the hopes of boosting their share prices, or perhaps of attracting the interest of a larger pharmaceutical company which would either buy out the biotech or participate in further clinical studies.
Some of the candidates in development by the small biotechs may indeed be highly promising, and among them may be the AD treatment successes of the future. At this point it is too early to say anything conclusive about most of them. As you may have gathered, Doc Gumshoe tends to be put off by what strikes him as inflated blather about these early results. But concealed in the blather there might be that elusive morsel of substance.
Big Pharma is less apt to engage in glowing statements about early results. In fact, you may have noticed that in many cases they play their cards close to the chest. They may not want to inflate their stock and have it come crashing down when there’s a bit of bad news. Or they may not want to incite pirates to launch efforts to copy their candidates. But my observation is that many of the most promising treatment prospects come from the bigger outfits.
One last small note on the amyloid beta (Aβ) versus tau protein “controversy:” I think this is inflated way out of proportion. To be sure, each conclave of researchers would like to be able to claim that they and only they are right. It is in their nature to make this claim and stick to it. But the likelihood is that both Aβ and tau have a significant role in the pathophysiology of AD. And it is also likely that, in some patients at least, other factors also play a role – genetic factors, certainly, and also perhaps infection, inflammation, life-style factors, and others. I revert to my point of departure for this Doc Gumshoe installment – borronilla en el mequembreque – an accumulation of crud in the system.
May we find a way to clear it out!
Happy Holidays to all from your correspondent, Michael Jorrin (aka Doc Gumshoe)
* * * * * * *