Posted 29 August 2006

E-mail discussion Printable version

Return to Discussion Text

Participants: Ruth Itzhaki, Einar Sigurdsson, David Morgan, Ken Ugen, Craig Atwood, Flavio Kamenetz, Jon Oscherwitz, Keith Crutcher, MaryAnn, Richard Bowen, Gabrielle Strobel, Juan Troncoso, Alexei Koudinov, Kiminobu Sugaya, Yasuji Matsuoka, John Travis, Glenda Bishop, Steve Snyder, Orest Hurko, Matthew Hejna, Giulio Pasinetti, Ed Spooner, Jodi Leverone, Jimin Zhang, Klee.

Note: The transcript has been edited for clarity and accuracy.

MaryAnn: My mother is only 62, and in the last stage of Alzheimer's. I want to find out more about early onset Alzheimer's. I'm 31 and already pretty forgetful and I get confused easily. I feel like I'm getting some sort of dementia. I'm looking for someone who does early onset Alzheimer's gene testing in my area. I live in Dayton Ohio. Can anyone help me?

Richard Bowen: Mary Ann: The ApoE genotype test is commercially available, but this is for late-onset. PS-1, PS-2, and AβPP are the primary ones for early onset.

Gabrielle: MaryAnn, unless you belong to a family with an inherited AD risk, which I hope not, there is no reliable gene test predicting your risk. We will be discussing Alzheimer vaccination at the top of the hour but feel free to contact me off-line after the chat to point you towards possible sites.

Richard Bowen: With so many folks looking into the genetics of AD it should be fairly easy. You can contact me at rbowenmd@voyagerpc.com and I can see if I can help.

MaryAnn: I want to learn more about the vaccine. I feel that I may need it someday.

Kiminobu Sugaya: MaryAnn, I am not sure if the vaccine is good or not. Hi! Gabrielle, thank you for the nice editing of the previous chat. May I introduce myself, I am Kiminobu Sugaya at the University of Illinois in Chicago. We have a cold day with snow here.

Gabrielle: Hello everyone and welcome. It is 12:00, so let's begin. I am Gabrielle Strobel, managing editor of the Alzheimer Research Forum, for those of you who do not know me yet. I suggest that perhaps everyone first state their full name so we know who's here. Meanwhile we are trying to figure out a technical problem we have with not enough people being able to log on.

Glenda Bishop: Craig Atwood can't log on.

Richard Bowen: Craig probably didn't have any opinions on amyloid anyway…

Gabrielle: Very funny Richard. Is there someone without an opinion on amyloid?

Dave Morgan: I am Dave Morgan in Tampa. Cool but sunny here.

Orest Hurko: My name is Orest Hurko. I am a neurologist in England.

Juan Troncoso: Hi Orest, you are missed here at Hopkins. I am Juan Troncoso from Neuropath at Johns Hopkins.

Yasuji Matsuoka: Hi, all. I'm Yasuji Matsuoka, Nathan Kline Inst/NYU.

Keith Crutcher: Keith Crutcher, logging in from cool sunny Cincinnati.

Ken Ugen: I am Ken Ugen from University of South Florida in Tampa

Steve Snyder: I am Steve Snyder in Bethesda

Glenda Bishop: My name is Glenda Bishop, I am a postdoc in Cleveland Ohio

Flavio Kamenetz: Flavio Kamenetz from Cold Spring Harbor, NY

John Travis: I'm John Travis of Science News in unexpectedly chilly Wash DC

Matthew Hejna: Matthew Hejna, Loyola Univ. Chicago

Einar Sigurdsson: I am Einar Sigurdsson from NYU in Manhattan.

Ed Spooner/Jodi Leverone: Ed Spooner and Jodi Leverone from the Center for Neurologic Diseases in Boston.

Ruth Itzhaki: Ruth Itzhaki is here (with Curtis Dobson and Matthew Wozniak). Manchester, UK

Jon Oscherwitz: I am Jon Oscherwitz from the University of Michigan

Jimin Zhang: Jimin Zhang, a scientist at Genetics Institute, just interested in this vaccine issue.

Richard Bowen: I am VP of scientific and medical affairs of Voyager Pharmaceutical

Glenda Bishop: I would like to give apologies for Steve Robinson from Monash University in Melbourne Australia. The time difference for this talk is a little incompatible for him, but he did submit a comment to the Forum.

Richard Bowen: I wanted to ask Dave Morgan why he thinks these problems were not seen in animals.

Dave Morgan: Richard, Many possibilities. We obviously were looking for it. My first guess is the mice were not old enough. My second is that the mouse immune system is not the same. My big concern is we still have no good idea what the problem in the clinical trial is.

Richard Bowen: Could it be that they don't have plaques?

Orest Hurko: The mice clearly had plaques but no good evidence of tangles.

Dave Morgan: Both mice and early stage humans have plaques (biopsy studies).

Richard Bowen: I would think that the problem is that you are making antibodies to a protein that has high concentrations in CNS. It would be understandable why this would cause encephalitis.

Glenda Bishop: One reason why problems may not have occurred in the mice is that they still have their own amyloid intact as well as having the mutant human AβPP.

Ruth Itzhaki: To Richard and to Gabrielle - We have been told that herpes simplex virus type 1 has been found in "some" CSFs of the patients with inflammation. It could be responsible for the damage. Mice are not natural hosts for this virus.

Richard Bowen: Ruth - good point. HSV can frequently be activated when the immune system is challenged.

Kiminobu Sugaya: General activation of the immune system may activate glia and produce oxidative stress to cause damage to the neurons.

Juan Troncoso: We do not know what the vaccine formulation was; whether patients develop antibodies; if so, what levels and what type of antibodies, etc.

Richard Bowen: It would be nice to know whether there was a correlation between Aβ titer and encephalitis. (Editor: See ARF news story, scroll down to Schenk, for comment regarding antibodies in patients with encephalitis.)

Gabrielle: Ruth: Assuming study subjects had a latent infection, like many of us, and that the vaccine somehow activated it, how can we find out? And what does that mean for any future vaccine trials? Antiviral drugs don't work so well against HSV in CNS, I believe.

Ruth Itzhaki: : HSV1 DNA could be sought in CSF by PCR, and antibodies to HSV1 proteins in CSF by immunological methods; N.B., HSV1 antibodies are detectable in CSF for very much longer periods than is viral DNA. Future vaccines could be used with antiviral agents, such as acyclovir, and these would stop virus replication. Acyclovir works very well against HSV encephalitis.

Juan Troncoso: Does anyone know how demented the immunized subjects were? Early AD, advanced?

Jon Oscherwitz: They were classified as mild to moderate AD in trial.

Kiminobu Sugaya: On top of that, AβPP may have physiological functions, which may be compromised by the autoimmune problem.

Alexei Koudinov: Good point on physiological role of AβPP.

Dave Morgan: Given that Elan has indicated they will not inject anyone else with this vaccine, I assume the problem is significant. My guess, however, is that passive immunization approaches might still be attempted. (See, for example, ARF news story on one-time antibody injection.)

Yasuji Matsuoka: Do they use adjuvant for vaccination?

Dave Morgan: Yes there was an adjuvant used, but I am not sure what was in it. An article in Nature Medicine indicates the acronym for it, but I was not familiar with it.

Jimin Zhang: Is it possible that the adjuvant caused the inflammation in the trial, not Aβ?

Giulio Pasinetti: {enters}

Dave Morgan: Hi Big Giulio.

Giulio Pasinetti: Hi Dave.

Gabrielle: Welcome, Giulio.

Orest Hurko: It is difficult to do anything but conjecture, since there is very little information available. But am I just not looking in the right places?

Gabrielle: Orest, nothing published in the scientific press that I know of. Elan/AHP is still conducting their own investigation.

Glenda Bishop: Why would you do passive immunization? You are removing the amyloid anyway.

Jon Oscherwitz: Passive immunization would certainly give more control.

Dave Morgan: The advantage of passive is if adverse events occur, they will abate over weeks as the antibodies clear. You cannot unvaccinate someone.

Giulio Pasinetti: Dave, excellent point, this is one of our greatest concerns.

Einar Sigurdsson: Passive immunization may be appropriate for acute disorders but is not very feasible for chronic diseases like AD. Repeated antibody injections may lead to serum sickness.

Alexei Koudinov: I am wondering how many immunologists we have in this auditorium?

Einar Sigurdsson: We need to keep in mind potential toxicity of Ab1-42. There was no reason to use the whole peptide in clinical trials.

Jon Oscherwitz: Why did they use whole peptide? Do we really know the antibody specificities associated with in vivo efficacy?

Glenda Bishop: Passive immunization will remove amyloid anyway, isn't that the problem?

Kiminobu Sugaya: Taking out Aβ is good, but do not forget about the physiological function of AβPP and Aβ.

John Travis: 15 out of 350 people is not stunningly rare. Why wasn't this seen in phase 1--the repeated dosing or a broader population sample that included a susceptible genotype?

Giulio Pasinetti: John, it may be that manipulation of amyloid load or the antibody target (as in the disease) is required to elicit an inflammatory response.

Yasuji Matsuoka: Do they see an elevation of plasma Aβ level after vaccination?

Giulio Pasinetti: Everybody: the reason we are here is to try to put in perspective the evidence that the clinical trials from Elan have been interrupted. What do you think we should do next?

Ruth Itzhaki: Guilio - Get Elan or AHP to seek HSV1 DNA by PCR or antibodies to HSV1 by immunological methods in CSF - or to reveal their results. They have said (in a personal communication) that HSV1 has been detected in "some" of the first four. Our source seems not contactable anymore.

Glenda Bishop: We need to determine why the vaccinations did not work to decide where to go next. Why were the side effects caused, etc?

Juan Troncoso: First of all, we should not abandon the immunotherapy approach because of an initial failure. We need to do more work in animal models, rigorous analyses of immune response in brain, try different adjuvants, and more emphasis on antibody transfer.

Giulio Pasinetti: Juan: this is a point we are trying to make. Far more exhaustive studies are required before we should go forward with clinical trials.

Dave Morgan: One interesting feature of the vaccination in mice that we have recently published is that initially mice have increased microglial activation, but after a number of months of repetitive inoculation it goes away. It will be very interesting to see if the CNS inflammation in these patients resolves over time, and whether they might still show benefit from the treatment.

Giulio Pasinetti: Dave, even if it does so in an experimental model, the cognitive and physiologic consequences in a human are not worth risking.

Gabrielle: If some microglial activation is necessary to remove amyloid, will it be possible to have a therapeutic effect without any inflammation? Can we separate "good" from "bad" inflammation?

Orest Hurko: Dave, isn't microglial activation an inflammatory response?

Glenda Bishop: Yes. Microglial activation is an inflammatory response.

Dave Morgan: I think it's problematic to equate microglia activation with the word inflammation. It depends how one defines inflammation. To me its massive cytokine release with ultimate tissue damage. Microglia have multiple states of activation, some of which are associated with inflammation as defined by these criteria. If microglia activation equals inflammation, it will always be considered bad.

Giulio Pasinetti: Inflammation is a vast term, we need to define features of these responses and relate these to consequences to specific phases of the disease.

Dave Morgan: My general reaction to microglia activation is that some is good and too much is bad. Kind of like red wine.

Giulio Pasinetti: Dave, wine has been shown to prevent AD, however this is not the case for microglial activation. Orest, in a paper in preparation we have been studying microglial activation in the early stages of AD (CDR 0.5 - 1) and we do not show evidence for a strong microglial response.

Juan Troncoso: Do we know whether patients who developed encephalitis after Ab immunization also have systemic immune problems?

Kiminobu Sugaya: The patient has more Aβ and may produce a higher level of antibody against AβPP and Aβ. If we can control the AbPP level rather than completely shutting down the system, it would be helpful.

Glenda Bishop: But if we control AβPP levels, this may affect the beneficial role of AβPP in the brain.

Kiminobu Sugaya: In my study, AβPP regulated stem cell migration and differentiation, which may be necessary for adult neurogenesis. High levels of AβPP cause glial differentiation of stem cells, while low levels of AβPP just increase general differentiation. Thus low level of AβPP is necessary, but high levels of AβPP may produce a problem like AD.

Jon Oscherwitz: Aβ42 generates quite a polyclonal response with many antibody specificities. We probably need to study immunization with more defined sequences.

Kiminobu Sugaya: Some level of microglial activation may be beneficial, but Ab immunization may exceed such a level and we don't have control over how strongly we activate microglia.

Jon Oscherwitz: Kiminobu, perhaps that argues for studying a passive approach to determine the association between antibody level and inflammation.

Kiminobu Sugaya: I think a low level of AβPP is necessary, because in AβPP knockout mice we have not seen any migration of stem cells to the cortex and the lesion did not get repaired. If we can control antibody concentration and determine the level of Ab deposition, we should continue immunization therapy. Otherwise it would be dangerous to do so.

Glenda Bishop: Would you feel comfortable if we tested it on you, Kiminobu?

Kiminobu Sugaya: I am worried about healthy people getting Aβ immunization, because they may continue to eliminate their AβPP, and without AβPP neurogenesis may not occur. Thus I do not want get any immunization in the mean time.

Alexei Koudinov: By the way, why we don't have anybody here from Elan today?

Gabrielle: We invited two representatives from Elan.

Glenda Bishop: When will Elan publish all of its data so that we can understand what is going on?

John Travis: What if Elan doesn't publish its data or conclusions? One could argue the company could discourage others from pursuing the approach and retain a competitive advantage if it doesn't explain the cause. This would only become dangerous/unethical if another company launched a vaccine clinical trial that Elan had reason to fear would be dangerous.

Gabrielle: All here are concerned about Elan not publishing, or at least not fast enough. This is only the latest, not the first, instance of the problem that companies - and academics - are reluctant to publish negative results.

Glenda Bishop: Surely given the topicality of this issue nearly any journal would accept a rapid publication from Elan on this matter.

Orest Hurko: Next Wednesday there is a symposium on Aβ immunization in Paris. I hope Elan will disclose some of the data in that Forum. (Editor: see ARF Ipsen Foundation coverage.)

Yasuji Matsuoka: How about another mode of vaccination therapy? Dave Morgan showed a significant elevation of plasma Aβ after vaccination. Do they see similar change in the clinical trial?

Dave Morgan: Yasuji, it wasn't our group, but Dave Holtzman's that showed the big increase in plasma Aβ. One outcome here is it suggests antibody does get in the brain.

Yasuji Matsuoka: David, excuse my poor memory. Do you see an elevation of plasma Aβ in mice that received passive immunization in your lab? Do you have any idea how much antibody enters the CNS?

Dave Morgan: Yasuji, we have not looked at either issue extensively. Most confirm the antibodies increase plasma Aβ; if we inject into brain we clear Aβ.

Gabrielle: Guilio and Ruth: How about tangle pathology and inflammation? If HSV and Il-6 increase tangle formation, then the vaccine could make the disease much worse. Any way to test this?

Ruth Itzhaki: Gabrielle - we think that HSV1 is normally latent and, as Richard said, can be reactivated by e.g., inflammation and stress. It then becomes an acute infection and causes damage mainly through inflammation. Incidentally, almost all cases of "natural" encephalitis are caused by herpes simplex viruses. The results of the vaccine test suggest that if HSV1 is involved it makes patients worse. We are currently examining the effects of HSV1 infection on cells in culture relating to plaques and NFT.

Alexei Koudinov: I would like to remind the issue that a top journal announced last year that we have to wait for the results of a vaccination study to know whether Aβ is good or bad. We have that now. So, do we limit the discussion to inflammation as a pitfall on the way to the gold, or talk about the whole approach being wrong?

Kiminobu Sugaya: My understanding is that AβPP is necessary for some physiological function of the brain and Aβ production is a byproduct. If it exceeds a certain level it may harm the brain but at low levels, it is not a bad guy.

Glenda Bishop: I don't think that Aβ production is necessarily a byproduct, considering that on its own it can have neurotrophic effects. Craig Atwood says the neurotrophic properties of Aβ have all but been forgotten. Fortunately, or unfortunately, the Elan trial now highlights the importance of this normal physiological property of Aβ.

Alexei Koudinov: And what about its antioxidant capacity at physiological concentrations?

Gabrielle: We are planning an upcoming chat on the physiological function of AβPP. Still a big open question.

Giulio Pasinetti: Do we want to consider combining antiinflammatory therapies with vaccination to keep in check responses coincidental to vaccination? The question is, however, what is the actual mode for clearance induced by vaccination, and what inflammatory cascades are necessary for this response?

Gabrielle: Exactly Guilio, I was going to ask do we have any antiinflammatories that can separate the amyloid-clearing action from the "bad' inflammation. Even experimental ones?

Dave Morgan: Giulio, you are right on. I will send the specific aims for Marcia Gordon's grant!

Gabrielle: Marcia Gordon just has a paper due out this week on one such NSAID. (See ARF news story.)

Giulio Pasinetti: Gabrielle, understanding of vaccination is still in its infancy, and we need to further characterize many facets of these responses before we can tailor antiinflammatory regimens to coincide with this therapy.

Kiminobu Sugaya: Most of the antibodies that worked to eliminate Aβ deposition recognized the N terminal of Aβ, which has a human-specific sequence. That is why in the mice we did not see the side-effect but do in the humans. In the mice, even if the transgene (AβPP) is eliminated by the antibody, they still have their own AβPP system. If inflammation were the only problem, combination of NSAID with vaccination could solve that.

Giulio Pasinetti: Dave, we actually have data demonstrating that both endogenous and exogenous AβPP expression were not influenced by COX-2, only processing of AβPP was altered.

Alexei Koudinov: Am I correct that anti-Aβ-immunization is grounded on the belief that clearing Aβ will cure the disease?

Glenda Bishop: Alex, yes, unfortunately.

Juan Troncoso: Aβ immunization will be grounded as an approach until we know what went wrong in the Elan trial, i.e. adjuvant, Aβ levels, antibody levels, CSF data, etc

Yasuji Matsuoka: Clearance of Aβ is a cure but I am not sure that clearance by microglia is a cure. David Holtzman showed another mode of vaccination therapy, the Aβ sink (see ARF news story). This has been confirmed in a different lab as well.

Yasuji Matsuoka: Dave, do you see more efficient clearance when you inject antibodies into the CNS directly?

Glenda Bishop: Maybe the inflammation side effect we are seeing after vaccination is due to removing the beneficial Aβ.

Orest Hurko: Glenda, do you mean that absence of Aβ would trigger inflammation or that this approach to its clearance results in secondary inflammation?

Glenda Bishop: Craig says to Orest: Yes, it's the removal of the amyloid that is the problem because more than 15 studies show that it has neurotrophic properties including antioxidant, metal-binding and sealing features to keep the blood-brain barrier intact. Remove that and you will engender inflammation. (see Atwood comment.)

Dave Morgan: I think the answer to Aβ physiology is partially addressed by the BACE knockouts. They have no Aβ, yet the phenotype is minor so far.

Glenda Bishop: The BACE knockouts also have not been put under stressful conditions, and it is likely that it is these conditions that will require the beneficial effects of Aβ.

Kiminobu Sugaya: Putting them under stress or lesioning the brain may show phenotype.

Kiminobu Sugaya: In the mouse model the phenotype may not show up because they only live 2 years at most. If somebody can give me the BACE knockout mice I can do the same kind of study I did with AβPP knockout mice to find out if AβPP is important.

Glenda Bishop: Kiminobu, Robert Vassar has the BACE KO mice, and he is aware of the need to put these mice under stress. I don't know if he has done it yet though.

Kiminobu Sugaya: Glenda: what kind of stress?

Glenda Bishop: Kiminobu, anything that causes trauma to the brain.

Kiminobu Sugaya: If he is checking only damage, he may not see the effect. Checking the recovery from the damage would be the target.

Glenda Bishop: Is there any other example of a physiological protein being used to treat a disease? Normally this leads to autoimmune disease.

Orest Hurko: Cancer vaccines.

Glenda Bishop: Against differentiated cells?

Steve Snyder: Insulin.

Kiminobu Sugaya: What happens when insulin was imunnosuppressed?

Giulio Pasinetti: Everybody: we all anticipate Elan's findings both of the inflammation and efficacy of the trials, and look forward to future characterization of vaccination in rodent models. Ciao, gotta go.

Richard Bowen: {enters} Sorry I'm late.

KLee: {enters}

Gabrielle: Welcome Richard! Hi KLee. Would you mind introducing yourself? I am managing editor of the Alzforum.

KLee: Hi, I'm an Information Specialist at the Alzheimer's Disease Education and Referral Center. Just trying to stay on top of the news for our callers.

Gabrielle: What about the blood-brain barrier? There are conflicting reports on whether antibodies cross it, on whether it is breached during AD to allow in peripheral inflammatory cells. Can any vaccination approach be safe in light of this uncertainty?

Craig Atwood: Good question! This point seems to be ignored, but obviously if amyloid is being removed, then the integrity of the BBB must be compromised.

Gabrielle: Craig, why? Could amyloid not be removed without injuring the BBB?

Craig Atwood: I don't know if amyloid removal will maintain vessels intact. Physiologically, amyloid probably deposits to maintain regional integrity given its binding to extracellular matrix proteins.

Kiminobu Sugaya: I am not sure if antibodies cross the BBB. It has to be tested in humans.

Steve Snyder: The amyloid "sink" work by the Holtzman group doesn't appear to involve BBB damage. Maybe it's a ligand and receptor story.

Glenda Bishop: Amyloid deposits during head injury when you expect to see BBB disruption.

Einar Sigurdsson: Numerous labs have shown that Aβ crosses the BBB bi-directionally.

Ruth Itzhaki: Regarding a point raised by Trey Sunderland. He suggested that a virus might be involved in AD--coming from the blood after inflammatory damage to the BBB. Elan's finding that the virus is HSV1 in CSF in the vaccine trials is not consistent with this, as HSV1 resides in nerve cells, not in blood cells.

Craig Atwood: Ruth, exactly! But how does the virus get in the nerve cells?

Ruth Itzhaki: The virus might travel from PNS to CNS by retrograde transport or possibly as a new infection via the olfactory tract. We have found the virus in most elderly people including AD patients. In general, most people now accept that HSV1 is present in brain. HSV1 is present in the PNS of nearly all elderly people. It is latent generally but can reactivate under stress or immunosuppression, and in some people causes cold sores (a relatively high proportion of whom have an apoE4 allele).

Orest Hurko: Glenda, (1) Isn't there normally some flux between systemic antibodies and CSF? (2) Could the antibody be depleting amyloid systemically, creating a sink? Is there traffic of CNS amyloid and the systemic circulation?

Keith Crutcher: Is anyone willing to sign on to the possibility that this is the first in a series of failed clinical trials based on an inappropriate drug target? Allen Roses was the most vocal, but not the only, opponent of targeting amyloid deposition to treat this disease. Will there be a point at which the hypothesis might be said to have failed?

Glenda Bishop: I agree with you.

Alexei Koudinov: I would say that the hypothesis was not proved. Ever. Of the 22 of us here, could any supporter of immunization provide me with a reference showing directly a bad role for Aβ in Alzheimer's?

Orest Hurko: Alexei, how about the Elan report itself. Isn't that the most straightforward explanation.

Ken Ugen: As an HIV vaccinologist now working in AD, I would like to point out that the vaccine approach should not be given up. Historically, vaccines against infectious diseases have saved more lives than all other medical interventions combined. If the smallpox vaccine were being tested now it would likely have been pulled because of adverse events. I believe that it may be possible to fix this problem whether it be untoward cellular immune responses, autoimmunity, or adjuvants.

Gabrielle: Ken, what do you say to the detractors (see also Stephen Robinson's comment) that until we know the physiological function of AβPP and Aβ, they are dangerous targets?

Ken Ugen: Gabrielle, nothing in life is perfectly safe.

Kiminobu Sugaya: If it is not perfectly safe I would not get the vaccine.

Kiminobu Sugaya: A vaccine to a virus like HIV is different; it just needs to eliminate the enemy. A vaccine to Aβ is eliminating the functional protein in our body.

Richard Bowen: Ken, have there been any other vaccines produced against ubiquitous proteins that have significant deposits in certain tissues? It seems to me that making a vaccine against Aβ for AD is similar to making a vaccine against uric acid to treat gouty arthritis. The expected result would be a much worsening of the arthritis.

Gabrielle: Ken and Dave, will the failure of this trial make investors and IRBs more hesitant and put a damper on new approaches? Do you feel any of that?

Dave Morgan: Gabrielle, I think all will have additional caution in moving forward with the approach, and reasonably so.

Kiminobu Sugaya: Once we have an antibody against one of our proteins it will stay there for a long time without control, it is not like a drug.

Gabrielle: We are nearing the end of the hour. If anyone wants to make a final statement on where we should go next, now is the time. I would love some suggestions by everyone as to what scientific questions they would like to see answered (We all hope for full disclosure from Elan.) To those who have to leave soon: thanks for joining us, and be back.

Orest Hurko: Goodbye. I hope Elan provides us more data -- some of these speculations could be tested against data that must already exist.

Kiminobu Sugaya: If antibodies cross the BBB and if we can find a good specific antibody against Aβ deposition without compromising the physiological function of AβPP, that would be nice.

Yasuji Matsuoka: If the periphery sink is a primary mode, then we don't have to stick with antibody/immune reaction.

Dave Morgan: Thanks for the chat.

Ruth Itzhaki: We hope that the possibility of HSV1 involvement in AD, further supported by Elan's trial, will encourage the companies to release more information. The trial results must be a great disappointment for patients and their carers, particularly in the absence of information. This is especially important given that safe (off-patent) antiviral drugs are currently available.

Glenda Bishop: I think that until Elan puts out specific information regarding the failure of the trial, it is not possible to decide what should be done next. We need to know whether failure was due to immunizing against Aβ, interactions with AβPP, or problems with the adjuvant. Until then it is dangerous to persist with any vaccination trials. Goodbye.

Alexei Koudinov: What we learn from this is that curing patients is more complicated than getting Aβ aggregated in the test tube.

Keith Crutcher: My concern is that this will go on for a very long time with refinements of the approach substituting for critical testing of the underlying hypotheses.

Gabrielle: Keith, do you have a suggestion for an experiment that would critically test the underlying hypothesis?

Keith Crutcher: I would say that once there is some neuropathology of patients being treated for amyloid deposition and they show loss of amyloid but no improvement (or worsening), that this is the time to seriously question the assumptions about the target.

Alexei Koudinov: The amyloid hypothesis has been on the market for 15 years and gradually become dogma. How long will it take to address Keith's point?

Gabrielle: Keith, this makes sense. It would also be fun to treat tau transgenic mice with the present vaccines.

Keith Crutcher: Yes it would.

Alexei Koudinov: It will not be fun, as tau has clear normal function.

Gabrielle: Alexei, Ruth and Giulio have written in their comments that viral infection, Il-6 expression, and tau pathology may be connected.

Ruth Itzhaki: Gabrielle, thanks for mentioning this. We hope to test this in in vitro shortly.

Alexei Koudinov: I see.

Keith Crutcher: I need to go but I am placing my bets on Ruth's work as an important clue to the etiology of this disease. Cheers to all!

Ruth Itzhaki: Thanks Keith. Your support is appreciated! Goodbye all!

Gabrielle: Ken, in your view, are we squabbling too much about safety problems that are to be expected in a radical new approach? Also, are there reliable methods to make passive immunization safe from serum sickness, i.e. humanize antibodies and the like? Dave Holtzman's "sink" approach sounds attractive.

Ken Ugen: I did not mean for my comments earlier to indicate that I do not think that safety is important. These are significant adverse effects. However, it is possible that further research may circumvent some of these problems. In particular it may be important to do studies in non-human primates. This approach has been fast-tracked in the clinic (much like HIV drugs and vaccine work) because of the serious nature of the disease. There are advantages and disadvantages in doing this. As for passive immune therapy, the way to go is to use human monoclonal antibodies. No serum sickness concerns.

Gabrielle: Ken, are you aware of Beka Solomon's recent Diagnostics paper in which she describes raising anti-aggregating Aβ antibodies with a filamentous phage displaying amino acids 3 to 6 of Aβ, the epitope said to regulate fibril formation? No adjuvant, high titres. What do you think of it? Would that be safer?

Ken Ugen: Yes, I think that this is a potentially good approach since the antibody is only against a fraction of the 42 amino acids in Aβ. This may cut down on adverse events. Also, a truncated peptide displaying only a fraction of the amino acids may be something to try, too, like Einar and Blas Frangione are doing.

Gabrielle: Have a good day everyone. We will fix the access problems we had today. Thank you everyone and goodbye for now.

Kiminobu Sugaya: See you everybody.

Einar Sigurdsson: Goodbye.

Juan Troncoso: Good-bye everyone.

Glenda Bishop: Bye.

Craig Atwood: Goodbye.

Richard Bowen: Goodbye.

Ken Ugen: Goodbye to everyone

Alexei Koudinov: Thanks for today's discussion and talk to you soon again.