Hopes of harnessing the immune system to fight Alzheimer disease took another hit with last month’s announcement that an adverse effect has forced pharmaceutical partners Elan and Wyeth to halt dosing in a Phase 2 clinical trial of their newest active vaccine for AD. A study published yesterday in the journal PLoS ONE may cushion the blow, but it also fuels budding questions in the field about how future vaccines could be geared toward AD prevention more than treatment. Southern California researchers have produced a DNA-based vaccine that prevents the buildup of toxic Aβ plaques in an Alzheimer’s mouse model by driving a strong anti-Aβ antibody response. Importantly, this protection came without the generation of self-reactive, Aβ-specific T cells—the suspected culprits behind vexing brain inflammation events that have plagued the development of safe vaccines for AD for some years. Though much work remains to move such a DNA-based vaccine toward human testing, scientists say the mouse study is a step in the right direction.

“This is a nice and logical advance in the AD immunotherapy field,” said Brian Bacskai of the Alzheimer’s Disease Research Unit, Massachusetts General Hospital, Boston.

The PLoS ONE study—by Michael Agadjanyan, David Cribbs, and colleagues at the University of California, Irvine, and the Institute for Molecular Medicine, Huntington Beach—extends recent work by the same scientists. Those earlier mouse studies (Mamikonyan et al., 2007; Petrushina et al., 2007) used an anti-Aβ peptide vaccine that seemed to avert the suspected T cell auto-reactivity woes that halted clinical trials of the first AD vaccine to reach human testing, back in 2002 (see ARF related news story).

Given that robust production of specific antibodies by B cells requires help from activated T cells, the question at the time was how to design a vaccine that would induce high levels of anti-Aβ antibodies without simultaneously raising an army of Aβ-specific T cells that could invade the brain and wreak havoc. The answer: summon help from irrelevant (i.e., non-Aβ-specific) activated T cells. So instead of using full-length Aβ42 peptide, which contains both B cell and T cell epitopes, Agadjanyan’s team fashioned a new immunogen. It used Aβ1-11 peptide to drive anti-Aβ antibody production and a synthetic T cell activating peptide, PADRE (pan HLA DR-binding epitope) (Alexander et al., 1994), to provide help without triggering production of Aβ-specific T cells (see ARF related news story). In adult AD mice, this vaccine cleared plaques and insoluble Aβ without inducing adverse events linked to Aβ-reactive T cells. However, it did not decrease soluble Aβ in animals with established plaque pathology. These results suggested to the authors that their vaccine might only work as a preventive tool in people expected to develop AD later.

Such a preventive vaccine would need to be given frequently, but producing large batches of peptide-based vaccine is expensive and difficult, Agadjanyan told this reporter. In addition, he noted that peptide-based vaccines require strong adjuvants, which have been blamed for the harmful cellular immune responses that suspended AN1792 clinical trials (Pride et al., 2008).

To avoid these problems, first author Nina Movsesyan and colleagues applied their peptide immunotherapy approach toward a DNA-based vaccine that could be produced more easily and cheaply. DNA vaccines have been tested as cancer and flu treatments in a number of human trials, though most have proved ineffective because the antibody responses they trigger are too weak, and none have gained FDA approval. Indeed, earlier efforts by Agadjanyan’s group and others to make DNA vaccines using full-length Aβ42 also met this ill fate (Ghochikyan et al., 2003; Schultz et al., 2004).

In the new study, the researchers engineered a DNA vaccine from a plasmid containing three copies of the gene encoding the B cell epitope Aβ1-11, and the gene for the synthetic T cell peptide, PADRE. Hoping to overcome the efficacy problems that have stymied DNA vaccine development, they added to this plasmid the gene for macrophage-derived chemokine (MDC/CCL22), which simultaneously activates the helper T cells that drive antibody production and tones down the T cell subset that mediates inflammatory responses.

In three- to four-month-old wild-type and 3xTg-AD transgenic mice given the DNA vaccine, Movsesyan and colleagues found strong anti-Aβ antibody responses. The resulting antibodies recognized monomeric, oligomeric, and fibrillar forms of human Aβ, as well as Aβ deposits in brain tissue from an AD patient. Consistent with results using the Aβ1-11/PADRE peptide vaccine, the DNA vaccine sparked strong anti-PADRE, but not Aβ-specific, T cell immunity in the 3xTg-AD mice.

Vaccinating transgenic mice that did not yet show AD-like pathology restored their spatial memory test performance to that of wild-type counterparts, with untreated and irrelevant plasmid-vaccinated transgenic animals lagging behind. Analyzing the brains of these older (~18-month-old) transgenic mice, the researchers saw significantly fewer Aβ plaques in the immunized animals. These mice also had sharply reduced levels of soluble Aβ42 and Aβ40 peptides, potentially toxic forms of brain amyloid. On the safety front, the DNA vaccine did not appear to induce T cell infiltration or brain hemorrhages in treated animals.

“The data was very thorough and well thought out,” said immunologist and CEO Marc Hertz of Pharmexa-Epimmune, a U.S. subsidiary of Denmark-based Pharmexa A/S that develops T cell-based immunotherapies and owns the patent on the PADRE epitope. “My concerns have more to do with the practicality of commercializing it.”

For starters, the vaccination approach presented in the PLoS ONE study is not a therapeutic model, but a preventive one. The immunized mice did not yet have AD pathology. The findings strengthened Agadjanyan’s view that prevention is the way of the future when it comes to AD immunotherapy. “DNA vaccination will be safer and much more effective if used as a preventive measure,” he said. More broadly, the issue of prevention versus treatment is coming up with increasing frequency in discussions about development of other experimental AD drugs, as well, and it will likely come to the fore as ongoing research validates candidate biomarkers for preclinical AD.

In an e-mailed response, Hertz pointed out that a DNA vaccine would in theory “be given rather frequently for an extended period of time—tens of years—and there would be significant safety concerns in such repeated administration of the vaccine in healthy individuals.” See below for an extended dialog between Hertz and Agadjanyan.

Hertz also noted that the DNA dose and number of administrations were extreme for a rodent model. In the new study, the mice received nine micrograms of DNA—equivalent to 4.5 times a typical human dose for a gene gun delivery approach—every two to six weeks for 16 months. Acknowledging delivery concerns, Agadjanyan said his team is gearing up for tests of their new vaccine using an electroporation system developed by San Diego-based Ichor Medical Systems, Inc.. This technology has been shown to boost DNA vaccine potency 10- to 1,000-fold compared to conventional injection methods. Furthermore, the studies are planned in monkeys—a key proof-of-principle, as DNA vaccines have a track record of poor antibody responses in larger animals.

This vaccine faces herculean challenges en route to the clinic, as many scientifically novel approaches initially do. Nevertheless, Agadjanyan envisions the day when a preventive AD vaccine might deploy pools of memory T cells generated from previous immunizations, such as tetanus and diphtheria, to drive production of Aβ antibodies. Take a look at his vaccination proposal, and join the discussion.—Esther Landhuis.

Esther Landhuis is a science writer in Dublin, California.

Discussion Between Marc Hertz, CEO of Pharmexa-Epimmune, San Diego, and Michael Agadjanyan, Institute for Molecular Medicine, Huntington Beach, California. (For related discussions, see also Agadjanyan proposal for a future preventive AD vaccine and Agadjanyan comment on Pride et al., 2008.)

Hertz: First, I should mention that my parent company (Pharmexa A/S) is involved in AD vaccine development and has a partnership with the Danish CNS pharmaceutical H. Lundbeck. I have no association with Michael’s research and only met him recently, but the approach my company is developing, and has been developing since early 2000, is very much along the lines of what Michael is testing. Although ours is not a DNA vaccine but a recombinant protein vaccine, it does incorporate the same concept of using a promiscuous foreign T helper epitope, like PADRE®, to drive the anti-Aβ humoral immune response.

I found the data in Movsesyan et al., 2008, convincing and the paper well written. The vaccine does appear to induce anti-Aβ antibodies and prevents AD-like pathology in the 3xTg-AD mice. Based on the data presented and data generated at my company with a similar product, I believe that the vaccine does not induce Aβ-specific T cells. From a research perspective, the paper adds to the AD vaccine field and opens the door for more research into developing AD therapies.

My concerns with the study have more to do with the practicality of the approach. The model is a preventive rather than therapeutic model, which would suggest that people should be receiving the vaccine prior to the onset of AD. I’m not sure this is a realistic goal for a DNA vaccine at this time. In theory, the vaccine would be given rather frequently for an extended period of time—tens of years—and there could be significant safety concerns in such repeated administration of a DNA vaccine in healthy individuals.

Agadjanyan: Marc’s concerns are well taken at the present time, but I respectfully submit that we should look ahead and prepare for a second phase of immunotherapy development for AD. I do believe protective vaccination is the future for AD, and there are several issues concerning protective versus therapeutic vaccination. First, data both from clinical trials of Elan/Wyeth’s peptide-based vaccine AN1792 (Bayer et al., 2005; Ferrer et al., 2004; Gilman et al., 2005; Masliah et al., 2005; Nicoll et al., 2003), and from mouse and dog models of AD (many papers, for recent ones see Head et al., 2008; Mamikonyan et al., 2007; Petrushina et al., 2007) indicate that therapeutic vaccines tested to date are not effective. Specifically, anti-Aβ antibodies (i) did not clear the most toxic pre-existing oligomeric deposits from the brains; (ii) did not disaggregate preformed oligomers in vitro; (iii) did not improve behavior in dogs. This is not surprising because the AN1792 vaccine also failed overall to improve cognitive function or prevent progressive decline (Bayer et al., 2005; Gilman et al., 2005). Only 59 out of 300 treated patients responded, and responders with low to moderate antibody titers (~1:2,200) showed no clinical treatment effect (Bayer et al., 2005; Gilman et al., 2005) even while they had remarkably improved Aβ plaque pathology (Boche and Nicoll, 2008; Nicoll et al., 2003). That neuropil threads and tangles are still observed in AN1792 responders (Boche and Nicoll, 2008; Nicoll et al., 2003) further supports my conclusion that antibodies cannot regenerate neurons.

A note about the side effects: in sharp contrast to animal models, the AN1792 vaccine induced meningoencephalitis in 22 percent of responders, whereas only 2 percent of the 241 non-responders experienced this adverse outcome (Patton et al., 2006). Others and I suggest that this side effect is connected to auto-reactive helper T (Th) cells; however, decisive data are unavailable and it remains possible that Th1 proinflammatory responses or anti-Aβ antibodies are involved in this process. In addition, studies have shown that antibodies generated by AN1792 formulated in QS21 adjuvant removed and/or solubilized Aβ, part of which subsequently ended up as vascular deposits (Patton et al., 2006) and may even have triggered hemorrhages (Ferrer et al., 2004). Finally, a potential problem of immunotherapy is that reduction of insoluble Aβ may lead to increased levels of soluble Aβ peptide (Patton et al., 2006), possibly oligomers, which are known to be more neurotoxic and can impair cognitive function (Cleary et al., 2005; Gong et al., 2003; Klein et al., 2001; Klyubin et al., 2005; Lesne et al., 2006). That is why, based on data generated in AN1792 trials, Alex Roher and colleagues suggested that it might be most effective to use an AD vaccine not as a therapeutic but as a prophylactic measure (Patton et al., 2006).

Our data both with peptide- and DNA-based epitope vaccines support this conclusion. They suggest that such a vaccine could be used as a safe, effective measure for treatment of people with early or preclinical AD, especially if they can be diagnosed by measuring CSF Tau/Aβ and pTau/Aβ ratios (de Jong et al., 2006; Fagan et al., 2007a; Fagan et al., 2007b) and/or by detecting Aβ accumulation in the brain using PIB-PET imaging (Klunk et al., 2004; HAI Conference series). The dosing and injection frequency of such a vaccine remain to be elucidated in human trials.

Of course, determining the efficacy of any vaccine is a lengthy process often requiring tens of years. I acknowledge that my proposals would require added research and a shift in the current strategic investments by the pharmaceutical and biotech industry. But I believe preventive immunotherapy is worth exploring. In animal models, immunotherapy at an early age can safely protect against Aβ accumulation, and it will be interesting to see if this can usefully be applied to humans to prevent AD. I believe that in time, the FDA will allow such clinical trials if the safety of the vaccines can be demonstrated and the requisite biomarkers for preclinical detection formally validated.

Hertz: I agree that the field will move in the direction of a preventive vaccine for AD. Another concern lies with the DNA vaccine. The dose of DNA and the number of administrations used in Movsesyan et al., 2008, were extreme for a rodent model. The human dose of DNA for a gene gun approach is limited to two micrograms per “shot” per 1 mg gold. They administered nine micrograms of DNA in a mouse—equivalent to 4.5 times the human dose—every two to six weeks. Of course, multiple doses can be given in humans, but for a DNA vaccine, doses are typically scaled on a mg/kg ratio. I’m not sure if the gene gun dose is scaled in the same manner, but even a nine-microgram dose in people would dictate a 4.5 mg dose of gold, making this an extremely expensive vaccine if the dosing regimen were extended beyond the two- to six-week schedule used in the mice.

Agadjanyan: I see this technical problem, but it can be overcome. Up to 8 μg (four delivery sites) of HBsAg DNA vaccine have been administered in pre-immune patients by gene gun (Fuller et al., 2006). Of note, 2 μg and 4 μg doses induced significant humoral and cellular responses in 100 percent of the immunized subjects, as well as in mice. We saw similar results in mice: a 2 μg/mouse DNA vaccine worked as well as a 9 μg vaccine. In addition, an 8 μg dose in humans was given by adding delivery sites without increasing the amount of DNA per cartridge, which is also possible (0.5 mg gold particles may be coated with up to 5 μg DNA, i.e., 5 μg/cartridge (Coligan et al., 2004). Thus, it is possible to administer up to 20 μg/subject using four delivery sites both in humans and mice. That said, we understand that dosing with the gene gun is limited by the capacity of gold particles for DNA. In fact, the gene gun from PowderMed/Pfizer Inc. that is currently being used for DNA immunization in humans has other limitations, as well. It is limited to dermal delivery, requires expensive gold particles, and the preparation of DNA-coated gold particles under good manufacturing practices (GMP) conditions is costly and requires significant labor.

Thus, toward bringing our vaccine to human testing, we are proposing to use the TriGrid Delivery System (TDS) from Ichor, Medical Systems, Inc. We recently compared Bio-Rad’s gene gun to Ichor’s TDS and concluded that the electroporation system is as good as the gene gun. Currently, we are comparing DNA immunization protocols using intramuscular versus intradermal routes of administration in large animals using Ichor’s TDS, which enables delivery of very high doses of DNA. (Of note, I have no commercial association with either of the companies whose products I have mentioned here.)

References:
Bayer AJ, Bullock R, Jones RW, Wilkinson D, Paterson KR, Jenkins L, Millais SB, Donoghue S. Evaluation of the safety and immunogenicity of synthetic Abeta42 (AN1792) in patients with AD. Neurology. 2005 Jan 11;64(1):94-101. Abstract

Boche D, Nicoll JA. The role of the immune system in clearance of abeta from the brain. Brain Pathol. 2008 Apr;18(2):267-78. Abstract

Cleary JP, Walsh DM, Hofmeister JJ, Shankar GM, Kuskowski MA, Selkoe DJ, Ashe KH. Natural oligomers of the amyloid-beta protein specifically disrupt cognitive function. Nat Neurosci. 2005 Jan;8(1):79-84. Abstract

Coligan JE, Kruisbeek AM, Margulies DH, Shevach EM, Strober W (2004): Gene Gun Inoculation of the DNA in Mice: "Current Protocols in Immunology." John Willey & Sons, pp 2.14.6.-2.14.12.

de Jong D, Jansen RW, Kremer BP, Verbeek MM. Cerebrospinal fluid amyloid beta42/phosphorylated tau ratio discriminates between Alzheimer's disease and vascular dementia. J Gerontol A Biol Sci Med Sci. 2006 Jul;61(7):755-8. Abstract

Fagan AM, Roe CM, Xiong C, Mintun MA, Morris JC, Holtzman DM (2007a): Cerebrospinal Fluid Biomarkers of Early Stage Alzheimer Disease: 8th International Conference: Alzheimer's and Parkinson's Disease: Progress and New Perspectives. Salzburg, Austria.

Fagan AM, Roe CM, Xiong C, Mintun MA, Morris JC, Holtzman DM. Cerebrospinal fluid tau/beta-amyloid(42) ratio as a prediction of cognitive decline in nondemented older adults. Arch Neurol. 2007b Mar;64(3):343-9. Abstract

Ferrer I, Boada Rovira M, Sánchez Guerra ML, Rey MJ, Costa-Jussá F. Neuropathology and pathogenesis of encephalitis following amyloid-beta immunization in Alzheimer's disease. Brain Pathol. 2004 Jan;14(1):11-20. Abstract

Fuller DH, Loudon P, Schmaljohn C. Preclinical and clinical progress of particle-mediated DNA vaccines for infectious diseases. Methods. 2006 Sep;40(1):86-97. Abstract

Gilman S, Koller M, Black RS, Jenkins L, Griffith SG, Fox NC, Eisner L, Kirby L, Rovira MB, Forette F, Orgogozo JM, AN1792(QS-21)-201 Study Team. Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology. 2005 May 10;64(9):1553-62. Abstract

Gong Y, Chang L, Viola KL, Lacor PN, Lambert MP, Finch CE, Krafft GA, Klein WL. Alzheimer's disease-affected brain: presence of oligomeric A beta ligands (ADDLs) suggests a molecular basis for reversible memory loss. Proc Natl Acad Sci U S A. 2003 Sep 2;100(18):10417-22. Abstract

Head E, Pop V, Vasilevko V, Hill M, Saing T, Sarsoza F, Nistor M, Christie LA, Milton S, Glabe C, Barrett E, Cribbs D. A two-year study with fibrillar beta-amyloid (Abeta) immunization in aged canines: effects on cognitive function and brain Abeta. J Neurosci. 2008 Apr 2;28(14):3555-66. Abstract

Klein WL, Krafft GA, Finch CE. Targeting small Abeta oligomers: the solution to an Alzheimer's disease conundrum? Trends Neurosci. 2001 Apr;24(4):219-24. Abstract

Klunk WE, Engler H, Nordberg A, Wang Y, Blomqvist G, Holt DP, Bergström M, Savitcheva I, Huang GF, Estrada S, Ausén B, Debnath ML, Barletta J, Price JC, Sandell J, Lopresti BJ, Wall A, Koivisto P, Antoni G, Mathis CA, Långström B. Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004 Mar;55(3):306-19. Abstract

Klyubin I, Walsh DM, Lemere CA, Cullen WK, Shankar GM, Betts V, Spooner ET, Jiang L, Anwyl R, Selkoe DJ, Rowan MJ. Amyloid beta protein immunotherapy neutralizes Abeta oligomers that disrupt synaptic plasticity in vivo. Nat Med. 2005 May;11(5):556-61. Abstract

Lesné S, Koh MT, Kotilinek L, Kayed R, Glabe CG, Yang A, Gallagher M, Ashe KH. A specific amyloid-beta protein assembly in the brain impairs memory. Nature. 2006 Mar 16;440(7082):352-7. Abstract

Mamikonyan G, Necula M, Mkrtichyan M, Ghochikyan A, Petrushina I, Movsesyan N, Mina E, Kiyatkin A, Glabe CG, Cribbs DH, Agadjanyan MG. Anti-A beta 1-11 antibody binds to different beta-amyloid species, inhibits fibril formation, and disaggregates preformed fibrils but not the most toxic oligomers. J Biol Chem. 2007 Aug 3;282(31):22376-86. Abstract

Masliah E, Hansen L, Adame A, Crews L, Bard F, Lee C, Seubert P, Games D, Kirby L, Schenk D. Abeta vaccination effects on plaque pathology in the absence of encephalitis in Alzheimer disease. Neurology. 2005 Jan 11;64(1):129-31. Abstract

Movsesyan N, Ghochikyan A, Mkrtichyan M, Petrushina I, Davtyan H, Olkhanud PB, Head E, Biragyn A, Cribbs DH, Agadjanyan MG. Reducing AD-like pathology in 3xTg-AD mouse model by DNA epitope vaccine-a novel immunotherapeutic strategy. PLoS ONE. 2008;3(5):e2124. Abstract

Nicoll JA, Wilkinson D, Holmes C, Steart P, Markham H, Weller RO. Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat Med. 2003 Apr;9(4):448-52. Abstract

Patton RL, Kalback WM, Esh CL, Kokjohn TA, Van Vickle GD, Luehrs DC, Kuo YM, Lopez J, Brune D, Ferrer I, Masliah E, Newel AJ, Beach TG, Castano EM, Roher AE (2006): Amyloid-beta peptide remnants in AN-1792-immunized Alzheimer's disease patients: a biochemical analysis. Am J Pathol 169:1048-63. Abstract

Petrushina I, Ghochikyan A, Mktrichyan M, Mamikonyan G, Movsesyan N, Davtyan H, Patel A, Head E, Cribbs DH, Agadjanyan MG (2007): Alzheimer's Disease Peptide Epitope Vaccine Reduces Insoluble But Not Soluble/Oligomeric A{beta} Species in Amyloid Precursor Protein Transgenic Mice. J Neurosci 27:12721-12731. Abstract

Utsuyama M, Hirokawa K, Kurashima C, Fukayama M, Inamatsu T, Suzuki K, Hashimoto W, Sato K. Differential age-change in the numbers of CD4+CD45RA+ and CD4+CD29+ T cell subsets in human peripheral blood. Mech Ageing Dev. 1992 Mar 15;63(1):57-68. Abstract

Comments

Make a Comment

To make a comment you must login or register.

Comments on News and Primary Papers

  1. I think this very well-performed study shows that in mice, an Aβ vaccination strategy using a DNA vaccine can generate strong anti-Aβ antibodies yet also a strong Th2 response, which should theoretically prevent a T cell response against Aβ. This is similar to what several drug companies are doing with non-DNA vaccine techniques. There may be some advantages to the DNA vaccine technique as pointed out in the manuscript. This type of approach should prevent a T cell response to Aβ, which may have caused the problems with encephalitis in the AN1792 study. However, a critical issue for humans studies in this area is whether all the toxicity issues encountered with AN1792 were due to an abnormal T cell response to Aβ. If so, this study, along with what is being done with non-DNA vaccines, is promising in that the T cell response to Aβ can probably be prevented. It is also possible that some of the encephalitis in AN1792 was not only caused by a T cell response but also by certain anti-Aβ antibodies to aggregated Aβ. This will need to be sorted out in the passive anti-Aβ antibody studies that are also moving along in clinical trials.

  2. This study is an important advance showing that gene vaccines delivered by gene gun may hold promise for future use in the fight against Alzheimer disease. In 2003, Ghochikyan et al. (1) constructed a DNA minigene with Aβ fused to mouse interleukin-4 (pAβ42-IL-4) as a molecular adjuvant to generate anti-Aβ antibodies and enhance Th2-type immune responses. The DNA minigene-induced anti-Aβ antibodies bound to Aβ plaques in brain tissue from an AD patient.

    In 2004, we showed that Aβ42 gene vaccination with gene gun in AD double transgenic mice (APPswe/PSEN1(A246E) produced anti-Aβ42 antibodies that were predominantly IgG1, reflecting a Th2 immune response (2). In 2006 (3) and 2007 (4), we reported for the first time that the Aβ42 gene vaccine administered with a gene gun produced an IgG1 (Th2) immune response and significantly reduced brain levels of Aβ42 in treated APPswe/PS1ΔE9 double transgenic mice. In the 2007 report (4), brain Aβ42 levels were decreased by 41 percent and increased in plasma by 43 percent in vaccinated compared with control mice, as assessed by ELISA analysis. Aβ42 plaque deposits in cerebral cortex and hippocampus of vaccinated animals were reduced by 51 and 52 percent, respectively, compared with controls. Glial cell activation was also significantly attenuated in vaccinated compared with control mice. The 2007 study (4) also described a vaccinated rhesus monkey that developed anti-Aβ42 antibodies.

    Our publications (3,4) provide the first direct immunological evidence suggesting that Aβ42 gene immunization delivered by gene gun effectively induces a Th2 immune response and reduces brain Aβ42 levels in APPswe/PS1ΔE9 mice. Our Aβ42 gene vaccine is also preventive. We immunized the mice beginning at three months of age, and the Aβ42 deposition in this double transgenic line begins around five to six months. We examined the brains at 14-15 months and showed the 41 percent reduction in Aβ42 peptide levels. As the induced immune response was predominately Th2, which has a low probability of producing an inflammatory response, Aβ42 gene vaccination may be a safe and efficient option for Alzheimer disease immunotherapy.

    View all comments by Roger N. Rosenberg
  3. I have a couple of quick points to make. First, Michael Agadjanyan is a brilliant immunologist. He and Dave Cribbs have been leaders in the development of safer and effective active immunization protocols against Aβ, both in this manuscript and others.

    Second, I reviewed the manuscript for PLoS (and signed the review), and the only concern I had was the absence of a control vaccine group. It is conceivable some of the effects were due to the fortnightly gene gun treatments and nonspecific immune activation rather than the specific effects against Aβ.

    Third, it is premature to consider prophylactic vaccination against Aß in the general population due to potential risks without any demonstrated benefit in man. However, carriers of dominant FAD mutations might consider the risk-benefit ratio to favor vaccination.

    Immunotherapy will likely be the first test of the amyloid hypothesis of AD pathogenesis, given its efficacy in clearing amyloid in animal models and the number of clinical trials underway. However, there are potential concerns associated with encephalitic reactions and development of micro-hemorrhage. A large number of studies have argued the absence of these reactions in their immunotherapy protocols means their approach is "safe". However, unless there are immunotherapy protocols tested in parallel which produce these toxic effects in the animal model used, it is uncertain that their protocol avoids the problem.

    We all anxiously await the results from the clinical trials using active and passive immunotherapy.

  4. The DNA epitope vaccine described by Movsesyan et al. has raised a discussion concerning preventative versus therapeutic strategies for the management of Alzheimer disease (AD). It goes without saying that before we can have this debate in earnest, the strategy at hand must be proven sufficiently safe in the therapeutic setting and even safer if it is to be applied in a preventative setting in people who are currently symptom-free. I do not believe that we are yet at that point with any anti-amyloid immunotherapy approach, so this discussion remains theoretical for the time being.

    Despite this, the discussion of when we need to intervene in AD remains relevant and important. The comments raised by Michael Agadjanyan on Alzforum echo the concerns previously voiced by many researchers and can be boiled down to this simple question: Can we achieve a meaningful impact of any therapy for AD if it is begun after the clinical symptoms become apparent? We do not yet know the answer to this question, but decades of therapeutic efforts with only modest success justify raising this concern.

    Every serious disease has a point of no return, and one has to at least entertain the possibility that this point may be prior to the onset of clinical symptoms in AD. Many postmortem studies have implied that the Aβ pathology characteristic of AD begins a decade or more prior to the clinical symptoms (Haroutunian et al., 1998; Price and Morris, 1999; Wolf et al., 1999). Today we have tools such as CSF determination of Aβ and tau markers and PET amyloid imaging that can indirectly (CSF) or directly (PET amyloid tracers) detect Aβ pathology in living patients.

    Several PET amyloid imaging studies have shown that about 25 percent of cognitively normal elderly show some evidence of early Aβ deposition (Mintun et al., 2006; Rowe et al., 2007; Aizenstein et al., 2008) and focal Aβ deposits are clearly present in some presenilin-1 mutation carriers at least a decade prior to their expected onset of clinical symptoms (Klunk et al., 2007). In time, we will learn if the presence of Aβ pathology in a cognitively normal person indicates preclinical AD or if it can be innocuous. We also will learn about the natural history of pathological changes in this pre-symptomatic period.

    While these imaging studies are ongoing, we will learn about the safety of several immunotherapy approaches, as well. We also will learn if these approaches are effective at: 1) removing Aβ from the human brain and 2) improving the clinical course of mild-moderate AD. It is entirely possible, and perhaps was foreshadowed by AN1792, that an immunotherapy (or any anti-amyloid therapy) can be relatively effective at removing Aβ, but show little or no clinical effects in mild to moderate AD. If such an anti-amyloid therapy also is safe, we, as a field, will face a very important decision. Do we abandon anti-amyloid therapy as clinically unproductive, or do we revise the trial design to focus on earlier stages of AD, including preclinical stages, i.e., embark on preventative anti-amyloid trials?

    The idea of preventative trial design presents considerable challenges to drug companies, and few, if any, are eager to pursue this approach at present. It will take a major paradigm shift and great patience and long-term thinking/commitment to contemplate a prevention trial that could last five to 10 years. At present, we have more questions than answers, but each should be given careful thought:

    • Could current economic forces that constrain trials to six to 24 months ensure failure in the search for an effective treatment for AD?
    • Could a focus on well-defined carriers of autosomal-dominant mutations for early-onset familial AD make prevention trials with anti-amyloid therapy feasible?
    • Is the mere presence of Aβ deposition in the brain of an asymptomatic individual a disease in the same sense that the presence of various amyloids in the periphery are considered pathologic?
    • Could removal of brain Aβ be a primary outcome measure?

    It may be that our thinking must change before our success at treating AD can change.

    View all comments by William Klunk
  5. This interesting paper by Movsesyan and coworkers describes a novel DNA-based Aβ vaccine that relies on amino acids 1-11 of the peptide in combination with the synthetic T cell peptide, PADRE, and the Th2-promoting chemokine CCL22. The authors have nicely shown that this vaccine reduces behavioral impairment and amyloid burden in brains of Frank LaFerla’s 3xTg-AD mice, but this was only appreciable when the vaccine was given in a prophylactic regimen to younger mice. This raises an important issue regarding timing of immunotherapy, which, if these results in mice translate to humans, suggests that treatment would need to begin early (likely in asymptomatic individuals) for it to be effective.

    I just wanted to raise one caveat for interpreting these results. Previous Aβ vaccination attempts by us and by other groups have failed to model the aseptic meningoencephalitis that occurred in about 5 percent of patients who received the Elan/Wyeth AN1792 vaccine. Further, we and others have not observed the auto-aggressive T cell response (presumed Th1 response that likely occurred in vaccinated patients) after vaccination of mice with the original Schenk et al. protocol or with other Aβ vaccines (unless pertussis toxin is co-administered, widely used to induce brain T cell penetration in experimental autoimmune encephalomyelitis). So it is not possible to conclude that a vaccine that does not induce auto-aggressive T cells in mice may be safe in humans unless detailed toxicology studies are performed in non-human primates.

    View all comments by Terrence Town
  6. The current discussion appears to suggest that immunotherapy has the most potential as a preventative approach to managing Alzheimer disease (AD). The development of biomarkers will be critical for the design of these studies, and there are many exciting avenues being explored (e.g., comment by Dr. Klunk, plasma profiling by Tony Wyss-Coray’s group [Ray et al., 2007], CSF measures of Aβ and tau; also see discussion on ARF). Further, several groups of individuals have been identified as being at high risk for developing dementia. Thus, recruiting members of families with familial AD or possibly individuals with ApoE4/4 for clinical trials would be options. An additional group of adults who are at high risk for developing AD are individuals with Down syndrome. Indeed, the first signs of β amyloid (Aβ) pathology can occur in the early thirties (Hof et al., 1995; Leverenz, 1998), clearly at least a decade (and sometimes two) before dementia may be detected. These vulnerable individuals may benefit greatly from a preventative approach using immunotherapy if started in middle age. Given that virtually all adults with DS will develop full-blown AD pathology by the time they are in their forties (Wisniewski et al., 1985), this would be a fascinating cohort to study and a group of individuals that remains relatively underrepresented in AD clinical trials.

    On the other hand, I would like to remain somewhat optimistic about a possible therapeutic approach. But a simple Aβ-targeted treatment may be insufficient without repairing secondary pathologies associated with chronic Aβ exposure. In other words, what if we could remove Aβ and follow up or, in parallel, repair remaining neurons? Might we predict a significant improvement in cognition? As others have suggested, immunotherapy may not be as efficacious once the disease has progressed (to what point is unknown, but at least moderate to severe dementia). But given that our clinicians can detect very early signs of cognitive decline and identify subjects with mild dementia, there may be treatment opportunities here.

    Another idea might be to use immunotherapy as a short-term treatment to clear Aβ pathology and subsequently follow up with a regimen that prevents new Aβ deposition, such as BACE inhibitors or compounds that increase α-secretase activity (see, e.g., ARF Keystone BACE story; ARF SfN BACE story; Fahrenholz, 2007). In this way, immunotherapy does not need to be continued for an extensive period of time, minimizing possible development of adverse events. The “one bullet” hypothesis may be too simple for such a complex disease.

    View all comments by Elizabeth Head
  7. Reply to comments above and on related Vaccine Page
    I agree completely with William Klunk that the discussion about preventive versus therapeutic vaccination is currently more theoretical than practical. This discussion will become more practical as scientists gain more knowledge on AD pathology and vaccination strategies. To draw a historic analogy, the approach of finding the right time when theory and knowledge intersect, has allowed the Manhattan Project to be successful. Thus, I agree with Klunk’s conclusion that “our thinking must change before our success at treating AD can change,” and this is why it is important to continue having these theoretical discussions.

    Let’s briefly consider the story of another historic example, the HIV vaccine. In 1997, President Clinton challenged the scientific community to create an HIV vaccine within a decade. As a result, NIH received extensive funds and announced a new vaccine lab, as well as formed an AIDS Vaccine Research Committee chaired by David Baltimore. During the first conference at NIH on the Innovation Grant for the HIV Vaccine Development Program in 1998, I mentioned that I found it hard to believe that a therapeutic HIV vaccine is realistic and that even the generation of a protective HIV vaccine will be unlikely because this virus is unbelievably variable and attacks/destroys the immune system. Baltimore, who headed this conference, asked me why I accepted this innovation grant if I do not believe in an HIV vaccine. My response was that this program should allow scientists from different disciplines to show that an HIV vaccine is not realistic, at least in the current state of our scientific knowledge. In 2006, Baltimore was quoted as saying: “We're going to live in a world without an HIV vaccine for at least another decade…and we've been saying it's going to be another decade for the last few decades” (see Discover story. Not surprisingly Merck’s latest HIV vaccine trial initiated in healthy volunteers failed, 25 years to the year when the first HIV-1 strain was isolated (see BBC news story).

    The situation with the AD vaccine is quite different from the one with the HIV-1 vaccine. I personally am very optimistic and think that a well-designed AD vaccine could safely induce the production of protective antibodies specific to β-amyloid. That’s because this protein is not changing (at least several N-terminal aa are available in any Aβ forms), is not destroying the immune system, and does not require generation of more complex cellular immune responses specific to this peptide. However, I also believe that this should be done before Aβ accumulation in the vasculature and parenchyma of brains induces an unalterable process.

    Specifically, the AN1792 vaccine is practically ineffective in AD patients (Patton et al., 2006), a peptide epitope vaccine in aged Tg 2576 mice as well as the Aβ42 vaccine in aged dogs are ineffective, but a DNA-based epitope vaccine works when applied as a preventive measure (Head et al., 2008; Mamikonyan et al., 2007; Movsesyan et al., 2008; Petrushina et al., 2007). As Liz Head notes above, it is possible that anti-Aβ42 antibodies could effectively remove toxic deposits of Aβ42 from parenchyma and vasculature, but this will not help to heal damaged neurons unless we could otherwise repair them or initiate neurogenesis. The removal of Aβ42 could be effective only when titers of antibodies are relatively high (Patton et al., 2006; Petrushina et al., 2007) and if they are present in the periphery for a rather long period of time.

    However, as David Holtzman notes, high titers of antibodies may also be detrimental because they may increase deposition of the toxic forms of Aβ and antibody-antigen complexes in the vasculature. Interestingly, we recently demonstrated that sub-stoichiometric concentrations of purified anti-Aβ antibody prevented Aβ42 aggregation and induced disaggregation of preformed Aβ42 fibrils down to a non-filamentous and non-toxic species. However, an anti-Aβ antibody could not disaggregate oligomers, although it did delay Aβ42 oligomer formation (Mamikonyan et al., 2007). These in-vitro observations suggest that therapeutic vaccination cannot disrupt toxic Aβ42 oligomers in vivo, and to the extent that AD is associated with accumulation of those oligomers in brain, I would expect therapeutic vaccination to be ineffective. Moreover, if these in vitro data mimic the situation in vivo, they suggest that even preventive vaccination could not protect elderly people from AD, although a therapeutic vaccine could delay its onset. Speaking with other scientists in the AD vaccine field, I know that many of them nine years after Dale Schenk and colleagues’ remarkable discovery believe that an AD vaccine can become a reality if it is used for protection of Aβ42 accumulation in the brains of healthy people. This raises other questions, including ones about the safety of a preventive vaccine, as noted by Holtzman and Terrence Town, and the cost of such a strategy, as noted by M. Paul Murphy.

    I’ll reply on safety first. The significant amount of data generated in AN1792 human trials as well as in many different mouse models of AD indicates that anti-Aβ antibodies can inhibit/clear Aβ deposits. Published data also suggest that although the Aβ42-based vaccine is safe in mice, it is not safe in AD patients (see Holtzman and Town comments), specifically when it is formulated into a strong Th1 adjuvant and may be in polysorbate B (Pride et al., 2008). As Holtzman notes, it is likely that activation of autoreactive Th cells (specific to Aβ42 peptide) and to a lesser extent proinflammatory cellular responses may induce adverse effects in AD patients. This is why we changed our first-generation DNA vaccine based on Aβ42 (Ghochikyan et al., 2003) to a DNA epitope vaccine (Ghochikyan et al., 2007; Movsesyan et al., 2008) that in theory should not induce autoreactive cellular responses in humans.

    Regarding Roger Rosenberg’s comment, a DNA vaccine expressing full-length Aβ42 could be potentially harmful to humans as was shown with AN1792 based on fibrillar Aβ42. Another difference between a DNA vaccine expressing Aβ42 and a DNA epitope vaccine as described by our groups is the potential of the latter for inducing strong cellular and humoral immunity not only in mice, but also in humans. We base this assumption on that fact that this DNA epitope vaccine is composed of a strong Th epitope that is proven not only in mice, rabbits, and monkeys but also in people expressing 14 different MHC (Alexander et al., 1994) and therefore has the potential to be highly immunogenic in all humans. We are almost certain that such a vaccine will not induce autoreactive T cells and proinflammatory cellular responses in humans (Monsonego et al., 2006; Pride et al., 2008), but its safety in rabbits, dogs, monkeys, maybe chimps, should be demonstrated first. We are planning immunological and toxicological studies in rabbits and monkeys immunized with a DNA epitope vaccine delivered by electroporation before we suggest clinical trials in people who have been identified as being high risk for developing dementia (see comments by Klunk, Head, and David Morgan).

    In these experiments we will use a control plasmid to demonstrate the specificity of vaccination. Of note, as was requested by the reviewer of Movsesyan et al., 2008, we have included data with control plasmid, encoding MDC fused with an irrelevant antigen, in the final version of the paper. Specifically, results with a control plasmid have been incorporated into Figures 2, 3, 4. These data demonstrate that the control vaccine does not induce anti-Aβ antibody in C57BL/6 (Figure 2) and 3xTg-AD (Figure 3) mice. Data in Figure 4 show that injections with an MDC-irrelevant control vaccine did not rescue aged 3xTg-AD mice from cognitive decline (see also reviewer comments and author responses in PLoS ONE.

    An economist could calculate the cost of a preventive vaccination strategy of healthy people better than biomed scientists. Into this calculation should go data as recently released by the Alzheimer’s Association: “…10 million baby boomers will get Alzheimer's disease in their lifetime. … today there are an estimated 5.2 million Americans living with Alzheimer's disease, which is the seventh-leading cause of death in the country and the fifth-leading cause of death for those over age 65” (2008 Alzheimer's Disease Facts and Figures. The calculation could compare the cost of treating all these people with, for example, the cost of treating AIDS patients in the U.S. ($13.1 billion in 2007). It would factor in that today there is no effective treatment for AD and so we often also have to treat the stressed and exhausted caregiver. In a Senate hearing earlier this month, former Speaker Newt Gingrich said: "Under current trends, federal spending on Alzheimer's will increase to more than $1 trillion per year by 2050 in today's dollars. That's more than one-tenth of America's current economy. With this amount of money at stake, the government simply will not be able to solve its looming fiscal problems if it fails to address the growing Alzheimer's crisis."

    Faced with such numbers, I think it is worth spending money and time (not only 10, but maybe 20 years) to study the efficacy of a preventive AD vaccine in humans who know they are at high risk for developing AD if we have enough data supporting such a strategy. That is why I call on the companies involved in passive and active AD vaccine clinical trials to provide detailed results on their clinical studies to the scientific community more quickly. This way, they will get rapid feedback from the community and, with their help, develop a safe and potent protective or therapeutic vaccine against this devastating disease (see comments).

    See also:

    Ghochikyan A, Movsesyan N, Mkrtichyan M, Petrushina I, Biragyn A, Cribbs DH, Agadjanyan MG (2007, March 14-18): DNA epitope vaccine induced strong anti-Aβ antibodies inhibiting AD like pathology in 3xTg-AD mice and protecting them from cognitive decline: 8th International Conference AD/PD, Salzburg, Austria.

    View all comments by Michael G. Agadjanyan
  8. Comment by Mark A. Smith, Rudy J. Castellani, Paula I. Moreira, Akihiko Nunomura, Hyoung-gon Lee, Xiongwei Zhu, George Perry

    No Justification in Moving from Treatment to Prevention
    The use of immunotherapy as a preventative measure for Alzheimer disease has little merit. First, is abject (Smith et al., 2002) or presumed failure in use as a treatment a good start? If so, would this also be true for other failed treatments? Second, success in “preventing” the pathology/deficits in transgenic mice seems to be driving some of this move toward prevention (Movsesyan et al., 2008). With this logic, the list of potential preventatives would likewise expand to everything that works in mice. Since there is a laundry list of drugs that work in mice but fail in treating the disease, we are left with a laundry list of drugs that we could justify as a valid preventative strategy.

    The field has bet the bank on amyloid as a therapeutic and seems determined to bet another bundle on amyloid as a preventative. In our opinion, the failure thus far of treatment strategies is more an indication of a focus on incorrect targets than of not starting early enough (Smith et al., 2002; Castellani et al., 2006).

    View all comments by Akihiko Nunomura

References

News Citations

  1. Trial Troika—Immunotherapy Interrupted, Lipitor Lags, Dimebon Delivers
  2. Trials and Tribulations—Autopsy Reveals Pros and Cons of AD Vaccine
  3. New Vaccine Divides T and B Cell Epitopes to Conquer Aβ

Paper Citations

  1. . Anti-A beta 1-11 antibody binds to different beta-amyloid species, inhibits fibril formation, and disaggregates preformed fibrils but not the most toxic oligomers. J Biol Chem. 2007 Aug 3;282(31):22376-86. PubMed.
  2. . Alzheimer's disease peptide epitope vaccine reduces insoluble but not soluble/oligomeric Abeta species in amyloid precursor protein transgenic mice. J Neurosci. 2007 Nov 14;27(46):12721-31. PubMed.
  3. . Development of high potency universal DR-restricted helper epitopes by modification of high affinity DR-blocking peptides. Immunity. 1994 Dec;1(9):751-61. PubMed.
  4. . Progress in the active immunotherapeutic approach to Alzheimer's disease: clinical investigations into AN1792-associated meningoencephalitis. Neurodegener Dis. 2008;5(3-4):194-6. PubMed.
  5. . Generation and characterization of the humoral immune response to DNA immunization with a chimeric beta-amyloid-interleukin-4 minigene. Eur J Immunol. 2003 Dec;33(12):3232-41. PubMed.
  6. . Antibodies from a DNA peptide vaccination decrease the brain amyloid burden in a mouse model of Alzheimer's disease. J Mol Med (Berl). 2004 Oct;82(10):706-14. PubMed.
  7. . Evaluation of the safety and immunogenicity of synthetic Abeta42 (AN1792) in patients with AD. Neurology. 2005 Jan 11;64(1):94-101. PubMed.
  8. . The role of the immune system in clearance of Abeta from the brain. Brain Pathol. 2008 Apr;18(2):267-78. PubMed.
  9. . Natural oligomers of the amyloid-beta protein specifically disrupt cognitive function. Nat Neurosci. 2005 Jan;8(1):79-84. PubMed.
  10. . Cerebrospinal fluid amyloid beta42/phosphorylated tau ratio discriminates between Alzheimer's disease and vascular dementia. J Gerontol A Biol Sci Med Sci. 2006 Jul;61(7):755-8. PubMed.
  11. . Cerebrospinal fluid tau/beta-amyloid(42) ratio as a prediction of cognitive decline in nondemented older adults. Arch Neurol. 2007 Mar;64(3):343-9. PubMed.
  12. . Neuropathology and pathogenesis of encephalitis following amyloid-beta immunization in Alzheimer's disease. Brain Pathol. 2004 Jan;14(1):11-20. PubMed.
  13. . Preclinical and clinical progress of particle-mediated DNA vaccines for infectious diseases. Methods. 2006 Sep;40(1):86-97. PubMed.
  14. . Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology. 2005 May 10;64(9):1553-62. PubMed.
  15. . Alzheimer's disease-affected brain: presence of oligomeric A beta ligands (ADDLs) suggests a molecular basis for reversible memory loss. Proc Natl Acad Sci U S A. 2003 Sep 2;100(18):10417-22. PubMed.
  16. . A two-year study with fibrillar beta-amyloid (Abeta) immunization in aged canines: effects on cognitive function and brain Abeta. J Neurosci. 2008 Apr 2;28(14):3555-66. PubMed.
  17. . Targeting small Abeta oligomers: the solution to an Alzheimer's disease conundrum?. Trends Neurosci. 2001 Apr;24(4):219-24. PubMed.
  18. . Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004 Mar;55(3):306-19. PubMed.
  19. . Amyloid beta protein immunotherapy neutralizes Abeta oligomers that disrupt synaptic plasticity in vivo. Nat Med. 2005 May;11(5):556-61. PubMed.
  20. . A specific amyloid-beta protein assembly in the brain impairs memory. Nature. 2006 Mar 16;440(7082):352-7. PubMed.
  21. . Abeta vaccination effects on plaque pathology in the absence of encephalitis in Alzheimer disease. Neurology. 2005 Jan 11;64(1):129-31. PubMed.
  22. . Reducing AD-like pathology in 3xTg-AD mouse model by DNA epitope vaccine - a novel immunotherapeutic strategy. PLoS One. 2008;3(5):e2124. PubMed.
  23. . Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat Med. 2003 Apr;9(4):448-52. PubMed.
  24. . Amyloid-beta peptide remnants in AN-1792-immunized Alzheimer's disease patients: a biochemical analysis. Am J Pathol. 2006 Sep;169(3):1048-63. PubMed.
  25. . Differential age-change in the numbers of CD4+CD45RA+ and CD4+CD29+ T cell subsets in human peripheral blood. Mech Ageing Dev. 1992 Mar 15;63(1):57-68. PubMed.

Other Citations

  1. extended dialog

External Citations

  1. Pharmexa-Epimmune
  2. Ichor Medical Systems, Inc.

Further Reading

No Available Further Reading

Primary Papers

  1. . Reducing AD-like pathology in 3xTg-AD mouse model by DNA epitope vaccine - a novel immunotherapeutic strategy. PLoS One. 2008;3(5):e2124. PubMed.