. 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.


Please login to recommend the paper.


  1. This comment concerns both this paper (Pride et al., 2008) and the recent news about the suspension of ACC-001 clinical trials by Elan Pharmaceutical /Wyeth Research (see ARF drug news story.

    Pride and colleagues analyzed Aβ-specific cellular immune responses in PBMCs obtained from patients enrolled in Phase 1 (Study 102) and Phase 2a (Study 201) AN1792 vaccine trials. These cells were re-stimulated with Aβ42 or overlapping 15-mer peptides from the carboxyl terminus of Aβ42). Specifically, scientists from Elan/Wyeth detected a number of cells secreting IL-2 and IFN-γ cytokines using the ELISPOT assay. This paper is important to the scientific community. It is therefore unfortunate that the authors did not present clear data on: (i) the number of patients involved in these immunological studies; (ii) the number of cells secreting these two cytokines; (iii) the results generated with PBMCs obtained from the placebo group; and (iv) the results on cytokine production by PBMCs re-stimulated in vitro with an irrelevant antigen. In this paper, the authors merely present data on the percentage of ELISPOT responses from AN1792-treated patients in a single table and conclude that:

    1. “T cell responses are specific to the carboxyl terminus of Aβ.”

    2. “Meningoencephalitis following AN1792 injection may be related to immune response differences driven by a formulation change,” e.g., adding Polysorbate 80 to the QS21 adjuvant/fibrillar Aβ42 vaccine.

    3. “ELISPOT responses to the various stimuli (e.g., Aβ activation) in Study 102 were generally IL-2-positive and IFN-γ negative, indicative of Class II MHC-restricted (CD4+) Th2-type profile, whereas in Study 201 most responders mounted both IL-2 and IFN-γ positive responses, indicative of a Class II (CD4+) Th1-type response.”

    The first conclusion about the specificity of the anti-Aβ Th cells reported by Pride et al. has been previously reported by others (Cribbs et al., 2003; Monsonego et al., 2003). This indicates that it is possible to dissociate antigenic determinants of Aβ42 that are in charge of humoral and cellular immune responses. In fact, this strategy has allowed us to prepare not only peptide-, but also DNA-based epitope vaccines composed of a foreign T cell epitope fused with an immunogenic B cell epitope of Aβ42 and demonstrate the feasibility of such an approach in wild-type and APP-transgenic mice (Agadjanyan et al., 2005; Mamikonyan et al., 2007; Movsesyan et al., 2008; Petrushina et al., 2007).

    In the present manuscript, Pride et al. also suggest that such an approach could be beneficial. In this regard, they discuss a vaccine used in the ACC-001 clinical trial. This vaccine is composed of an Aβ N-terminal peptide conjugated with a carrier; three doses of vaccine (3 μg, 10 μg, and 30 μg) are formulated in one dose of QS21 adjuvant (50 μg) or in PBS. An additional control group of patients is injected with QS21 alone (the active-to-placebo ratio is 3/1/1 for ACC-001+ QS21/ACC-001/QS21). Unfortunately, the publication of this paper coincides with information that the ACC-001 trial has been temporarily suspended because of vasculitis detected in one patient injected with the antigen-adjuvant. Even though both the information from Reuters as well as the data presented by Pride et al. are not very specific, I want to share my views about the ACC-001 trial as well as the role of Polysorbate 80 and Th2/Th1 types of cellular response in the AN1792 trials.

    This website has reported (see ARF related news story) that the ACC-001 vaccine is a 7-amino acid N-terminal fragment from Aβ42 conjugated to a mutated diphtheria toxin protein (CRM 197). Although this conjugate vaccine lacks the T helper epitope of Aβ42 and therefore cannot induce activation of anti-self CD4+T cell responses, it could still induce Th1-type proinflammatory cellular responses to the CRM 197 carrier, because it is formulated in a strong Th1 adjuvant, QS21.

    I would like to point out that there is no strong evidence that in AN1792 studies 102 and 201, Pride et al. have actually detected Th2- and Th1-biased immune responses in vaccinated patients. The study is technically weak on this point. Here is why: it is true that the majority of published papers indicate that IFN-γ is solely a Th1 cytokine. Likewise, for a long time human IL-2 has also been considered to be a Th1 cytokine. That is why the FDA has approved a recombinant IL-2 (Proleukin) for the treatment of cancers and chronic viral infections as a booster for vaccines to induce proinflammatory cellular responses. Interestingly, it was also shown that IL-2–producing mouse or human CD4+T lymphocytes are uncommitted cells and may differentiate into Th1 or Th2 phenotypes later during immune responses (Wang and Mosmann, 2001; Divekar et al., 2006).

    Finally, current data demonstrate that IL-2 plays an important role not only in the development of effector T cells, but also in the maintenance of CD4+CD25+FOXP3+ regulatory T cells (Treg) and in the establishment/maintenance of immune tolerance. These findings on IL-2 cytokine have become widely accepted in immunology (see Janeway’s Immunobiology, 7th Edition, Garland Science, NY, London, 2008, pp352-353).

    For these reasons, IL-2 is not a suitable cytokine for the determination of Th1 and Th2 responses in PBMCs obtained from AN1792 trial participants. Instead, Th1 and Th2 types of cellular responses are more rigorously measured by quantifying the numbers of CD4+T cells producing IFN-γ (Th1) or IL-4 (Th2) cytokines in FACS assays. It is important to use the FACS method instead of an ELISPOT assay for the detection of CD4+T cell responses, because it is well known that CD8+T cells can also produce different cytokines, including IFN-γ. Thus, the conclusion in Pride et al., that “patients’ PBMCs in an earlier multiple dose study (Study 102) were Th2 biased, while those from Study 201 were biased toward a proinflammatory Th1 response,” is premature based on the data provided.

    Secondly, even though there is no evidence about the involvement of QS21 in vasculitis, the patient with inflammation of the blood vessels was in the antigen-adjuvant group of the ACC-001 trial. Thus, it is important to discuss the adjuvant used in both AN1792 and ACC-001 studies. It is known that Th1-type proinflammatory immune responses that are important for generating protection against viral infections and cancer are also implicated in autoimmune disorders, whereas Th2-type anti-inflammatory responses have generally been shown to inhibit autoimmune disease. The role of proinflammatory Th1 responses in the inhibition/clearance of Aβ pathology is not obvious. Moreover, passive vaccination with monoclonal anti-Aβ antibodies inhibits/clears Aβ deposits in the brain without any help from Th1 cytokines. Therefore, the main goal of any active AD vaccine is the generation of high titers of anti-Aβ antibodies that can alter or block AD pathology (Patton et al., 2006; Petrushina et al., 2007). Of course, the stimulation of antibody responses against any T-dependent protein/peptide vaccine requires the formulation of the vaccine in a strong adjuvant. Whether or not an Aβ-based vaccine should be formulated in a Th1 or a Th2 adjuvant in order to be effective remains to be elucidated. It has been suggested that autoreactive Th1 cells and/or proinflammatory cytokines may be harmful to AD patients vaccinated with AN1792 formulated in QS21 adjuvant (Schenk et al., 2004), which is known as a Th1 cell adjuvant (Kensil et al., 1991).

    In this regard, as well, data presented by Pride et. al. is somewhat confusing. The authors mention “In Study 102, patients received up to 8 doses of AN1792. The only apparent differences in the two trials was the addition of Polysorbate 80 to the formulation used throughout Study 201; this excipient was added to the formulation used in Study 102 beginning with dose 5. ME was diagnosed only in one patient (1/64, 1.6%) at autopsy. This patient had received 5 doses of AN1792 with QS21, the last containing Polysorbate 80.” This section indicates that individuals from both Studies 102 and 201 in fact received Polysorbate 80, but participants in Study 102 received it up to four times (Bayer et al., 2005) , while participants in Study 201 received it one to three times (Gilman et al., 2005). Thus, it is not entirely clear how the authors investigated the differences between groups immunized with a vaccine formulated in QS21 and QS21 plus Polysorbate 80, and attributed differences in their immune responses to Polysorbate 80.

    It will indeed be imperative to analyze Th1 (IFN-γ) or Th2 (IL-4, IL-5, but not IL-2) cellular responses in AD patients who received Aβ antigen formulated in QS21 or QS21 plus Polysorbate 80. These results should help the scientific community understand the differences between T cell responses measured in PBMCs obtained from participants immunized with a vaccine formulated in an adjuvant alone vs. in an adjuvant plus Polysorbate 80.

    Thirdly, regarding meningoencephalitis, it has been suggested that it is connected with autoreactive Th cells and/or Th1 proinflammatory cytokines (Ferrer et al., 2004; Nicoll et al., 2003). This is quite possible. First off, I should mention that low/moderate titers of “free” anti-Aβ antibodies (~1:2200) have been generated in only 59 out of 300 individuals immunized with AN1792 (“responders”), so about 80 percent of the immunized subjects were “non-responders” (Gilman et al., 2005; Patton et al., 2006). In immunological vaccination trials, this would be an inadequate response rate, although we should take into account that some participants of the trial received only one injection. The paper does not spell out how many patients received how many injections. Although titers of free Aβ-specific antibodies did not correlate with the incidence of adverse events, they did correlate with a positive clinical outcome (Bayer et al., 2005; Ferrer et al., 2004; Gilman et al., 2005; Masliah et al., 2005; Nicoll et al., 2003). Both these findings do not necessarily prove that these antibodies are harmless. It is worth mentioning that 22 percent of patients from this responsive subgroup developed meningoencephalitis, whereas only five (2 percent) of the 241 “non-responders” did (Patton et al., 2006). These data suggest that the side effect may be connected to responders and therefore to autoreactive Th cells and/or to a lesser extent to Th1 proinflammatory responses, as well as to anti-Aβ antibodies. For example, it is possible that anti-Aβ antibodies could remove solubilized Aβ peptide that in part ends up as vascular deposits; this happened with the AN1792 (Patton et al., 2006) and may have triggered hemorrhages (Ferrer et al., 2004). I hope that data from ongoing late-stage trials with bapineuzumab, as well as the ACC-001 trials, will help us clearly distinguish positive and negative effects of self and non-self Th cell responses, proinflammatory cytokines, adjuvants, and antibody responses in vaccinated AD patients.

    As an immunologist I believe that the vasculitis reported in the ACC-001 trial is unlikely to be connected to a Th1-type adjuvant, autoreactive T cells, or proinflammatory responses. If this patient did not develop anti-Aβ antibodies, then the lesions are likely unrelated to immune responses generated against the vaccine. Finally, I call on the companies involved in AD clinical trials to provide detailed results on their clinical studies to the scientific community quicker. That way, they’ll get rapid feedback from the community and, with their help, develop a safe and potent protective or therapeutic vaccine against this devastating disease.


    . Prototype Alzheimer's disease vaccine using the immunodominant B cell epitope from beta-amyloid and promiscuous T cell epitope pan HLA DR-binding peptide. J Immunol. 2005 Feb 1;174(3):1580-6. PubMed.

    . Adjuvant-dependent modulation of Th1 and Th2 responses to immunization with beta-amyloid. Int Immunol. 2003 Apr;15(4):505-14. PubMed.

    . Protein vaccines induce uncommitted IL-2-secreting human and mouse CD4 T cells, whereas infections induce more IFN-gamma-secreting cells. J Immunol. 2006 Feb 1;176(3):1465-73. PubMed.

    . Evaluation of the safety and immunogenicity of synthetic Abeta42 (AN1792) in patients with AD. Neurology. 2005 Jan 11;64(1):94-101. PubMed.

    . Neuropathology and pathogenesis of encephalitis following amyloid-beta immunization in Alzheimer's disease. Brain Pathol. 2004 Jan;14(1):11-20. PubMed.

    . Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology. 2005 May 10;64(9):1553-62. PubMed.

    . Separation and characterization of saponins with adjuvant activity from Quillaja saponaria Molina cortex. J Immunol. 1991 Jan 15;146(2):431-7. PubMed.

    . 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.

    . Abeta vaccination effects on plaque pathology in the absence of encephalitis in Alzheimer disease. Neurology. 2005 Jan 11;64(1):129-31. PubMed.

    . Increased T cell reactivity to amyloid beta protein in older humans and patients with Alzheimer disease. J Clin Invest. 2003 Aug;112(3):415-22. PubMed.

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

    . Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat Med. 2003 Apr;9(4):448-52. PubMed.

    . 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.

    . 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.

    . 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.

    . Current progress in beta-amyloid immunotherapy. Curr Opin Immunol. 2004 Oct;16(5):599-606. PubMed.

    . In vivo priming of CD4 T cells that produce interleukin (IL)-2 but not IL-4 or interferon (IFN)-gamma, and can subsequently differentiate into IL-4- or IFN-gamma-secreting cells. J Exp Med. 2001 Oct 15;194(8):1069-80. PubMed.

Make a Comment

To make a comment you must login or register.