Immune tolerance in multiple sclerosis: the holy grail or not?

Research news - how important are papers describing new auto-antibodies in people with MS and new ways of inducing immune tolerance to auto-antigens?

The holy grail in treating autoimmunity is resetting the immune system to become tolerant to the proteins or antigens driving the disease. Since I started working in the field of multiple sclerosis (MS) in the early 90s, immunologists have been working on immune tolerance as a treatment for MS. To date, no antigen-specific immune tolerance therapies have worked. However, many immunologists argue that immune system rebooters or immune reconstitution therapies, such as alemtuzumab and AHSCT, work by immune tolerance mechanisms. I am not convinced.

Image from Wikipedia.

I have always questioned the underlying assumption that MS is an autoimmune disease. The reasons are mainly epidemiological, which tells us that Epstein Barr Virus plays a causal role in MS. Immunologists assume EBV is triggering autoimmunity and is not directly causing the disease. Despite identifying many putative autoantigens, with or without accompanying evidence of molecular mimicry to viral or other proteins, including EBV antigens, none adequately explains the disease. However, the one that intrigues me most is the alpha-B-crystallin protein because it is upregulated in B-cells by EBV and is also expressed in myelin in the MS brain but not in healthy controls. A phase 2 trial of trying to induce tolerance to alpha-B-crystallin gave lukewarm results (paper 3). I know some immunologists have questioned the protocol used in this study. It is a pity the alpha-B-crystallin was taken forward in further trials. Sadly, the patent life of alpha-B-crystallin as a tolerance therapy in MS is too short. Therefore, it is unlikely to be taken forward. Often, potential MS therapies get discarded because there is too little time to test them and recoup the development costs for pharma companies. 

People recently got very excited by the paper below that used a molecular trick to direct proteins to the liver. The liver is an interesting organ in relation to immunology. The liver is designed to dampen inflammation and not drive immune responses to proteins. It receives venous blood via the enterohepatic (gut-liver) portal system. Food and microbiome-related proteins will be in the blood, draining from the gut. Mounting and immune response to these might set up autoimmune diseases in the gut. This may be what happens with inflammatory bowel and coeliac disease. In coeliac disease, the immune response is to gluten, but patients with coeliac disease also mount an immune response to transglutaminase. Antibodies to tissue transglutaminase are a practical diagnostic test for coeliac disease but are not pathogenic in coeliac disease. The anti-transglutaminase antibody is as an example that the finding of an autoantibody, which is strongly associated with a disease, may not necessarily mean that that specific protein or antigen is driving the disease. Please consider this example uppermost when interpreting any paper describing the putative autoantigens in MS.  

The researchers in paper-1 below show that by targeting proteins and antigens to the liver, they can induce immune tolerance to myelin oligodendrocyte protein (MOG) and myelin proteolipid protein (PLP) in animals with EAE (experimental allergic encephalomyelitis). The MOG model was an adoptive transfer model using autoreactive T-cells, and the PLP model was an active immunisation model. The tolerance paradigm worked in that relapses in EAE were suppressed. Will this translate into MS? I suspect not. There is little evidence that an autoimmune response to MOG drives MS. However, there is a different disease called MOGAD (MOG-associated demyelinating diseases) with anti-MOG antibodies that look and behave differently to MS. Similarly, the evidence that PLP is the autoantigen in MS needs strengthening. The recent study by Gregory Owens and colleagues has shown that many pwMS have antibodies to conformational proteolipid protein-1 complex in membranes. However intriguing the results of this paper seem, we need to wait for the results to be reproduced and for more research to be done on the pathogenicity of these anti-PLP-1-membrane-complexed antibodies. 

The question is whether or not the findings in these two studies can come together and lead to a clinical trial in pwMS is an important one. The aim would be to select pwMS with anti-PLP antibodies and to use a similar protocol to design tolerogenic peptides couples to N-acetylgalactosamine (pGal) for intravenous administration. The task is not trivial but worth a try.

However, I would start by tackling autoimmune diseases with known and relatively well-characterised autoantigens, for example, type 1 diabetes, myasthenia gravis, Grave’s disease and anti-aquaporin-4-associated NMOSD. Testing this strategy in MS is high risk. I say this because in many diseases with end-organ damage, the immune system responds to organ-specific proteins as part of the damage, and hence the immune responses are secondary. This has been shown to occur in MS. Many people with progressive MS have anti-neurofilament antibodies, particularly against neurofilament heavy chain, which is not a primary but a secondary response. Therefore, we don’t know if the anti-PLP-1-membrane-complex response is primary or secondary. This is why more work needs to be done on the anti-PLP-1-membrane-complex response. Does the presence of these antibodies precede the development of MS? Do these antibodies predict the development of MS in the future? 

If I were a chief scientific officer of a large drug company doing due diligence on using this immune tolerance strategy, I would want to see much more basic science done before investing in an MS clinical trial. I would want more convincing data in a disease with a well-defined auto-antigen. 

All these studies bring back memories for me. Almost twenty years ago, Prof. David Baker and I tried translating his immune tolerance work in mice into humans. We elected to use anti-interferon-beta neutralising antibodies for the autoimmune disease. The idea was to use mitoxantrone immune depletion, followed by intravenous antigen (interferon-beta) administration. The problem is that we had to pull the plug on the study after we treated only one subject. The study was too difficult to recruit because mitoxantrone was shown to cause leukaemia in about 1 in 200 treated subjects, and most potential study subjects didn’t want to take on this risk to remove their NABs. However, the strategy worked in our n=1 trial, and the patient lost her NABs. It is a pity we didn’t use the strategy in other autoimmune diseases. 

We are now asking ourselves whether this new way of delivering tolerogenic antigen must also be coupled with immunodepletion or will work without depletion. The theory behind immune depletion is that newly divided cells are more sensitive to tolerogenic signals than older established memory cells.

I have received many emails and direct messages about these new studies (papers 1 & 2) on my social media channels. Yes, they are interesting, but they are a long way from delivering a treatment for MS. A development programme of intravenous immune tolerance strategy for MS will take 10-15 years to deliver results. By then, we will likely have the first read-outs of the EBV vaccination programme, hopefully showing that preventing IM prevents downstream events such as MS, SLE, other autoimmune diseases and the many EBV-associated cancers. 

Please note that anti-CD40L (frexalimab), which blocks necessary costimulatory signals between B-cells, professional antigen-presenting cells and T-cells, is going forward into phase 3  trials in MS. This pathway may induce tolerance in an antigen-agnostic way. If I were to invest in an expensive drug development programme, I would use an antigen-agnostic approach. Please note I am biased as I am actively involved with the frexalimab development programme. You can read my MS-Seflie newsletter on frexalimab for more information. 

Frexalimab breaking news: we have a new therapeutic target in MS (31-May-2023)

Paper 1

Tremain et al. Synthetically glycosylated antigens for the antigen-specific suppression of established immune responses. Nat Biomed Eng. 2023 Sep;7(9):1142-1155.

Inducing antigen-specific tolerance during an established immune response typically requires non-specific immunosuppressive signalling molecules. Hence, standard treatments for autoimmunity trigger global immunosuppression. Here we show that established antigen-specific responses in effector T cells and memory T cells can be suppressed by a polymer glycosylated with N-acetylgalactosamine (pGal) and conjugated to the antigen via a self-immolative linker that allows for the dissociation of the antigen on endocytosis and its presentation in the immunoregulatory environment. We show that pGal-antigen therapy induces antigen-specific tolerance in a mouse model of experimental autoimmune encephalomyelitis (with programmed cell-death-1 and the co-inhibitory ligand CD276 driving the tolerogenic responses), as well as the suppression of antigen-specific responses to vaccination against a DNA-based simian immunodeficiency virus in non-human primates. Our findings show that pGal-antigen therapy invokes mechanisms of immune tolerance to resolve antigen-specific inflammatory T-cell responses and suggest that the therapy may be applicable across autoimmune diseases.

I would be interested to know if you find these research-related Newsletters of interest?

Paper 2

Owens et al. Pathogenic myelin specific antibodies in multiple sclerosis target conformational proteolipid protein 1 anchored membrane domains. J Clin Invest. 2023 Aug 10;e162731.

B cell clonal expansion and cerebrospinal fluid (CSF) oligoclonal IgG bands are established features of the immune response in multiple sclerosis (MS). Clone-specific IgG1 monoclonal recombinant antibodies (rAbs) derived from MS patient CSF plasmablasts bound to conformational proteolipid protein 1 (PLP1) membrane complexes and, when injected into mouse brain with human complement, recapitulated histologic features of MS pathology: oligodendrocyte cell loss, complement deposition, and CD68+ phagocyte infiltration. Conformational PLP1 membrane epitopes were complex and governed by the local cholesterol and glycolipid microenvironment. Antibodies against conformational PLP1 membrane complexes targeted multiple surface epitopes, were enriched within the CSF compartment, and were detected in most MS patients but not in inflammatory and non-inflammatory neurologic controls. CSF PLP1 complex antibodies provide a pathogenic autoantibody biomarker specific for MS.

Paper 3

van Noort et al. Therapeutic Intervention in Multiple Sclerosis with Alpha B-Crystallin: A Randomized Controlled Phase IIa Trial. PLoS One. 2015 Nov 23;10(11):e0143366. 

As a molecular chaperone and activator of Toll-like receptor 2-mediated protective responses by microglia and macrophages, the small heat shock protein alpha B-crystallin (HspB5) exerts therapeutic effects in different animal models for neuroinflammation, including the model for multiple sclerosis (MS). Yet, HspB5 can also stimulate human antigen-specific memory T cells to release IFN-γ, a cytokine with well-documented detrimental effects during MS. In this study, we explored in a Phase IIa randomized clinical trial the therapeutic application of HspB5 in relapsing-remitting MS (RR-MS), using intravenous doses sufficient to support its protective effects, but too low to trigger pathogenic memory T-cell responses. These sub-immunogenic doses were selected based on in vitro analysis of the dose-response profile of human T cells and macrophages to HspB5, and on the immunological effects of HspB5 in healthy humans as established in a preparatory Phase I study. In a 48-week randomized, placebo-controlled, double-blind Phase IIa trial, three bimonthly intravenous injections of 7.5, 12.5 or 17.5 mg HspB5 were found to be safe and well tolerated in RR-MS patients. While predefined clinical endpoints did not differ significantly between the relatively small groups of MS patients treated with either HspB5 or placebo, repeated administration especially of the lower doses of HspB5 led to a progressive decline in MS lesion activity as monitored by magnetic resonance imaging (MRI), which was not seen in the placebo group. Exploratory linear regression analysis revealed this decline to be significant in the combined group receiving either of the two lower doses, and to result in a 76% reduction in both number and total volumes of active MRI lesions at 9 months into the study. These data provide the first indication for clinical benefit resulting from intervention in RR-MS with HspB5.

Paper 4

Marta et al. Antigen-specific tolerization in human autoimmunity: Inhibition of interferon-beta1a anti-drug antibodies in multiple sclerosis: A case report. Mult Scler Relat Disord. 2021 Nov;56:103284. 

Background: Antigen-specific tolerance in auto-immune diseases is the goal for effective treatment with minimal side-effects. Whilst this is achievable in animal models, notably via intravenous delivery of the model-specific autoantigen following transient CD4 T cell depletion, specific multiple sclerosis autoantigens remain unproven. However, anti-drug antibodies to human therapeutic proteins represent a model human autoimmune condition, which may be used to examine immune-tolerance induction. Some people with MS (PwMS) on interferon-beta1a (IFNβ1a) develop neutralizing antibodies to IFNβ1a that do not disappear in repeated tests over years.

Methods: One PwMS was recruited, as part of a planned phase IIa trial (n = 15), who had developed neutralizing antibodies to subcutaneous IFNβ1a. Mitoxantrone (12 mg/m2) was administered as a lymphocyte depleting agent followed by four days of (88 μg/day + three 132 μg/day) intravenous IFNβ1a. Subcutaneous IFNβ1a three times a week was maintained during follow-up. IFNβ1a neutralizing antibody responses in serum were measured during treatment and three-monthly for 12 months.

Findings: One participant was recruited and, within 6 months of tolerization, the neutralizing antibodies were undetectable. The tolerization treatment was well tolerated. However, the study was terminated after the first enrolment, on ethical grounds, as treatment alternatives became available and the potential risks of mitoxantrone use increased.

Interpretation: The data suggest that it may be possible to induce antigen-specific tolerance by providing tolerogenic antigen following transient immune depletion. Further studies are warranted.

Subscriptions and donations

MS-Selfie newsletters and access to the MS-Selfie microsite are free. In comparison, weekly off-topic Q&A sessions are restricted to paying subscribers. Subscriptions are being used to run and maintain the MS Selfie microsite, as I don’t have time to do it myself. You must be a paying subscriber if you ask questions unrelated to the Newsletters or Podcasts. If you can’t afford to become a paying subscriber, please email a request for a complimentary subscription (ms-selfie@giovannoni.net).

Important Links

MS-Selfie microsite

Donations to MS-Selfie

Prof. G’s Backstory and CV

Prof. G’s MS Blog Archive

Conflicts of Interest

Twitter / LinkedIn / Medium

General Disclaimer

Please note that the opinions expressed here are those of Professor Giovannoni and do not necessarily reflect the positions of Queen Mary University of London or Barts Health NHS Trust. The advice is intended as general and should not be interpreted as personal clinical advice. If you have problems, please tell your own healthcare professional, who will be able to help you.