New Advances in the Treatment of Neurological Diseases Using High Dose Intravenous Immunoglobulins

Abstract

Since the incidental discovery in 1981 that intravenous immunoglobulins (IVIg) are immunomodulatory, they have been investigated in a large number of putative autoimmune diseases. This has led to licensing for idiopathic thrombocytopenic purpura, Kawasaki disease, and in neurological disorders for Guillain-Barré syndrome (GBS). Although not licensed, randomized controlled trials have also shown IVIg efficacy in other neuroimmunological diseases such as multifocal motor neuropathy (MMN), chronic inflammatory demyelinating neuropathy (CIDP), myasthenia gravis, dermatomyositis, and stiff-person syndrome. However, other indications are currently being explored including Alzheimer's disease, postpolio syndrome, and narcolepsy. There are even reports from experimental studies in stroke. The results of recently published clinical trials in both the classical neuroimmunological disorders as well as for new indications are reported and their role in clinical practice is discussed.

Introduction

Intravenous immunoglobulins (IVIg) consists of pooled polyclonal immunoglobulins derived from several thousand healthy donors. In 1981, Imbach and colleagues treated children with hypogammaglobulinaemia and coincident idio-pathic thrombocytopenic purpura (ITP) with IVIg as a substitution therapy [Imbach et al. 1981]. The observation that platelet counts increased after each IVIg administration was the first description of an immunomodulatory effect of IVIg. Randomized trials could clearly establish the potential of IVIg to modulate the course of diseases with a putative autoimmune pathophy-siology. This led to the licensing of IVIg (beside substitution therapy in immunodeficiency disorders) for Kawasaki disease, ITP, and Guillain-Barré syndrome. However, particularly in neuroimmunological disorders there are several conditions where double-blind randomized, placebo-controlled, class I evidence clinical trials have clearly established the efficacy of IVIg. For some conditions, although off-label, IVIg are even recommended as a first-line treatment. This review concentrates on clinical trials of IVIg in neurological disorders. As well as a summary of the clinical indications and an update on new trials in classical neuroimmunolo-gical disorders where IVIg have already been used for more than a decade, an overview of possible new indications for IVIg is given.

Indications for IVIg in established neuroimmunological disorders

Guillain-Barré syndrome (GBS)

The efficacy of IVIg in GBS has been established in two large trials comparing IVIg with the standard therapy of plasma exchange (PE) [Plasma Exchange/Sandoglobulin Guillain-Barré Syndrome Trial Group, 1997; van der Meche et al. 1992]. In general it is accepted that both PE and IVIg have an equal therapeutic efficacy [Hughes et al. 2007]. Since IVIg is more readily available and more convenient to use, this is in most centres the treatment of choice. This has also led to the licensing of IVIg in GBS in some countries as the only neurological disorder for which IVIg is approved.

There are only limited data on the dosage of IVIg in GBS. The standard dose consists of 0.4 g/kg bodyweight on five consecutive days. This total dose of 2.0g/kg may also be given on two days (two times 1 g/kg), but there are no controlled data. A small trial comparing 0.4 g/kg on three days vs six days found the higher dose to be slighly, but not significantly, more effective [Raphael et al. 2001].

The addition of intravenous steroids (500 mg methylprednisolone per day for five days) did not show a significant improvement compared to IVIg alone [Van Koningsveld et al. 2004] and therefore the use of steroids is not generally recommended.

There are no evidence-based studies on which treatment to recommend in cases where there is no response to IVIg. Another course of IVIg may be given or alternatively PE could be performed.

Chronic inflammatory demyelinating neuropathy (CIDP)

The therapeutic efficacy of IVIg in the treatment of CIDP has been demonstrated in several randomized controlled trials that compared IVIg to placebo, PE, or steroids [Hughes et al. 2001; Mendell et al. 2001; Hahn et al. 1996]. These trials have mostly investigated treatment initiation, while there were no data on long-term treatment in this chronic disease. In a recent study, 117 patients were randomized in a response-conditional crossover trial. IVIg (2g/kg loading dose and 1g/kg maintenance) or placebo were given every 3 weeks for an initial treatment period of up to 24 weeks [Hughes et al. 2008]. Patients crossed to the other treatment arm when they were nonresponders (as judged by a deterioration of at least one point on the INCAT disability score or if the score was stable until week six). Patients who improved in the first treatment period were re-randomised after 24 weeks to another 24-week treatment period with either IVIg (1 g/kg every three weeks) or placebo. If a patient worsened they were considered to have relapsed and were withdrawn from the study. There was a highly significant improvement in favour of the IVIg treatment for both the first treatment period and the crossover. During the extension study the time to relapse was significantly longer in the IVIg group as compared to placebo. This is the first study that clearly demonstrates the long-term effect of IVIg in CIDP. However, the best dose for maintenance treatment is still not clear. This study used a rather high dose with short intervals and there was no dose titration as is done in clinical practice.

Although there are now strong arguments in favour of a long-term treatment of CIDP with IVIg, they are not yet licensed for this indication and due to their lower cost, steroids will probably remain the first-line treatment in many patients. However, the manifold side effects from long-term steroid treatment may change this in the future.

Multifocal motor neuropathy (MMN)

The overall efficacy of IVIg in patients with MMN has been established in a number of randomized controlled trials [van den Berg et al. 1995; Federico et al. 2000; Leger et al. 2001]. Although patient numbers are in all studies rather low due to the low prevalence of the disease (between 5 and 18 patients), there is consensus that IVIg are the first-line treatment [Joint Task Force of the EFNS and the PNS, 2006]. The response to IVIg may also be of value for a differential diagnosis. Several investigations aimed to define predictors for IVIg responsiveness. However, this proved to be extremely difficult and neither the presence of conduction block nor anti-GM1 ganglioside antibody status proved to be reliable [Leger et al. 2008; Slee et al. 2007; Strigl-Pill et al. 2006]. Although the short-term effect of IVIg is unquestioned the long-term response is less clear. Despite continuous treatment new conduction blocks may occur and patients are usually dependent on periodic IVIg infusions [Leger et al. 2008; Terenghi et al. 2004; Vucic et al. 2004]. The frequency and dose of the repeated IVIg infusions remains unclear. In clinical practice, the initial dose consists of 0.4 g/kg bodyweight on five consecutive days. In the case of treatment response, a dose titration to the minimal effective dose and an extension of the interval until shortly before the effect weans off is recommended. Only where there is an insufficient response or where the IVIg effect wears off should other immunotherapies such as cyclophosphamide, azathioprine, cyclosporine, rituximab, or interferon-beta be considered [Joint Task Force of the EFNS and the PNS, 2006]. In particular with cyclopho-sphamide the safety profile needs to be kept in mind.

Myasthenia gravis

The first case reports and case series describing a beneficial effect of IVIg in myasthenia gravis were published in 1984, shortly after the discovery of the immunomodulatory effect of IVIg [Fateh-Moghadam et al. 1984; Gajdos et al. 1984]. It took more than a decade until the first controlled trial was published, comparing IVIg and PE in patients with myasthenic crisis [Gajdos et al. 1997]. The effectiveness has been confirmed by another controlled trial comparing IVIg with placebo in patients with worsening weakness due to myas-thenia gravis [Zinman et al. 2007]. These results have led to the recommendation that IVIg may be used as an alternative to PE in patients with severe worsening. The data do not support the superiority of one treatment over the other. However, the optimal dose still remains unclear. The study by Gajdos et al. (2005) suggests that a total dose of 1 g/kg bodyweight is sufficient.

The data are less clear for the long-term treatment of myasthenia. Standard immune therapies are steroids, azathioprine, and cyclosporine. There is no controlled trial and the evidence is not sufficient to recommend IVIg as maintenance therapy in the management of myasthenia [Skeie et al. 2006]. However, in single cases IVIg may be helpful when other immune therapies are ineffective or if there are contraindications.

Inflammatory myopathies

The inflammatory myopathies are usually divided into polymyositis, dermatomyositis, and inclusion body myositis.

For dermatomyositis a placebo-controlled crossover trial clearly demonstrated the efficacy of IVIg 15 years ago [Dalakas et al. 1993]. However, since most patients respond to steroids, azathioprine, methotrexate, or cyclophospha-mide, IVIg has remained a second or third-line treatment in this disorder. In cases where these immunosuppressive treatments fail, the dose and frequency of IVIg have to be evaluated individually. Some experts think that IVIg may even be used as first-line treatment in juvenile dermatomyositis to avoid the adverse effects of immunosuppressive agents in children and adolescents [Breems et al. 1993].

In polymyositis, case series suggest that IVIg efficacy is similar to that in dermatomyositis [Saito et al. 2008; Cherin et al. 2002; Moriguchi et al. 1996; Cherin et al. 1994; Jann et al. 1992; Cherin et al. 1991]. There is no controlled trial available and some of these studies included a mixed population of patients with der-matomyositis and polymyositis. Nevertheless, in patients not responsive to steroid treatment IVIg was often effective [Saito et al. 2008; Cherin et al. 1991]. Due to some similarities of the two diseases, expert panels have recommended IVIg as second or third-line treatment in patients with polymyositis when other immune therapies fail [Feasby et al. 2007; Stangel and Gold, 2004; Wiles et al. 2002].

There have been three controlled trials of IVIg in inclusion body myositis (IBM) [Dalakas et al. 2001b; Walter et al. 2000; Dalakas et al. 1997]. There was uniformly only a small clinical effect of IVIg treatment, which was overall not statistically significant. However, in the study with the longest duration [Walter et al. 2000] 18 of 22 patients (90%) had no disease progression, suggesting a stabilizing effect of IVIg. Most prominently, there was a significant effect of IVIg on swallowing muscles with improvement of swallowing [Walter et al. 2000]. Overall, the use of IVIg in IBM is controversial, but due to the lack of alternative treatments it seems justified to explore the effect of IVIg for individual patients, in particular when there is life-threatening dysphagia [Dalakas, 2006].

Stiff-person syndrome

There is one placebo-controlled study that clearly demonstrated that IVIg are effective in the treatment of stiff-person syndrome with anti-GAD65 antibodies [Dalakas et al. 2001a]. Nevertheless, most experts consider IVIg as a second-line treatment in patients where treatment with corticos-teroids fails.

Multiple sclerosis (MS)

Several clinical trials have evaluated IVIg during different stages of MS [Stangel and Gold, 2005]. Unfortunately, many results did not meet the expectations given from preliminary studies.

Relapsing-remitting MS (RRMS) Four placebo-controlled randomized trials have suggested the efficacy of IVIg on relapse rate and possibly also on disease progression [Lewanska et al. 2002; Achiron et al. 1998; Sorensen et al. 1998; Fazekas et al. 1997]. There was some criticism of these studies due to the lack of MRI data in some studies, the small patient numbers in others, the different dosing regimens, and the duration of treatment. A meta-analysis of these four trials [Sorensen et al. 2002] showed a superiority of IVIg compared to placebo for yearly relapse rate, relapse-free patients, and EDSS change. This led to the recommendation to use IVIg as a second-line therapy in patients where first-line treatments like beta-interferons or glatiramer acetate are contraindicated or have failed [Rieckmann and Toyka, 1999]. However, another well-designed study, that included MRI, in 127 RRMS patients did not show a statistically significant reduction of the relapse rate between the two IVIg treatments (0.2 g/kg and 0.4 g/kg monthly) and placebo [Fazekas et al. 2008]. Thus the potential of IVIg in RRMS has possibly been overestimated. Due to these most recent negative results it seems rather unlikely for IVIg to become a first-line treatment in RRMS.

Secondary progressive MS (SPMS) Two large trials have investigated the effect of IVIg on disease progression in progressive MS patients. In the first trial (ESIMS) including 318 patients, 1 g/kg IVIg monthly vs placebo, did not show a difference on clinical parameters [Hommes et al. 2004]. With regard to MRI, there was a delayed reduction of brain volume in the IVIg group [Filippi et al. 2004], but the relevance of this finding is not clear. In the second trial, 231 patients with primary progressive MS (PPMS) and SPMS were treated with either 0.4 g/kg IVIg per month or placebo for two years [Pöhlau et al. 2007]. There was a trend towards a prolonged time until disease progression that was significant in the PPMS group. Although the authors concluded that IVIg could delay disease progression in PPMS patients, 120 patients withdrew prematurely from the treatment and thus only 18 PPMS patients received the per protocol treatment. These data do not allow for a recommendation of IVIg in progressive MS and it can be concluded that IVIg is not effective in delaying disease progression in SPMS.

Relapse treatment Two studies have investigated whether IVIg can improve recovery after relapse [Sorensen et al. 2004; Visser et al. 2004]. Both used IVIg or placebo as an add-on before treatment with corticosteroids. There was no statistically significant difference between the treatment groups. A third study investigated the administration of IVIg (0.4 g/kg bodyweight on days 0, 1, 2, 30, and 60) vs placebo within four weeks after the onset of acute optic neuritis [Roed et al. 2005]. There was no long-term effect of IVIg on visual function or visual evoked potentials. Thus, the treatment of acute relapses in MS with IVIg can currently not be recommended. Future studies may investigate the effect of IVIg in relapses that do not respond to steroid treatment, where plasma exchange was shown to be effective, particularly in patients with a B-cell mediated patho-physiology [Keegan et al. 2005].

Clinically isolated syndrome (CIS) There is one single-centre trial available that explored the capacity of IVIg to delay clinically definite MS after a first demyelinating event [Achiron et al. 2004b]. Ninety-one patients were randomized to receive either placebo or IVIg (2 g/kg loading dose followed by 0.4 g/kg every six weeks) for one year. There was a significantly lower probability of developing clinically definite MS during the study in the IVIg group; the number of T2-weighted lesions and the volume of Gd-enhancing lesions were also less in the IVIg treated patients. Despite these promising data this study needs confirmation and IVIg is not used as a routine treatment in CIS patients where beta-interferons are licensed in some countries.

Postpartal treatment It is well known that the relapse rate is lowered during pregnancy and the risk of relapse is increased in the postpartum period [Confavreux et al. 1998]. It is thus desirable to prevent these relapses, but the licensed therapies for RRMS are not recommended during lactation. Several uncontrolled case series have reported that the administration of IVIg after child delivery may prevent postpartal relapses [Achiron et al. 2004a; Haas, 2000; Achiron et al. 1996]. A randomized study, comparing 60 g vs 10 g within three days after delivery followed by monthly infusions of 10 g for five months in both arms, did not show a difference between these two treatment regimens [Haas and Hommes, 2007]. The postpartal relapse rate returned to pre-pregnancy levels and was not increased as would have been expected [Confavreux et al. 1998]. However, this kind of control may not be appropriate. Despite the lack of a decisive placebo-controlled trial IVIg can be considered an optional treatment to reduce post-partal relapses due to the lack of alternative treatments with proven efficacy.

Induction of repair The rationale for the design of clinical studies to induce remyelination and repair by IVIg originates from a number of experimental studies in the animal model of Theiler's virus mouse encephalitis (TMEV) [Warrington et al. 2000; Bieber et al. 2001]. In patients with long-standing clinical deficits in MS, three studies could not demonstrate a clinical or neurophysiological improvement induced by IVIg [Stangel et al. 2000; Noseworthy et al. 2000; Noseworthy et al. 2001]. The reasons for this failure may be manifold [Stangel and Hartung, 2002], e.g., because remyelination of chronically demyelinated and damaged axons may not restore function. Furthermore, in the animal model the best effect was achieved by IgM antibodies that are not contained in the IVIg preparations used in the clinical trials. Since there are also human monoclonal IgM identified that can achieve remyelination in TMEV [Mitsunaga et al. 2002], it remains to be tested if this is also true in patients with MS.

New indications

Alzheimer's disease (AD)

Animal experiments have shown that administration of antibodies against Amyloid-ß (Aß) pep-tide, which is thought to be one of the main pathogenic elements in plaque formation in AD, can reduce plaques in the CNS [Bard et al. 2000]. Naturally occuring antibodies against Aß are contained in IVIg and a first preliminary study giving 0.4g/kg IVIg on three consecutive days led to a decrease of Aß in the cerebrospinal fluid [Dodel et al. 2002]. The treatment of five patients with IVIg (0.4 g/kg on three consecutive days every month) over six months led to an improvement in a cognitive score [Dodel et al. 2004]. In another uncontrolled trial in eight patients, IVIg was administered for six months, discontinued, and then resumed for another nine months [Relkin et al. 2008b]. Aß levels in the cerebrospinal fluid decreased, returned to baseline during washout, and decreased again during the re-treatment phase. Similarly, mini-mental status examination (MMSE) score increased by 2.5 points, returned to baseline after washout and remained stable during subsequent IVIg treatment. A first small placebo-controlled study treating probable AD patients (MMSE 14-26) with various dosages of IVIg (n=16, 0.2 kg/2 weeks to 0.8 kg/month) for six months has been reported at scientific meetings [Relkin et al. 2008a]. The global outcome was significantly better for the whole IVIg group, but this study was too small to give a clear answer. The coming years and clinical phase III trials will clarify the potential of IVIg in the treatment of AD. More specific antibody-based treatments including monoclonal antibodies against Aß are also being followed [Lichtlen and Mohajeri, 2008].

Postpolio syndrome (PPS)

Studies of cytokine levels have suggested that there may be an immunological component in the pathogenesis of PPS [Gonzalez et al. 2004]. A placebo-controlled study in 142 PPS patients receiving 90 g IVIg, repeated after three months, led to an 8.3% increase in strength in selected muscles, while there was no change in the placebo group [Gonzalez et al. 2006]. In particular, patients with pain seemed to benefit, but there was no change in quality-of-life questionnaires. Another small placebo-controlled study with 20 PPS patients also found an effect on pain, but there was no difference in muscle strength and fatigue [Farbu et al. 2007]. Thus, there is currently no recommendation to treat PPS with IVIg on a routine basis. Future studies need to clarify if there are subgroups of PPS patients that benefit from IVIg.

Fibromyalgia and other pain syndromes

Goebel and coworkers have invesigated the effect of IVIg in 130 patients with pain syndromes [Goebel et al. 2002]. Pain reduction of 470% was observed in 20% of patients and another 27% reported pain reduction of 20-70% after IVIg administration. Early treatment (less than two years after the beginning of symptoms) was associated with better treatment response. However, this study included very heterogenous patients with a variety of pain syndromes including fibromyalgia, spinal pain, complex regional pain syndrome, peripheral neuropathic pain, and atypical facial pain. Furthermore, there was no control group and thus further controlled studies with predefined pain syndromes need to be carried out.

A recent study in patients with fibromyalgia has postulated that a considerable subset have electrodiagnostic findings suggestive of CIDP and were thus treated with IVIg (0.4 g/kg on five days). Both fatigue and stiffness had a trend towards improvement [Caro et al. 2008]. These findings need to be confirmed in a controlled clinical trial.

Narcolepsy

Because narcolepsy has a strong linkage to human leucocyte antigens, an autoimmune mechanisms has been hypothesized. The treatment of four patients within a few months after the onset of narcolepsy led to an improvement of frequency and severity of cataplexy [Dauvilliers et al. 2002]. In three of these patients, the reduction of cataplexy lasted for at least two years without any additional IVIg treatment [Dauvilliers, 2006]. In contrast, a recent report that was published as an abstract could not confirm these findings. Of four patients treated with 1g/kg IVIg on two consecutive days, repeated three times at five week intervals, only two patients had a transient subjective improvement. There were no persistent and objective improvements in narcoleptic symptoms [Valko et al. 2008]. Thus, again, a placebo-controlled randomized trial is required to determine the efficacy of IVIg in narcolepsy.

Stroke

There is experimental evidence that IVIg may be beneficial in stroke. Mice who were administered 2 g/kg IVIg 30 min before or three hours after transient middle cerebral artery occlusion exhibited a 50-60% reduction of brain infarct size and a 2–3-fold inprovement in functional outcome [Arumugam et al. 2007]. The authors suggest that the mechanism may be mediated by inhibition of complement-mediated neuronal injury. There are no studies in stroke patients available yet, and it has to be kept in mind that IVIg increases in blood viscosity and stroke have been reported as a possible side effect of IVIg [Caress et al. 2003]. Nevertheless, stroke may become an area of future clinical trials.

Conclusion and future prospects

Many studies investigating the clinical effects of IVIg in a large number of different neurological disorders have been published in recent years. In the treatment of GBS, CIDP, and MMN IVIg are considered as first-line treatments. IVIg efficacy has been demonstrated in randomized controlled studies for a number of other diseases such as stiff-person syndrome, dermato-myosits, or myasthenic crisis. Due to the efficacy of alternative treatments, IVIg are used here rather as second-line treatment. There are several new indications being explored, all of which need to be tested by large randomized controlled trials before IVIg are used on a routine basis.

With the better understanding of the mechanism of action of IVIg [for recent reviews see Nimmerjahn and Ravetch, 2008; Clynes, 2007a,b; Nimmerjahn and Ravetch, 2007; Stangel and Pul, 2006; Misra et al. 2005] more specific treatments may also be developed. The recent finding that the anti-inflammatory activity (at least in a rheumatoid arthritis model) seems to depend on the sialylation of the N-linked glycan of the IgG Fc fragment [Kaneko et al. 2006] has led to the design of a recombinant sialylated IgG1 Fc fragment that represents the anti-inflammatory activity [Anthony et al. 2008]. Such a potent specific and recombinant compound could also solve the problems of a shortage of IVIg [Bayry et al. 2007]. However, it remains to be tested if such a sialylated Fc fragment is also effective in human disease and if its anti-inflammatory activity is comparable in the various diseases currently treated with IVIg. This highlights the poor understanding of the mechanism of action of IVIg, in particular because many proposed mechanisms are derived from experimental studies lacking its proof of principle in humans.

Another future prospect may be an alternative route of administration. Subcutaneous IgG (scIg) can be administered via a minipump by patients and may thus be more convenient and cost effective [Kittner et al. 2006]. First case reports in CIDP and MMN suggest that scIg is effective [Lee et al. 2008; Köller et al. 2006]. The results of randomized trials with scIg are expected to be published next year [Harbo et al. 2008].

Conflict of interest statement

M.S. has received honoraria for lectures from Baxter, Biotest, CSL Behring, Talecris.