Dermatological Uses of High-Dose Intravenous Immunoglobulin FREE

There are few indications licensed by the Food and Drug Administration for high-dose intravenous immunoglobulin (hdIVIg) therapy, and these vary from product to product but include primary immunodeficiencies; secondary immunodeficiencies such as chronic lymphocytic leukemia, graft vs host disease, and post–bone marrow transplantation; idiopathic thrombocytopenic purpura; pediatric human immunodeficiency virus; and Kawasaki syndrome. Unlicensed use of hdIVIg for the treatment of autoimmune conditions of many types is increasing rapidly. Intravenous immunoglobulin is a blood product prepared from the pooled plasma of between 10000 and 20000 donors per batch by cold ethanol fractionation.¹ Several measures are used to ensure the safety of the product: (1) careful selection of donors, with importance placed particularly on voluntary, unpaid donations; (2) screening of every donation for hepatitis B surface antigen, anti–hepatitis C virus antibodies, anti–human immunodeficiency virus 1 and 2 antibodies, syphilis serology, and normal liver function; and (3) use of viral inactivation procedures in addition to the already high viral inactivation afforded by cold ethanol fractionation. These procedures vary among manufacturers and include the use of trace pepsin, low pH, solvent/detergent, and pasteurization. These antiviral steps have been validated by "spiking" preparations with known amounts of different viruses and screening the resulting treated product by polymerase chain reaction for these agents.² Concern has recently been raised regarding hepatitis G, but this is an enveloped flavivirus with homology to hepatitis C virus, and current procedures, it is hoped, will limit any possible risks. Nevertheless, it is possible that an as yet unidentified pathogen might be transmitted.³ Some patients were infected with hepatitis C virus in the 1980s before introduction of the antiviral steps, but human immunodeficiency virus has never been transmitted. The World Health Organization, also provides criteria for hdIVIg therapy, eg, preparations should contain at least 90% intact IgG with a normal IgG subclass distribution, as small an amount of IgA as possible, and be free from fragments and aggregates. Most manufacturers clearly surpass these criteria.²

Thus, the availability of a safe, aggregate-free preparation for IV use has vastly improved our ability to deliver adequate blood levels to antibody-deficient patients with fewer adverse reactions. Treatment of antibody-deficient patients is often achieved using IVIg doses of 0.2 g/kg every 2 weeks, but in the context of autoimmune and inflammatory diseases the dose used is much greater, usually 2 g/kg per month. High-dose IVIg may be divided into 2 doses of 1 g/kg per day or 5 doses of 0.4 g/kg per day and is believed to play an immunomodulatory role through a number of proposed mechanisms.

The immunomodulatory mechanisms of hdIVIg are mediated via the Fc portion of IgG (which interacts with Fc receptors and complement) or the antigen binding site and the variable regions of the antibody molecule F(ab′)₂. There are 5 main nonexclusive mechanisms that have been proposed⁴ : (1) functional blockade of Fc receptors on splenic macrophages, (2) inhibition of complement-mediated damage, (3) modulation of the production of cytokines and cytokine antagonists, (4) neutralization of circulating autoantibodies by anti-idiotypic antibodies in hdIVIg, and (5) neutralization of pathogens involved in the proposed etiology of the autoimmune disease. Other immunologically active substances present in hdIVIg, eg, class II human leukocyte antigens, soluble CD4, and interferon γ, have also been suggested as mechanistic candidates.⁵

Blockade of Fc receptors, especially on splenic macrophages, by hdIVIg has been shown to reduce the clearance of autoantibody-coated targets, eg, platelets, erythrocytes, and neutrophils. The most important example is the use of hdIVIg to treat idiopathic thrombocytopenic purpura, which results in an increase in the platelet count. This can also be achieved by giving anti-Fc receptor antibodies or by infusing Fc fragments.⁴ This saturation of splenic Fc receptors plays a critical role in the mechanism of hdIVIg in the treatment of peripheral autoimmune cytopenias.⁶ The effect should be of rapid onset and transient, although other mechanisms such as idiotype−anti-idiotype interactions between autoantibodies and hdIVIg may operate simultaneously.

The Fc region of IgG is able to bind C3b and C4b complement components and prevent the deposition of activated C3 fragments. This is believed to be due to interference in the formation of the membrane attack complex at the level of the C5 convertase C4b2a3b as IgG will reduce the available C3b. This has been demonstrated in vitro and in vivo in patients with dermatomyositis (DM).^{7 - 8} The onset of this effect is likely to be rapid and of a duration within the half-life of hdIVIg.

The effects of alterations in individual cytokines are hard to evaluate in vivo because they may have multiple effects, which may be different at different sites and be effected by cytokine antagonists. In vitro studies indicate modulation of cytokine production by hdIVIg in T cells, B cells, and monocytes and macrophages, showing down-regulation of interleukin (IL)–1, IL-2, IL-3, IL-4, IL-5, IL-10, tumor necrosis factor α, and granulocyte-macrophage colony-stimulating factor with variable effects on interferon γ and up-regulation of IL-1 receptor antagonist.⁹ Thus, an overall down-regulation of proliferative and proinflammatory cytokines such as tumor necrosis factor α and IL-1 is observed. Little is known about the ability of hdIVIg to influence the TH1/TH2 cytokine balance, and perhaps this information will become more readily accessible with techniques able to measure intracellular cytokines at the single-cell level. This may be achieved using anti-cytokine antibodies that enter permeabilized cells stained simultaneously for membrane markers by flow cytometry to allow identification of both the cell and the cytokine it is producing.¹⁰ Proliferation responses to a range of polyclonal mitogens (phytohemagglutinin, concanavalin A, and pokeweed mitogen) have been shown to be down-regulated by IVIg in vitro.¹¹ These effects were observed at supraphysiological doses, which may be achieved in vivo with hdIVIg therapy. High-dose IVIg also suppresses the production of IgM by Epstein-Barr virus–transformed B cells in a dose-dependent manner that can be prevented by pretreatment with anti-Fc receptor antibodies or Fc fragments, suggesting dependence on the Fc portion of IgG.¹²

High-dose IVIg contains anti-idiotypic antibodies that are able to interact with the binding sites of autoantibodies, which may result in their neutralization and modulation of autoantibody synthesis by binding to autoreactive B cells. The best characterized example is that of hdIVIg therapy for antibodies to factor VIII, in which within 36 hours after hdIVIg therapy, 95% of these antibodies had disappeared from the serum. This observation was confirmed to be dependent on variable region interactions using F(ab′)₂ preparations of immunoglobulin that were still able to neutralize the anti–factor VIII antibodies or to remove them when the F(ab′)₂ was bound to sepharose columns.^{6 ,13}

These findings have been extended to other autoantigens, including thyroglobulin, DNA, intrinsic factor, and antineutrophil cytoplasmic antigens. However, correlations between autoantibody titers and disease activity should be interpreted with caution because the target antigen may have multiple epitopes, some of which may be more important in pathogenesis than others, and in certain situations other mechanisms may be playing a role.⁶

High-dose IVIg variable region interactions are not restricted to those with autoantibodies but may involve interaction with other functionally important molecules such as CD4, human leukocyte antigens, and the T-cell receptor or with infectious agents and superantigens involved in disease pathogenesis.^{4 ,14} Several possible mechanisms may be involved, and some will operate within the half-life of the hdIVIg while others may have profound effects on the immunological repertoire, resulting in long-term remissions.

Adverse effects of hdIVIg therapy are generally mild and self-limiting, often occurring 30 to 60 minutes after onset of the infusion and including flushing, myalgia, headache, fever, chills, lower backache, nausea or vomiting, chest tightness, wheezing, changes in blood pressure, and tachycardia. These reactions are thought to be due to aggregated immunoglobulin, antibody-antigen complex formation with subsequent complement activation, or stabilizing carbohydrates used during manufacture. They are generally easily managed by slowing or stopping the infusion or by administering hydrocortisone or an antihistamine before the infusion.¹⁵

Rare episodes of anaphylaxis have occurred, particularly in IgA-deficient patients with anti-IgA antibodies. Deficiency of IgA occurs in approximately 1 of 700 of the population and should be screened for before hdIVIg therapy; if present, anti-IgA antibodies should be excluded, but even if detected, it is possible to use an IgA-depleted preparation of IVIg.^{15 - 16} Hematologic complications of hdIVIg therapy include Coombs'-positive hemolysis reported in 2 patients associated with autoantibodies to blood group antigens of the ABO and Rh systems.^{17 - 18} The risk of this rare complication may be minimized by using a 5-day infusion protocol for the first infusion (rather than 2 days) and measuring hemoglobin and haptoglobin levels early in the course of therapy. Falls in haptoglobin levels associated with mild reticulocytosis have been described in volunteers receiving hdIVIg, suggesting minor compensated hemolysis.¹⁶ Transient neutropenia also has been reported in some patients, occurring maximally on day 4 of a 5-day infusion regimen and returning to normal after about 5 days.^{19 - 20} The infusion of a high solute load such as hdIVIg poses a theoretical risk of increasing plasma viscosity and producing a prothrombotic tendency, but clinical data to support this are lacking.¹⁵ Acute renal failure also has been reported and is believed to be related to a high solute load–induced injury to the proximal tubule, but this is virtually always reversible.^{21 - 22} More rapid and severe renal injury was noted in a patient with underlying mixed cryoglobulinemia after a single hdIVIg infusion. The mechanism is likely to have been related to deposition of antigen-antibody complexes such as endogenous rheumatoid factor combining with the Fc region in hdIVIg.²³ Neurologic complications such as aseptic meningitis are occasionally seen 10 hours to 7 days after hdIVIg therapy, with complete recovery; the mechanism is unclear. Dermatological adverse effects are restricted to anecdotal case reports of eczema, alopecia, and erythema multiforme.^{24 - 26}

In the United States, a number of hdIVIg products are available, including Veno-I, Veno-S, Gammagard SD, Sandoglobulin, Cytogam, WinRho SD, Iveegam, Gammar-P, Gamimune N, and Polygam. These products are individually licensed and therefore have different licensed indications, and some are available at different concentrations of IgG. Prices vary between products but will also depend on local contracts with individual companies for each hospital. Considering the drug costs alone in treating a 70-kg man with DM at a dose of 2 g/kg per month, this would amount to 1680 g per year at an average cost of $25 per gram, amounting to a drug bill of $42000 per year. In addition, the inpatient costs of at the minimum a 2-day admission per month need to be added. This must be balanced against improvement of symptoms and quality of life, reduced costs and adverse effects of conventional therapy, fewer admissions due to disease flares, and less time lost from work. It is thus clear that the time and inconvenience of infusions and the overall cost necessitate careful patient assessment before embarking on a therapeutic trial of hdIVIg so that those most likely to benefit are given that opportunity.

Drug costs may be reduced by closely monitoring disease indexes and adjusting dose or dosage interval on an individual basis. Successful therapy resulting in a reduction in steroid dose often also leads to a reduction in weight and, therefore, in the overall dose of hdIVIg required. Inpatient costs may be reduced in 2 ways: by using day care facilities and by using an existing IVIg home therapy training program in the hospital. Patients are trained in product reconstitution, venipuncture, and the management of adverse reactions so that they are empowered to play a greater role in managing their own disease. This type of program has been successful in the management of primary antibody deficiencies with replacement doses and chronic inflammatory demyelinating polyneuropathy with hdIVIg therapy.

An essential aspect of cost containment before embarking on long-term hdIVIg therapy is a therapeutic trial to document objective benefit. It is vital to fully explain to the patient at this stage the concept of a therapeutic trial and the need for objective criteria of improvement so that if therapy needs to be discontinued it is done with the understanding and prior consent of the patient. Although there is some variation in price between products, there is no current experimental data to suggest that any particular product is superior to another in the treatment of dermatological conditions. It is well recognized, however, that individual tolerance to different products may vary.

The previously mentioned Food and Drug Administration–licensed indications for hdIVIg use account for only 40% of hdIVIg use; thus, 60% is used in a broad range of unlicensed applications, including Guillain-Barré syndrome, parvovirus infection, intractable childhood epilepsy, chronic inflammatory demyelinating polyneuropathy, systemic vasculitis, multiple sclerosis, inflammatory myopathies, asthma, and autoimmune cytopenias.

The following list includes all skin conditions in which hdIVIg has been used as treatment. In theory, patients with these diseases might benefit from hdIVIg therapy, but there is no adequate evidence to support clinical efficacy and use in patients with these conditions. With the exception of DM, use of hdIVIg should be restricted to controlled clinical trials. The reported dermatological uses of hdIVIg therapy are (1) DM,^{27 - 37} (2) pemphigus vulgaris and foliaceus,^{38 - 42} (3) bullous pemphigoid,^{38 - 39 ,43} (4) epidermolysis bullosa acquisita,^{44 - 46} (5) atopic dermatitis,^{47 - 49} (6) pyoderma gangrenosum,⁵⁰ (7) erythema multiforme,⁵¹ (8) pemphigoid gestationis,⁵² and (9) chronic urticaria (Malcolm Greaves, FRCP, oral communication, 1996).

In DM, there is increasing evidence that the membrane attack complex of complement is important in both muscle and skin inflammation.⁷ Autoantibodies also have been detected in certain clinical subsets of DM to histidyl transfer RNA synthetase, Jo-1.⁵³ Treatment is intended to control skeletal muscle and skin inflammation and to prevent necrosis; it usually involves (1) steroids alone or in combination with an assortment of other immunosuppressive agents, including azathioprine, methotrexate, cyclophosphamide, and cyclosporin, (2) lymphopheresis, and (3) total body irradiation. No agent is uniformly effective, and all have significant adverse effects.

Dermatomyositis represents the most extensively studied dermatological condition with regards to hdIVIg therapy. The literature now contains almost 100 patients in trials of hdIVIg (Table 1).^{27 - 37} This includes a mixture of case reports, uncontrolled trials, and a placebo-controlled crossover trial by Dalakas et al.³² This trial, although it included only 15 patients, provides the most compelling evidence for a benefit from hdIVIg therapy in patients with refractory DM, with assessment of muscle strength, ability to perform activities of daily living, skin inflammation, and multiple muscle biopsies. Of the 12 patients receiving hdIVIg, 11 improved compared with 3 of 11 patients taking placebo (dextrose in half-normal saline solution), and these findings were reinforced following crossover. One report of juvenile DM³⁶ has been omitted from the list; although results were generally promising, it is difficult to draw any conclusions because periods between infusions varied with clinical assessment and at no time was a trial of therapy given continuously for 4 months. Some authors did note fewer adverse effects (headache, nausea, and gastrointestinal tract disturbance) with the 5-day than with the 2-day regimen.

The trials listed in Table 1 must be interpreted in light of several points: (1) only 1 study was a randomized placebo-controlled trial; (2) some authors^{27 ,30 - 31} included polymyositis in their studies, which is reported to respond less favorably to hdIVIg therapy; (3) pretreatment regimens varied; (4) dose regimen and duration of hdIVIg therapy varied; (5) duration of disease before hdIVIg therapy varied; (6) 4 patients for whom treatment failed had polymyositis or DM secondary to malignancy; and (7) concurrent medications varied. Having accepted the shortcomings of the data, an overall response rate of 80% at about 2 months becoming maximal at 4 months is encouraging. However, the effects were long lasting in only a few patients, the majority requiring further hdIVIg therapy, which allowed reduction of concomitant steroid and other immunosuppressive therapies, and, importantly, response rates were clearly higher in those receiving hdIVIg as adjunctive treatment. Duration of effect of hdIVIg in the majority of patients was short-lived, and additional infusions were required within weeks to several months. There was no clear evidence of a difference in outcome between a 2-day or a 5-day regimen, which has been observed in the context of Kawasaki syndrome when a single dose of 2 g/kg is compared with 4 doses of 0.4 g/kg.⁵⁴ In the patients in whom treatment failed, half were not receiving other therapy and the majority of treatment failures occurred in uncontrolled trials where polymyositis was the predominant diagnosis. It also seems that treatment successes are more commonly described as case reports.

The nature of the hdIVIg preparation used was rarely described, and thus any questions as to possible differences between preparations remain unanswered, although the overall safety seemed excellent, with few adverse effects and only 1 patient worsening during therapy. In patients with juvenile DM, hdIVIg was also well tolerated. The place of hdIVIg in the treatment of DM seems to be as adjunctive therapy to establish a remission, allowing dosage reduction of other agents such as steroids or cyclosporin. Patients who have failed conventional therapy or are suffering unacceptable adverse effects may therefore be considered for a therapeutic trial of hdIVIg, although this is not yet a licensed indication. The trial should probably be of 4 months' duration to detect late responders with objective biochemical and clinical monitoring of response. If a response is observed, dosage regimens of conventional therapies may be cautiously reduced; on the other hand, if no improvement is observed after 4 months, hdIVIg therapy should be discontinued.

To date, 10 patients—9 with pemphigus vulgaris and 1 with pemphigus foliaceus—have been treated with hdIVIg: 7 patients improved, 2 were unchanged, and in 1 the disease progressed.^{38 - 42} None of these reports are controlled studies, which makes interpretation extremely difficult. A randomized controlled trial of hdIVIg plus conventional therapy vs conventional therapy alone is required in patients with pemphigus, and only then will it be possible to determine the value of adjunctive hdIVIg therapy in patients with this disorder.

A beneficial response to hdIVIg therapy was observed in 10 of 15 patients with bullous pemphigoid from 1 uncontrolled study of 11 patients and 4 case reports.^{38 - 39 ,43}

There are 3 case reports describing the use of hdIVIg in patients with epidermolysis bullosa acquisita^{44 - 46} ; 2 reported benefit, although again no controlled trials exist.

Three patients with atopic dermatitis have been treated in a small, open study, and in addition there are several anecdotal reports of hdIVIg therapy that suggest possible benefit. Other single case reports exist describing the use of hdIVIg in patients with pyoderma gangrenosum,⁵⁰ erythema multiforme,⁵¹ and pemphigoid gestationis,⁵² but no firm conclusions can be drawn from these data.

The single, adequately controlled trial of hdIVIg in dermatological disease clearly indicates that hdIVIg is an effective adjunctive therapy in patients with DM. This trial provides sufficient justification for hdIVIg therapy in patients with DM who are responding poorly to conventional therapy or developing adverse effects. This trial has not established the duration of effect or the required dose and frequency of maintenance hdIVIg therapy.

Except in DM, published evidence for the efficacy of hdIVIg in the treatment of dermatological conditions is at present restricted largely to case reports. This makes it difficult to unravel the possible roles of hdIVIg in cutaneous disease; nonetheless, it is possible to draw some conclusions and offer some guidelines. First, the safety profile seems to be excellent, with few adverse effects and only very rare disease exacerbations during hdIVIg therapy. Adverse effects can be further minimized by taking the precautions outlined in the checklist. The effective dose of hdIVIg is between 1 and 2 g/kg per month, with no clear difference in efficacy between the 2-day and the 5-day regimen. A long-lasting remission with a single course of therapy was only reported in a minority of patients (3 with DM and 1 with bullous pemphigoid), and, therefore, the required dose and frequency of further hdIVIg after the establishment of a remission has yet to be determined. The rate of response in these reports ranged from a few days to 6 months, perhaps reflecting different mechanisms of action of hdIVIg; it seems that DM responds more slowly to hdIVIg therapy than pemphigus, although there were exceptions in all groups. Owing to lack of numbers and reporting failure, it was not possible to determine any differences between preparations of hdIVIg. Taking all the reported cases of hdIVIg use in dermatological disease together, efficacy was much greater when hdIVIg was used as adjunctive therapy, with a response rate of 88% compared with 46% when used as a single agent. This finding is still more pronounced when DM is examined in isolation. This observation must be interpreted with caution as the data include numerous case reports and uncontrolled trials with a reporting bias for therapeutic successes. However, this suggests that hdIVIg in addition to existing therapy, rather than as the sole therapy, may be more likely to succeed even if the patient's disease is not controlled with existing therapy. Thus, patients who have responded poorly to or are suffering unacceptable adverse effects from conventional therapy may be the best candidates for hdIVIg therapy.

The proposed pathogenesis of autoimmune bullous disease makes hdIVIg therapy an atttractive option. Unfortunately, there are few adequately controlled therapeutic trials in patients with autoimmune blistering diseases, and a controlled trial of hdIVIg therapy is certainly required in patients with pemphigus, who often respond poorly to conventional therapy and may develop significant adverse effects. However, until a controlled trial has demonstrated a significant benefit in patients with pemphigus, adjunctive hdIVIg cannot be considered established therapy. Pemphigoid, on the other hand, is usually well controlled with immunosuppressive agents and often spontaneously remits after several years, thus reducing the risks of adverse effects. There is also an attractive rationale for using hdIVIg therapy in patients with epidermolysis bullosa acquisita, but again controlled trials are required, which would also establish whether responders are restricted to the inflammatory autoimmune or the mechanobullous forms of epidermolysis bullosa acquisita. However, given the rarity of such cases, an adequate study may not be easily achieved. In atopic dermatitis, a controlled trial in patients with severe unresponsive disease is essential before hdIVIg can be used on a widespread basis for treatment of this disease. If it were demonstrated that hdIVIg therapy down-regulated TH2-type cytokines (eg, IL-4 and IL-5) in patients with atopic dermatitis, this would be of considerable interest in the fields of asthma, allergic rhinitis, and other diseases believed to be mediated via a TH2-type cytokine mechanism.

High-dose IVIg is an effective adjunctive therapy for patients with DM who fail to respond to conventional therapy or who experience unacceptable adverse effects. A therapeutic trial of hdIVIg should be considered only in rare selected patients with pemphigus and other autoimmune bullous diseases who have responded poorly to conventional agents or are troubled by serious adverse effects. It should be introduced as an adjunctive therapy at a dose of 2 g/kg per month using either a 2-day or a 5-day regimen with objective clinical and laboratory measures of efficacy before and during a 4-month period. If during this time a response is obvious, it may be possible to taper preexisting therapy before assessing the frequency and dose requirement of further hdIVIg therapy. Patients who respond to and require ongoing hdIVIg therapy may be assessed for training for home therapy, which has improved quality of life for patients and has significant implications for reducing inpatient costs.

High-dose IVIg, therefore, offers a potential therapeutic avenue for a number of dermatological conditions, but it should be further assessed using double-blind placebo-controlled trials.

Accepted for publication May 20, 1997.

We are indebted to Vanessa Lipton and Marva Brown for their secretarial assistance and to David Webster, Carrock Sewell, and Sarah Deacock for their expert comments during the preparation of the manuscript.

Reprints: Stephen Jolles, MSc, MRCP, Department of Immunology, Royal Free Hospital, Pond Street, London, England NW3 2QG.