High-dose intravenous immunoglobulin (hdIVIg) in the treatment of autoimmune blistering disorders

Intravenous immunoglobulin (IVIg) is a blood product prepared from the serum of between 1000 and 15 000 donors per batch. It is the treatment of choice for patients with antibody deficiencies where a ‘replacement dose’ of 200–400 mg/kg body weight is given approximately 3-weekly. In contrast, ‘high dose’ IVIg (hdIVIg), given most frequently at 2 g/kg/month, is used as an ‘immunomodulatory’ agent in an increasing number of immune and inflammatory disorders, despite the limited number of double-blind, randomized, placebo-controlled trials [1].

The blistering disorders are all autoantibody-mediated diseases in which many of the target antigens in the skin have been well defined. The two main circumstances in which hdIVIg has been used are when multiple conventional treatments have been ineffective or unacceptable side effects occur. The evidence for using hdIVIg in treatment-resistant autoimmune blistering disorders (pemphigus vulgaris, pemphigus foliaceus, bullous pemphigoid, pemphigoid gestationis, mucous membrane pemphigoid, epidermolysis bullosa acquisita and linear IgA disease) is based on small controlled trials, uncontrolled trials and case reports with a total of 158 reported patients.

IVIg may, for clarity, be considered to have four separate mechanistic components: (1) actions mediated by the antibody variable regions F(ab′)₂, (2) actions of Fc on a range of Fc receptors (FcR), (3) actions mediated by complement binding within the Fc fragment and (4) immunomodulatory substances other than antibody in the IVIg preparations. It is probable that these components act concurrently and different mechanisms may be important in different settings. A summary of the mechanisms of action is shown in Table 1[2]. Although the precise mechanisms of action in the blistering disorders are not understood it is probable that effects on the production, catabolism and effector functions of the autoantibodies involved play a major role and it has been established in numerous studies that hdIVIg reduces the titre of autoantibodies (Table 2).

Table 1.

Immunomodulatory category effects

Immunomodulatory category	Effects
Effects due to F(ab′)2	Anti-proliferative effects
	Modulation of apoptosis and cell cycle
	Activation of specific cells
	Effects on cell adhesion
	Antibodies to pathogens and superantigens
	Anti-idiotypes
	Antibodies to immunoregulatory molecules (cytokines, TCR, CD4, CD5)
	Effects on cytokine levels
Effects due to Fc receptors	Inhibition of phagocytosis
	Inhibition of ADCC
	Effects on antibody production and recycling
	Effects on glucocorticoid receptor binding affinity
Effects due to complement-Fcbinding	Inhibition of deposition ofactivated complement
Effects due to substances otherthan Ab within IVIg	IVIg contains cytokines, cytokinereceptors, CD4, MHC Class II,and stabilizing agents, mainlysugars

ADCC: antibody-dependent cellular cytotoxicity.

Table 2.

HdIVIg in the autoimmune blistering diseases

Disease	No. of patients	Demographics	Dose and frequency	Preparation	Additional treatment	Outcome	Response time	Duration	Reference
Pemphigus vulgaris	42	23–83 years 19M, 23F	2 g/kg/month in 37	Sandoglobulin in 10	Adjunctive in 39	Improved in 38/39†	Days to months	Weeks to 3 months with a minority having very long remissions	[6,4,5,7–13]
			0.3 g/kg/d in 4 0.25 g/kg/d in 1	Puimmun in 1 N/A in 31	Monotherapy in 3	No change in 2, worse in1	N/A
Pemphigus foliaceous	28	27–79 years 12M, 16F	1–2 g/kg/month 0.27 g/kg/monthin 1	N/A	Adjunctive in all	Improved	Rapid in some, months in others	Weeks to >5 months	[6,14–17]
Bullous and nodular pemphigoid	34	61–89 years 20M, 14F	2 g/kg/month in 32	Veinoglobulin in 11	Adjunctive in 22	19/22 improved*	Rapid in some and 2-4 monthsin others	2 weeks to 14 months	[4,6,10,18,19
			0.1 g/kg/d for 5 d in 1	Sandoglobulin in 4 and N/A in 23	Monotherapy in 12	8/12 improved
			0.3 g/kg/d for 5 d in 1
Mucous membrane pemphigoid	43	36–77 years 17M, 26F	1–3 g/kg over 3–5 d 2–4 weekly	SNBTS in 1 and N/A for 42	Adjunctive 28	Improved	2–6 months	1–3 months
					Monotherapy 15	Improved	4–6 months	Gradual increase in treatment interval	[20–25]
Epidermolysis bullosa acquisita	7	16–59 years 7M	0.4 g/kg/d for 5 d, 2–6 weekly (4)	Polyglobin in 1 Sandoglogulinin 3	Adjunctive 3	3 improved	Rapid – 4 months	10 d – 4 months	[10,27–30,32]
			2 g/kg/d 2- weekly	N/A for 2	Monotherapy 3	2/3 improved	1 week – many months		[31]
			40 mg/kg/d for 5 d 3–4 weekly × 4
			1.2 g/kg/month	N/A	UV protection 1	Improved	Gradual	Follow-up9 months
Linear IgA disease	3	45–67 years 2M, 1F	0.4 g/kg/d – 5 d	Sandoglobulin	Methylpred 10 mg/d	Improved	Days	4 weeks	33–35
			2 g/kg over 3 d repeated 2- weekly	N/A	Pred and antibiotics	Improved	Complete at 2 months	8 weeks 10–12 d initially
			4 g/kg/month	N/A	Monotherapy	Improved	2–4 months	longer in remission
Pemphigoid gestationis	1	17 years, F	0.4 g/kg/d for 5 d, two cycles	Sandoglobulin	Prednisolone 20 mg/d	Improved	Rapid	5 weeks	[36]

^†Three patients in the group reported by Harman et al. [10] experienced transient benefit from hdIVIg with subsequent courses being less effective.

^*There are two treatment failures reported by Godard et al.[18] in the adjunctive group, both of whom received doses of IVIg lower than 0·4 g/kg/day for 5 d and were also on low-dose prednisolone.

Pemphigus vulgaris (PV)

In pemphigus vulgaris circulating IgG autoantibodies have been shown to be pathogenic. The target antigen is desmoglein 3, a 130-kDa cadherin expressed on basal keratinocytes [3]. Forty-two patients with PV have been treated with high dose IVIg; overall 38 patients improved, three failed to respond and one progressed. There are no controlled studies, which makes interpretation extremely difficult; however, looking more closely at the reports [4–13], some conclusions may be drawn. All patients except one treated with 2 g/kg/month of IVIg responded and produced clinical benefits lasting weeks to months, often allowing a reduction in other therapies. The patient who failed to respond to 2 g/kg/month given adjunctively died subsequently of sepsis [9]. One case report described, using a slightly lower dose of 0·4 g/kg/day for 3 days per month, took 4 months to respond, but yielded a long-lasting effect. All responders used hdIVIg treatment as an adjunctive therapy. Of the four treatment failures three received hdIVIg alone and were deemed to have failed if no response was observed after 5 days; they were then commenced on conventional therapy of prednisolone and azathioprine resulting in a complete remission of disease in all patients [4]. The interpretation of these responses is difficult, because none received an adequate therapeutic trial and indeed it is unclear whether the prior hdIVIg enhanced the effect of prednisolone and azathioprine. Reductions in second-line therapies were achieved in the majority of the responders and decreases in autoantibody titre are reported in 31 patients. Monotherapy given to three patients was unsuccessful. Three responders had only transient improvement [10].

Pemphigus foliaceus (PF)

In pemphigus foliaceous the autoantibody target is desmoglein 1 on the keratinocyte surface [3]. Twenty-eight patients have been treated with adjunctive hdIVIg, all of whom improved [6,14–17]. Twenty-seven received 1–2 g/kg/month of hdIVIg and one 0·27 g/kg/month. In one controlled study eight patients with features of PV and PF were given hdIVIg as monotherapy, but had all received prolonged treatment with multiple immunosuppressive agents prior to this [14]. In most but not all patients autoantibody titres fell with successful treatment.

Bullous pemphigoid (BP)

Bullous pemphigoid (BP) is characterized by the linear deposition of IgG and C3 at the epidermal basement membrane, the targets being a 180-kDa BPAg2 and a 230-kDa BPAg1 within hemidesmosomes [3]. Thirty-four patients treated with hdIVIg have now been reported, two uncontrolled studies totalling 26 patients and eight case reports. A response to hdIVIg was noted in 27 patients (79%) [4,6,10,18,19]. Interpretation of the data is complicated by its heterogeneity. Of the seven non-responders, four had monotherapy, two received adjunctive therapy at doses of hdIVIg lower than 2 g/kg (0·1 g/kg/day and 0·3 g/kg/day for 5d) and two had nodular type pemphigoid. There was a dramatic response in some of these patients to conventional therapy following hdIVIg. In the 27 responding patients eight were treated with monotherapy and had responses lasting, on average, 2 weeks with one long-lasting response [18]. The remaining patients with adjunctive treatment had responses of 2–14 months’ duration and generally other therapies could be successfully reduced or withdrawn. The time to response was generally rapid but in some occurred over 2–4 months. Changes in autoantibody titres when reported did not correspond uniformly with clinical improvement.

Mucous membrane pemphigoid (MMP)

MMP is an uncommon autoimmune blistering disease of skin and mucosal surfaces in which blistering may be followed by scarring. Conjunctival scarring may lead to blindness. Six reports describe a total of 43 patients [20–25], 28 treated adjunctively and 15 with monotherapy, all of whom responded to hdIVIg. Twenty-six of the 28 patients treated adjunctively [20,21,23,25] had disease at multiple mucosal sites and received doses of hdIVIg ranging from 1 to 2 g/kg/month to 2–3 g/kg every fortnight. In most cases it was possible to reduce the concomitant doses of second-line agents and where documented autoantibody titres declined. Two further studies analyse the use of hdIVIg as monotherapy in severe MMP restricted to the oral cavity [22,24]. In the first, a retrospective study of eight patients treated with hdIVIg (1–2 g/kg/month) with 12 controls [22], disease remission and no progression to other sites was noted in the hdIVIg group compared with 58% progression in the 12 receiving conventional therapy. In the second study published in this issue [24] the authors describe a controlled study using hdIVIg as monotherapy in seven patients with oral MMP alongside seven conventional treatment controls (although the protocol permitted the use of intralesional steroid in both groups). One of the autoantibody targets in MMP is the α6 component of α6/β4 integrin within the hemidesmosome, which mediates binding to laminin anchoring the epidermis to the basement membrane. Blockade of α6 integrin with a monoclonal antibody has been shown to interfere with Langerhans cell migration from the epidermis [26], although the role of this in the pathogenesis of MMP is not understood. Titres of anti-α6 integrin antibody correlated with disease activity and clinical and serological remission was achieved in the patients treated with hdIVIg.

Epidermolysis bullosa acquisita (EBA)

EBA is a chronic bullous disease characterized by mechanically induced detachment of the epidermis from the dermis after minor trauma. Type VII collagen within the dermo-epidermal junction appears to be the target antigen [3]. There are seven case reports of the use of high dose IVIg to treat EBA [10,27–32] six of seven patients improved following hdIVIg. Three patients received adjunctive therapy and all improved and were able to reduce other second-line medication, while two of three given monotherapy improved and one further patient with UV-induced blistering was given UV protection (sunblock and beta-carotene) with hdIVIg improved. Response time varied from 1 week to many months and again autoantibody titres did not always reflect improvements in the disease. The duration of action of hdIVIg was up to 4 months and repeated doses would be required to maintain remission.

Linear IgA disease

There are two reports of adjunctive hdIVIg and one of monotherapy used in linear IgA disease [33–35], all improved. The response time was 12 days–2 months. It was possible to reduce second-line therapies in both patients treated adjunctively and duration of effect was 4–8 weeks. Autoantibody titres were reported in two patients and one declined with therapy.

Pemphigoid gestationis (PG)

PG is an autoimmune blistering disease specific to pregnancy, which usually presents in the second or third trimester. There is a single report of PG responding to adjunctive hdIVIg [36] allowing prompt disease control and steroid withdrawal. Remission was maintained on cyclosporin as hdIVIg was effective for only 5 weeks and autoantibody titres fell after the first course only.

DISCUSSION

The number of reported patients with autoimmune blistering diseases who have been treated with hdIVIg has almost tripled in the past 2 years to 158, with 92% of patients improving overall. Adjunctive therapy was slightly more successful than monotherapy with 97% and 76% improving, respectively, and treatment was well tolerated with few side-effects. The data in these largely uncontrolled and heterogeneous studies must be interpreted with caution in view of the likely reporting bias for favourable outcomes, differences in IVIg preparations, dosing schedules, use of concurrent therapy, severity of disease as well as previous exposure to immunosuppressive agents (ISAs). The controlled study by Sami and co-authors in patients with severe oral MMP strengthens the evidence for hdIVIg, as it was used as monotherapy or as close to this as is possible in this patient population accepting prolonged previous exposure to ISAs and concurrent intralesional steroids. The reduction in anti-α6 integrin autoantibodies and ability to reduce the IVIg requirement by increasing the interval between cycles once clinical remission had been attained is also important. It seems counterintuitive, however, that monotherapy will be the way forward in the highly selected treatment resistant group of patients who might be considered for hdIVIg, particularly as IVIg has been shown to synergise with steroids [37]; adjunctive therapy seems more successful overall, and response was more gradual in oral MMP treated with monotherapy compared with multiple mucosal site MMP treated adjunctively.

What does the future hold for hdIVIg in the blistering disorders? It does seem that with the increased numbers of reported patients with successful outcomes that a ‘critical mass’ has been reached to justify a double-blind placebo-controlled randomized multi-centre study to clearly define the role of IVIg, an increasingly expensive and scarce resource, due to the world plasma shortage, which needs to be used appropriately. The trial co-ordinated by a national hdIVIg study panel should perhaps be carried out first in therapy-resistant PV and the data suggest features important in the design of such a study. Clearly defined entry criteria end-points and outcome measures need to be established, including clinical disease severity scores and photographs, serological and quality of life measures. A dose of 2 g/kg/month of adjunctive hdIVIg should be used with sufficient follow-up to allow assessment of gradual dose reduction strategies. There are insufficient data to choose the other agents although steroids, in view of their synergy with IVIg, and mycophenolate because of its effects on B cells, are potential candidates. Pharmaco- economic considerations need to be assessed realistically in the light of the significant expenditure of health resources on the small cohort of patients requiring repeated admissions for disease flares, complications and side effects of conventional therapies, particularly as it may be possible to reduce hdIVIg.

In the longer term the information learned about pathogenesis of disease and mechanism of action of hdIVIg may allow the development of cheaper and more specific treatments.