Severe pemphigus vulgaris: successful combination therapy of plasmapheresis followed by intravenous high-dose immunoglobulin to

ARTICLE

Auteur(s) : Yumi Aoyama, Chikako Nagasawa, Miki Nagai, Yasuo Kitajima

Department of Dermatology, Gifu University School of Medicine, Yanagido 1-1, Gifu City, 501-1194, Japan

accepté le 27 Mai 2008

Pemphigus cannot be controlled in some cases, even with very high doses of systemic corticosteroids and immunosuppressive agents. Nowadays, adjuvant therapies are needed that will more reliably minimize the side effects of corticosteroids and various immunosuppressive agents (azathioprine, cyclophosphamide, mycophenolate mofetil, mizoribin and rituximab). Plasmapheresis is one of the most effective therapeutic methods to deplete sera of immunoglobulins (Ig), including pathogenic autoantibodies, in pemphigus and other autoimmune diseases [1]. However, it is not always easy to prevent feedback-induced rebound with rapid pathogenic antibody synthesis after plasmapheresis [2, 3]. Therapy with high intravenous immunoglobulins (IVIg) appears to provide a partial solution to this problem. High-dose IVIg works rapidly and selectively, lowering serum levels of autoantibodies [4]. We report here our experience with a patient who had severe pemphigus vulgaris (PV), and who was successfully treated with a combination of plasmapheresis, immediately followed by high-dose IVIg and systemic corticosteroids accompanied with immunosuppressive drugs, eventually leading to a complete suppression of the rebound increase in pathogenic PV-IgG, as monitored by weekly enzyme-linked immunosorbent assay (ELISA) for desmoglein (Dsg) 1 and Dsg3. The results of weekly Dsg-ELISA scores demonstrated a distinct difference in the alteration curves of serum levels of pathogenic IgG after plasmapheresis with and without high-dose IVIg. We employed double filtration plasmapheresis (DFPP), because this method requires only albumin solution as a substitution fluid, which minimizes the risk of infection and anaphylactic shock, in contrast to plasma exchange, which requires fresh-frozen plasma.

Report of a case

The patient was initially treated with oral prednisolone (60 mg/day, i.e., approximately 1.0 mg/kg/day), when the Dsg1 and Dsg3 ELISA titres were 180 and 345, respectively, and after one month, the patient experienced a remission with almost normal ranges (below approximately 20) for Dsg1 and Dsg3 ELISA titres (not shown. This is before day 0 in figure 1). However, when the corticosteroid dosage was progressively downscaled, an abrupt recurrence was seen, with increases in ELISA titres and extensive generation of new blisters. For treatment of this severe flare-up of the disease, the patient was treated by pulse therapy with intravenous (IV) methylprednisolone at 1,000 mg/day for 3 days. This therapy, however, appeared to be effective only for a few days after the last IV of methylprednisolone (not shown. This is before day 0 in figure 1). Therefore, the patient underwent 4 sessions of plasmapheresis (DFPP) for 14 days, followed by low-dose IVIg (3 g/day, 4 days). DFPP led to a temporary clinical improvement and reduction to normal ranges in both Dsg1 and Dsg3 ELISA titres (below 20), as shown in figure 1. Prior to DFPP, all counter-indications were excluded. Although short-term dizziness and nausea were seen, no severe side effects were observed.

In order to prevent feedback-induced rebound with rapid pathogenic antibody synthesis after plasmapheresis [2, 3], DFPP was combined with oral azathioprine (150 mg/day). However, the ELISA titres increased abruptly (198 for Dsg1 and 340 for Dsg3), accompanied by a severe recurrence of blistering, despite the addition of oral 50 mg/day prednisolone (figure 1). This increase in ELISA titres was much higher than that of total IgG in serum (figure 1). Because prompt remission was required, 8 sessions of DFPP were again performed in combination with two cycles of high-dose of IVIg (20 g/day) for 5 days with a one-month interval in order to prevent the rebound increase in pathogenic IgG via feedback suppression of IgG production. These IVIg treatments increased total serum IgG from 235 mg/dL to 2,858 mg/dL after the first IVIg and from 1,500 to 3,450 mg/dL after the second IVIg. This supply of IgG prevented the rebound increase in both Dsg1 and Dsg3 ELISA titres for around three months after DFPP. During and after this combination therapy, no evident side effects were observed.

However, at 2 months after the final high-dose IVIg (approximately 100 days after the last DFPP), the Dsg3-ELISA titre increased to 138 as the total serum IgG decreased from 3,450 mg/dL to normal ranges (approximately 1,100 mg/dL), associated with a prominent clinical recurrence evidenced by new blister formation. Therefore, the patient was treated with a 3^rd round of DFPP and high-dose IVIG, followed 2 months later by DFPP without a high-dose IVIg (figure 2). The disease has since been well controlled by low doses of oral corticosteroids (prednisolone, 10-15 mg/day) with mizoribin (100 mg/day), azathioprine (75-100 mg/day) or mycophenolate mofetil (1,000-1,500 mg/day) and the patient is doing well after 25 months of follow up.

Discussion

The pathogenicity of pemphigus IgG is now widely accepted, as experimental results have shown that passive transfer of IgG from patients with active pemphigus induces pemphigus in neonatal mice [12], and that adsorption of IgG by recombinant Dsg3 (rDsg3) eliminates IgG pathogenicity from active PV patients [13, 14]. More recently, it was shown, using rDsg3 produced by baculovirus, that conformational epitopes are important in the pathogenicity of pemphigus IgG [14]. Therefore, depletion or reduction of IgG, which include pathogenic anti-Dsg IgGs, from sera is considered to be the principal therapy for pemphigus.

In the patient described here, typical therapies for pemphigus, including oral corticosteroids, pulse therapy with methylprednisolone, followed by immunosuppressive agents, such as mizoribin (100 mg/day), azathioprine (75-100 mg/day) and mycophenlate mofetil (1,000-1,500 mg/day) were ineffective before we started a series of treatments as shown in figures 1 and 2. Furthermore, an alternative therapy, first DFPP followed by azathioprine (100 mg/day; Japanese are generally more susceptible to this drug than Caucasians) from day 10 to day 40 failed to prevent the feedback rebound synthesis of pathogenic PV-IgGs after removal of serum IgG, which led to an acute recurrence, although a remission was briefly obtained. As shown in figure 1, an abrupt increase in anti-Dsg antibody synthesis, particularly anti-Dsg3, which was much more extensive than that of total serum IgG synthesis, was induced during the 40 days (day 20 to day 60) until the second DFPP, as monitored by weekly ELISA examination. Therefore, we employed high-dose IVIg immediately after the second DFPP therapy performed from day 60 to day 81, in order to suppress de novo serum IgG synthesis. In contrast, the combination therapy of DFPP plus high-dose IVIg accompanied by systemic steroids and immunosuppressive drugs (mizoribin or mycophenolate mofetil), prevented this rebound increase in pathogenic PV-IgG, anti-Dsg1 and anti-Dsg3 antibodies, as shown in figure 2. The comparison between DFPP with and without high-dose IVIg (figure 1) clearly demonstrates that the intensive increase in total serum IgG caused by high-dose IVIg prevents the generation of anti-Dsg1 and anti-Dsg3 antibodies, thus suggesting feedback suppression. However, the principal therapies of corticosteroids and immunosuppressive agents are typically better with regard to long-term efficacy.

High-dose IVIg therapy is an effective alternative-treatment modality in patients with PV, who require systemic corticosteroids. High-dose IVIg therapy decreases serum levels of autoantibodies rapidly and selectively [15]. Possible mechanisms of action include 1) IVIg blocks the synthesis of pathogenic autoantibodies, 2) IVIg preparation contains blocking factors to inactivate the reactivity of autoantibodies, such as anti-id antibodies, 3) IVIg increases catabolism of autoantibodies [4]. 4) IVIg has a corticosteroid-sparing effect [16] and less adverse effects [17]. In contrast, plasmapheresis physically removes circulating antibodies, including anti-Dsg antibodies. Although IVIg therapy itself also works similarly to plasmapheresis in terms of the reduction of autoantibodies, the combination of DFPP and IVIg appears to have a major advantage that almost all the patient Igs, including pemphigus antibodies, are removed and Igs that do not contain pemphigus antibodies, are refilled, resulting in a selective depletion of the pemphigus antibody [18]. This is the first case report of PV in which high-dose IVIg associated with plasmapheresis (DFPP in the present case) protracted remission of PV blistering by preventing the feedback rebound of pathogenic PV-IgG synthesis. The present case also confirmed that weekly ELISA examination is valuable in evaluating the relationship between antibody clearance and clinical improvement in pemphigus patients with anti-Dsg1 and anti-Dsg3 antibodies. The clinical improvement trends roughly paralleled the reductions in Dsg1 and Dsg3 ELISA values.

Conclusion

Acknowledgements

References

2 Bystryn JC, Rudolph JL. Pemphigus. Lancet 2005; 366: 61-73.

3 Bystryn JC, Steinman NM. The adjuvant therapy of pemphigus. An update. Arch Dermatol 1996; 132: 203-12.

4 Bystryn JC, Rudolph JL. IVI g Treatment of pemphigus: how it works and how to use it. J Invest Dermatol 2005; 125: 1093-8.

5 Eyre RW, Stanley JR. Identification of pemphigus vulgaris antigen extracted from normal human epidermis and comparison with pemphigus foliaceus antigen. J Clin Invest 1988; 81: 807-12.

6 Hashimoto T, Ogawa MM, Konohana A, Nishikawa T. Detection of pemphigus vulgaris and pemphigus foliaceus antigens by immunoblot analysis using different antigen sources. J Invest Dermatol 1990; 94: 327-31.

7 Koulu L, Kusumi A, Steinberg MS, Klaus-Kovtun V, Stanley JR. Human autoantibodies against a desmosomal core protein in pemphigus foliaceus. J Exp Med 1984; 160: 1509-18.

8 Kowalczyk AP, Anderson JE, Borgwardt JE, Hashimoto T, Stanley JR, Green KJ. Pemphigus sera recognize conformationally sensitive epitopes in the amino-terminal region of desmoglein-1. J Invest Dermatol 1995; 105: 147-52.

9 Amagai M. Autoimmunity against desmosomal cadherins in pemphigus. J Dermatol Sci 1999; 20: 92-102.

10 Kitajima Y. Current and prospective understanding of clinical classification, pathomechanisms and therapy in pemphigus. Arch Dermatol Res 2003; 295(Suppl 1): S17-S23.

11 Kottke MD, Delva E, Kowalczyk AP. The desmosome: cell science lessons from human diseases. J Cell Sci 2006; 119: 797-806.

12 Anhalt GJ, Labib RS, Voorhees JJ, Beals TF, Diaz LA. Induction of pemphigus in neonatal mice by passive transfer of IgG from patients with the disease. N Engl J Med 1982; 306: 1189-96.

13 Amagai M, Karpati S, Prussick R, Klaus-Kovtun V, Stanley JR. Autoantibodies against the amino-terminal cadherin-like binding domain of pemphigus vulgaris antigen are pathogenic. J Clin Invest 1992; 90: 919-26.

14 Amagai M, Hashimoto T, Shimizu N, Nishikawa T. Absorption of pathogenic autoantibodies by the extracellular domain of pemphigus vulgaris antigen (Dsg3) produced by baculovirus. J Clin Invest 1994; 94: 59-67.

15 Bystryn JC, Jiao D, Natow S. Treatment of pemphigus with intravenous immunoglobulin. J Am Acad Dermatol 2002; 47: 358-63.

16 Sami N, Qureshi A, Ruocco E, Ahmed AR. Corticosteroid-sparing effect of intravenous immunoglobulin therapy in patients with pemphigus vulgaris. Arch Dermatol 2002; 138: 1158-62.

17 Gurcan HM, Ahmed AR. Frequency of adverse events associated with intravenous immunoglobulin therapy in patients with pemphigus or pemphigoid. Ann Pharmacother 2007; 41: 1604-10.

18 Tan-Lim R, Bystryn JC. Effect of plasmapheresis therapy on circulating levels of pemphigus antibodies. J Am Acad Dermatol 1990; 22: 35-40.