You are on page 1of 10



From American Journal of Clinical Dermatology

Pemphigus - A Treatment Update

Sue Jessop; Nonhlanhla P. Khumalo Authors and Disclosures Posted: 05/06/2009; Am J Clin Dermatol. 2009;9(3):147-154. 2009 Adis Data Information BV

Pemphigus is an uncommon but potentially life-threatening chronic autoimmune bullous disorder. Antibodies are directed against antigens (desmoglein 1 and 3) in the desmosomes linking keratinocytes and against acetylcholine receptors. Conventional treatment with high-dose corticosteroids, sometimes with adjuvant immunosuppressive agents, may be associated with very serious adverse effects. There is an urgent need to establish the evidence for the safest and most effective form of treatment. A literature review has revealed 11 controlled (9 randomized) trials of treatment for pemphigus. The numbers of participants in the individual trials are small and the data cannot be pooled as they evaluate different forms of treatment. The results of these trials suggest that very high doses of corticosteroids, either as pulse therapy or in daily dosage, are not superior to moderate daily doses. Based on evidence from the available trials, addition of an immunosuppressive agent generally does not appear to offer substantial benefit in terms of clinical response. However, a recent study demonstrated a significant reduction in corticosteroid requirements among patients receiving immunosuppressive agents. Newer therapies, such as biologic agents (in particular rituximab), calcineurin inhibitors, or immunoadsorption appear promising but there are inadequate controlled trials to establish their role clearly. Initial openlabel studies suggest that specific peptide immunotherapy may offer a safe and novel approach to the treatment of pemphigus in the future. At present, treatment of an individual patient with pemphigus requires clinical judgment and should not be based purely on guidelines or on the inadequate available evidence alone. There is an urgent need for large randomized, controlled, multicenter trials of treatment in patients with pemphigus.

Pemphigus is an uncommon, autoimmune blistering disorder characterized by the presence of flaccid blisters of the skin and/or mucous membranes. It may occur at any age (peak 5060 years).[1] In the newborn it is thought to result from passive placental transfer of pathogenic antibodies from the maternal circulation. Neonatal pemphigus is usually self-limiting, resolving with the disappearance of maternal antibodies.[2,3] The recognized forms of this disease are pemphigus vulgaris (PV), pemphigus vegetans, pemphigus foliaceus (PF), IgA pemphigus, and paraneoplastic pemphigus. This article reviews

only the literature relating to the two most common types (PV and PF including endemic forms).
Epidemiology and Etiology

Accurate prevalence figures are limited but it has long been recognized that PV has a relatively high prevalence in people from the Mediterranean area and among those of Jewish ancestry.[1] The two common forms, PV and PF, show a variation in relative prevalence from one geographic area to another. PF is known to have an extraordinarily high prevalence in parts of Tunisia and Brazil (the endemic, so-called 'fogo selvagem,' form of the disorder).[4] A genetic predisposition (HLA-DRB1*0102) has been identified.
Clinical Features

PV is characterized by the presence of flaccid blisters that rupture easily, leading to erosions and crusts. Without treatment the blisters tend to spread, showing little tendency to heal spontaneously. Skin and mucosal surfaces are often both involved, although pemphigus may initially remain at one site for months to years.[5] PF is a more superficial condition in which the cleft occurs in the outer layers of the epidermis and frequently leads to erosive or crusted lesions rather than blisters.[6]
Impact of Disease and Natural History

Lesions in pemphigus can be anything from uncomfortable to extremely painful, particularly when mucosae are involved. Secondary candidal infection of the mucous membranes is common and contributes to discomfort. Secondary bacterial infection is a major hazard in patients with impaired barrier function, and the risk is enhanced by the use of high-dose corticosteroids and immunosuppressive drugs. Many of the symptoms experienced by patients with pemphigus are secondary to the therapy. The psychosocial impact of pemphigus can be extreme, with painful mucosae and frequently visible skin lesions and changes in body image resulting from corticosteroid effects.[7] Reported mortality rates for pemphigus vary but recent figures indicate an average of around 6%, with many of the deaths being due to sepsis.[8,9] Severe disease and death appear to be more prevalent during the first year of the disease.[9,10] Retrospective analysis of mortality in historical cohorts suggests that the prognosis has improved since the introduction of corticosteroids, even though these agents may enhance the risk of infection.[11] In a review of the results of 77 published studies, Carson et al.[12] highlighted the changing mortality pattern and explored the possible relationship between mortality and corticosteroid therapy. However, improved survival may have complex underlying causes, including improved infection control and intensive medical care. Herbst and Bystryn[9] reviewed the pattern of remission in 40 patients with PV. They identified four patterns of illness in their cohort: rapid responders, slow responders, patients with persistent intermittent disease, and patients with refractory disease.


Loss of adhesion of keratinocytes is the fundamental abnormality in both PV and PF. The pathogenic role of pemphigus antibodies has been well established both in humans with the disorder and in animal models.[13,14] Antibodies (generally IgG) are usually directed against transmembrane glycoproteins of the keratinocyte membrane, desmoglein (Dsg) 1 or 3, or against acetylcholine receptors.[15] These components of the desmosomes are critical to cell-to-cell adhesion in the epidermis. The antibodies are thought to produce acantholysis by inducing apoptotic and oncotic pathways, as demonstrated in in vivo and in vitro studies.[16] The cleft forms in the superficial levels of the epidermis in PF and suprabasally in PV, with resultant separation of epidermal cells. Clinical reports of the successful use of nicotine and nicotinamide have suggested a possible role for the acetylcholine axis in epidermal cell adhesion.[17] This finding is supported by work in experimental animals, where carbachol inhibited acantholysis.

A diagnosis of PV or PF should be considered in patients presenting with non-healing superficial blisters, erosions, or crusts. Light microscopy of involved skin (preferably from a biopsy of a small intact blister) characteristically shows a suprabasal cleft, with acantholysis. Immunofluorescent studies of adjacent skin identify the target antigens in the desmosomes, displaying pericellular epidermal IgG fluorescence. Indirect immunofluorescence studies of serum can detect circulating antibodies against Dsg-1 and/or Dsg-3 and confirm the diagnosis.

Activity and Severity Scores

Several clinical scoring systems have been used in the assessment and monitoring of oral and cutaneous pemphigus. Previous scoring systems used in autoimmune blistering disorders have been reviewed by Pfutze et al.,[20] who have also proposed a new combined score, the Autoimmune Bullous Skin Disorder Intensity Score (ABSIS).
Treatment Aims of Review

This article examines the evidence for the best treatment of pemphigus. As treatment for pemphigus is frequently associated with potentially severe adverse effects, the safety profile of all treatments is considered an important outcome. Oral or parenteral corticosteroids have long been considered first-line treatment for pemphigus and are thought to have had a favorable impact on prognosis, given the decline in mortality rate since the introduction of corticosteroid therapy.[11,12,21] The clinical experience that corticosteroids are effective treatment for pemphigus renders any placebo-controlled study unethical for this potentially life-threatening disorder. This view is supported by the group who drew up the

British guidelines for the treatment of PV.[22] Thus, placebo-controlled evidence for the effectiveness of corticosteroids is not available and is not likely to become available. However, in some patients, corticosteroid treatment does not adequately control the disease and in all patients taking prolonged and often high doses there is concern about adverse effects. There is thus a need to identify the safest and most effective adjunctive forms of therapy. This article examines the evidence with the aim of answering the following questions:
1. What is the safest and most effective dose, method of delivery, and weaning schedule of corticosteroids in the treatment of PV and PF? 2. Does addition of immunosuppressive agents or other modalities of treatment improve remission rate or time or reduce adverse events related to corticosteroid therapy? 3. What is the safest and most effective adjunctive therapy? Search Methods

The following search terms were used to search the MEDLINE database: 'pemphigus, treatment'; 'pemphigus vulgaris, treatment'; 'pemphigus foliaceus, treatment'; 'RCT'; 'trial'; 'controlled trial'; 'humans, pemphigus', 'TNF blockers'; 'pemphigus, new agents'; 'pemphigus, biological agents'; 'pemphigus, evidence'; 'pemphigus, severity'; 'pemphigus, score'; 'pemphigus, activity'; 'pemphigus, outcomes'. Abstracts and full-text articles were read to identify controlled trials and adverse events. Reference lists were scrutinized for further relevant studies.

Systemic Agents. One systematic review of pemphigus treatment has been published.[7] All reported trials involved small numbers of participants, probably because of the relative rarity of pemphigus. These low numbers may have obscured small differences between treatment groups in some instances. Participants with both PV and PF were included in most trials, again because of the rarity of the two disorders. Nine controlled trials, including seven randomized controlled trials (RCTs), have evaluated treatment of PV or PF[23-31] (Table I). These trials, which included a total of 344 participants, recorded responses to 19 interventions, dosages, or routes of delivery. The heterogeneity of these studies precludes pooling of data. Outcomes reported were generally similar, including time to response, time to remission off treatment, time to relapse, total dose of oral corticosteroid required, and adverse events. Some of the reports did not describe the method of randomization or allocation concealment. Many individual cases, retrospective studies, and case series have also been reported.

The optimal dose, route, and weaning schedule for corticosteroids have been explored in three small trials involving 62 participants.[23-25] In one of these trials, Ratnam et al.[23] randomized 22 patients with PV (19 patients) or PF (3 patients) to treatment with either low-dose oral prednisolone (4560 mg/day) or high-dose oral prednisolone (120150 mg/day). Following remission, prednisolone was weaned and all participants received adjuvant therapy (also randomized). The time needed to induce remission did not differ significantly between the two groups (24 days in the low-dose group vs 19.6 days in the high-dose group). Similarly, there was no significant difference in the relapse rate, time to relapse, or adverse effects of therapy between the two groups. Thus, high-dose corticosteroid therapy did not appear to offer any benefit over low-dose therapy in pemphigus. However, the small number of participants may have disguised any small differences between the groups. In an open-label, controlled trial, Femiano et al.[24] treated 20 patients with oral PV with oral prednisone (starting dose 125 mg/day) or prednisone (50 mg/day) alternating with 4-day pulses of intravenous betamethasone (20 mg/day). Individual figures were not given in the report but the mean time to resolution of mouth lesions was 30 days in the prednisone group and 25 days in the pulse group (p = 0.0001). Adverse events included hyperglycemia, hypertension, weight increase, and mood change and were present in both groups but more marked in those receiving oral treatment only. Corticosteroid pulse therapy was the subject of another multicenter trial reported by Mentink et al.[25] Participants all took oral prednisolone (80 mg/day initial dose) and azathioprine and were then randomized to receive a monthly pulse of oral dexamethasone (300 mg) or placebo. Time to remission, mean cumulative prednisolone dose, and duration of remission were the same in the two groups. Adverse events were more common among participants receiving the dexamethasone pulse. Five RCTs have examined the benefit of additional immunosuppressive agents in PV or PF.[26-30] In 1994, Chryssomallis[26] investigated the addition of immunosuppressants to oral prednisone in the treatment of oral pemphigus. Twenty-eight participants were randomly assigned to three groups: prednisone alone, prednisone plus cyclophosphamide, and prednisone plus cyclosporine (ciclosporin). Times to remission and relapse were equal in the three groups and adverse events were more frequent in patients taking the additional agents. Rose et al.[27] reported a multicenter study comparing outcomes in a group of patients given oral methylprednisolone 2 mg/kg plus azathioprine with a group given pulse intravenous dexamethasone 100 mg and cyclophosphamide 500 mg intravenously every 24 weeks, followed by oral cyclophosphamide daily. A higher number of patients receiving oral therapy achieved remission within 2 years (9/11 vs 5/11), and the oral therapy group experienced fewer adverse events; however, these differences were not significant, possibly because of the small numbers of participants. The investigators concluded that their study did not support the concept that pulse therapy with corticosteroids and cyclophosphamide is superior to oral prednisolone therapy in PV. In the study by Beissert et al.,[28] participants received methylprednisolone 2 mg/kg/day plus either azathioprine (18 patients) or mycophenolate mofetil (21 patients). No benefit was demonstrated for one agent over the other. Complete resolution of the lesions was observed in 13/18 patients taking azathioprine after a mean of 74 days, compared with 20/21 patients after a mean of 91 days in the mycophenolate mofetil group. These differences were not significant. Total cumulative corticosteroid dose and the incidence of adverse events were also similar in the two groups. Ioannides[29] compared oral methylprednisolone (initial dose 1 mg/kg) with oral methylprednisolone plus cyclosporine (5 mg/kg) in 33 patients. Time to remission, cumulative corticosteroid dose, and relapse rate were similar in the two groups. Adverse events

(hypertension and reduced renal function) were more common in the group treated with cyclosporine. A large, non-blinded RCT (120 patients) investigated the benefit of adjuvant immunosuppression in pemphigus.[30] The trial compared four treatments, namely prednisolone alone, prednisolone plus azathioprine, prednisolone plus mycophenolate mofetil, and prednisolone plus intravenous cyclophosphamide. The percentages of patients showing complete response at 1 year were similar in the four groups (7080%), but the mean total corticosteroid dose required (a surrogate measure of response) was significantly greater in patients taking prednisolone alone. Adverse events were similar in the four patient groups. A large retrospective study that compared the efficacy and safety of cyclophosphamide, azathioprine, and cyclosporine suggested that cyclophosphamide was superior with respect to time to remission.[34] However, these findings have not been confirmed in a prospective randomized trial. Plasma exchange was the subject of an RCT performed by Guillaume[31] in 1988. A group of 34 patients with PV or PF were randomized to receive either prednisone (starting dose 0.5 mg/kg) or prednisone plus plasma exchange (three per week for 2 weeks, then two per week for 2 weeks). The cumulative dose of corticosteroids and time to disease control were found to be the same in the two groups but adverse events, in particular infection, were more frequent in the patients treated with plasma exchange (four participants in this group died from infections). The results of a refinement of plasma exchange involving immunoadsorption of pemphigus antibodies have been reported in four small uncontrolled series.[35-38] A total of 29 patients were treated, with good to excellent clinical response. One patient developed anaphylaxis, one a deep vein thrombosis, and one hypotension. Shimanovich et al.[38] recommend that plasma exchange with immunoadsorption be used in combination with immunosuppressive therapy to induce long-term remission. A large RCT is required to confirm the results of these small uncontrolled studies. Immunoadsorption removes specific immunoglobulins during plasma exchange. Being a more selective approach, this method may offer a safer and more effective form of plasmapheresis in autoimmune disorders. In summary, several small clinical trials have not demonstrated a consistent benefit for high-dose or pulse corticosteroid treatment or immunosuppressive agents over moderate doses of oral corticosteroids in the treatment of PV and PF. There is some evidence that addition of immunosuppressive agents may reduce the total dose of corticosteroids required to induce remission. However, results from some of these trials have also suggested that adjuvant therapy may increase adverse events and, at least in the case of plasma exchange, may be associated with increased deaths. Topical Agents. There have been two reported trials of topical agents (one using pilocarpine[32] and the other epidermal growth factor[33]) [Table I]. Both trials reported benefit from the test agent in terms of more rapid healing of erosions, suggesting that these topical agents may be useful when used as an adjunct to systemic therapy in pemphigus. No adverse effects were described with either of these agents. There have also been case reports but no controlled trials reporting the use of other topical treatments, such as calcineurin inhibitors.[39]

It also seems wise to promote good oral hygiene in patients with chronic oral ulcers, recommend use of a soft toothbrush to reduce trauma, and identify and treat complicating infections such as mucosal candidiasis.[22] Therapies of Possible Benefit. There are no controlled trials of intravenous immunoglobulins in the treatment of pemphigus. A review of the literature up to 1999 identified various case reports and small clinical series describing use of intravenous immunoglobulin in 21 patients with PV.[21] These investigators concluded that a multicenter RCT was required to evaluate the effectiveness of this therapy. In 2003, Ahmed and Dahl,[40] on behalf of the Consensus Development Group, reviewed reports on use of immunoglobulins in pemphigus and outlined guidelines on the safe use of intravenous immunoglobulin therapy for the treatment of autoimmune mucocutaneous blistering diseases. Biologic agents have been reported to be effective for pemphigus in several case reports but have not been subject to rigorous investigation in an RCT. Successful outcomes with etanercept[41] and infliximab[42] have been reported in single cases or small case series. Anti-CD20 monoclonal antibody (rituximab), in single or repeated cycles, has also been used in a larger number of patients.[43-54] Results reported from these case series and case reports suggest a very favorable response, with sustained remission in some patients. In the largest series, Joly et al.[53] observed remissions in 18 of 21 patients with refractory disease after treatment with a single cycle (four once-weekly infusions) of rituximab. However, two patients developed severe infections and the investigators suggested that repeat cycles of rituximab should be used with caution. Further RCTs are urgently required to establish the effectiveness and safety profile of these agents in the treatment of pemphigus. Other published case reports or small series have suggested possible benefits from topical tacrolimus,[39] dapsone,[55] and stem-cell transplantation.[56] Use of a proteomics-defined peptide in a topical formulation in a patient with recalcitrant PV appeared beneficial,[57] and an openlabel study of specific peptide immunotherapy, with intravenous infusions of peptide homologous to a portion of the Dsg-3 molecule, suggested that this approach is well tolerated and may be helpful.[58] Adverse Events. The adverse events associated with corticosteroids reported in controlled trials of pemphigus treatment are similar to those described elsewhere (Table I). Weight gain with Cushingoid features, hypertension, disturbances of glucose and calcium metabolism, myopathy, mood disorder, and sepsis are frequent adverse events with use of corticosteroids that were all reported in these studies. While one trial suggested that pulse corticosteroids might reduce total corticosteroid dose,[27] other investigators have not confirmed this finding in patients with pemphigus.[25] Addition of adjuvant immunosuppressive agents does not appear to reduce the total dose of corticosteroid and may actually increase the risk of toxicity.[26] In all trials, the dominant adverse events appeared to be related to high-dose corticosteroid therapy and patients' subjective views of these adverse effects have recently been reported.[59] One of the arguments for use of immunosuppressive agents is the wish to reduce the cumulative dose of corticosteroids. However, the trials reviewed in this article did not generally demonstrate any impact of adjuvant therapy on cumulative corticosteroid dose.

The risks of plasma exchange, such as hypotension, fever, and increased sepsis (sometimes with fatal outcome) have been highlighted by Guillaume.[31] However, use of more selective immunoadsorption techniques may minimize these risks.[35-38] Adverse events associated with immunoglobulin infusion include headache, backache, flushing, fever, nausea and vomiting, anaphylaxis, aseptic meningitis, renal failure, and hypertension. Some of these effects appear to be related to the speed of infusion and others may be avoided if patients are selected carefully.[40] The risk of major infection appears to be increased among patients taking biologic agents.[53,54]

Pemphigus is a serious chronic illness associated with considerable risks of morbidity and mortality. Current clinical wisdom recommends the use of potentially toxic agents for the treatment of this condition, while recognizing that the treatment itself may pose serious threats to the patient. Several retrospective studies have suggested that modern treatment (at least since the start of the corticosteroid era) has improved prognosis in patients with pemphigus.[9,11,12] It has generally been assumed that corticosteroid therapy is responsible for improved survival; however, there may be other contributing factors, such as improved identification of patients and control of serious infection. UK guidelines strongly recommend the use of corticosteroids, based on mortality data, but observe that selection of dosage and route of administration is largely empiric.[22] These guidelines 'reflect the best data available' and warn that current recommendations may well need revision when further data become available.[22] This literature review has revealed a very limited body of evidence for identifying the best treatment for PV and PF; only 11 controlled clinical trials were identified,[23-34] of which 9 were randomized.[25-34] These studies do not demonstrate a clear benefit for higher doses or pulses of corticosteroid therapy. Among the 11 controlled trials reviewed in this study, 6 employed immunosuppressive agents.[2530] Apart from the corticosteroid-sparing effect demonstrated by Chams-Davatchi et al.,[30] these trials did not demonstrate significant benefit from the addition of any immunosuppressive agent. [25-29] The well established immunologic basis for pemphigus makes the use of immunosuppressive agents very attractive as a treatment option and selective use of these agents may still be of great value in individual patients. However, the evidence at this time does not support this practice as a routine approach in the management of PV or PF. Although the findings of the plasma exchange trial by Guillaume[31] provide reason for caution about the use of plasma exchange in pemphigus, more selective immunoadsorption methods may offer benefit.[35-38] Unfortunately, there are no RCTs to provide evidence of a benefit for plasma exchange at this time.

Similarly, there is no RCT evidence to support the use of immunoglobulin infusion or newer therapies such as calcineurin inhibitors or biologic agents. However, these modalities (particularly rituximab) may indeed offer significant benefit, as suggested by case reports and case series,[41-54] and more rigorous evidence of their effectiveness is urgently needed. It has long been recognized that high-dose corticosteroid and immunosuppressive treatment carries a significant risk of serious, even life-threatening, adverse events and patients using these agents require careful monitoring. A great source of concern is the high level of adverse events with all established forms of treatment recorded in most of the published studies.[8,22,23,40] Longterm effects on bone were not specifically mentioned in the studies, but an increased risk of fracture and osteonecrosis are definite hazards of corticosteroid use. It is noteworthy that most of the adverse events were those related to corticosteroid therapy, and that these effects were not reduced by the use of immunosuppressive, 'corticosteroid-sparing' agents. Use of the latter may actually increase the risks of certain adverse events, such as infection, without any evidence of benefit, and practitioners might be wise to exercise caution when using immunosuppressive agents rather than just assuming that their use will reduce corticosteroid toxicity. At the same time, serious concerns about according undue weight to the reviewed evidence for treatment in PV and PF arise because most of the studies have marked limitations. The greatest weakness in most published RCTs involving patients with pemphigus is the small size of the study groups. As pemphigus is an uncommon disorder, recruitment of large numbers of participants in clinical trials is likely to be an ongoing problem. Large international, multicenter trials would seem to offer the best prospect for obtaining highly powered evidence. A further limitation is the lack of consistent and comparable outcome measures, although many investigators have included time to remission, time to relapse, cumulative dose of corticosteroid required, adverse events, and death in their trials. In summary, current evidence does not lend powerful support to the common practice of using high initial doses of corticosteroids (orally or by intravenous pulse), followed by the addition of (or substitution with) immunosuppressive agents or newer agents. However, while the evidence may not support this approach, it can be argued that the level of evidence is not adequate to guide treatment of the individual patient with PV or PF. There is, therefore, a pressing need for evidence to guide the selection of treatment modalities, drug dosages, routes of administration, and weaning programs.
Conclusion and Recommendations

At present, based on the available evidence, the following recommendations can be made regarding treatment of patients with PV and PF:
1. Selection of treatment needs to be made on an individual basis and current guidelines should not be rigidly applied. 2. All patients using any of the agents currently recommended for PV and PF require careful monitoring and health promotion, including advice about exercise, diet, and avoidance of smoking, to prevent and/or identify drugrelated adverse events.

3. There is an urgent need for large multicenter, international RCTs to establish best treatment of pemphigus. 4. ABSIS[20] and patients' subjective assessment should be included in the outcome measures of such studies. Reprint Address

Dr. Sue Jessop, Division of Dermatology, Ward G23, Groote Schuur Hospital and University of Cape Town, Observatory 7925, Cape Town, South Africa. E-mail: