Rituximab in autoimmune bullous diseases: mixed responses and adverse effects


  • E. Schmidt,

    1. Department of Dermatology, University of Würzburg, Josef-Schneider-Strasse 2, D-97080 Würzburg, Germany
      *Department of Dermatology, University of Heidelberg, University Medical Center Mannheim, Mannheim, Germany
    Search for more papers by this author
  • C.S. Seitz,

    1. Department of Dermatology, University of Würzburg, Josef-Schneider-Strasse 2, D-97080 Würzburg, Germany
      *Department of Dermatology, University of Heidelberg, University Medical Center Mannheim, Mannheim, Germany
    Search for more papers by this author
  • S. Benoit,

    1. Department of Dermatology, University of Würzburg, Josef-Schneider-Strasse 2, D-97080 Würzburg, Germany
      *Department of Dermatology, University of Heidelberg, University Medical Center Mannheim, Mannheim, Germany
    Search for more papers by this author
  • E.B. Bröcker,

    1. Department of Dermatology, University of Würzburg, Josef-Schneider-Strasse 2, D-97080 Würzburg, Germany
      *Department of Dermatology, University of Heidelberg, University Medical Center Mannheim, Mannheim, Germany
    Search for more papers by this author
  • M. Goebeler

    1. Department of Dermatology, University of Würzburg, Josef-Schneider-Strasse 2, D-97080 Würzburg, Germany
      *Department of Dermatology, University of Heidelberg, University Medical Center Mannheim, Mannheim, Germany
    Search for more papers by this author

  • Conflicts of interest one declared.

Enno Schmidt.
E-mail: schmidt_e@klinik.uni-wuerzburg.de


Background  Intolerably high doses of systemic corticosteroids and additional immunosuppressants may be required to control disease activity in autoimmune bullous skin diseases. New therapeutic options are needed for such patients.

Objectives  To determine the efficacy and adverse effects of adjuvant rituximab.

Methods  Seven patients with refractory autoimmune blistering diseases (pemphigus vulgaris, PV, n = 4; bullous pemphigoid, BP, n = 2; mucous membrane pemphigoid, MMP, n = 1) were treated four times with rituximab at an individual dose of 375 mg m−2 at weekly intervals.

Results  All lesions cleared in three patients (two PV, one BP), while they were reduced by more than 50% in three others (two PV, one BP). The concomitant immunosuppressive medication was reduced in five patients (four PV, one BP). The patient with MMP developed bilateral blindness while nasopharyngeal lesions resolved. Three patients (two BP, one PV) experienced severe adverse events including fatal pneumonia.

Conclusions  Adjuvant B-cell depletion by rituximab is effective in otherwise therapy-resistant bullous autoimmune disorders but may be associated with substantial adverse effects including fatal outcomes.

To control disease activity in pemphigus and other autoimmune bullous skin diseases such as bullous pemphigoid (BP) and mucous membrane pemphigoid (MMP) intolerably high doses of systemic corticosteroids and additional immunosuppressants may be required which bear the risk of severe adverse reactions.1,2 New therapeutic options are needed for such patients.

Rituximab, a humanized monoclonal antibody directed against the cell surface glycoprotein CD20, leads to a transitory depletion of B lymphocytes. Originally developed to treat B-cell malignancies, it is increasingly used in autoimmune diseases including autoimmune bullous disorders.3–9

Patients and methods

Seven patients with pemphigus vulgaris (PV, n = 4), BP (n = 2) and MMP (n = 1) were treated off-label with intravenous rituximab (MabTheraTM; Roche, Basel, Switzerland) at a dose of 375 mg m−2 at weekly intervals after written consent. Patients’ characteristics prior to initiation of rituximab including age, sex, preceding medication, disease duration and clinical presentation are detailed in Table 1. Six patients received four infusions at weekly intervals; in one patient with PV (patient 2) rituximab was administered only twice because of her advanced age, and in the child with BP (patient 5) four additional rituximab infusions were given 8–10 months after the first dose. All patients had previously not responded to at least one systemic immunosuppressive regimen.

Table 1.   Patient characteristics prior to administration of rituximab
PatientDiagnosisSex Agea (years)Disease durationa (months)Previous medicationClinical presentation
  1. BP, bullous pemphigoid; IVIG, intravenous immunoglobulins 2 g kg−1 monthly; MMP, mucous membrane pemphigoid; PA-IA, protein A immunoadsorption;14 PV, pemphigus vulgaris. aAt the time of rituximab administration. bDetailing the medication in the last 6 months prior to initiation of rituximab. cPreviously published at the beginning of his disease.15

1PVF3979IVIG, azathioprine, prednisolone; last 6 months:b azathioprine + methylprednisolone 10–50 mg dailyBuccal mucosa, gingiva, trunk
2PVF81 7Last 6 months: 4-weekly and finally 2-weekly dexamethasone (3 × 100 mg)–cyclophosphamide (750 mg) pulsesAbdomen, oral and nasal mucosa, lips
3PVF6864Dexamethasone–cyclophosphamide pulses, azathioprine, prednisolone, IVIG, PA-IA, mycophenolate mofetil; last 6 months: mycophenolate mofetil 1·5 g daily + prednisolone 100–140 mg dailyOral and nasal mucosa
4PVF1730IVIG, methotrexate, azathioprine, mycophenolate mofetil, methylprednisolone; last 6 months: mycophenolate mofetil 2 g daily + methylprednisolone 48 mg dailyBuccal mucosa, tongue, gingiva
5BPM2 (9 kg body weight)18Prednisolone pulses (3 × 150 mg per pulse), dapsone, mycophenolate mofetil, cyclophosphamide, IVIG; last 6 months: cyclophosphamide 500 mg m−2 monthly + mycophenolate mofetil 320 mg daily + prednisolone 7·5 mg daily +  dapsone 22 mg dailyBuccal mucosa, tongue, arms, legs, trunk
6BPF63 6Last 6 months: prednisolone 100 mg daily +  dapsone 125 mg daily; 3-monthly dexamethasone (3 × 100 mg)–cyclophosphamide (500 mg) pulsesArms, legs, trunk
7MMPcM7864Dexamethasone–cyclophosphamide pulses, dexamethasone pulses, IVIG; last 6 months: 4-weekly and finally 2-weekly dexamethasone (3 × 100 mg)–cyclophosphamide (750 mg) pulsesLarynx, nasopharynx, conjunctivae


Clinical response

In two of four patients with PV, lesions healed completely within 2 and 4 months, respectively, after the first rituximab infusion (patient 2 in Fig. 1a,b). In the other two patients oral erosions regressed by 90% and 50% while skin lesions resolved completely (patient 1 in Fig. 1c,d). Skin and oral lesions in the boy with BP improved slowly and, 13 months after initiation of rituximab, healed completely leaving generalized erythematous slightly elevated papules. Nasopharyngeal lesions in the patient with MMP healed after 4 months, whereas conjunctival scarring progressed, finally resulting in bilateral blindness despite administration of additional intravenous dexamethasone–cyclophosphamide pulses. Subsequently, only dexamethasone pulses were given at prolonged intervals while 13 months after the first rituximab infusion the immunosuppressive medication was withdrawn. Currently, an orthokeratoprothesis is planned. In the adult patient with BP, blister formation stopped and erosions started to heal 4 weeks after rituximab initiation. At that time, she developed a Clostridium difficile enteropathy and, 2 weeks later, died due to a hospital-acquired bacterial pneumonia while still receiving oral prednisolone 10 mg daily.

Figure 1.

 Patients 2 (a, b) and 1 (c, d) with pemphigus vulgaris prior to (a, c) and 4 (b) and 5 months (d) after the first rituximab infusion, respectively.

Adjuvant immunosuppressive therapies, clinical outcomes 3, 6 and 12 months after initiation of rituximab and the patients’ current state with follow-up periods ranging from 7 to 21 months are documented in Table 2.

Table 2.   Patient characteristics after administration of rituximab
PatientDiagnosisAdjuvant therapybAdverse effectsClinical responseaCurrent therapycAutoantibody levels prior to rituximab/currentlycFollow upd (months)
After 3  monthsAfter 6  monthsAfter 12  monthsTo datec
  1. BP, bullous pemphigoid; Con, conjunctivae; Dsg, desmoglein; LNP, larynx and nasopharynx; MMP, mucous membrane pemphigoid; neg., negative; PV, pemphigus vulgaris. aCliR, clinical remission (healing of all lesions); CR, complete remission (clinical remission and no further therapy needed); PD, progressive disease (> 25% more lesions); PR, partial remission (healing of > 50% of lesions); NC, no change (< 25% increase and < 50% healing of lesions). bAt the time of rituximab infusions; doses were then tapered according to clinical improvement. In patient 2, only a single pulse of dexamethasone–cyclophosphamide was given. cAt the time shown as follow up. dAfter the first rituximab infusion. eMeasured by commercial enzyme-linked immunosorbent assays according to the manufacturer's instructions (Medical and Biological Laboratories, Nagoya, Japan); normal values for Dsg 1 were < 14 U mL−1, for Dsg 3 < 7 U mL−1, and for BP180 NC16A < 9 U mL−1, fOnly two rituximab infusions were administered because of the patient's advanced age. gFour additional rituximab infusions were given 8–10 months after the first four rituximab infusions. hBy indirect immunofluorescence microscopy on 1 mol L −1 NaCl-split human skin.

1PVAzathioprine 150 mg daily +  methylprednisolone 16 mg dailyNoneNCPRPRPRAzathioprine 150 mg daily + methylprednisolone 12 mg dailyDsg 3e 806/108 U mL−121
2PVfSingle pulse dexamethasone (3 × 100 mg)– cyclophosphamide (500 mg)Bacterial pneumonia, pulmonary embolismPRCliRCRCRNoneDsg 3e 222/neg. U mL−1
Dsg 1e 985/neg. U mL−1
3PVMycophenolate mofetil 1·5 g daily +  prednisolone 50 mg dailyNonePRCliRCliRMycophenolate mofetil 1·5 g dailyDsg 3e 877/27 U mL−1 9
4PVMycophenolate mofetil 2 g daily +  methylprednisolone 30 mg dailyNonePRPRPRMycophenolate mofetil 2 g daily +  methylprednisolone 8 mg dailyDsg 3e 7708/517 U mL−1 7
5BPgMycophenolate mofetil 320 mg daily +  prednisolone 7·5 mg daily + dapsone 22 mg dailyVaricella sepsis, parainfluenzal pneumonia, exudative enteropathy, hypogammaglobulinaemiaNCNCPRCliRPrednisolone 3·5 mg dailyBP180 NC16Ae 3528/77 U mL−126
6BPPrednisolone 10 mg dailyClostridium difficile enteropathy, fatal nosocomial bacterial pneumoniaThe patient died 6 weeks after the first rituximab infusion due to a nosocomial bacterial pneumonia. She was in PR; her BP180 NC16A reactivity decreased from 17 736 to 8486 U mL−1      
7MMPDexamethasone (3 × 100 mg)– cyclophosphamide (500 mg) pulses at 2- to 12-weekly intervalsNoneLNP: PR
Con: PD
Con: PD
NoneEpidermal binding, titre 1:20/neg.h14

Adverse reactions

No adverse reactions were observed in four of seven patients. Patient 2, an 81-year-old woman with PV, developed a community-acquired bacterial pneumonia and pulmonary embolism 8 and 10 weeks, respectively, after her first rituximab infusion and 3 weeks after a single intravenous dexamethasone (100 mg on three consecutive days)–cyclophosphamide (500 mg) pulse. Five months after the initiation of rituximab, the boy with BP (patient 5) developed a noninfective secretory enteropathy requiring parenteral nutrition 7 months later. During his second course of four rituximab infusions, parainfluenzal pneumonia and, 1 month later, varicella-zoster virus sepsis with pulmonary and meningeal involvement occurred. Furthermore, the child developed hypogammaglobulinaemia requiring monthly immunoglobulin treatment for 30 months. Currently the boy, now aged 4 years, has been receiving parenteral nutrition for 13 months, weighs 10·6 kg and is 85 cm tall (both below the third percentile).


Here, we present clinical data on the long-term follow-up of seven patients with refractory autoimmune bullous diseases who underwent treatment with rituximab. Of the four patients with PV, one attained complete remission, one clinical remission (defined as healing of all lesions but further requirement for therapy) and two partial remission (defined as healing of > 50% of lesions). These results are in line with the outcome observed in 24 patients with pemphigus, in whom rituximab led to at least a partial remission in all but one patient: clinical remission was achieved in about a third of patients and complete remission not requiring further immunosuppression in about a quarter of them (reviewed in Schmidt et al.4). Although no control group was included in the present study, the overall dose of adjuvant immunosuppressive medication could generally be reduced by the use of rituximab.

In addition, we report a patient with treatment-resistant MMP, in whom conjunctival inflammation further progressed, resulting in bilateral blindness. It may be speculated that his conjunctival disease had already reached such an advanced state that rituximab, which begins to control autoimmune bullous disease often not before 2–3 months after initiation,6–8 could not prevent further deterioration of vision.

A partial remission and a clinical remission were observed in the two patients with BP, the latter in the young boy, although only after a second course of four rituximab infusions and in combination with high-dose adjuvant immunosuppression. The only other patient with BP so far reported to have received rituximab, a 10-year-old boy with graft-versus-host disease after cord blood transplantation due to B-precursor acute lymphocytic leukaemia, died from bacterial sepsis while in clinical remission; however, he additionally received the anti-CD25 antibody daclizumab.9

In a considerable number of our patients severe adverse events occurred which were mostly related to infection. All systemic infections were seen within 9 months after the first rituximab infusion when B cells were depleted from the circulation. Our child with BP also developed a persistent hypogammaglobulinaemia requiring regular immunoglobulin infusion. A long-term hypogammaglobulinaemia not requiring immunoglobulin supplementation has previously been noted in a 14-year-old boy with PV treated with rituximab. In this boy, however, no infections were observed.8 Neither hypogammaglobulinaemia nor a higher risk of infections compared with adults was reported in 36 children with chronic immune thrombocytopenic purpura receiving rituximab.10 In contrast, both hypogammaglobulinaemia and severe systemic infections were seen in two of 11 girls with systemic lupus erythematosus who also received, like our patients, high doses of immunosuppressants.11 Severe hypogammaglobulinaemia and high-dose immunosuppression may therefore be possible risk factors for systemic infections in children receiving rituximab. The secretory enteropathy observed in the boy with BP (patient 5) has not yet been reported as an adverse effect of rituximab.

The frequency of serious adverse events observed in our case series (in three of seven patients) was higher compared with the rate of eight severe adverse events reported in 28 anecdotally published patients with autoimmune bullous disorders who had received rituximab.4 In these patients, severe adverse reactions appeared to be associated with underlying malignancies and high-dose adjuvant immunosuppression. The incidence of serious adverse events in patients with autoimmune bullous diseases receiving rituximab (31% in 35 patients4 including this case series) is higher than the rates observed in patients with systemic lupus erythematosus (17% in 18 patients) or rheumatoid arthritis (9% in 161 patients) who also received adjuvant immunosuppression.12,13 Controlled studies are thus necessary to identify risk factors for severe adverse events in patients with autoimmune bullous diseases.

Taken together, our data indicate that B-cell depletion by rituximab may be a valuable approach for treating otherwise therapy-resistant bullous autoimmune disorders. However, use of rituximab may be associated with substantial adverse effects including fatal outcome which should especially be considered in very young and very old individuals.


We are grateful to Drs Tina Wachter, Martina Auth and Gregor Wienrich, Department of Dermatology, and Drs Hans W. Kreth and Hermann Girschik, Department of Paediatrics (all Würzburg), who were involved in the clinical care of the patients. We thank Silvana Noll, Andrea Fuss and Christa Knaus, Würzburg, for excellent technical assistance.