Intralesional Rituximab in the Treatment of Refractory Oral Pemphigus Vulgaris

Importance  Oral lesions of pemphigus vulgaris are usually recalcitrant and respond slowly to treatments. Corticosteroid injection is considered to be the most effective local treatment in oral pemphigus vulgaris. However, intralesional corticosteroids are not effective in all remnant lesions. In 3 such patients with pemphigus vulgaris, we evaluated the utility of 2 injections (on days 1 and 15) of intralesional rituximab, 5 mg/cm², in terms of accelerated healing, limitation of the use of systemic immunosuppressants, and reduction of their adverse effects.

Observations  Three patients (1 man and 2 women) received 2 doses of intralesional rituximab in March and April 2013. All 3 patients responded to the treatment. In patients 1 and 2, the objective severity score was reduced to 0 at the final visit from a baseline score of 4 and 5, respectively (range, 0-11). The subject severity score in these patients was reduced to 1.0 and 0 from a baseline score of 22.0 and 22.5, respectively. After clinical remission was achieved, patient 3 developed a relapse of mucosal lesions. At the final visit, all of the patients were satisfied with the treatment, with a mean satisfaction score of 8 (maximum score, 10). We found a marked decline in the CD19 cell count from a pretreatment mean count of 287 cells/µL to 6 cells/µL on day 15 after a single intralesional rituximab injection. Adverse events were limited to local pain in 1 patient.

Conclusions and Relevance  Intralesional rituximab administration lacks the adverse effects of intravenous administration. This method reduces the amount of drug administered and therefore is less expensive. Encouraging results from our study should prompt further evaluation of this novel route of rituximab administration in patients with refractory oral pemphigus vulgaris.

Treatment of oral pemphigus vulgaris (PV) remains a challenge because of the chronic, persistent nature of this disease. The healing process of oral ulcers is much slower than that of cutaneous lesions, which require slow tapering of corticosteroid therapy.¹ Various topical therapies for oral lesions in PV, including corticosteroids, cyclosporine, and tacrolimus, have been used without any significant outcome.¹

Rituximab, a chimeric monoclonal antibody to CD20, has been approved by the US Food and Drug Administration to treat CD20-positive non–Hodgkin lymphoma. Intralesional rituximab treatment has been used successfully in a few studies of primary cutaneous B-cell lymphoma.^2,3 Because previous studies have proven the efficacy of intravenous rituximab in the treatment of the pemphigus group of disorders,^4,5 the same efficacy might be obtained by intralesional administration of rituximab in selected cases of oral PV. In this case series, we evaluated for the first time, to our knowledge, the usefulness of intralesional rituximab administration in patients with refractory oral PV.

After providing written informed consent, 3 patients with oral PV who satisfied the inclusion criteria were recruited in March 2013. The study was performed in accordance with the Declaration of Helsinki and was approved by the ethics committee at Postgraduate Institute of Medical Education and Research, Chandigarh, India.

All of the patients fulfilled the following inclusion criteria:

1. Pemphigus vulgaris diagnosed by means of clinical, histologic, and immunological features and refractory oral mucosal lesions or remnant lesions after treatment with conventional immunosuppressive agents, which qualified as resistant disease by the standard definition⁶;

2. Inadequate response to treatment with intralesional corticosteroids, which was defined as failure of established lesions to begin to heal even after 3 doses of intralesional triamcinolone acetonide (40 mg/mL; diluted 1:1 with normal saline) at a dosage of 0.1 mL/cm² every 2 weeks; and

3. No new mucosal or cutaneous lesions for at least 12 weeks.

Patients with cutaneous lesions of PV; with active disease, which was defined by continuing extension of old lesions or development of new lesions in the preceding 12 weeks; or with known standard contraindications for rituximab administration⁷ were excluded from the study.

The patients underwent a pretreatment workup as standard practice.⁷ We assessed a severity index for PV subjectively and objectively using the method of Saraswat et al (Saraswat severity score).⁸ We calculated the size of the oral erosions by measuring the largest diameter using a clinical thermometer as a probe. Baseline enzyme-linked immunosorbent assay (ELISA) index values for IgG anti–desmoglein 1 (Dsg1) and Dsg3 antibodies were determined and defined as positive (>20), borderline (10-20), and negative (<10). The CD19 cell count was performed on fresh unclotted blood samples using flow cytometry at baseline and day 15 to study the systemic effect of intralesional rituximab therapy.

Intralesional rituximab was administered in an outpatient setting by 2 of us (K.V. and A.M.) in 5-mg/cm² injections on days 1 and 15 (ie, 2 injections per patient). Patients 1 and 2 received intralesional rituximab on March 13 and 27, 2013, and patient 3 received intralesional rituximab on March 20 and April 3, 2013. After premedication with intravenous hydrocortisone (100 mg) and pheniramine maleate, a stock solution of rituximab (10 mg/mL) was administered intralesionally. The systemic treatments that were given to the patients at the time of recruitment were continued. Patients who developed any new cutaneous or mucosal lesions during the study period were released from the study and given standard treatments.

The patients were followed up at days 15, 30, and monthly thereafter until their final visit at 6 months or the day of release from the study, whichever was earlier. At each visit, the treatment response was assessed using the Saraswat severity score for oral mucosal PV.⁸ Clinical remission was defined as complete healing of oral lesions with an objective Saraswat severity score of 0. We performed ELISAs for IgG anti-Dsg1 and Dsg3 antibodies again during the final visit. At the final visit, all patients were asked to grade their satisfaction with the treatment on a scale ranging from 1 to 10, where 1 represented no response and 10, a complete response.

Three middle-aged patients (2 women and 1 man) were included in the study. All 3 patients had a long duration of disease (mean, 48 months) and had received multiple immunosuppressants in the past, including azathioprine sodium (2 mg/kg/d), corticosteroids (0.5-1.0 mg/kg/d), dapsone (100 mg/d), dexamethasone cyclophosphamide pulse therapy, mycophenolate mofetil (3 g/d), methotrexate sodium (15 mg/wk), and intravenous rituximab according to the rheumatoid arthritis protocol (Table).⁹ All patients had resistant disease defined as the failure of established lesions to begin to heal despite 3 weeks of therapy with a prednisone equivalent, 1.5 mg/kg/d, with or without any of the following agents: cyclophosphamide, 2 mg/kg/d for 12 weeks; azathioprine sodium, 2.5 mg/kg/d for 12 weeks; methotrexate sodium, 20 mg/wk for 12 weeks; or mycophenolate mofetil, 3 g/d for 12 weeks.⁶ After intravenous rituximab therapy, patient 2 experienced only partial improvement, whereas patients 1 and 3 achieved clinical remission but showed relapse of oral lesions. In all of the patients, at least 12 months elapsed between intravenous rituximab administration and recruitment to the present study. The Saraswat severity scores at baseline and the dose of rituximab administered in each patient are depicted in the Table.

Response to treatment was seen in all 3 patients, with a progressive decline in the objective and subjective Saraswat severity scores (Figure 1). Clinical remission of oral lesions was seen at 12 weeks in patient 3 and 16 weeks in patients 1 and 2 (Figure 2). After a clinical remission of 8 weeks, patient 3 experienced a relapse of oral mucosal lesions at month 5 of follow-up and was released from the study. The patient subsequently was treated with oral corticosteroids and azathioprine. Concomitant therapies included oral prednisolone acetate at a dosage of 0.5 mg/kg/d in patients 1 and 2 and azathioprine sodium at a dosage of 1.5 mg/kg/d in patient 3. Prednisolone therapy was tapered and stopped in patients 1 and 2 by the end of month 5 (Figure 3). At the final visit, all patients were satisfied with the treatment (Table). Baseline ELISA index findings of IgG anti-Dsg1 antibodies were negative in all patients, and those of IgG anti-Dsg3 antibodies were positive only in patient 3. However, 2 weeks after the first intralesional rituximab injection, the CD19 cell count declined in all 3 patients (Table). Adverse effects were limited to pain during intralesional injection in patient 1 (subjectively graded as severe by the patient), lasting for 2 to 3 days. No serious or long-term adverse effects were noted.

Oral mucosal involvement is an initial manifestation in most patients with PV.¹ However, the chronic nature of the oral lesions and tough oral environment (ie, poor oral hygiene, prosthesis, restorations, smoking, and alcohol consumption) make the treatment of oral lesions difficult.¹⁰ Corticosteroid injection is commonly used for the treatment of resistant and recurrent oral lesions and is considered to be the most effective local treatment.¹ An open-label trial of intralesional triamcinolone for oral PV lesions showed earlier and higher rates of complete clinical remission and a smaller cumulative dose of corticosteroids.¹⁰ The major drawbacks of intralesional triamcinolone were pain and mucosal atrophy.^1,10 However, some of the remnant lesions do not respond to the therapy, thus necessitating continuation of systemic immunosuppressant therapy. All of the patients recruited in this study had refractory oral PV lesions that had failed to respond to systemic immunosuppressants, topical coritcosteroids, oral antibiotics, antivirals, and maintenance of good oral hygiene. In such patients, we evaluated the usefulness of intralesional rituximab injection in terms of accelerated healing, reduction of the dose of systemic immunosuppressants, and the avoidance of adverse effects.

The previous retrospective analysis of 35 patients with primary cutaneous B-cell lymphoma who received intralesional rituximab treatment³ showed improvement in 94% of cases and a complete cure rate of 71%. In that study, the treatment regimen for most patients (74%) was a course of 3 injections in a single week at monthly intervals. The dose per day for a single lesion was 10 mg in 71% of patients. The median dose of cumulative rituximab per lesion was 60 mg.³ Compared with the previous study,³ the mean treated area in our patients ranged from 10 to 15 cm², and we arbitrarily selected a dose of 5 mg/cm² to achieve a minimal total cumulative dose of 60 to 100 mg/cm².

At present, consensus is lacking on the optimum dose and schedule of intravenous rituximab administration in autoimmune bullous disorders. The 2 main intravenous regimens used are the lymphoma protocol and the rheumatoid arthritis protocol. As highlighted in a previous report,⁴ for an autoimmune disease such as PV, the fixed-dose rheumatoid arthritis protocol is pharmacokinetically better than the body weight–based lymphoma protocol. In addition, because of previous experience with the intravenous rheumatoid arthritis protocol,^4,9 we limited administration of intralesional rituximab to 2 doses on days 1 and 15.

All of the patients responded to this novel therapeutic intervention, as made evident by the progressive decline in the clinical severity score and by the satisfaction scores of the patients (mean score, 8) at the end of the study. The mode of action of intralesional rituximab in oral PV is speculative. We believe that the therapeutic effects were exerted by the local action of rituximab because local factors play an important role in the persistence of the oral lesions of PV. Patient 2, who had mucosal PV limited to the oral cavity, had only partial improvement with intravenous rituximab therapy but showed complete healing of oral lesions after intralesional rituximab therapy (Figure 2). This response may be owing to the local but unknown mechanisms of rituximab in addition to its systemic effect that caused the CD19 cell count to decline. On the other hand, repeated administration of rituximab can induce remission in patients whose conditions have failed to respond to the first infusion.¹¹ The ELISA index values for IgG anti-Dsg1 and Dsg3 antibodies were negative in 2 patients at baseline. This finding may further support the fact that local factors are pathologically more important than circulating antibody levels in the persistence and slow healing of oral lesions.

Another interesting immunological finding was the dramatic decline in the CD19 cell count after intralesional administration of small doses of rituximab, similar to the results in previous pharmacokinetic studies of systemic rituximab administration.¹² Similar findings were also reported by Roguedas et al¹³ and Heinzerling et al² in their patients with primary cutaneous B-cell lymphoma who were treated with intralesional rituximab. Few of these patients also showed clinical response in distant untreated lesions of primary cutaneous B-cell lymphoma.^2,13 These findings suggest that a small dose of intralesional rituximab may show a systemic effect and deplete cells positive for CD19. However, future large-scale studies should assess whether this B-cell depletion is maintained because a previous study⁴ found that the B-cell depletion achieved by low-dose rituximab (500 mg) treatment was short lived compared with that achieved by high-dose (1000 mg) treatment.

Newer methods of drug administration that are convenient for patients and offer potential improvements in quality of life and treatment adherence are under constant exploration. Intravenous infusion of monoclonal antibodies faces some challenges, such as the need for trained personnel and infusion facilities, long infusion times, and the risk for infusion-related reactions and complications.¹⁴ Subcutaneous formulations of rituximab in combination with hyaluronidase are being developed to overcome such limitations without compromising efficacy.¹⁴

Intralesional rituximab administration has many advantages compared with the conventional intravenous administration. Intralesional administration allows local delivery of the drug, lacks the adverse effects of intravenous administration, reduces the amount of drug administered (<10% of the intravenous dose), and therefore is less expensive. Being an outpatient procedure also offers an advantage. Our study is limited by the small sample size. Currently, the mechanism of action of intralesional rituximab is not fully understood; it appears to act by a local immunomodulatory effect and at least partially by causing a systemic decline in the CD19 cell count.

Intralesional rituximab administration is an unexplored area in the therapeutics of PV. The outcomes of our case series should encourage further investigation of this modality in selected cases of oral PV. Future large-scale studies are required to confirm our initial findings and to assess the role of local factors in persistent oral PV lesions and their modulation by intralesional rituximab therapy.

Accepted for Publication: September 5, 2014.

Corresponding Author: Amrinder J. Kanwar, MD, FRCP, Department of Dermatology, Venereology, and Leprology, Postgraduate Institute of Medical Education and Research, Sector 12, Chandigarh 160 012, India (ajkanwar1948@gmail.com).

Published Online: December 23, 2014. doi:10.1001/jamadermatol.2014.3674.

Author Contributions: Drs Vinay and Kanwar had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Vinay, Dogra.

Acquisition, analysis, or interpretation of data: Vinay, Kanwar, Mittal, Minz, Hashimoto.

Drafting of the manuscript: Vinay, Mittal, Dogra, Hashimoto.

Critical revision of the manuscript for important intellectual content: Kanwar, Minz.

Statistical analysis: Vinay, Mittal.

Administrative, technical, or material support: Kanwar, Dogra, Hashimoto.

Study supervision: Dogra, Minz.

Conflict of Interest Disclosures: None reported.

Correction: This article was corrected on April 13, 2015, to fix Figure 2.