The Use of Biologic Agents in the Treatment of Oral Lesions due to Pemphigus and Behçet’s Disease: A Systematic Review

Gerald E. Davis II1,2, George Sarandev1, Alexander T. Vaughan1, Kamal Al-Eryani3, Reyes Enciso4* 1Advanced graduate, Master of Science Program in Orofacial Pain and Oral Medicine, Herman Ostrow School of Dentistry of USC, Los Angeles, California, USA 2Assistant Dean of Academic Affairs, Assistant Professor, Restorative Dentistry, Meharry Medical College, School of Dentistry, Nashville, Tennessee, USA 3Assistant Professor of Clinical Dentistry, Division of Periodontology, Dental Hygiene & Diagnostic Sciences, Herman Ostrow School of Dentistry of USC, Los Angeles, California, USA 4Associate Professor (Instructional), Division of Dental Public Health and Pediatric Dentistry, Herman Ostrow School of Dentistry of USC, Los Angeles, California, USA


Introduction
Pemphigus, Behçet's disease, mucous membrane pemphigoid, oral lichen planus (OLP), and recurrent aphthous ulcers are immunemediated disorders, which commonly have oral manifestations. Though there are variations in the intraoral presentation of these conditions, they often present with ulcerations 1 .
Pemphigus is an autoimmune condition with three major variants: Pemphigus Vulgaris (PV), Pemphigus Foliaceous (PF), and Paraneoplastic Pemphigus (PNP) 2 . Only PV and PNP have common oral involvement 3 . In PV, IgG antibodies targeted against desmoglein-3 (Dsg-3), a protein responsible for cell-cell adhesion, lead to loss of adhesion (acantholysis) and blister formation 4 . In addition to Dsg-3, patients with PNP also have antibodies against Dsg-1 and plakin family 3 . Moreover, PNP is usually associated with underlying malignancy, such as non-Hodgkin lymphoma 3 .
Although PV and PNP share many clinical findings with other vesiculobullous lesions, it is still very possible to differentiate between them using their unique clinical and histopathological features 5 .
Behçet's disease is an autoimmune disease that targets small blood vessels, causing vasculitis. This leads to clinical effects such as uveitis and oral and genital ulcerations 6 . Currently, there is no cure for Behçet's disease; however, it has been managed using immunosuppressive therapy. The cause of Behçet's disease is unknown, but there are genetic predispositions for the disease, namely, in the Middle East along the path of the original Silk Road 7 . HLA-B51 proteins help aid in the precipitation of Behçet's disease due to chemotaxis hyper-function 7 .
Mucous Membrane Pemphigoid (MMP) is an autoimmune condition that targets hemidesmosomes; lichen planus is a chronic condition with both extraoral and intraoral effects.
Recurrent Aphthous Stomatitis (RAS) are non-specific ulcers of unknown etiology and may be present secondary to systemic diseases 8 . Thalidomide is considered to be an effective therapy to treat RAS, and it can be used in refractory cases with complete remission in 85-90 % of patients 9 .
In addition to palliative therapy such as topical anesthetics, primary therapy for these immune-mediated intraoral lesions consists of glucocorticosteroids, either given topically, as is the case in MMP, Behçet's disease, and OLP or systemically, as in pemphigus for example [10][11][12][13] . When lesions in MMP and OLP are recalcitrant to topical glucocorticoids, systemic glucocorticoids are often attempted 12 . However, long-term management with systemic glucocorticoids may present with minor side effects (such as acne) to major (glaucoma, cataract, adrenal atrophy, Cushing's syndrome, increased glucose levels, or cerebral atrophy) outcomes 14 .
Although steroids significantly reduced the mortality associated with pemphigus in one retrospective multicenter cohort study, serious side effects are a common cause of morbidity and mortality 15 . It has recently been proposed that treating pemphigus with a combination of rituximab and short-term prednisone is safer and more effective than high doses of corticosteroids 16 . The use of biologic agents has emerged as an additional line of treatment for ineffective immunosuppressive therapy with corticosteroids. Infliximab and rituximab are chimeric monoclonal antibodies, while etanercept is a chimeric fusion protein. Infliximab is a drug that was developed in 1993, and it functions by acting as a TNF-α antagonist 17 . Rituximab conversely targets CD-20 18 . Etanercept also inhibits TNF-α 19 .
The objective of this systematic review was to evaluate the efficacy of biologic agents (biopharmaceuticals manufactured via a biological source) on the treatment of intraoral lesions associated with pemphigus and Behçet's disease compared to glucocorticoids or placebo.

Materials and Methods
This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement 20

Inclusion and exclusion criteria
Studies were limited to randomized controlled trials comparing the efficacy of biologic agents to placebo or corticosteroids to treat oral lesions associated with pemphigus vulgaris/foliaceus, mucous membrane pemphigoid, Behçet's disease, oral lichen planus, or recurrent aphthous ulcers. Articles not available in English were excluded.

Search methods for identification of studies
Four electronic databases (MEDLINE via PubMed, EMBASE, Web of Science, and Cochrane Library) were searched on 2/27/2018 using the strategies described in Table 1. We re-ran the search for the four databases on 1/20/2019, and no relevant results were found.

Data collection and analysis
Three review authors (G.E.D., G.S., A.T.V.) screened titles and abstracts of the search results for inclusion/exclusion with a fourth author (R.E), making the final decision in case of disagreement. The same three review authors also independently extracted the data from the full-text articles meeting the inclusion criteria. Data extraction included the number of participants and their demographics, inclusion/ exclusion criteria, interventions, and outcome data. The assessment of the risk of bias in the included studies was undertaken independently by the three authors (G.E.D., G.S., A.T.V.) in accordance with the approach described in the Cochrane Handbook 21 . The three assessments are low risk, unclear risk, and high risk for each of the following: 1) random sequence generation; 2) allocation concealment; 3) blinding; 4) incomplete outcome; 5) selective reporting; 6) other potential bias.

Levels of evidence and summary of the review findings
Quality of evidence assessment and summary of the review findings were conducted following the Cochrane Collaboration and GRADE Working Group recommendations 21 . The GRADE Working Group grades of evidence are: • High quality: Further research is very unlikely to change our confidence in the estimate of effect • Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
• Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
• Very low quality: We are very uncertain about the estimate.

Results of the search
The initial search strategy through database searching yielded 721 references plus 19 additional records identified through other sources (scanning of the reference section of included studies). 740 records were assessed independently by three review authors (G.E.D., G.D., A.T.V), and based on the abstracts and titles, these were reduced to 27 relevant manuscripts. All 27 manuscripts identified were analyzed for inclusion independently by the same three review authors. Four manuscripts were assessed as relevant for inclusion. The main reasons for exclusion are presented in the PRISMA flowchart ( Figure 1). Table 1 provides a summary of the search strategy.

Study Design:
A single open-label prospective multicenter parallel-group RCT 16 and three double-blinded RCTs 22-24 were eligible for qualitative analysis (  Davis GE, Sarandev G, Vaughan AT, Al-Eryani K, Enciso R. The Use of Biologic Agents in the Treatment of Oral Lesions due to Pemphigus and Behçet's Disease: A Systematic Review. J Anesthesiol & Pain Therapy. 2020;1(1): [14][15][16][17][18][19][20][21][22][23] Population: A combined total of 158 adults were enrolled in these four RCTs (138 pemphigus and 20 Behçet's disease patients). Patients' diagnosis of pemphigus vulgaris/ foliaceus was confirmed via direct immunofluorescence and histology 16,22 or via disease activity score 24 . Behçet's disease was confirmed via monosodium urate (MSU) testing and pathergy 23 . All four RCTs included only adults; three of the RCTs enrolled both males and females 16,22,24 , while one, by design, included only male participants 23 . The average age of the participants ranged from 28.5 years 23 to 54.5 years 24 . The inclusion criteria, as well as the average age and gender distribution of the included studies, are presented in Table 2.
Interventions: Two RCTs randomized participants to receive etanercept 22,23 , one utilized infliximab 24 , and another utilized rituximab 16 as their biologic intervention. In two trials 16,24 , the intervention included a corticosteroid, prednisone, along with the biologic medication. In one study 16 , prednisone dosing was based on disease severity (moderate versus severe), while the other study 24 allowed the investigator to adjust the dose using the best medical judgment. Three RCTs utilized fixed doses of their respective intervention 22-24 , while one 16 utilized a larger loading dose over two weeks followed by maintenance doses at 12 and 18 months.
Comparison group: Patients in the comparison groups were labeled as controls and received placebo injections 22,23 or prednisone 16 or prednisone plus placebo 24 .
Primary outcomes: Melikoglu et al. 23 reported a primary endpoint of the amount of suppression of the pathergy response (reported as the number of positive responses out of total tested) and MSU test (reported as the area of erythema in mm 2 ). In Hall et al. (2015), the primary efficacy endpoint was a response to treatment at week 18. Joly et al. 16 reported the proportion of patients  who achieved a complete remission at month 24. Finally, Fiorentino et al. 22 reported the meantime to achieve a 50% reduction in the number of active lesions as the primary endpoint.

Risk of bias, outcomes and adverse events
Overall, all four included studies were considered at a high risk of bias. (Table 3). Table 4 presents the outcomes of the four included RCTs. Table 5 presents the adverse events reported in the four RCTs.

Summary of the evidence and quality of the findings
The quality of evidence was very low due to the high risk of bias, heterogeneity of the outcomes (only one study reported each outcome of interest), and the small number of studies with a small sample size. Very low evidence grading indicates that we are very uncertain about the estimate of effect.

Summary of the results
A single open-label prospective multi-center parallelgroup RCT 16 and three double-blinded RCTs 22-24 were included in this review. A total of 158 subjects in the studies were treated for pemphigus vulgaris/foliaceus (138 patients) 16,22,24 or Behçet's (20 patients) 23 .
Pemphigus vulgaris/foliaceus: Patients were treated either by IV infliximab plus prednisone 24 , IV rituximab plus prednisone 16 , or subcutaneous etanercept 22 . The results were inconclusive when etanercept alone was the active treatment 22 . When infliximab was added to prednisone, a statistically significant improvement was recorded with the Pemphigus Disease Area Index, the Dermatology Quality of Life Index, and Skindex scores; however, this was not the primary endpoint and was evaluated on a perprotocol basis 24       Median delay to complete remission off-therapy and median duration of complete remission off-therapy were more than ½ lower and 7x higher respectively in the treatment group than the prednisone alone group (P<.0001). Two-years disease-free survival was twice as high for the rituximab group (P=0.0191). Cumulative dose for prednisone needed was close to 1/3 lower in the active group (p<.0001). Total and desmoglein-specific B-lymphocytes disappeared in rituximab plus short-term prednisone group of patients, and remained unchanged in the prednisone-alone group. Anti-desmoglein antibodies decreased for the active group and increased after the 12 th month for the controls. Prednisone cumulative dose post treatment was 3x higher in the control group (P<.0001). At month 24, more relapse cases were recorded in the control group. Behҫet's Disease Melikoglu et al. (2005) Group 1: 25mg of etanercept subcutaneous injection (n=20)

Pemphigus Vulgaris & Pemphigus Foliaceus
The frequency of pathergy was not different between the two groups at any time. The area of erythema by MSU was less in the placebo group at Week 1 and no different than the etanercept group at Week 4 and at the third month after the end of the study. MSU was lower for both study arms at the end of the third month. The numbers of oral ulcers were less in the treatment group at weeks 1, 2, 3 and 4. Post-study, both groups returned to pretreatment levels. median duration of complete remission, and two years of disease-free survival were all higher with IV rituximab plus prednisone compared to prednisone alone control group 16 . Less prednisone was needed to achieve the desired Davis GE, Sarandev G, Vaughan AT, Al-Eryani K, Enciso R. The Use of Biologic Agents in the Treatment of Oral Lesions due to Pemphigus and Behçet's Disease: A Systematic Review. J Anesthesiol & Pain Therapy. 2020;1(1):14-23  outcomes with the addition of rituximab 16 or infliximab 24 to prednisone. IgG anti-Dsg1 and Dsg3 antibodies 24 or total and desmoglein-specific B-lymphocytes 16 were reduced or disappeared in infliximab plus prednisone or rituximab plus prednisone groups of patients compared to prednisone alone. Based on these results, IV rituximab plus prednisone or IV infliximab plus prednisone may be a superior treatment compared to prednisone alone.

Adverse Events in the Placebo group Pemphigus Vulgaris & Pemphigus Foliaceus
Behҫet's Disease: Etanercept was more effective in reducing the number of oral ulcers and achieving complete remission than placebo 23 .

Adverse effects
Infliximab with prednisone 24 and rituximab with shortterm prednisone 16 showed higher safety than prednisone alone in the treatment of pemphigus. The higher safety is likely delivered by the lower dose of prednisone when infliximab or rituximab is added to the treatment. More adverse events were reported when etanercept was utilized for the treatment for pemphigus vs. placebo controls 22 . The number of severe complications was lower, and the overall complications were higher in the etanercept group compared to the placebo in the treatment of Behçet's disease. The active treatment group reported abdominal pain, bloody diarrhea, and an ileocecal ulcer 23 .

Agreements and disagreements with other studies or reviews
Only RCTs were included in this systematic review.
Five clinical trials were excluded from this review for lack of control group, two were retrospective studies 25,26 , and three were uncontrolled prospective studies [27][28][29] . Three clinical trials found rituximab to be beneficial in the treatment of recalcitrant pemphigus 25,26,28 with early treatment within six months of diagnosis being more beneficial 26 . One of the uncontrolled studies with 49 MMP patients (24 receiving rituximab as adjuvant therapy and 25 using conventional therapy), concluded that the addition of rituximab to conventional therapy resulted in faster and more sustained disease control with potentially fewer adverse events 29 . An open-label pilot study with four patients with erosive lichen planus treated via subcutaneous efalizumab showed that two patients experienced serious adverse events 27 . One patient experienced urticaria and a staphylococcal abscess of an artificial hip joint. The second patient had drug-induced subacute cutaneous lupus 27 .
The findings of this systematic review are in agreement with other reviews; that biologic agents offer either an alternative treatment to or adjuvant treatment with glucocorticoids; however, not without risks and significant adverse effects 30,31 . A recent systematic review also identified favorable results when administering rituximab for treating pemphigus 32 . Two reviews concluded that rituximab and a short course of corticosteroids are more effective and safer than standard corticosteroids treatment, leading to its approval as first-line therapy in pemphigus by the FDA in 2018 33,34 . According to one review 35 , and a systematic review 36 , MMP is less responsive to rituximab than pemphigus. Further studies are needed.
Only one RCT with etanercept has shown beneficial results for Behçet's disease (included in this review 23 ) with few adverse events 37 . Infliximab and adalimumab are increasingly used for various refractory Behçet's syndrome manifestations despite the lack of controlled studies, according to one review 38 .

Quality of the Evidence
The overall quality of the evidence (according to the GRADE system) was very low due to the risk of bias (which was high in all studies), a small number of randomized controlled trials for each outcome (n=1), a small sample size (less than 400 total number of participants), heterogeneity of the outcomes reported and quality of study designs. While three DBRCT were included in this review [22][23][24] , one included study 16 was an open-label randomized trial. Additionally, one trial was a pilot study 22 . Further randomized controlled trials of a larger scale are needed to improve the quality of the data.

Implications for research and clinical practice
Infliximab plus prednisone and rituximab plus prednisone were more effective and safer compared to prednisone alone for the treatment of moderate to severe pemphigus. The reduced prednisone dose may be the contributing factor for the lower side effects achieved with the addition of the biologic agents. Those results make these treatments a viable alternative for refractory cases of pemphigus or cases where higher doses of prednisone are contraindicated. In June 2018, the US Food and Drug Administration approved rituximab for the treatment of adults with moderate to severe pemphigus vulgaris. Both systemic steroids and rituximab are considered first-line treatment 39 . According to the European Dermatology Forum, rituximab is indicated in patients who remain dependent on more than 10 mg prednisolone combined with an immunosuppressive adjuvant 40 . Using biologics, in general, allows for more rapid tapering of systemic steroid doses and a major steroid-sparing 39 . Research related to maintenance infusions of rituximab or infliximab is indicated as infusion may keep B-lymphocytes and IgG anti-Dsg1 and Dsg3 antibodies at the desired levels and reducing relapse. Similarly, etanercept is showing a reduction of relapses in refractory Behçet's disease 41 .
Infliximab binds to soluble and transmembrane forms of TNF-α, which prevents TNF-α from interacting with its receptors. The therapeutic role of infliximab is an anti-inflammatory role by downregulation of local and systemic pro-inflammatory cytokines (i.e., IL-1, IL-6), and reduction of lymphocyte and leukocyte migration to sites of inflammation 17 . On the other hand, desmoglein antigens, which play an essential role in the pathogenesis of pemphigus, are produced by plasma cells developed from autoreactive B-cells; Rituximab may act by binding to the CD20 cell surface receptor of B-cells and destroy them 18 . Thus the therapeutic role of rituximab is suppressing the autoimmune reaction against the skin or/and mucosa. The long term safety, cost, and efficacy of the infusions need to be further investigated.
Etanercept, which is a receptor fusion protein that inhibits TNF-α, was an effective treatment in reducing the number of oral ulcers and achieving complete remission in Behçet's disease in one study compared to placebo 23 . However, the effect of etanercept on women with Behçet's needs to be investigated, as the study did not include women. Etanercept was more effective in the treatment of oral ulcers than ulcers in other body locations in patients with Behçet's disease. Although the effectiveness of etanercept for pemphigus patients is heterogeneous, it has a favorable outcome with Behçet's patients 19 . Why etanercept was effective for Behçet's while ineffective for pemphigus may be due to the difference in pathogenesis between these conditions and/or inadequate dosing/ administration.
Further research in optimal dosing, selection, and administration of biologic treatments is necessary. This is a relatively new therapy with an unknown cost. Finally, well designed and high-quality studies limited to oral manifestations of these conditions will allow for improved clinical application and usefulness.
In conclusion, this systematic review found the use of biologic agents as an adjunct to prednisone may reduce the amount of required steroid, thus reducing the adverse effects of steroid usage in pemphigus patients. However, more studies are needed to confirm the benefits of biologic agents, considering the heterogeneity of the included research (conditions, interventions, outcomes, comparison groups, and adverse events). Although the addition of biologic agents to prednisone may show promise, the current evidence is inadequate to support the routine administration of biologic medications as first-line therapy for the prevention or treatment of Behçet's disease.