ORAL MYTH SERIES
Urban legend series: mucous membrane pemphigoid
G Di Zenzo1, M Carrozzo2, LS Chan3
1Molecular and Cell Biology Laboratory, Istituto Dermopatico dell’Immacolata, IDI-IRCCS, Rome, Italy;2Department of Oral Medicine, Centre for Oral Health Research, Newcastle University, Newcastle upon Tyne, UK;3Department of Dermatology and Immunology/Microbiology, University of Illinois College of Medicine, Chicago, IL, USA
Mucous membrane pemphigoid (MMP) is a heteroge- neous group of autoimmune subepithelial blistering dis- eases affecting primarily mucous membranes showing marked degree of clinical and immunological variability.
We investigated four controversial topics: (i) Does oral pemphigoid (OP) really exist as a separate entity? (ii) Is mucous membrane pemphigoid curable? (iii) What is the best therapeutic option for MMP? (iv) Does exclu- sive oral IgA dermatitis exist as a distinct entity from MMP? Results from extensive literature searches sug- gested that (i) it is still unclear whether patients with OP could be considered as a distinct subset of MMP with specific clinical and immunological features; (ii) it is uncertain whether treatment regimens that get MMP under control can be eliminated to allow patients to be in drug-free remission or they should be continuously administered in some capacities; (iii) there is a concern- ing paucity of good-quality trials on MMP and available recommendations are solely based on generally small patients’ cohorts or case series. Some of the 2002 con- sensus experts’ opinions should be possibly updated, particularly regarding the safety of sulfa drugs; (iv) we did not find any strong evidence to support an exclusive oral (and perhaps also mucosal) form of LAD as a separate entity.
Oral Diseases (2014) 20, 35–54
Keywords: mucous membrane pemphigoid; oral pemphigoid;
treatment; mucosal IgA linear dermatitis
Introduction
Broadly defined, mucous membrane pemphigoid (MMP) is a heterogeneous group of putative autoimmune subepithelial
blistering diseases affecting primarily mucous membranes, with or without some degrees of skin involvement (Chan et al, 2002). Scarring is the clinical hallmark; however, this is not always obvious, particularly in the oral mucosa (Chan et al, 2002). As a heterogeneous group of diseases, patients affected by MMP can develop autoantibodies that target a variety of known epithelial basement membrane structure components, bullous pemphigoid antigen 2 (BP180), alpha6 integrin, beta4 integrin, laminin-332, laminin-331, and type VII collagen (Bernard et al, 1992;
Domloge-Hultsch et al, 1992; Chan et al, 1997; Chan et al, 1999; Bhol et al, 2000; Bhol et al, 2001; Chan et al, 2002; Malik et al, 2007; Letko et al, 2007). In some patients, the antigens of their autoantibodies targeted are not defined (Chan et al, 1991, 1993). The relative frequency of mucous membrane location affected is esti- mated to be oral > ocular > nasal > nasopharyngeal >
anogenital> laryngeal > esophageal (Chan et al, 2002).
The first consensus supported by 26 international experts in the field recommended that the diagnosis of MMP should be established by both clinical morphology and a direct immunofluorescence (DIF) finding of linear deposi- tion of IgG, IgA, or C3 at the epithelial basement mem- brane zone (Chan et al, 2002). In this chapter of the Urban Legends series on controversial topics in oral medi- cine (Carrozzo, 2011), we focused on four questions about MMP: (i) Does oral pemphigoid really exist as a separate entity? (ii) Is mucous membrane pemphigoid curable? (iii) What is the best therapeutic option? (iv) Does exclusive oral IgA dermatitis exist as a distinct entity from MMP?
All along the text, the terms ‘oral pemphigoid’ (OP) and ‘ocular pemphigoid’ (OCP) have been used to indicate patients with exclusive lesions in the oral cavity and the eyes, respectively. The term ‘MMP’ has been used to define patients with predominant involvement of any mucosal areas either with or without any skin lesions.
Does oral pemphigoid really exist as a separate entity?
As yet said, there is a marked degree of variability in the clinical and immunological features of MMP, suggesting the existence of several phenotypic variants. From a
Correspondence: Marco Carrozzo, MD, DSM (Turin 1995), Professor of Oral Medicine, School of Dental Sciences, University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne NE2 4BW, UK.
Tel: +0044 191 222 7818, Fax: +0044 191 222 6137, E-mail: marco.carrozzo@ncl.ac.uk
Received 17 July 2013; revised 17 September 2013; accepted 2 October 2013
All rights reserved www.wiley.com
diagnostic and therapeutic perspective, it might be of significant benefit to attempt to distinguish whether a clinical subset of MMP termed oral pemphigoid (OP) in which the disease is limited to the oral cavity really exists, mainly because it has been suggested that OP has a better prognosis compared to other MMP variants (Chan et al, 2002).
From a therapeutic point of view
In patients with MMP, scarring and the associated loss of function are the major complications, except usually for some patients in whom the disease is restricted to the oral mucosa. Interestingly, both IL-4 and IL-13 are thought to be involved in cicatricial scarring process in MMP (Bho- gal et al, 2005; Giomi et al, 2005). Very recently, it has been suggested that the interleukin-4 receptor A (IL- 4RA)-1902 A/A, a genotype that has been found in 90%
of patients with OP, is associated with a reduced response to IL-4 and thus may explain a better clinical outcome for this group of patients (Carrozzo et al, 2013).
OP was originally reported to be associated with a bet- ter prognosis and to be more amenable to medical treat- ments (Chan et al, 2002). However, there is a paucity of long-term follow-up studies on MMP, and several case reports and cohort series report the difficult treatment for OP (Bohn et al, 1999; Ahmed and Colon, 2001; Sacher et al, 2002; Canizares et al, 2006; Segura et al, 2007;
Carrozzo et al, 2008; Kasperkiewicz et al, 2011; Le Roux-Villet et al, 2011). At least 21 cases of OP recalcitrant to even doses such as 100 mg per daily of prednisolone (Pred) and other immunosuppressive/immu- nomodulant drugs, including intravenous cyclophosphamide (CYC), have been published (Table 1). In those patients, modalities such as plasma exchange, tumor necrosis factor alpha antagonists, intravenous immunoglobulins (IVIg), and even rituximab (RTX) were used to control, some- times only temporarily, the oral lesions (Table 1). Because the limited number of reports existed in the literature, at the present time it is not possible to definitively determine whether the exclusive oral involvement may account for a significant difference in the response to therapy and more research is needed to identify the most effective treatment options.
From a clinical presentation point of view
Because MMP is not a single entity, it does not have a unified and predictable natural history. In some patient, the disease is localized and has a slowly progressive course without complications; in others, it is devastating, with severe morbidity. At the present time, it is clear that only in a subset of the total MMP patients studied, the disease remains localized to the oral cavity. An important point to address is whether the exclusive oral involvement is only a stage of MMP course, often presents at disease onset, or represents the phenotype of a distinct clinical entity.
Mobini and co-workers have studied 29 MMP patients with disease confined to the oral cavity in which a long- term follow-up study showed that no other mucosae or the skin was involved (mean length of follow-up was
6.7 2 years) (Mobini et al, 1998). Furthermore, a Table1Treatmentforrecalcitrantoralmucousmembranepemphigoid(OP) Author(year)CountryPatientsTreatmentDurationoftreatment(months)PrevioustreatmentResponse 1 Bohnetal(1999)Germany1PlasmapheresisplusCYCIV(12mgkg)followedNADDS,Pred,topicalsteroidsCR 1byoralCYC(150–200mgday) 1 AhmedandColonUSA8IVIg1–2mgkgpercycleMean:32.9(range:26–42)DDS,topicalsteroids100%CR (2001) 11AhmedandColonUSA3Pred(40–80mgday)plusMTX(25mgweek)or 1(2001)AZA(150mgday)orAZAplusMTX 11(20mgday)plusCYC(150mgday)
Mean:38.8(range:33–45)DDS,topicalsteroids67%CR Sacheretal (2002)Germany1Etanercept25mgpersubcutaneouslypertwiceweekly plusPred(60mgday1NAPred,AZA,MMFCR Canizaresetal (2006)USA1Etanercept25mgpersubcutaneouslypertwiceweeklyNATopicalsteroidsCR Seguraetal (2007)Spain2IVIg2mgkg1 percyclealoneorplusPredand CYCNADDS,Pred,AZA,CYCPR Carrozzoetal (2008)Italy1MMF(2gday1)plusMino(200mgday1)6Pred,AZA,topicalsteroids, topicaltacrolimusCR Kasperkiewicz etal(2011)Germany1RTX(375mgm2perweekfor4consecutiveweeks)NADDS,PredCR LeRoux-Villet etal(2011)France3RTX(375mgm2perweekfor4consecutiveweeksNADDS,CYC,IVIg,Pred, Aza,MMF100%CR CYC,cyclophosphamide;NA,notavailable;DDS,dapsone;CR,completeresponse;PR,partialresponse;IVIg,intravenousimmunoglobulin;MTX,methotrexate;AZA,azathioprine;MMF,mycopheno- latemofetil;Mino,minocycline;RTX,rituximab.
36
long-term follow-up study of a large cohort of 70 patients with OP (comprising 51 patients already analyzed in previous studies) showed that no other mucosae or the skin was involved during the course of the disease (mean length of follow-up was 9.1 years) (Malik et al, 2007).
By contrast, two independent groups reported that MMP patients with exclusive oral involvement at disease onset could later on develop ocular lesions with an inci- dence ranging from 0.03 to 0.07 per person-year over 5 years of follow-up (Thorne et al, 2004; Higgins et al, 2006).
Recently, 20 patients with OP followed up for almost 3 years were immunologically characterized. All patients included in this study had had exclusively oral lesions without scarring during the entire follow-up period (Calab- resi et al, 2007).
Notably, a recent research suggests that patients with OP, with antibodies to integrin alpha6 (see below), may have a possible reduced relative risk for developing cancer compared to anti-laminin 332-positive patients (Egan et al, 2001; Matsushima et al, 2004; Malik et al, 2007).
Although in several long-term follow-up studies, MMP with exclusive oral involvement does not develop lesions in other sites during the course of the disease and also seems to be often associated with a good prognosis, the categorization of OP as a separate entity should be restricted on the homogeneous immunological and immu- nopathological features of a specific MMP subset. Notably, the identical genetic predisposition of having HLA- DQB1*03:01 allele (formerly known as DQB1*0301) could lead to OP phenotype as well as ocular MMP (OCP) and other MMP (Delgado et al, 1996; Carrozzo et al, 2001; Carrozzo et al, 2013).
From an immunological point of view
Circulating autoantibodies (IgG and rarely IgA) can be detected in MMP by indirect immunofluorescence (IIF) on normal or salt-split human skin. Patients with MMP con- fined to the oral cavity often do not have circulating IgG antibodies (Scully et al, 1999; Chan et al, 2002). Immu- nostaining results assessed by DIF and IIF appeared simi- lar between MMP with the involvement of multiple sites and those with exclusive oral involvement (Carrozzo et al, 2004).
Immunoblotting (IB) and enzyme-linked immunosorbent assay (ELISA) have simplified the diagnostic process and have identified novel protein targets recognized by autoan- tibodies of MMP (Zillikens et al, 1997; Schumann et al, 2000; Bhol et al, 2001; Lee et al, 2003; Mariotti et al, 2004). However, the characterization of target antigens of humoral immune response in OP has showed some discor- dant results.
A first study in a group of six MMP patients with disease limited to the oral cavity (not in all) showed circu- lating autoantibodies reacting against a 168-kDa oral mucosal protein (Ghohestani et al, 1996). Further studies neither confirmed this finding nor characterized the unknown target antigen.
Several recent studies, all performed in the same labora- tory, subsequently demonstrated that OP circulating autoantibodies target the a6 integrin subunit in a region
between fibrinogen repeats III and IV and are capable of inducing a separation of the epithelium from the basement membrane of normal human buccal mucosa in organ cul- ture (Bhol et al, 2001; Rashid et al, 2006a,b; Mignogna et al, 2006). Absorption studies showed that OP sera reacted exclusively against a6 integrin subunit and from bovine or human gingiva. The OP autoantibody anti-a6 integrin was absent in patients in a prolonged remission and not detected in the sera of patients with MMP involv- ing multiple mucosal membranes. In addition, no serologic reactivity to BP antigens or to other currently recognized MMP antigens has been reported (Chan et al, 1993;
Mobini et al, 1998; Rashid et al, 2006b). A further follow-up study indicates that the extent and severity of the oral disease correlates with the titer of antibody against a6 integrin (Sami et al, 2002b). Finally, an impor- tant limitation of all these studies remains the lack of in vivo functional data in inducing blisters in living animals.
In contrast to these results, two independent studies on MMP and OP show that their sera contain IgG antibodies to the two major BP antigens, BP180 and BP230, regard- less of their distinct clinical presentations (Egan et al, 1999; Carrozzo et al, 2004). In addition, a more recent study showed that 75% of a cohort of 20 OP patients without scarring phenotype had circulating autoantibodies against BP180 antigen, supporting the notion that this molecule represents a major target antigen in patients with OP (Calabresi et al, 2007). Similar results were also obtained in a study on a large cohort of MMP with scar- ring phenotype and involvement of various mucosal sites (Oyama et al, 2006). Most of the autoantibodies from OP patients were directed against epitopes in the BP180 ecto- domain, similarly to what reported for MMP with multiple mucosal sites (Balding et al, 1996; Bedane et al, 1997;
Murakami et al, 1998). In addition, OP sera displayed a low frequency of reactivity against the immune-dominant region of BP180 termed NC16A (45%), as previously reported for patients with MMP (Murakami et al, 1998;
Schmidt et al, 2001). In contrast with previously reported data, none of OP sera reacted against a 120-kDa protein by immunoblotting on keratinocyte extract, suggesting the absence of autoantibodies against the a6 integrin subunit (Carrozzo et al, 2004; Egan et al, 1999; Calabresi et al, 2007). However, the lack of correlation analysis between disease severity and anti-BP180 reactivity during the course of OP disease and experiments on mouse model to establish the possible pathogenic role of these specific autoantibodies are the major limitations of these studies.
Altogether these data show that the IgG reactivity against a6 integrin in oral mucosa could be a key immu- nological feature of patients with OP never detected in other MMP clinical subsets. However, the reactivity against BP180 is shared by both OP and MMP patients, regardless of whether they have the cicatricial phenotype.
In this context, a possible role of anti-a6 integrin autoanti- bodies in the pathogenesis of OP has been postulated.
Because integrin a6b4 interacts with laminin through the binding in the region between fibrinogen repeats III and IV to stabilize skin BMZ (Kikkawa et al, 2000), autoanti- bodies from patients with OP may perturb this binding possibly leading to BMZ separation.
37
Some controversial points of these studies remain to clarify. Firstly, although in both oral mucosa and skin there is the same variant of integrin subunit (alfa6A), OP sera reactivity against a6 integrin subunit was always detected by IB on extracts from a human prostate cancer cell line (DU145), human and bovine gingival, while it was never detected on keratinocyte extracts. Secondly, because a6 integrin is a hemidesmosomal component of mucosa and skin, autoantibodies against integrin could induce lesions both in mucosa and in skin. In fact, a6b4 gene-knockout mice die shortly after birth having an extensive blistering of the skin and other stratified squa- mous epithelium (Georges-Labouesse et al, 1996), and patients with junctional epidermolysis bullosa carrying mutations in the integrin a6b4 genes manifest cutaneous blistering and pyloric atresia (Ashton et al, 2001). Con- trary to what is expected for a humoral response against a6 integrin, patients with OP possess mucosal lesions without skin involvement and organ culture model shows that OP sera were able to separate basement membrane zone in normal buccal mucosa showing no effect on cultured human skin.
Last but not least, the results on integrin a6 have to be duplicated by other independent groups.
Conclusion
In conclusion, it is still unclear whether patients with OP could be considered as a distinct subset of MMP with spe- cific clinical and immunological features such as (i) exclu- sive involvement of oral mucosa even after a long-term follow-up study, rarely scarring and typically associated with a good prognosis, and (ii) specific recognition of a6 integrin subunit. As discussed above and also below, long-term follow-up and therapeutic studies are scarce and available evidence seems controversial. Moreover, the lack of well-verified serologic markers and the absence of in vivo studies to definitively assess pathogenic activity of OP autoantibodies do not lead to an unequivocal answer to the original question, and further studies are warranted.
Is mucous membrane pemphigoid curable?
If curability is defined as the patient will be in total remis- sion without the need for continuous treatment, the answer should be possibly no. Theoretically, MMP, as an autoim- mune disease, cannot be cured, as autoreactive T cells can be persisted in the patients’ lymphoid system for indefinite period of time and can always activate B cells to produce autoantibodies that cause the disease.
However, it has been shown that complete and long- lasting remission without treatment can be induced in a more serious than MMP blistering disease such as pem- phigus vulgaris (PV), in up to 75% of patients after 10 years (Herbst and Bystryn, 2000). Thus, a comparable control of the disease could be likely achieved in MMP.
Indeed, several studies report complete response of patients with MMP to treatment (see below Table 4).
However, available evidence provides information on remission at only a single point and they do generally not allow to determine how often remissions occur, how long it takes to achieve those, and how long remissions last.
Only 4 small studies with an average follow-up of 67 months (Table 2) show long-term outcome of patients with MMP, and three of them are from the same center (Ahmed and Colon, 2001; Sami et al, 2002a,b); thus, a single-center bias cannot be excluded.
In at least one of these trials, 15 MMP patients with multiple mucosal involvement experienced complete remission off therapy for an average of only 24 months (Sami et al, 2002a).
Thus, it is still uncertain whether treatment regimens that get MMP under control can be eliminated to allow patients to be in drug-free remission or they should be continuously administered in some capacities. Well-con- trolled, long-term studies, enrolling larger cohort of patients, are clearly warranted to better clarify the actual prognosis of MMP and to ascertain the pattern of remis- sion, if any, for this group of diseases.
What is the best therapeutic option for MMP?
Introduction
Because of the rarity of the disease (Bernard et al, 1995;
Zillikens et al, 1995; Rauz et al, 2005; Bertram et al, 2009), clinical trials of treatments for MMP are scarce and often enroll only a limited number of patients with hetero- geneous entities. Indeed, MMP is highly variable and does not have a predictable natural history. In some patients, the disease is localized and has a slowly progressive course without complications; in others, it causes severe morbidity (Scully et al, 1999).
A wide range of treatment modalities has been employed in MMP (Table 3), but randomized controlled trials are scarce (Kirtschig et al, 2003).
In 2002, an international consensus statement on treat- ment for MMP developed by a group of experts, mainly from dermatology field, issued its opinion on therapeutic approaches (Chan et al, 2002). If the mucosal lesions are localized to oral cavity, topical corticosteroid and dapsone (DDS) should be the first line of medications. If more control is needed, a low dose of systemic corticosteroid or immunosuppressive agent (such as azathioprine [AZA] or mycophenolate mofetil [MMF]) could be added. If other mucosae are affected, more aggressive treatment options are needed, including immunosuppressive agents such as cyclophosphamide (CYC), AZA, and MMF (Chan et al, 2002). In cases of rapidly progressed diseases, particularly ocular disease, CYC was considered the best choice (Chan et al, 2002; Thorne et al, 2008). However, a Cochrane systematic review on MMP treatment, first published in 2003 (Kirtschig et al, 2003) and lastly updated in April 2005, found only limited evidence that MMP involving the eyes responds best to treatment with CYC combined with corticosteroids.
To address the above cited question, we have conducted a review of studies reporting medical intervention for MMP from the date of last update of the Cochrane review.
Methods
A search of the pertinent literature was performed by two authors (L.S.C and M.C.) using Medline/PubMed, limiting
38
the search to human clinical trials published in any lan- guage from May 1, 2005, to March 23, 2013, using the following terms: ‘mucous membrane pemphigoid OR cicatricial pemphigoid AND treatment OR Therapy’. Eligible studies were randomized controlled trials (RCTs) and uncontrolled and controlled, not randomized, therapeutic studies of MMP involving 5 or more participants who received medical intervention for MMP. The diagnosis should be confirmed in all cases by immunofluorescence study findings. The studies included in the last published version of the Cochrane review were also checked, as well as the reference lists from identified studies. When selective reporting was evident, studies were excluded.
When possible, the studies were then rated based on quality and potential for bias according to Richards (2009).
Results
Three RCTs (Foster, 1986; El-Darouti et al, 2011), all of uncertain quality (Richards, 2009), and 42 non-random- ized trials on the treatment for MMP were identified (Tables 4 and 5). Sixteen of these 45 studies, including all the 3 RCTs, investigated patients with predominantly OCP, 10 patients with predominantly OP, and the other 19 mixed patients. Sixteen of these studies, including one RCT (Foster, 1986) commented on sulfa drugs (DDS, sulfapyridine, sulfamethoxypyridazine, sulfasalazine), sometimes associated with other medications. Whereas most of the patients benefited from these medications to some extent, complete and permanent remissions were rare, the patients experienced commonly adverse effects (AE), and up to 33% discontinued the drug due to those AE.
Sixteen trials, including a RCT (Foster, 1986), employed different modalities of prednisolone, whereas another RCT (El-Darouti et al, 2011) used intravenous methylprednisolone. Generally, systemic corticosteroids were successful, mainly with various adjuvant drugs, although some studies reported the lack of efficacy of prednisolone alone (Nisengard and Rogers, 1987; Nayar and Wojnarowska, 1993). Systemic corticosteroids can commonly cause side effects but they are generally man- ageable and rarely need complete drug discontinuation.
Nine studies, including 2 RCTs (Foster, 1986; El-Daro- uti et al, 2011), used oral or intravenous cyclophospha- mide (CYC) in various dosages and with several other drugs. CYC seems particularly effective for aggressive OCP or recalcitrant MMP, particularly associated with prednisolone, but it causes AE in up to 77% of patients and leads to high rates of discontinuation.
Eight trials commented on azathioprine (AZA). In all but two (Bialasiewicz et al, 1994; Pasadhika et al, 2009), AZA was used as steroid-sparring agent and the results were usually positive. There is a scarcity of data on AE of AZA for MMP, but according to a large study on ocular inflammatory diseases (Pasadhika et al, 2009), the drug was discontinued for AE at a rate of 0.16 per person-year (95% CI, 0.11–0.22).
Seven trials assessed intravenous immunoglobulins (IVIg) also as a monotherapy, and the overall response rate was 100%. Impressively, IVIg were apparently never
Table2Long-termremissioninmucousmembranepemphigoid(MMP) Author(year)NSitesofinvolvementTreatmentMeanduration oftreatment(months)Durationof remission(months)Totalduration offollow-up(months)Maintenancetreatment Ahmedand Colon(2001)10Mouth(N=10)IVIg,MTX,Pred,AZA33Mean:14 (range:11–18)61(range:51–74)Unclear Samietal(2002a)15Mouth(N=14); eyes(N=13); nose(N=7); esophagus(N=6); pharynx(N=3); vagina(N=2),larynx (N=2);anal(N=1)
Pred,DDS,AZA,MTX,CYC, Doxy,Tacro,Tetra,IVIg103Mean:24 (range:12–72)127(range:50–248)None Samietal(2002b)7Mouth(N=7)IVIg,Triamc27Mean:22 (range:15–36)59(range:48–80)Unclear LeRoux-Villet etal(2011)13UnclearRTX,DDS,SAZ,MMF,Pred3Mean:17 (range:3–46)22(range:5–49)DDS,SAZ CYC,cyclophosphamide;DDS,dapsone;IVig,intravenousimmunoglobulin;MTX,methotrexate;AZA,azathioprine;MMF,mycophenolatemofetil;Doxy,doxycycline;SAZ,sulfasalazine;Tacro,tacroli- mus;Tetra,tetracycline;Triamc,triamcinoloneacetonide;RTX,rituximab.
39
discontinued due to side effects and they could control MMP better than other conventional therapies (Letko et al, 2004).
Various topical corticosteroids (clobetasol propionate, fluocinonide, fluocinolone acetonide) have been primarily used in five studies, mostly involving patients with OP, and with apparently very positive results and virtually no drop-out (Table 4).
Five studies (Nayar and Wojnarowska, 1993; Poskitt and Wojnarowska, 1995; Reiche et al, 1998; Carrozzo et al, 2009; Chaidemenos et al, 2011) commented on cycline family of drugs (mainly minocycline) with and without nicotinamide, the results of which are controver- sial with a discontinuation rate up to 67%.
Five trials analyzed mycophenolate mofetil (MMF) effi- cacy with general positive results. MMF has also been used without concurrent systemic corticosteroids in OCP and OP (Zurdel et al, 2001; Ingen-Housz-Oro et al, 2005;
Carrozzo et al, 2008) with promising results, and it is apparently safer than CYC.
Two trials used rituximab (RTX) in particular recalci- trant MMP cases, and this drug showed encouraging results. It should, however, be emphasized that two patient died as a result of RTX treatment (Le Roux-Villet et al, 2011).
Treatment with colchicine (Chaidemenos et al, 2011), cyclosporine (Kacßmaz et al, 2010), and methotrexate (Gangaputra et al, 2009) has been described in single trials (Table 4).
Discussion
Comparing to the latest version of the Cochrane review on treatment for MMP, we found 9 more studies but the amount of evidence to determine the best treatment for this disease remains scarce. Only three small RCTs were found, and in all of them, allocation concealment was unclear. Many of the non-randomized studies are small, retrospective case series combining a wide range of
medications at different dosages. The lack of uniform outcome measures was another drawback. The largest cohorts are from ophthalmologists but these are mainly focused on ocular effects and commonly exclude other mucosal or skin lesions, thus making the applicability very limited.
Anti-inflammatories
Some medications such as tetracycline or sulfa drugs that were deemed promising (Kirtschig et al, 2003) are now under serious scrutiny for their safety profile (Wertheim et al, 2006; Hegarty et al, 2010; Carrozzo et al, 2009).
Cycline group of medications can rarely cause clinical remission, have little effect on ocular disease, and can cause serious and frequent adverse effects, particularly minocycline (Carrozzo et al, 2010). Sulfa drugs, particu- larly dapsone (DDS), have been widely employed in MMP, but still their efficacy is unclear because of the lack of good-quality RCTs. DDS therapy may cause a variety of adverse effects, which may be categorized as pharma- cologic, dose dependent, and allergic, or idiosyncratic reactions (G€urcan and Ahmed, 2009). Some degree of anemia is common using DDS but also severe adverse effects such as meta-hemoglobinemia (Kirtschig et al, 1998), agranulocytosis (Raizman et al, 1994), DDS hyper- sensitivity syndrome (Risse et al, 1994), and peripheral neuropathy (Foster, 1986) have been reported in patients with MMP. AE are supposed to be dose related and mostly not serious at daily dose below 100 mg, but the evidences are controversial (Table 3). In a recent compre- hensive review on DDS in bullous disorders, 41% of the patients with MMP experience AE and overall 14% of the treated patients discontinued DDS due to AE (G€urcan and Ahmed, 2009). Other sulfa drugs such as sulfamethoxypy- ridazine (SMXP) have been reported to be of value and better tolerated than DDS in the treatment for MMP, but they also can cause potentially fatal AE such as allergic alveolitis (Steinfort et al, 1989; McFadden et al, 1989).
Table 3 Treatment modalities for mucous membrane pemphigoid (MMP)
Surgical Topical Systemic
Low-energy
laser phototherapy Amniotic membrane transplantation Cryotherapy
Cultivated oral mucosal
epithelial transplantation (COMET) Keratoprosthesis
Corticosteroids
Triamcinolone acetonide Betamethasone valerato Beclomethasone dipropionate Budesonide
Clobetasol propionate Fluocinolone acetonide Fluocinonide Calcineurin inhibitors
Tacrolimus Cyclosporine Antibiotics
Mitomycin C
Corticosteroids Methylprednisolone Prednisolone
Other immunosuppressants Azathioprine
Cyclophosphamide Cyclosporine Leflunomide Methotrexate Mycophenolate mofetil Biologics
Etanercept/
infliximab Daclizumab Rituximab
Sulfonamides Dapsone Sulfapyridine
Sulfamethoxypyridazine Sulfasalazine
Tetracyclines Tetracycline Doxycycline Minocycline Other
Colchicine Interferona-2b Intravenous Ig Nicotinamide Immunoadsorption Pentoxifylline Plasmapheresis Thalidomide 40
Table4MMP:non-randomizedstudies Author(year)NTreatmentsTrialdurationOutcomeandadverseevents(AE)Comments Rogersetal (1982)24(oral[OP]and ocular[OCP])75–200mgday1 DDSplusPred/Aza/CYC atvariousdosagesin14patientsNA20(83%)responded;2(8%)inprolonged remissionbut1recurredandtreatedwith Pred+AZAAEin9(37.5%); discontinuedin4(16.7%) In4patients,DDSwasnot effectiveandthey respondedtoPred+Aza Fosteretal (1982)26OCP1–2mgkgday1CYC+20–80mgday1 Pred(Group1)vsPred(Group2)vs topicals(Group3)
3months14(78%)inGroup1responded;allpatients inGroups2–3hadprogressionofthe disease.AEin3/18(17%)discontinued becauseofgastrointestinalupset Norandomizedcontrolled study Nisengardand Rogers(1987)44OP(allwith desquamative gingivitis)
Fluo0.05%orFluo0.01%(N=17);DDS (25–150mgday1 )orSulfa (0.5–1.5gday1)(N=16);Pred(unclear dosage)(N=3);notreatment(N=8) NA31(70%)responded:15(88%)ontopical steroids,15(94%)onDapor sulfapyridine,and1(12%)withno treatment.NoresponsewithPred AE:NA NotstatedPreddosage Rogersand Mehregan(1988)77(16OP,30OCP, and31MMP)150mgday1 DDSor1.5–3gday1 Sulfa12weeks (minimum)15OP(94%),21MMP(68%),and26 (87%)improved;unclearifonremission AE:NA
Follow-upofRogersetal (1982) McFaddenetal (1989)15OP,MMP500–1500mgSMXP3months10(67%)responded 1(7%)inremission AE:1(allergicalveolitis)
Allergicalveolitiscanbe life-threateningandhas beenreportedbeforeasa resultofSMXPtherapyin MMP Mondino(1990)139OCP1.5mgkgday1CYC(N=13); 1.5mgkgday1 CYC+20mgday1 Pred(N=17);1.5mgkgday1Aza (N=10);60–80mgday1Pred(N=11); combinedtreatment(N=51);notreatment (N=35)
NAPredbetteroptionforstages1–2disease; Pred+CYCoverallbetterthanother modalities.Non-treatedseemstohave moreprogression AE:NA
Uncleardifferencebetween severalmodalities Tauberetal (1991)117OCP2mgkgday1DDS(N=59)plusPred (N=8)vs2mgkgday1CYC(N=25, additionalPredin23?)vs2mgkgday1 Aza(N=23,additionalPredin2?)
NADDSwasthemosteffectiveinitialagentfor modestlyactiveOCP;Cyclowasthemost effectiveinitialchoiceforhighlyactive cases AE:NA
Nosignificantdifferences werefoundcomparing progressionrates Lameyetal (1992)50(40OP,9OCP, and1MMP)Fluo(N=19)vsFluo+othertopicalsvs systemicsincludingPred 40mgday1+topicals(N=14)vs Pred+Aza50mgday1(N=3)vsDDS 100mgday1(N=2)vsPred+CYC 50mgday1(N=1)
1–48weeksAll(100%)respondedtoFluo,Fluo+other topicals,AzaandPred+Cyclo;allbut onerespondedtoPred+topicals(93%) andhalf(50%)toDDS+topicals.Unclear remissionratesasjustmentioned ‘asymptomatic’AE:NA Patientstreatedin2different countries(UKandUSA). Nofurtherdetailsprovided Fernetal(1992)5OCP(2withoral lesions)50–150mgday1 DDSNAAll(100%)respondedandunderwent remissionbutrelapsedoffDDS AEin2(40%)(jaundiceandhemolysis)
Onepatientneededsystemic steroidswhensevere cornealandscleral involvementoccurred.Some patientshadalsotopical antibioticand corticosteroids (continued)
41
Table4(continued) Author(year)NTreatmentsTrialdurationOutcomeandadverseevents(AE)Comments Nayarand Wojnarowska (1993) 48MMP40–60mgday1Pred(N=15); 50–150mgday1DDS(N=14); 100–150mgday1Aza(N=9); 50–100mgday1Mino(N=10); 0.5–1gday1SMXP(N=1)
NA5(30%)onPred,7(50%)onDDS/SMXP, and2(20%)onMinoresponded AE:unclear;6(43%)discontinuedDDS; manydiscontinuedAza
Inallgroups,additional topicalorlow-doseoral corticosteroidsmayhave beenused Bialasiewiczetal (1994)9OCP120–150mgday1 Aza+nasalmucosal graftNA9(100%)responded;unclearifinremission AE:NARecurrenceofsymblepharon in2whenAZAstopped Poskittand Wojnarowska (1995)
7(6withorallesions; 5MMPand2OP)50–100mgday1 Mino3–39months (average: 10months) 6(86%)responded(symptomatic improvement);noneinremission AE:6(86%),mainlypigmentation Patientswerealsotaking topicalandsystemic steroids Carrozzoetal (1997)11(allwithoral lesions;9OPand2 MMP)
Clob0.05%in4%hydroxyethylcellulose gel(N=8);Clob+prednisone 25–100mgday1(N=3) 2–27monthsAll(100%)responded;6(54%)complete remissionafter5.7monthsonaverage AE:4(36%);nonediscontinued
PatientstakingClobwere alsogivenchlorhexidine 0.12%andmiconazolegel Reicheetal (1998)8MMP(1OP)100mgday1Mino+nicotinamide 2.5–3gday1NA7(87%)responded;nocompleteremission AEin3(37%);1(12.5%)discontinuedOnepatientused oxytetracycline1gday1 . Durationoftetracycline treatmentunclear Carboneetal (1998)6(allwithoral lesions;3OPand3 MMP)
50–100mgPred20–80daysAll(100%)responded;2(33%)complete remissionAE:4(67%);nonediscontinuedTwocompleteremissions wereachievedusingPred 100mgday1 Ciarroccaand Greenberg (1999)
20MMP(15OP)Fluo0.05%(N=9);Fluo+50– 175mgday1DDS(N=11)NA20(100%)responded;9(45%)complete remission:2(10%)onFluoaloneand7 (35%)onDDS+FluoAE:unclear;2 (10%)discontinuedDDS 11patientswhofailtobe controlledbytopical steroidsweresuccessfully treatedaddingDDS FosterandAhmed (1999)10(6OCP,4MMP, 3withorallesions)IVIg2–3gkg1percycleover3days, every2–6weeks,increasedbyresponse (meanof19cycles)
18monthsAll(100%)responded;9(90%)in remissionbutundertreatmentAE:noneAllpatientsdidnotrespond toDDS,Pred,Aza,Cyclo, MTX,topicals.Allcontinue totakePred+ immunosuppressivesduring trial Thornhilletal (2000)25MMP(8OP)1gday1SMXP0.1–26months22(88%)responded;5(20%)complete remissionAEin6(24%),3(12% discontinuedSMXP)
Nointention-to-treatanalysis. Threepatientswerealso takingPredorPred+Aza AhmedandColon (2001)20OPIVIg1–2mgkg1percycleover3days every3–4weeks,increasedto6,8,10, 12,14,16weeks(mean:18cycles)(Group 1,N=8)vsPred0.5–1mgkgday1 plus immunosuppressants(Group2,N=12)
26–42months (Group1)33– 51months (Group2) AlleightpatientsinGroup1(100%) completeremissionvs6(50%)inGroup2 AE:3(37%)Group1;12(100%)inGroup 2;nonediscontinuedinbothgroups
Allpatientswere unresponsivetoDDSand topicalcorticosteroids Doanetal(2001)9OCP(3withoral and1withskin lesions)
Sulfa1–4gday1NA6(66%)responded;noneonremissionAE: 3(33%),alldiscontinuedAllwithDDS-relatedAE2 patientshadalsoCYC (continued) 42
