The goal of clinical research is to answer questions that affect the way in which we care for our patients. This process involves studying data from large numbers of patients with a particular disease and then applying the conclusions to an individual patient.
The path of this deductive journey is fraught with many land mines. We must identify these obstacles and continue forward to make the conclusions as applicable as possible to our patient.
The work by Abasq et al from Rouen University Hospital in France has assessed the predictive values of anti–desmoglein 1 and 3 (Dsg1 and Dsg3) antibody enzyme-linked immunosorbent assay (ELISA) values for the occurrence of relapses in the management of pemphigus. This work has provided significant answers that help the practitioner in treating a patient with pemphigus.
Predictably, it has created new questions. In this editorial, I review the basic questions that exist regarding the clinical uses of pemphigus serum antibody tests and assess how our current understanding of these questions affects the care of the individual patient. Lastly, I review how the work by Abasq et al has helped answer these questions.
WHAT ARE THE CLINICAL, HISTOLOGICAL AND IMMUNOPATHOLOGICAL CHARACTERISTICS OF PEMPHIGUS?
Patients with pemphigus foliaceus (PF) have cutaneous erosions and blisters without mucosal involvement and a histological split that occurs in the superficial epidermis predominantly in the granular layer. In contrast, patients with pemphigus vulgaris (PV) have mucosal erosions and may also have cutaneous lesions in approximately 50% of the cases. The histological split in the epidermis occurs in the suprabasalar area. The histopathological hallmark of pemphigus is acantholysis or separation of the epidermal cells, one from another. The disease can progress to large areas of involvement and be fatal secondary to fluid loss or secondary infection.
Beutner and Jordon initially described the presence of IgG that bound to the cell surface of keratinocytes as observed in serum samples taken from patients with pemphigus. Subsequently, the same authors showed that IgG was bound in vivo to the cell surface of keratinocytes in patients with pemphigus. Since that time, there has been an assumption that these antibodies are responsible for the histological acantholysis that produces clinical blistering in pemphigus. A number of subsequent studies using indirect immunofluorescence suggested that pemphigus serum antibody titers correlated with disease activity as reviewed by Beutner et al. This implied that studies of pemphigus antibody in the serum would allow clinicians to more accurately monitor the status of the underlying pathogenic process and predict remissions and exacerbations of disease as well as impending clinical response to therapy.
Over the past 20 years, our knowledge of the pathogenesis of pemphigus has steadily improved, and we are now at the point where we may be able to use serum studies in a patient with pemphigus as an essential component of management.
ARE ANTIBODIES THAT BIND TO THE CELL SURFACE OF KERATINOCYTES PATHOGENIC?
In 1982, Anhalt et al performed landmark experiments and were able to passively transfer the IgG fraction of plasma from patients with pemphigus to neonatal mice and produce the binding of IgG to the keratinocytes cell surface with resultant acantholysis and blistering of mouse skin. Solid scientific experimentation indicated that cell surface antibodies produced acantholysis. Because Anhalt et al included the entire IgG fraction of plasma from patients with pemphigus; they could not tell if individual clones of cell surface antibodies were pathogenic, just that the IgG fraction as a whole was pathogenic. Thus, the answer to this question is that some antibodies that bind to the cell surface of keratinocytes are pathogenic but some may not be.
Subsequently, immunochemical characterization of antibodies in pemphigus showed that PF and PV antigens were 160-kDa and 130-kDa glycoproteins, respectively. These proteins were found to be members of the cadherin family and were termed desmogleins. The PF and PV antigens are termed Dsg1 and Dsg3, respectively. These transmembrane proteins are responsible for the attachment of epidermal cells to each other.
ARE ANTIBODIES TO Dsg1 AND Dsg3 PATHOGENIC?
Amagai et al have shown that antibodies to Dsg1 and Dsg3 are necessary for passive transfer of pemphigus to neonatal mice. If antibodies to Dsg1 and Dsg3 are removed from the serum samples by immunoabsorption, sera from patients with PF and PV will not produce acantholysis in mouse models. Elegant studies using autoantigen knockout mice have shown that Dsg3 antibodies are pathogenic. In these studies, Dsg3–/– mice (which do not have self tolerance to Dsg3) were immunized with Dsg3. Lymphocytes from the immunized Dsg3–/– mice were then passively transferred to immunodeficient mice. This produced the phenotype of PV. There is little question that IgG antibodies to Dsg1 and Dsg3 can produce the phenotype of PF and PV.
It is likely that all antibodies to Dsgs are not equally pathogenic and that different clones of antibodies to various epitopes on Dsg molecules vary in their pathogenicity. It would be extremely difficult to clinically evaluate serum samples for antibodies to various Dsg antigens because serum from an individual patient contains the antibody product of many clones directed against a number of different epitopes on Dsg molecules. Amagai and coworkers have been able to analyze the pathogenicity of a number of clones derived from their mouse model of pemphigus and have established that there is a huge variance in the pathogenicity of antibodies to various epitopes on Dsgs. Some clones are highly pathogenic, while others have no identifiable pathogenicity. They have shown that the most potent pathogenic antibodies are located on the EC1 domain of the adhesive interface of the Dsg molecules.
These findings clearly indicate that the pathogenicity of antibodies to Dsg is related to the site of antigenic specificity on the Dsg molecule.
Other studies have examined the IgG subclass distribution in pemphigus. IgG4 subclass antibodies are believed to be pathogenic. Generally, IgG4 subclass antibody levels are elevated in patients with active disease, while IgG1 subclass antibodies are found in patients in remission as well as healthy relatives. IgG4 antibodies to Dsg may be a better method of monitoring patients. The answer to the current question is that some but not all antibodies to Dsg1 and Dsg3 are pathogenic.
ARE THERE OTHER ANTIBODIES THAT COULD BE PATHOGENIC IN PEMPHIGUS AND NOT BE MEASURED IN THE CURRENT Dsg1 AND Dsg3 ASSAYS?
Just because antibodies to Dsg are pathogenic does not mean that antibodies to other antigens on the cell surface of keratinocytes might not also be pathogenic. Grando has accumulated considerable evidence that indicates that antibodies to the acetylcholine receptor on the keratinocyte cell surface may be pathogenic in patients with pemphigus. The Dsg1 and Dsg3 assays would obviously not measure those antibodies. If antibodies to acetylcholine receptors alone were responsible for some cases of pemphigus, one would expect to see a certain number of cases of pemphigus with cell surface antibodies by indirect immunofluorescence and negative Dsg1 and Dsg3 assay results. This was not the case in the study by Abasq et al, and other studies have shown a high sensitivity for pemphigus for the Dsg1 and Dsg3 assays. Nonetheless, a role for antibodies to the acetylcholine receptor cannot currently be ruled out.
In addition, in IgA pemphigus, reactivity with other cell surface desmosomal cadherins termed desmocollins has been identified. This may well be the case in some cases of IgG pemphigus. Lastly, the recombinant Dsg protein used in the ELISA assay may not contain all of the epitopes present on Dsg in vivo. Conformational epitopes as well as epitopes formed during posttranslational modifications may be lost in the process of producing recombinant protein. Thus, the answer to this question is clearly that other potentially pathogenic antibodies may not be measured by the Dsg1 and Dsg3 assays.
WHAT IS THE UTILITY OF THE CURRENT ELISA ASSAY FOR Dsg1 AND Dsg3?
The currently used ELISA assay tests for reactivity with the Dsg molecules without differentiating between pathogenic and nonpathogenic epitopes. It also tests for all subclasses of IgG. Therefore, one would expect that nonpathogenic antibodies would be detected and mute but not eliminate the ability of the assay to monitor improvement and relapses in the disease. This was clearly the case in the article by Abasq et al. As Dsg antibody levels fluctuate or unidentified reasons or with therapy, it may be that there are few nonpathogenic antibodies present or that levels of pathogenic and nonpathogenic antibodies fluctuate in concert, in which case one would see that Dsg antibody levels as measured by ELISA correlate well with disease activity and provide all of the desired results, including predicting relapses and effectively monitoring disease activity.
In the case of patients with PF, the data from Abasq et al suggest that antibodies to Dsg1 have fewer of the confounding factors reviewed in the previous section and therefore correlate with relapses.
However, it may be that many nonpathogenic antibodies are present or that various clones are differentially expressed or suppressed, in which case we may see an individual patient with a high antibody level and little disease activity as Abasq et al have described in some of the PV cases.
Confounding factors have diluted the utility of the assay. A superior assay would measure antibodies limited to pathogenic epitopes and limited to pathogenic IgG subclasses. However, the technology for ELISA testing has not yet reached that degree of sophistication.
IS THE CURRENT ELISA ASSAY USEFUL IN MANAGING PEMPHIGUS IN AN INDIVIDUAL PATIENT?
The answer is yes in some cases. Falling antibody levels are usually a very good sign. Rising antibody levels are usually a bad sign. In some patients, such as the patients with PF noted by Abasq et al, there may not be many confounding factors, and the assay may be very useful. However, failure of change of antibody levels coincident with clinical improvement or worsening suggests that the assay may be measuring nonpathogenic antibodies or not be measuring all of the pathogenic antibodies. This was noted in some of the patients with PV in the study by Abasq et al.
My recommendation is that serial antibody testing to Dsg1 and Dsg3 by ELISA be performed in the individual patient with pemphigus and that the factors reviewed previously be taken into account when interpreting the result. I am confident that in the future there will be refinements of these assays and they will become progressively more useful in the individual patient.
Correspondence: Dr Zone, Department of Dermatology, 4A330 School of Medicine, University of Utah, 30 N 1900 E, Salt Lake City, UT 84132-2409 ( firstname.lastname@example.org). Financial Disclosure: None reported.
Abasq C, Mouquet H, Gilbert D; et al.
ELISA testing of anti–desmoglein1 and 3 antibodies in the management of pemphigus.
Beutner EH, Jordon RE.
Demonstration of skin antibodies in sera of pemphigus vulgaris patients by indirect immunofluorescent staining.
Proc Soc Exp Biol Med.
Beutner EH, Chorzelski TP, Kumar V.
Immunopathology of the Skin.
New York, NY: John Wiley & Sons; 1987.
Anhalt GJ, Labib RS, Voorhees JJ, Beals TF, Diaz LA.
Induction of pemphigus in neonatal mice by passive transfer of IgG from patients with the disease.
N Engl J Med.
Cell adhesion molecules as targets of autoantibodies in pemphigus and pemphigoid, bullous diseases due to defective epidermal cell adhesion.
6. Amagai M, Klaus-Kovtun V, Stanley JR.
Autoantibodies against a novel epithelial cadherin in pemphigus vulgaris, a disease of cell adhesion.
Amagai M, Hashimoto T, Green KJ, Shimizu N, Nishikawa T.
Antigen-specific immunoadsorption of pathogenic autoantibodies in pemphigus foliaceus.
J Invest Dermatol.
Amagai M, Hashimoto T, Shimizu N, Nishikawa T.
Absorption of pathogenic autoantibodies by the extracellular domain of pemphigus vulgaris antigen (Dsg3) produced by baculovirus.
J Clin Invest. 1994;94(1):59-67.
Pemphigus vulgaris and its active disease mouse model.
Curr Dir Autoimmun.
Tsunoda K, Ota T, Aoki M; et al.
Induction of pemphigus phenotype by a mouse monoclonal antibody against the amino-terminal adhesive interface of desmoglein 3.
Bhol K, Mohimen A, Ahmed AR.
Correlation of subclasses of IgG with disease activity in pemphigus vulgaris.
Bhol K, Natarajan K, Nagarwalla N, Mohimen A, Aoki V, Ahmed AR.
Correlation of peptide specificity and IgG subclass with pathogenic and nonpathogenic autoantibodies in pemphigus vulgaris: a model for autoimmunity.
Acad Sci U S A.
Ayatollahi M, Joubeh S, Mortazavi H, Jefferis R, Ghaderi A.
IgG4 as the predominant autoantibody in sera from patients with active state of pemphigus vulgaris.
J Eur Acad Dermatol Venereol.
Pemphigus in the XXI century: new life to an old story.
Yasuda H, Kobayashi H, Hashimoto T, Itoh K, Yamane M, Nakamura J.
Subcorneal pustular dermatosis type of IgA pemphigus: demonstration of autoantibodies to desmocollin-1 and clinical review.
Br J Dermatol.
ELISA Testing of Anti–Desmoglein 1 and 3 Antibodies in the Management of Pemphigus
Claire Abasq, Hugo Mouquet, Danièle Gilbert, François Tron, Vanessa Grassi, Philippe Musette, and Pascal Joly