|
|
Original article
Annales de dermatologie et de v?n?r?ologie 2001; 128: 220-223
? Masson, Paris, 2001Detection of clonal T-cell receptor gamma gene rearrangement with the use of PCR-DGGE technique in the diagnosis of erythroderma
N.?Cordel?(1), B.?Lenormand?(3), P.?Courville?(2), P.?Lauret?(1), P.?Joly?(1)
(1)Clinique Dermatologique, Unit? Inserm U519, H?pital Charles Nicolle, 76031 Rouen Cedex. Groupe Fran?ais d'Etude des Lymphomes Cutan?s (GFELC).
(2)Laboratoire d'Anatomo-Pathologie, Unit? Inserm U519, H?pital Charles Nicolle, 76031 Rouen Cedex. Groupe Fran?ais d'Etude des Lymphomes Cutan?s (GFELC).
(3)Laboratoire d'H?matologie, Unit? Inserm U519, H?pital Charles Nicolle, 76031 Rouen Cedex. Groupe Fran?ais d'Etude des Lymphomes Cutan?s (GFELC).Tir?s ? part : P.?JOLY , at the above address. E-mail: Pascal.Joly@chu-rouen.fr
SUMMARY
Background
It is often difficult to establish the etiological diagnosis of erythroderma because clinical findings and immunohistology cannot always distinguish between lymphomatous erythroderma and inflammatory erythroderma. The purpose of this work was to assess the contribution of PCR-DGGE for detecting clonal T-cell receptor gamma gene rearrangement to the etiological diagnosis of erythroderma.
Patients and methods
The following inclusion criteria were used: patient with erythroderma; skin biopsy for histologic study, immunophenotyping and molecular biology; minimal follow-up of 12 months after initial diagnosis. Thirty patients were included from May 1, 1995 to November 30, 1998. Histology slides were reread by one of the authors blinded to other data who classed them in three categories: probable lymphoma, probable inflammatory disease, uncertain diagnosis. Molecular data were also analyzed in the same blinded manner. Immunohistology diagnosis was compared with the molecular data and the final diagnosis retained from clinical, histological and molecular findings as well as the disease course to last follow-up (November 1, 1999) after a mean 12???18 months follow-up.
Results
Eight biopsies were classed as probable lymphomas; a T-cell clonal rearrangement of the TCR genes was detected in 7/8 cases. The one sample with no detectable T clone was a drug-induced S?zary pseudolymphoma. The histologial classification identified 16 cases of probable inflammatory disease; no clonal rearrangement of the TCR genes was found in these cases. One of these patients had fungoid mycosis treated with caryolysin for three months and developed treatment intolerance at the time of the skin biopsy. For six biopsies the histological diagnosis was "uncertain"; a clonal rearrangement of the TCR genes was found in 2/3 of the fungoid mycosis cases and in none of the three cases of toxic dermal reactions.
Discussion
This study demonstrated the contribution of genotypic analysis with PCR-DGGE to the diagnosis of erythroderma. Monoclonal TCR gene rearrangement was detected in 9/11 (82 p. 100) of the patients with lymphoma and in 0/19 of the patients with an inflammatory dermatosis. The etiological diagnosis of erythroderma is an excellent indication for molecular stud of skin biopsies with PCR-DGGE.
R?SUM?
R?arrangement des g?nes du r?cepteur des lymphocytes T cutan?s par la technique de PCR-DGGE pour le diagnostic ?tiologique des ?rythrodermies.
N.?Cordel, B.?Lenormand, P.?Courville, P.?Lauret, P.?JolyIntroduction
Le diagnostic ?tiologique des ?rythrodermies est souvent difficile car les donn?es des examens clinique et immuno-histologique ne permettent pas toujours de distinguer les ?rythrodermies lymphomateuses des ?rythrodermies inflammatoires. L'objectif de ce travail a ?t? d'?valuer l'int?r?t de l'?tude du r?arrangement des g?nes du r?cepteur des lymphocytes T cutan?s par la technique de PCR-DGGE dans le diagnostic ?tiologique des ?rythrodermies.
Malades et m?thodes
Les crit?res d'inclusion dans l'?tude ont ?t? les suivants: malade ayant une ?rythrodermie; mat?riel de biopsie cutan?e disponible pour une ?tude histologique, immunoph?notypique et mol?culaire; suivi minimum de 12 mois apr?s le diagnostic initial. Trente malades ont ?t? inclus entre le 1er mai 1995 et le 30 novembre 1998. Les lames histologiques ont ?t? relues en insu par l'un des auteurs et class?es en trois groupes: lymphome probable; pathologie inflammatoire probable; diagnostic incertain. Les donn?es mol?culaires ont ?galement ?t? analys?es en insu. Les diagnostics immunohistologiques ont ?t? confront?s aux r?sultats mol?culaires et au diagnostic d?finitif pos? ? partir des donn?es cliniques, histologiques, mol?culaires et ?volutives au moment de l'?tude (1er?novembre 1999) apr?s un suivi moyen de 12???18 mois.
R?sultats
Huit biopsies ont ?t? class?es histologiquement comme probablement lymphomateuses; un r?arrangement clonal des g?nes du TCR a ?t? d?tect? dans 7/8 cas. Le seul pr?l?vement ne comportant pas de clone T d?tectable correspondait ? un pseudolymphome m?dicamenteux ? type de pseudo syndrome de S?zary. Seize biopsies ont ?t? class?es histologiquement comme "pathologie inflammatoire probable"; aucun r?arrangement clonal des g?nes du TCR n'a ?t? mis en ?vidence. Un de ces malades avait un mycosis fongo?de trait? par caryolysine depuis 3 mois et avait une intol?rance au traitement au moment du pr?l?vement cutan?. Six biopsies avaient un "diagnostic histologique incertain"; un r?arrangement clonal des g?nes du TCR a trouv? dans 2/3 des cas de mycosis fongo?de et dans aucun des 3 cas de toxidermie.
Discussion
Cette ?tude souligne l'int?r?t de l'analyse g?notypique par la technique de PCR-DGGE pour le diagnostic ?tiologique des ?rythrodermies puisqu'un r?arrangement monoclonal des g?nes du TCR a ?t? d?tect? chez 9/11 malades (82 p. 100) atteints de lymphome et chez 0/19 malades ayant une dermatose inflammatoire. Le diagnostic ?tiologique des ?rythrodermies est une excellente indication de l'?tude mol?culaire cutan?e en PCR-DGGE.
The etiological diagnosis of erythroderma and in particular the distinction between a lymphomatous and an inflammatory disease is often difficult. Clinical examination generally reveals little information, immunohistological analysis is not always contributory, and the majority of cutaneous T-cell lymphomas express a mature phenotype without loss of markers. Conversely, loss of surface antigens such as the CD7 may be observed in inflammatory infiltrates [1, 2, 3, 4, 5, 6, 7, 8, 9]. Genotype analysis is a supplementary criterion of diagnosis, since identification of a monoclonal rearrangement of the genes encoding the T-cell receptor (TCR) confirms the presence of a potentially neoplastic monoclonal lymphoid population within the proliferation studied [10, 11, 12, 13, 14, 15]. Several studies have prompted interest in genotype analysis of skin infiltrates using the polymerase chain reaction/denaturing gradient gel electrophoresis (PCR-DGGE) technique to distinguish erythroderma due to lymphoma from inflammatory erythroderma, but these studies only evaluated patients suffering from S?zary's syndrome or actinic pseudolymphoma [16, 17, 18]. The aim of this study was to assess the benefit of studying TCR gene rearrangement using the PCR-DGGE technique for the diagnosis of erythroderma of varying etiology.
Patients and methods
The criteria for inclusion in this study were: 1) clinical diagnosis of erythroderma reported in the case report form and validated by retrospective examination of the photographs; 2) skin biopsies available for histological, immunophenotyping, and molecular studies; and 3) a 12-month follow-up after diagnosis. Thirty patients from one center were studied for a period of 42 months between May 1, 1995 and November 30, 1998.
Histological and immunostained slides were re-evaluated by a blinded author (Ph.C.) in November 1999 and assigned to one of three groups (probable lymphoma, probable inflammatory disease, and uncertain diagnosis) based on immunohistological characteristics. The molecular study was conducted on frozen skin samples according to standard PCR-DGGE techniques with a sensitivity threshold of 1 to 3?p.?100 [19]. Molecular data were analyzed and interpreted blindly by another author (B.L.). Immunohistological diagnoses were then compared with the molecular results, and a final diagnosis was made at the time of the study (November 1, 1999) based on all data and on evolution of the disorder (mean follow-up 12???18 months).
Results
Table?I presents the results of molecular analysis and the final diagnosis made at the time of the study for each of the histological groups (probable lymphoma, probable inflammatory disease, and uncertain diagnosis). The biopsies of 8 patients were classified histologically as probable lymphoma; the pathologist proposed a diagnosis of epidermotropic T-cell lymphoma in 7 cases and in one case pleomorphic intermediary and large T-cell lymphoma. Analysis of monoclonal T-cell receptor gene rearrangement (TCR) was positive in 7/8 cases. The final diagnosis retained in these seven patients was T-cell lymphoma, with 6 cases of epidermotropic T-cell lymphoma and 1 case of leukemia cutis with T pro-lymphocytes. In 7 cases a T-cell clone was detectable by PCR. In one patient the pathologist had classified the lesion histologically as probable lymphoma, but no cutaneous T-cell clone was detected by PCR. The diagnosis retained in this patient was drug-induced pseudo-S?zary's syndrome. The erythroderma had developed after introduction of allopurinol and fosinopril. It was accompanied by hypereosinophilia at 5,000/mm3, and after suspension of the imputed drugs the eruption cleared. No relapse was reported after 36 months.
Sixteen biopsies were histologically classified as "probable inflammatory dermatitis". The diagnosis proposed by the pathologist was drug-induced eruption in 8 cases and eczema in the other 8 cases. Study of TCR gene rearrangement was negative in all cases. The final diagnosis retained at the time of the study based on the clinical evolution was inflammatory dermatitis in 15/16 cases and mycosis fungoides (MF) in one case. The patient suffering from MF had been treated for 3 months with topical nitrogen mustard (Caryolysine?) and presented with intolerance to topical treatment at the time of the biopsy.
The pathologist distinguished six biopsies that did not permit clear immunohistological diagnosis. Study of TCR gene rearrangement in the skin was positive in 2 of these cases and negative in four. The final diagnosis retained was MF in 3 cases and drug-induced eruption in 3 cases. Comparison of these diagnoses with the molecular data revealed that 2 of the 3 cases of MF had T-lymphocyte clones detectable by PCR whereas none of the 3 cases of drug-induced eruption had detectable clonal TCR rearrangement.
Discussion
The etiological diagnosis of erythroderma is often difficult for the clinician and the pathologist because of the lack of specificity of histological and clinical signs [20]. Several studies have highlighted the use of genotype analysis to distinguish the erythroderma of inflammatory origin from those of lymphomatous origin, but these essentially concerned patients with S?zary's syndrome. Our study underscores the benefits of the PCR-DGGE technique for etiological diagnosis of various erythrodermal disorders. Indeed, in the group of biopsies classified histologically as probable lymphomas, molecular study confirmed the diagnosis of lymphoma in 7 of 8 cases.
For the single patient classified histologically as probable lymphoma and whose skin did not contain a detectable clonal T-cell population by PCR, the diagnosis finally retained was drug-induced pseudolymphoma. This example emphasizes the fact that diagnosis of lymphoma in cases of dissociation between immunohistological and molecular results must be reconsidered. In the group of biopsies classified as probably inflammatory, molecular analysis confirmed the diagnosis in all cases except in one patient with MF, who was treated with nitrogen mustard for 3 months and who presented with intolerance to the treatment at the time of the biopsy. The value of molecular studies is particularly evident in the group of patients with uncertain immunohistological diagnosis. In this group, molecular clonality studies were positive in 2 of the 3 skin biopsies that were ultimately diagnosed with T-cell lymphoma and negative in the 3 cases of drug-induced eruption
The sensitivity of molecular analysis was 82 p. 100 in this series, with 100 p. 100 specificity. The moderate sensitivity of the PCR-DGGE technique used in this study emphasizes the question of the sensitivity threshold required for the clinicians in dermatology practice. In general, the clinician will employ a study that offers a compromise between sensitivity and specificity. A molecular biological technique such as the Southern blot is obsolete because of its lack of sensitivity. On the other hand, techniques such as nested PCR are not well adapted to routine diagnosis of lymphocytic infiltrates of the skin because of their low specificity, since they detect clones during oligoclonal inflammatory diseases [21]. It is therefore most important that the clinician know the sensitivity of the molecular biology test used in order to interpret the results correctly.
In conclusion, etiological diagnosis of erythroderma is an excellent indication for molecular PCR-DGGE skin analysis. Nevertheless, it is essential to approach these molecular data with the clinical, histological, and immunophenotyping data as well as with the clinical evolution in order to establish an accurate diagnosis.
To cite the present paper, use exclusively following reference: Cordel N, Lenormand B, Courville P, Lauret P, Joly P. R?arrangement des g?nes du r?cepteur des lymphocytes T cutan?s par la technique de PCR-DGGE pour le diagnostic ?tiologique des ?rythrodermies (full text in english on e2med.com/ad). Ann Dermatol Venereol 2001; 128: 220-3.
REFERENCES
[1] Willemze R, Kerl H, Sterry W, E Berti, L Cerroni, S Chimenti, et al. EORTC classification for primary cutaneous lymphomas. A proposal from the cutaneous lymphomas study group of the European organization for research and treatment of cancer. Blood 1997;90:354-71.
[2] Rijlaarsdam JU, Willemze R. Diagnostic et classification des pseudo lymphomes cutan?s. Ann Dermatol Venereol 1993;120:100-6.
[3] Wood GS, Weiss LM, Warnke RA, Sklar JL. The immunopathology of cutaneous lymphomas: immunophenotypic and immunogenotypic characteristics. Semin Dermatol 1986;5:334-5.
[4] Picker LJ, Weiss LM, Medeiros LJ, Wood GS, Warnke RA. Immunophenotypic criteria for the diagnosis of non Hodgkin's lymphomas. Am J Pathol 1987;128:181-201.
[5] Willemze R, Beljaards RC, Meijer CJLM. Classification of primary cutaneous T-cell lymphomas. Histopathology 1995;24:405-15.
[6] Van der Putte SCJ, Toonstra J, Van Wicken DF, Van Unnick JAM, Van Vloten WA. Aberrant immunophenotypes in mycosis fungoides. Arch Dermatol 1988;124:373-80.
[7] Smith NP. Histologic criteria for early diagnosis of cutaneous T-cell lymphomas. Dermatol Clin 1994;12:315-22.
[8] Ralfkiaer R. Controversies and discussion on early diagnosis of cutaneous T-cell lymphomas. Dermatol Clin 1994;12:329-34.
[9] Rijlaarsdam JV, Boorsma DM, De Haan P. The significance of leu 8 negative T-cells in lympho?d skin infiltrates: malignant transformation selective homing or T-cell activation? Br J Dermatol 1990;123:587-93.
[10] Bertness V, Kirsch IR, Hillis G, Johnson B, Bunn PA. T cell receptor gene rearrangements as clinical markers of human T-cell lymphomas. N Engl J Med 1985;313:534-8.
[11] Waldmann TA, Davis MM, Bongiovanni KF, Korsmeyer SJ. Rearrangements of genes for the antigen receptor on T-cells as markers of lineage and clonality in human lympho?d neoplasms. N Engl J Med 1985;313:776-83.
[12] Trainor K, Brisco M, Wan J, Neoh S, Grist S, Morley A. Gene rearrangement in B and T lymphoproliferative disease detected by the polymerase chain reaction. Blood 1991;78:192-6.
[13] Lessin S, Rook AH, Rovera G. Molecular diagnosis of cutaneous T-cell lymphoma: polymerase chain reaction amplification of T-cell antigen receptor beta chain rearrangements. J Invest Dermatol 1991;96:299-302.
[14] Kneba M, Bolz I, Linke B, Bertram J, Rothaupt P, Hiddeman W. Characterization of clone specific rearrangement T-cell receptor gamma chain genes in lymphomas and leukemias by the polymerase chain reaction and DNA sequencing. Blood 1994;84:574-81.
[15] Wood GS, Haeffner A, Dummer R, Crooks CF. Molecular biology techniques for the diagnosis of cutaneous T-cell lymphoma. Dermatol Clin 1994;12:231-41.
[16] Wood GS, Tung R, Haeffner, Crooks CF, Liao S, Orozco R, et al. Detection of clonal T-cell receptor gamma gene rearrangements in early mycosis fungoides/S?zary syndrome by polymerase chain reaction and denaturating gradient gel electrophoresis (PCR-DGGE). J Invest Dermatol 1994;103:34-41.
[17] Bachelez H, Bioul L, Flageul B, Baccard M, Moulonguet-Michau I, Verola O, et al. Detection of clonal T-cell receptor gamma gene rearrangements with the use of the polymerase reaction in cutaneous lesions of mycosis fungoides and Sezary syndrome. Arch Dermatol 1995;131:1027-31.
[18] Bakels V, van Oostveen JW, Preesman AH, Meijer CJ, Willemze R. Differenciation between actinic reticuloid and cutaneous T cell lymphoma by T cell receptor gamma gene rearrangement analysis and immunophenotyping. J Clin Pathol 1998;51:154-8.
[19] Theodorou I, Delfau-Larue MH, Bigorgne C, Lahet C, Cochet G, Bagot M, et al. Cutaneous T-cell infiltrates: analysis of T-cell receptor gamma gene rearrangement by polymerase chain reaction and denaturating gradient gel electrophoresis. Blood 1995;86:305-10.
[20] Botella-Estrada R, Sanmartin O, Oliver V, Febrer I, Aliaga A. Erythroderma: a clinicopathologic study of 56 cases. Arch Dermatol 1994;130:1503-7.
[21] Dippel E, Assaf C, Hummel M, Schrag HJ, Stein H, Goerdt S, et al. Clonal T-cell receptor gamma-chain gene rearrangement by PCR-based GeneScan analysis in advanced cutaneous T-cell lymphoma: a critical evaluation. J Pathol 1999;188:146-54.
Rechercher dans
? Masson, Paris, 2001
?Résumé
?Texte intégral PDF
Texte intégral précédent ? ? ? Texte intégral suivant
| Page d'Accueil | A Propos | S'Abonner | Recherche | Actualité | Infos légales English | Français |
