- Email: [email protected]
Exploring Primary Cutaneous B-Cell Lymphoma by Microarray Technology
Exploring Primary Cutaneous B-Cell Lymphoma by Microarray Technology
COMMENTARY See related article on page 865
Reinhard Dummer, Mirjana Urosevic, Keith Hoek, Andreas H¡ner and Gˇnter Burg Department of Dermatology, University Hospital Zurich, Zurich, Switzerland
nization for Research and Treatment of Cancer (EORTC) classi¢cation distinguishes between indolent CBCLs (follicle center cell lymphoma, immunocytoma, marginal zone B-cell lymphoma), CBCLs with intermediate clinical behavior (large B-cell lymphoma of the leg), and the provisional entities (intravascular large B-cell lymphoma and plasmocytoma) (Santucci et al, 1991; Willemze et al, 1997). This classi¢cation scheme is still under debate. In this issue of the Journal of Investigative Dermatology, there is an interesting article from Storz et al giving us for the ¢rst time the opportunity to look into gene expression pro¢les of primary CBCLs (Storz et al, 2003). Using the unsupervised, hierarchical clustering method the authors explore the relationship among lymphoma, normal skin and tonsil samples, trying to identify the molecular signatures in lymphoma specimens that might relate to their normal counterparts in skin and tonsil. Clustering analysis reveals that the follicular center cell lymphoma (FCCL) samples group more or less together, implying that these tumors share a similar overall gene expression pattern. It is of note that two of the FCCL samples from the same patient taken two years apart did not cluster next to each other, although they remained in the same branch. Follow-up studies will elucidate whether this phenomenon is due to the changes of the malignant clone over time or to the sensitivity of the array technique itself. Di¡use large cell lymphoma (DLCL) and marginal zone lymphoma (MZL) samples fail to demonstrate a speci¢c clustering pattern. Primary DLCL, however, cluster adjacent to FCCL samples, whereas secondary DLCL cases are removed from these two groups, suggesting a biological and molecular distinctiveness. Subsequently, the authors focus on gene clusters associated with T- and B-cell lineages and development, delineating T-cell, Bcell germinal center (GC) and plasma cell ‘‘signatures,’’ respectively. Storz et al detected characteristic B-cell GC signature in tonsil, primary and secondary FCCL, and in primary DLCL, suggesting that these diseases are more closely related than expected from recent histological and immunohistological analyses. This may indeed be the case in some clinical situations where DLCL of the skin react similarly to FCCL. We have recently described a large cell lymphoma responsive to antibiotic treatment (Hofbauer et al, 2001), as are some primary cutaneous FCCL (Cerroni et al, 1997; Hofbauer et al, 2001). Interestingly, secondary DLCL lacked the expression of B-cell GC signature. Plasma cell signature on the other side could be identi¢ed in two out of ¢ve MZL with prominent plasmacytoid di¡erentiation. Although the study only comprises a small number of cases, Storz et al provide the ¢rst insight into the molecular pro¢les of most common subtypes of CBCL. As with nodal lymphomas, there is hope that with the help of high-throughput microarray technology, we will be able to characterize the nature of di¡erent primary cutaneous lymphoma entities and subsequently improve existing lymphoma classi¢cations.
B-cell lymphomas (BCLs) represent the most frequent lymphoid neoplasms. They include rare precursor lymphoblastic leukemia/ lymphomas, B-cell chronic lymphocytic leukemias, mantle zone lymphomas, marginal zone lymphomas, follicular lymphomas and most of the Hodgkin’s lymphomas (Sha¡er et al, 2002). Only a minority of B-cell lymphomas actually occurs in the skin (Burg et al, 1997; Dummer et al, 2000). The origin of human BCLs has been studied extensively using the various approaches of histology, immunohistochemistry, karyotyping, single cell analysis and somatic mutation study of variable-region genes (Sha¡er et al, 2001; Sha¡er et al, 2002). In order to understand the biology of BCLs, it is reasonable to classify them according to their presumed normal counterpart in the B-cell di¡erentiation process. Despite their origin in human B-cell development, BCLs still remain a controversial topic in terms of their precise classi¢cation and diagnosis. Standard immunohistochemistry and molecular biology has often failed to identify the malignant ‘‘clonal’’ B-cell populations. Even in Southern blot-positive cases, the origin or relationship of a malignant cell to any particular stage in B-cell development remains unclear. The germinal center cell reaction is essential for the education of a B lymphocyte since it modi¢es naive B-cells through somatic hypermutation, class switching and receptor editing, and is thus critical for an accurate classi¢cation of BCLs (Kuppers et al, 1999; Sha¡er et al, 2002). Based on the molecular status of the immunoglobulin gene, a B lymphocyte can be characterized into a so-called ‘‘pregerminal center’’ B cell that carries no mutations in the variable-region genes (Kuppers et al, 1999; Sha¡er et al, 2002). This may be distinguished from a ‘‘postgerminal center’’ B cell that presents somatic mutations in the variable-region genes on a single cell level (Kuppers et al, 1999; Sha¡er et al, 2002). Recent studies using gene expression pro¢ling have shown that lymphomas and leukemias which are histologically di⁄cult to distinguish can nevertheless be molecularly distinct diseases (Golub et al, 1999; Alizadeh et al, 2000), providing strong evidence that tumor expression pro¢les can be used for cancer classi¢cation. Using genes that de¢ne germinal center B-cell signatures, Alizadeh et al could show that di¡use large B-cell lymphomas consist of two distinct groups, carrying either an activated peripheral blood B-cell or a normal germinal center B-cell phenotype (Alizadeh et al, 2000). Most interesting is the fact that the group with an activated B-cell signature had signi¢cantly poorer survival compared with the center type (Alizadeh et al, 2000). This and the work of Shipp et al (2002) demonstrates that in addition to resolving classi¢cation problems, expression pro¢ling can add value to existing clinical prognostic parameters. Primary cutaneous B-cell lymphomas (CBCLs) are rare diseases (Dummer et al, 2000). They account for approximately 25% of all cutaneous lymphomas (estimated annual incidence 0.2/100,000/year) (Burg et al, 1997). Even for true ‘‘lymphomaniacs,’’ a convincing classi¢cation for primary cutaneous B-cell lymphomas is still far from being de¢ned. The European Orga-
0022-202X/03/$15.00 Copyright r 2003 by The Society for Investigative Dermatology, Inc.
vii
viii
DUMMER ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
REFERENCES Alizadeh AA, Eisen MB, Davis RE, et al: Distinct types of di¡use large B-cell lymphoma identi¢ed by gene expression pro¢ling. Nature 403:503^511, 2000 Burg G, Kempf W, Hae¡ner AC, et al: Cutaneous lymphomas. Curr Prob Dermatol 9:137^204, 1997 Cerroni L, Zochling N, Putz B, Kerl H: Infection by borrelia burgdorferi and cutaneous B-cell lymphoma. J Cutan Pathol 24:457^461, 1997 Dummer R, Willers J, Kamarashev J, Urosevic M, D˛bbeling U, Burg G: Pathogenesis of cutaneous lymphomas. Semin Cutan Med Surg 19:78^86, 2000 Golub TR, Slonim DK, Tamayo P, et al: Molecular classi¢cation of cancer: Class discovery and class prediction by gene expression monitoring. Science 286: 531^537, 1999 Hofbauer GF, Kessler B, Kempf W, Nestle FO, Burg G, Dummer R: Multilesional primary cutaneous di¡use large B-cell lymphoma responsive to antibiotic treatment. Dermatology 203:168^170, 2001 Kuppers R, Klein U, Hansmann ML, Rajewsky K: Cellular origin of human B-cell lymphomas. N Engl J Med 341:1520^1529, 1999
Santucci M, Pimpinelli N, Arganini L: Primary cutaneous B-cell lymphoma. A unique type of low-grade lymphoma. Clinicopathologic and immunologic study of 83 cases. Cancer 67:2311^2326, 1991 Sha¡er AL, Rosenwald A, Hurt EM, Giltnane JM, Lam LT, Pickeral OK, Staudt LM: Signatures of the immune response. Immunity 15:375^385, 2001 Sha¡er AL, Rosenwald A, Staudt LM: Lymphoid malignancies: The dark side of Bcell di¡erentiation. Nat Rev Immunol 2:920^932, 2002 Shipp MA, Ross KN, Tamayo P, et al: Di¡use large B-cell lymphoma outcome prediction by gene-expression pro¢ling and supervised machine learning. Nat Med 8:68^74, 2002 Storz MN, van de Rijn M, Kim YH, Mraz-Gernhard S, Hoppe TH, Kohler S: Gene expression pro¢les of cutaneous B-cell lymphoma. J Invest Dermatol in press, 2003 Willemze R, Kerl H, Sterry W, et al: EORTC classi¢cation for primary cutaneous lymphomas: A proposal from the Cutaneous Lymphoma Study Group of the European Organization for Research and Treatment of Cancer. Blood 90: 354^371, 1997
COMMENTARY See related article on page 865
Reinhard Dummer, Mirjana Urosevic, Keith Hoek, Andreas H¡ner and Gˇnter Burg Department of Dermatology, University Hospital Zurich, Zurich, Switzerland
nization for Research and Treatment of Cancer (EORTC) classi¢cation distinguishes between indolent CBCLs (follicle center cell lymphoma, immunocytoma, marginal zone B-cell lymphoma), CBCLs with intermediate clinical behavior (large B-cell lymphoma of the leg), and the provisional entities (intravascular large B-cell lymphoma and plasmocytoma) (Santucci et al, 1991; Willemze et al, 1997). This classi¢cation scheme is still under debate. In this issue of the Journal of Investigative Dermatology, there is an interesting article from Storz et al giving us for the ¢rst time the opportunity to look into gene expression pro¢les of primary CBCLs (Storz et al, 2003). Using the unsupervised, hierarchical clustering method the authors explore the relationship among lymphoma, normal skin and tonsil samples, trying to identify the molecular signatures in lymphoma specimens that might relate to their normal counterparts in skin and tonsil. Clustering analysis reveals that the follicular center cell lymphoma (FCCL) samples group more or less together, implying that these tumors share a similar overall gene expression pattern. It is of note that two of the FCCL samples from the same patient taken two years apart did not cluster next to each other, although they remained in the same branch. Follow-up studies will elucidate whether this phenomenon is due to the changes of the malignant clone over time or to the sensitivity of the array technique itself. Di¡use large cell lymphoma (DLCL) and marginal zone lymphoma (MZL) samples fail to demonstrate a speci¢c clustering pattern. Primary DLCL, however, cluster adjacent to FCCL samples, whereas secondary DLCL cases are removed from these two groups, suggesting a biological and molecular distinctiveness. Subsequently, the authors focus on gene clusters associated with T- and B-cell lineages and development, delineating T-cell, Bcell germinal center (GC) and plasma cell ‘‘signatures,’’ respectively. Storz et al detected characteristic B-cell GC signature in tonsil, primary and secondary FCCL, and in primary DLCL, suggesting that these diseases are more closely related than expected from recent histological and immunohistological analyses. This may indeed be the case in some clinical situations where DLCL of the skin react similarly to FCCL. We have recently described a large cell lymphoma responsive to antibiotic treatment (Hofbauer et al, 2001), as are some primary cutaneous FCCL (Cerroni et al, 1997; Hofbauer et al, 2001). Interestingly, secondary DLCL lacked the expression of B-cell GC signature. Plasma cell signature on the other side could be identi¢ed in two out of ¢ve MZL with prominent plasmacytoid di¡erentiation. Although the study only comprises a small number of cases, Storz et al provide the ¢rst insight into the molecular pro¢les of most common subtypes of CBCL. As with nodal lymphomas, there is hope that with the help of high-throughput microarray technology, we will be able to characterize the nature of di¡erent primary cutaneous lymphoma entities and subsequently improve existing lymphoma classi¢cations.
B-cell lymphomas (BCLs) represent the most frequent lymphoid neoplasms. They include rare precursor lymphoblastic leukemia/ lymphomas, B-cell chronic lymphocytic leukemias, mantle zone lymphomas, marginal zone lymphomas, follicular lymphomas and most of the Hodgkin’s lymphomas (Sha¡er et al, 2002). Only a minority of B-cell lymphomas actually occurs in the skin (Burg et al, 1997; Dummer et al, 2000). The origin of human BCLs has been studied extensively using the various approaches of histology, immunohistochemistry, karyotyping, single cell analysis and somatic mutation study of variable-region genes (Sha¡er et al, 2001; Sha¡er et al, 2002). In order to understand the biology of BCLs, it is reasonable to classify them according to their presumed normal counterpart in the B-cell di¡erentiation process. Despite their origin in human B-cell development, BCLs still remain a controversial topic in terms of their precise classi¢cation and diagnosis. Standard immunohistochemistry and molecular biology has often failed to identify the malignant ‘‘clonal’’ B-cell populations. Even in Southern blot-positive cases, the origin or relationship of a malignant cell to any particular stage in B-cell development remains unclear. The germinal center cell reaction is essential for the education of a B lymphocyte since it modi¢es naive B-cells through somatic hypermutation, class switching and receptor editing, and is thus critical for an accurate classi¢cation of BCLs (Kuppers et al, 1999; Sha¡er et al, 2002). Based on the molecular status of the immunoglobulin gene, a B lymphocyte can be characterized into a so-called ‘‘pregerminal center’’ B cell that carries no mutations in the variable-region genes (Kuppers et al, 1999; Sha¡er et al, 2002). This may be distinguished from a ‘‘postgerminal center’’ B cell that presents somatic mutations in the variable-region genes on a single cell level (Kuppers et al, 1999; Sha¡er et al, 2002). Recent studies using gene expression pro¢ling have shown that lymphomas and leukemias which are histologically di⁄cult to distinguish can nevertheless be molecularly distinct diseases (Golub et al, 1999; Alizadeh et al, 2000), providing strong evidence that tumor expression pro¢les can be used for cancer classi¢cation. Using genes that de¢ne germinal center B-cell signatures, Alizadeh et al could show that di¡use large B-cell lymphomas consist of two distinct groups, carrying either an activated peripheral blood B-cell or a normal germinal center B-cell phenotype (Alizadeh et al, 2000). Most interesting is the fact that the group with an activated B-cell signature had signi¢cantly poorer survival compared with the center type (Alizadeh et al, 2000). This and the work of Shipp et al (2002) demonstrates that in addition to resolving classi¢cation problems, expression pro¢ling can add value to existing clinical prognostic parameters. Primary cutaneous B-cell lymphomas (CBCLs) are rare diseases (Dummer et al, 2000). They account for approximately 25% of all cutaneous lymphomas (estimated annual incidence 0.2/100,000/year) (Burg et al, 1997). Even for true ‘‘lymphomaniacs,’’ a convincing classi¢cation for primary cutaneous B-cell lymphomas is still far from being de¢ned. The European Orga-
0022-202X/03/$15.00 Copyright r 2003 by The Society for Investigative Dermatology, Inc.
vii
viii
DUMMER ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
REFERENCES Alizadeh AA, Eisen MB, Davis RE, et al: Distinct types of di¡use large B-cell lymphoma identi¢ed by gene expression pro¢ling. Nature 403:503^511, 2000 Burg G, Kempf W, Hae¡ner AC, et al: Cutaneous lymphomas. Curr Prob Dermatol 9:137^204, 1997 Cerroni L, Zochling N, Putz B, Kerl H: Infection by borrelia burgdorferi and cutaneous B-cell lymphoma. J Cutan Pathol 24:457^461, 1997 Dummer R, Willers J, Kamarashev J, Urosevic M, D˛bbeling U, Burg G: Pathogenesis of cutaneous lymphomas. Semin Cutan Med Surg 19:78^86, 2000 Golub TR, Slonim DK, Tamayo P, et al: Molecular classi¢cation of cancer: Class discovery and class prediction by gene expression monitoring. Science 286: 531^537, 1999 Hofbauer GF, Kessler B, Kempf W, Nestle FO, Burg G, Dummer R: Multilesional primary cutaneous di¡use large B-cell lymphoma responsive to antibiotic treatment. Dermatology 203:168^170, 2001 Kuppers R, Klein U, Hansmann ML, Rajewsky K: Cellular origin of human B-cell lymphomas. N Engl J Med 341:1520^1529, 1999
Santucci M, Pimpinelli N, Arganini L: Primary cutaneous B-cell lymphoma. A unique type of low-grade lymphoma. Clinicopathologic and immunologic study of 83 cases. Cancer 67:2311^2326, 1991 Sha¡er AL, Rosenwald A, Hurt EM, Giltnane JM, Lam LT, Pickeral OK, Staudt LM: Signatures of the immune response. Immunity 15:375^385, 2001 Sha¡er AL, Rosenwald A, Staudt LM: Lymphoid malignancies: The dark side of Bcell di¡erentiation. Nat Rev Immunol 2:920^932, 2002 Shipp MA, Ross KN, Tamayo P, et al: Di¡use large B-cell lymphoma outcome prediction by gene-expression pro¢ling and supervised machine learning. Nat Med 8:68^74, 2002 Storz MN, van de Rijn M, Kim YH, Mraz-Gernhard S, Hoppe TH, Kohler S: Gene expression pro¢les of cutaneous B-cell lymphoma. J Invest Dermatol in press, 2003 Willemze R, Kerl H, Sterry W, et al: EORTC classi¢cation for primary cutaneous lymphomas: A proposal from the Cutaneous Lymphoma Study Group of the European Organization for Research and Treatment of Cancer. Blood 90: 354^371, 1997