Cutaneous Presentation of Follicular Lymphomas Renato Franco, M.D., Amalia Ferna´Ndez-Va´Zquez, M.D., Manuela Mollejo, M.D., Miguel A
Total Page:16
File Type:pdf, Size:1020Kb
Cutaneous Presentation of Follicular Lymphomas Renato Franco, M.D., Amalia Ferna´ndez-Va´zquez, M.D., Manuela Mollejo, M.D., Miguel A. Cruz, M.D., Francisca I. Camacho, M.D., Juan F. García, M.D., Mercedes Navarrete, B.SC., Miguel A. Piris, M.D. Department of Biomorphological and Functional Science, Federico II University of Napoli, Italy (RF); Department of Molecular Pathology, Centro Nacional de Investigaciones Oncologicas Carlos III, Majadahonda-Madrid (AFV, FIC, JFG, MN, MAP), Madrid, Spain; and Departments of Pathology (MM) and Oncology (MAC), Virgen de la Salud Hospital, Toledo, Spain The morphological features and clinical behavior of The description of primary cutaneous follicular lym- B-cell lymphomas presenting in the skin are phoma has raised interest in the differential diagnosis scarcely recognized in the lymphoma classification of this versus disseminated follicular lymphoma in- currently in use, the REAL classification (1), which volving the skin. We report here on four cases of Stage has led to the publication of a first proposal of IV follicular lymphoma, diagnosed in skin biopsy, in cutaneous lymphoma by the EORTC group (2). One which cutaneous lesion was the most noticeable fea- of the groups included in this category, so-called ture of clinical presentation. In all cases, the morpho- cutaneous follicular lymphoma (CFL), has fueled a logical features were superimposed over typical nodal debate on the features and differential nature of follicular lymphoma. Apart from classic B-cell mark- this entity. Although the molecular, phenotypical, ers, they were characterized by CD10 and bcl6 posi- tivity, markers of follicle germinal center cells; and and morphological traits of these tumors have yet bcl2 expression, with a corresponding t(14;18) trans- to be fully defined, the concept that some bona fide location in three of three cases examined. In all four B-cell lymphomas originating in the lymphoid fol- cases, bone marrow study and clinical staging re- licle may clinically present as cutaneous tumors vealed disease that had disseminated since diagnosis. seems to be clearly established (2–4). Thus the skin Follow-up showed relapsing cutaneous and nodal dis- could be involved by follicular lymphoma (FL), in ease in two cases. The only difference observed with a the form of a secondary lesion (sCFL), the expres- control group of 10 cases of primary cutaneous follic- sion of disseminated disease, or as a primary lesion ular lymphoma was the absence in this group of t(14; (pCFL), localized initially in the skin and usually for 18). Disseminated classical follicular lymphoma has to a long period of time (2–5). be considered in the differential diagnosis of follicular Systemic FL is usually a disseminated neoplasm, lymphoma presenting in the skin. This series of cases with over two thirds of cases being in Stage III or IV suggests that the presence of t(14;18) could imply the at diagnosis (2). It recapitulates both the cytological existence of disease that has disseminated beyond the and immunohistochemical features of secondary skin and that cases harboring this translocation could follicles, and neoplastic cells frequently show a spe- be candidates for systemic polychemotherapy. cific translocation, t(14;18). It may also involve the skin in the form of specific lesions, sCFL, and does KEY WORDS: bcl2, bcl6, CD10, Follicular lym- so in almost 4% of cases (4). phoma, Primary cutaneous follicular lymphoma, The existence of primary cutaneous B-cell lym- Translocation (14;18). phoma, defined as a B-cell non-Hodgkin’s lym- Mod Pathol 2001;14(9):913–919 phoma primarily occurring in and remaining con- fined to the skin for a period of at least 6 months, has now been widely accepted (2, 5, 7–21). Accord- ing to the EORTC, primary cutaneous follicular Copyright © 2001 by The United States and Canadian Academy of lymphoma (pCFL) shows a relatively good out- Pathology, Inc. come, rarely disseminating to extracutaneous sites. VOL. 14, NO. 9, P. 913, 2001 Printed in the U.S.A. Date of acceptance: March 30, 2001. This could support the selection of local treatment This work was supported by Grant IFD97-0431 from the Ministerio de of the tumor (2, 4, 19, 22, 23). The real frequency of Educacion y Ciencia, Spain. Address reprint requests to: Miguel A. Piris, Centro Nacional de Investi- pCFL and its morphological and molecular charac- gaciones Oncologicas Carlos III, Carretera Majadahonda-Pozuelo, km. 2, teristics are, however, a matter of controversy, as 28220 Majadahonda, Madrid, Spain; e-mail: [email protected]; fax:-34-91-509-70-55. are its differential diagnosis versus marginal zone 913 lymphoma in the skin (7, 24) and disseminated FL TABLE 1. Characteristics of Antibodies involving the skin. Antigen Clone (Manufacturer) Incubation Time Dilution In the course of a review of cases of cutaneous CD20 L26 (DAKO) 40 min 1:50 B-cell lymphoma, we noticed the existence of a CD79a JCB117 (DAKO) 40 min 1:25 CD23 MHM6 (DAKO) 40 min 1:25 group of four cases of FLs that, diagnosed on the CD10 56C6 (Novocastra) Overnight 1:25 basis of skin biopsies, have shown widespread dis- CD3 Polyclonal (DAKO) 40 min 1:50 ease since diagnosis, with clinical signs of progres- CD43 DFT1 (DAKO) 40 min 1:50 CD5 4C7 (Novocastra) 40 min 1:10 sion in at least two cases. Because this suggests that CiclD1 DC56 (DAKO) 40 min 1:100 these tumors must be recognized independently of P53 DO7 (DAKO) 40 min 1:50 pCFL, we reviewed the main features of this group Bcl2 124 (DAKO) 40 min 1:25 Bcl6 PG-B6P (DAKO) 40 min 1:10 of FLs presenting in the skin. MIB1 MIB1 (Immunotech) 40 min 1:50 MATERIALS AND METHODS The presence of positivity was quantified as a Patients and Tissue Samples percentage. In particular, CD20, CD10, bcl6, and Tumor specimens of four cutaneous biopsies bcl2 were scored as positive if the majority of tu- with FL were obtained from routine files of the moral cells showed a distinct reactivity. MIB1 ex- Virgen de la Salud Hospital, Toledo, Spain, corre- pression was scored as low Ki67 when the percent- Ͻ sponding to a period from 1990 to 2000. All patients age of positive tumoral cells was 30%; moderate Ͼ were under study for cutaneous lesions that at bi- when between 30% and 60%; and high when 60% opsy showed FL histology. From the onset of the of tumoral cells were positive. disease, the staging of all the patients included in the series showed widespread involvement, includ- Polymerase Chain Reaction ing bone marrow infiltration. All specimens were DNA isolated by conventional methods from pe- reevaluated, including initial cutaneous lesions, the ripheral blood (Case 3), bone marrow and periph- staging bone marrow biopsies, secondary cutane- eral blood (Case 2), and cryopreserved (Ϫ80°C) tis- ous lesions, or restaging bone marrow biopsies per- sue of skin biopsies was analyzed for t(14;18) by formed during the follow-up. Cases were classified polymerase chain reaction (PCR). PCR amplifica- into three grades, based on numbers of large cells tion of p53 Exon 8 (249 bp) was used to assess the (centroblasts), according to the criteria of Mann suitability of DNA extracts, using a volume of 100 and Bernard (25) ng. PCR was performed to detect the t(14;18) at the major breakpoint region (MBR). PCR was per- Immunohistochemistry formed with 0.1 g of purified DNA, which was All immunostaining techniques were performed subjected to 35 cycles of PCR amplification using in paraffin-embedded tissue sections, using a pre- MBR and JH consensus primer. The MBR and JH vious step of heat-induced antigen retrieval tech- primers have been described elsewhere (21). Ten nique for all antibodies. Thus, before incubation percent of PCR products was size-fractionated on with the primary antibody, slides were heated in a 1.5% agarose gel. Specimens were observed in ul- pressure cooker for 3 minutes in a solution of 0.01 traviolet B light. The presence of a band between mol/L sodium citrate. 100 and 250 base pairs (bp) was considered positive After incubation with the antibody, immunode- for translocation. tection was performed with biotinylated antimouse immunoglobulins, followed by peroxidase-labeled streptavidin (LSAB-DAKO, Glostrup, Denmark) Controls with diaminobenzidine chromogen as substrate. All A group of 10 cases of pCFL was used as the immunostaining was performed using the Tech- control group for the study of t(14;18). These cases mate 500 (DAKO) automatic immunostaining de- were selected according to clinical findings (com- vice. The antibodies used for immunohistochemi- puted tomography scan, ultrasonography, and cal study were directed against B-cell markers CD20 bone marrow biopsies without signs of lymphoma and CD79a, T-cell marker CD3; specific markers of involvement for at least 6 months after diagnosis), mantle-cell lymphoma CD5, cyclin D1, and IgD; morphological signs (follicular pattern of growth, germinal center markers such as CD10 and bcl6; germinal center cytology), immunophenotypical marker of dendritic cells CD23; oncoproteins p53 criteria (bcl6 and CD10 positivity and the presence and bcl2; and proliferative marker MIB1. The data of follicular dendritic cells), and suitability of the corresponding to the antibodies used are shown in DNA extracted from paraffin-embedded tissue for Table 1. PCR analysis. Samples were from a larger collection 914 Modern Pathology of cutaneous pCBCL obtained from the files of the lateral region of the left leg. Radiotherapy was Spanish hospitals integrated in the Spanish Lym- performed. phoma Club. These cases of pCFL belong to a larger Eight years after diagnosis, the patient is alive, series, whose results concerning the frequency of with disease restricted to the skin. IgH rearrangement and t(14;18) translocation have already been published (26).