Bone marrow and peripheral blood involvement in mantle cell lymphoma

Transcription

1 British Journal of Haematology, 1998, 101, Bone marrow and peripheral blood involvement in mantle cell lymphoma P. L. C OHEN, P.J.KURTIN, K.A.DONOVAN AND C. A. HANSON Divisions of Hematopathology and Medical Genetics, Mayo Clinic, Rochester, Minnesota, U.S.A. Received 11 August 1997; accepted for publication 2 February 1998 Summary. The peripheral blood smears, bone marrow aspirates and biopsies of 46 patients with mantle cell lymphoma were reviewed. The diagnosis of mantle cell lymphoma was established in all cases on extramedullary tissue samples using standard morphologic, phenotypic and molecular genetic criteria. 27/35 patients (77%) had circulating lymphoma cells (median 20% of all circulating white blood cells; range 5 90%) identified by morphology at some point during the course of their disease. No statistical difference in survival was detected in patients with or without peripheral blood involvement. Lymphoma was identified in bone marrow aspirate specimens from 33/40 patients (83%) and in bone marrow biopsy specimens from 39/43 patients (91%). The pattern of marrow biopsy involvement was nodular (31 cases; 82%), interstitial (19 cases; 50%), paratrabecular (17 cases, 45%) and diffuse (12 cases; 32%). Although the median survival of patients with 50% bone marrow involvement was 13 months, and the median survival of patients with 50% was 49 months, no statistically significant differences between these small subgroups were observed. Mantle cell lymphoma frequently involves the peripheral blood and bone marrow. Its appearance is distinctive but variable, and immunophenotypic studies as well as morphologic confirmation by a biopsy of tissue other than bone marrow is still required for diagnosis. Keywords: mantle cell lyphoma, peripheral blood, bone marrow, bcl-1, polymerase chain reaction. Mantle cell lymphoma (MCL) is a relatively recently recognized entity among malignant lymphomas (Banks et al, 1992). Numerous studies have been devoted to the morphology, natural history and molecular biology of this disease (reviewed in Shivdasani et al, 1993). Most of these studies have focused upon the morphology of mantle cell lymphoma in extramedullary sites. The peripheral blood and bone marrow manifestations of MCL have been the central focus of fewer studies (Argatoff et al, 1997; Pittaluga et al, 1996), and, in many of these, strict criteria for the diagnosis of MCL, including immunophenotypic and molecular characterization, were not available for all cases. The current study was undertaken to define the peripheral blood and bone marrow manifestations of MCL in a rigidly defined group of cases. MATERIALS AND METHODS All cases diagnosed as mantle cell lymphoma, mantle zone lymphoma, or intermediately differentiated lymphocytic Correspondence: Dr Paul J. Kurtin, Department of Laboratory Medicine and Pathology, Hilton 1156, Mayo Clinic, Rochester, MN 55905, U.S.A. lymphoma between 1984 and 1996 at the Mayo Clinic were reviewed. For inclusion in the study, both bone marrow aspirate and/or biopsy as well as tissue biopsy specimens had to be available for review. The morphologic criteria for diagnosis of mantle cell lymphoma in tissue biopsies were as previously described (Banks et al, 1992; Harris et al, 1994). In all cases the standard immunophenotype of mantle cell lymphoma (a monoclonal B-cell tumour with co-expression of CD5, and, in more recent cases, absence of CD23 expression) was established by either flow cytometry or frozen section immunoperoxidase stains on a tissue sample other than the bone marrow. In addition, the neoplastic cells from 33/38 cases (87%) that had available paraffin blocks were positive for bcl-1 protein by paraffin section immunohistochemistry. Immunohistochemistry. Immunohistochemical studies using the labelled streptavidin biotin method were performed on frozen tissue sections. In brief, frozen tissue specimens were sectioned on a cryostat, fixed briefly in acetone and air dried. After rehydration in buffer, the sections were incubated with a primary mouse monoclonal antibody, followed by incubation with a biotin-conjugated goat anti-mouse Fab fragment secondary antibody. The sections were incubated in a peroxidase-labelled streptavidin solution, then exposed to an amino ethyl carbazole and hydrogen peroxidase solution and Blackwell Science Ltd

2 lightly counterstained with haematoxylin. Monoclonal antibodies from the following sources were used at the dilutions specified: kappa (Becton Dickinson, 1:500 and 1:1000), lambda (Becton Dickinson, 1:20 and 1:40), CD5 (Becton Dickinson, 1:200), CD19 (Dako, 1:25), CD20 (Dako, 1:40), CD23 (Dako, 1:50), CD3 (Becton Dickinson, 1:20). Immunoperoxidase stains for bcl-1 on paraffin sections were performed by the method of Vasef et al (1997). Flow cytometry. Flow cytometric immunophenotyping was performed on a FACScan Analyser using standard techniques (Witzig et al, 1990). Monoclonal antibodies directly labelled to a fluorochrome were utilized and included CD2, CD3, CD5, CD7, CD10, CD11c, CD19, CD20, CD22, CD23, CD45, kappa and lambda, and HLA-DR (Becton Dickinson). Molecular genetics. High molecular weight DNA was extracted from all cases with available frozen tissue, using standard proteinase K digestion followed by phenol chloroform extraction. The polymerase chain reaction (PCR) technique was used to amplify DNA using a consensus immunoglobulin heavy-chain joining region gene ( JH) primer and a primer for the bcl-1 major translocation cluster (MTC), as previously described (Williams et al, 1993). In brief, standard PCR conditions were used with 5 0 primer CCTCTCTCCAAATTCCTG and 3 0 primer ACCTGAGGAGAC- GGTGAC. After 35 cycles of enzymatic amplification in an automated Perkin-Elmer DNA Thermal Cycler, with 30 s denaturation at 94 C, 30 s annealing at 60 C, and 1 min extension at 72 C, PCR products were electophoresed on a 2% ethidium bromide stained agarose gel. The specificity of the PCR product was confirmed by a Southern blot of a PCR product gel which was blotted to a nylon membrane and hybridized with a radioactively labelled internal oligonucleotide probe (CAACCACTGATTATTTTAGCT). Southern analysis was performed for immunoglobulin heavy chain gene rearrangement with probes for the constant region of the heavy chain gene (C m ) and the joining region of the kappa light chain gene ( J k ) after digestion with restriction endonucleases Bgl II and BamH 1, as previously described (Lust, 1996). Southern analysis for the cyclin D1 gene translocation was performed with the probe p94ps (a gift from Dr Timothy C. Meeker) after DNA digestion with restriction endonucleases EcoR1 and BamH1. Morphology. After confirmation of the diagnosis of mantle cell lymphoma by tissue section morphology, immunophenotype and genotype, the Wright-Giemsa stained slides of the peripheral blood and bone marrow and the haematoxylin and eosin stained bone marrow biopsy sections were reviewed to determine the Wright-Giemsa stained cytology of the mantle cells, the presence or absence of circulating lymphoma cells, and the pattern and extent of bone marrow involvement. Clinical features. Patient medical records were reviewed to determine patient age, stage of disease, absolute lymphocyte count, point of disease course, and disease sites. Because the study was designed to evaluate the spectrum of bone marrow involvement in MCL, cases were not excluded if the initial diagnostic tissue and staging bone marrow were not available for review. Statistical analysis. Survival times were calculated by Mantle Cell Lymphoma in Bone Marrow 303 Kaplan-Meier curve (Kaplan & Meier, 1958), with length of survival defined as time of initial diagnosis until death from any cause. Differences in survival times were calculated by univariate log-rank test (Cox, 1970). RESULTS Clinical features Forty-six patients with MCL meeting the above criteria were identified: 36 patients were male and 10 patients were female. The median age at diagnosis was 62 years, with a range of years. 36 patients were stage IV at diagnosis, four were stage III, two were stage II, and two patients were stage I; in two patients the initial stage could not be determined. Nodal based disease was identified in 37 patients. 14 patients required splenectomy for diagnosis or symptomatic relief; an additional 13 patients had splenomegaly as assessed by computerized tomography or physical examination. Gastrointestinal tract involvement was identified in 12 patients, and liver involvement was seen in seven patients. Three patients had orbital involvement, and two had conjunctival disease. An additional two patients each had tonsillar and lung disease. Pathologic and laboratory features of diagnostic specimens The diagnosis of mantle cell lymphoma was established by review of 69 tissue samples, not representing peripheral blood or bone marrow, from the 46 patients. The histologic pattern was diffuse in 42 samples (91%), and four cases (9%) had a mantle zone pattern. A blastoid cytology was seen in six tissue samples (13%); in three of these cases the blastoid cytology represented progression from a non-blastoid cytology 4 5 years after initial diagnosis. In another case, simultaneous biopsies at separate sites showed blastoid cytology in one site and non-blastoid in another. Seven cases had a cytology similar to monocytoid B cells or marginal zone B cells, the neoplastic cells containing a moderate amount of pale cytoplasm evident in tissue sections. The immunophenotype was determined by flow cytometry in 23 cases, by frozen section immunoperoxidase study (FSIP) in 13 cases, and by both modalities in 10 cases. As entry criteria required, all cases were B-cell tumours with immunoglobulin light chain restriction, and demonstrated aberrant co-expression of CD5. CD23 expression was assessed in 10 cases (five by flow cytometry, four by both flow cytometry and FSIP, and one by FSIP alone). The neoplastic cells were CD23 negative in all cases. Lambda light chain and kappa light chain restriction were each found in 23 cases. Paraffin-embedded tissue samples for bcl-1 immunohistochemistry were available in 38 cases. The lymphoma was bcl-1 positive in 33 (87%). Frozen tissue was available in 27 cases; the DNA was extracted from these. All cases showed immunoglobulin gene rearrangement by Southern analysis. PCR for the major translocation cluster amplified bands of the appropriate size, indicating bcl-1/igh gene rearrangements in 10/27 cases (37%) (Fig 1). Only one of the 27 cases (4%) showed cyclin-d1 gene rearrangement with the probe p94ps using Southern analysis technique (Fig 2).

3 304 P. L. Cohen et al Fig 1. Polymerase chain reaction detection of bcl-1. The upper panel shows positive PCR products for the bcl-1 major translocation cluster (MTC) electrophoresed on an ethidium bromide stained agarose gel. Lane 1, DNA ladder, lane 2, no DNA; lane 3, negative control; lanes 4 10, patient samples. The bottom portion of the photograph shows a Southern blot of the same gel blotted to a nylon membrane and then hybridized with an internal oligonucleotide probe, confirming that these bands are from the MTC area. Fig 2. Southern analysis with p94ps probe. The germline band is at 18 5 kb; the arrow shows the single case (1/27) with a detectable gene rearrangement. Peripheral blood findings Thirty-five peripheral blood smears were available for review, 25 from the time of initial diagnosis. 27/35 patients (77%) had circulating lymphoma cells at some point during the course of their disease as judged by morphology. In one additional case, judged negative by morphology, a small monoclonal B-cell population was detected in the blood by flow cytometry. Positive cases had a median of 20% lymphoma cells in the blood, with a range from < 5% to > 90% of the white blood cell population. Circulating lymphoma cells were identified by morphology in 20/25 (80%) of cases at initial diagnosis. A complete blood count was available at initial diagnosis in 36 patients; 10/36 (28%) had an absolute lymphocytosis (> /l) at diagnosis. The circulating lymphoma cells had a polymorphous appearance in the Wright-Giemsa stained smears (Fig 3). Some circulating lymphoma cells had very round nuclei with scant basophilic cytoplasm, similar in size and shape to the circulating cells in chronic lymphocytic leukaemia (CLL), but with a more reticular or lacy chromatin pattern. The condensed, clumped chromatin pattern of CLL was not seen. In many cases at least some of the cells in the blood had nuclei that were irregular, angulated or cleaved; others were bi- or trinucleated. However, multinucleated cells were a small minority of the neoplastic cells in all cases. In some cases ample pale basophilic cytoplasm was present in many of the ciruculating cells, and in 11/36 cases (31%) the neoplastic cells contained small, discrete cytoplasmic vacuoles. In most cases pale small nucleoli were visible in at least some of the cells in the peripheral blood or bone marrow aspirate specimens. Bone marrow findings Lymphoma was identified in the bone marrow aspirate specimens from 33/40 patients (83%). A median of 20% (range 5 80%) of the bone marrow aspirate cellularity consisted of lymphoma. The cytology of the lymphoma cells in the bone marrow aspirate smear recapitulated that seen in the peripheral blood. Lymphoma was identified by morphology in the bone marrow biopsy specimens from 39/43 patients (91%). A median of 15% (range < 5% to > 90%) of the biopsy consisted of lymphoma in the involved cases. The pattern of marrow biopsy involvement was nodular in 31 cases (82%), interstitial in 19 cases (50%), paratrabecular in 17 (45%) and diffuse in 12 (32%), with most cases showing a combination of these patterns (Figs 4 6). The cytology of the lymphoma cells in the bone marrow biopsy specimens tended to follow the distinctive subtypes noted in the peripheral blood smears and aspirates, with a dispersed chromatin pattern in the blastic variant cases, and increased cytoplasm present in the monocytoid B variants; this latter feature was not always as distinctive as in the Wright-Giemsa smears or in tissue biopsies (Figs 7 and 8). Nuclear irregularities were more prominent in the nonblastic variants. A discrepancy of > 20% lymphomatous involvement between the bone marrow aspirate and biopsy specimens was seen in six cases (16%). In five of these cases more lymphoma was present in the biopsy specimen than in the aspirate smear, and in four of those five cases the biopsy showed paratrabecular bone marrow involvement. No significant difference was found in degree of bone marrow involvement by mantle cell lymphomas of varying morphologic and cytologic subtypes, with a median lymphomatous involvement of 10% in cases with a mantle zone pattern, 15% in diffuse cases, 10% in blastoid variant, and 15% in the monocytoid variant. No significant difference was seen in degree of bone marrow involvement in cases with (7 5%) and without (15%) a detectable bcl-1 gene rearrangement by PCR.

4 Mantle Cell Lymphoma in Bone Marrow 305 Fig 3. Peripheral blood smear in mantle cell lymphoma. A polymorphous range of lymphoma cells can be seen, with a moderately fine chromatin pattern and occasional conspicuous nucleoli. Scant to ample basophilic cytoplasm is present. Binucleate cells are occaionally seen, and cytoplasmic vacuoles are not uncommon. Fig 4. Paratrabecular pattern of bone marrow involvement. Fig 5. Interstitial pattern of bone marrow involvement. Fig 6. Nodular pattern of bone marrow involvement. Granular eosinophilic macrophages, commonly found in lymph node biopsies of mantle cell lymphoma, were sometimes seen. Fig 7. Blastic variant of mantle cell lymphoma in bone marrow biopsy. Note the open chromatin. Fig 8. Monocytoid-B cell variant of mantle cell lymphoma in bone marrow biopsy. Note the abundant pale-staining well-defined cytoplasm.

5 306 P. L. Cohen et al Fig 9. Kaplan-Meier plot of overall survival in the studied patient population with mantle cell lymphoma and bone marrow involvement. Fig 12. Kaplan-Meier survival of patients with > 50% bone marrow involvement at initial diagnosis (dotted line) and with < 50% bone marrow involvement at diagnosis (solid line). Fig 10. Kaplan-Meier plot of survival of patients with a positive peripheral blood smear (solid line) and negative peripheral blood smear (dotted line). Fig 11. Kaplan-Meier plot of survival of patients with (dotted line) and without (solid line) a peripheral blood absolute lymphocytosis of /l. Survival and prognosis Median survival (based on the Kaplan-Meier survival curve) of all patients was 49 months (range months) (Fig 9). Median survival of patients with positive peripheral smears identified by morphology at initial diagnosis was 49 months (range months); median survival of patients with negative or equivocal peripheral blood smears at diagnosis was also 49 months (range 2 117þ months) (Fig 10). Median survival of patients with an absolute lymphocytosis > /l at diagnosis was 198 months (range months) (Fig 11). Median survival of patients without an absolute lymphocytosis was 49 months (range 1 117þ months). The median survival of patients with 50% bone marrow involvement at diagnosis was 13 months (n ¼ 5; range 2 90 months); the median survival of patients with 50% bone marrow involvement was 49 months (n ¼ 21; range 1 136þ months) (Fig 12). Because the numbers of patients in each of the subgroups described above were small, no statistically significant differences between them were observed. DISCUSSION Mantle cell lymphoma is a B-cell lymphoma with distinctive morphologic, phenotypic and molecular genetic features. Its differentiation from other B-cell lymphomas has been a gradual and relatively recent phenomenon. In the mid-1970s the morphology of mantle cell lymphoma was described by Lennert et al (1975) in the Kiel classification system (centrocytic lymphoma) and by Berard & Dorfman (1974) in the United States (intermediately differentiated lymphocytic lymphoma). In the 1980s and early 1990s the immunophenotype of this lymphoma was delineated by many investigators, revealing a B-cell lymphoma with tumour cell expression of IgM, IgD, CD5 and CD43, and usually lacking CD10 (Cossman et al, 1984; Harris et al, 1984; Lardelli et al, 1990; Pileri et al, 1985; Spier et al, 1986; Strickler et al, 1988; Swerdlow et al, 1983; van den Oord et al, 1986; Weisenburger et al, 1987; Zukerberg et al, 1993). In the early 1990s the association between the chromosomal translocation t(11;14)(q13;q32) and this lymphoma was noted (Frizzera et al, 1991; Leroux et al, 1991; Medeiros et al, 1990; Raffeld & Jaffe, 1991; Rosenberg et al, 1991; Vandenberghe et al, 1991; Williams et al, 1990) and in 1992 a group of American and European pathologists proposed the term mantle cell lymphoma for these non- Hodgkin s lymphomas (Banks et al, 1992). Although well described in lymph nodes and other tissues, relatively few studies have focused upon bone marrow and

6 peripheral blood involvement in mantle cell lymphoma (Daniel et al, 1995; Pittaluga et al, 1996; Pombo De Oliveira et al, 1989; Vadlamudi et al, 1996; Vasef et al, 1997; Wasman et al, 1996). Most of these studies involve a small number of patients, and most larger studies do not limit the study population to cases in which the initial diagnosis of mantle cell lymphoma was made by a combination of morphologic, immunophenotypic and molecular genetic techniques. To our knowledge, this is the largest study in which mantle cell lymphoma in bone marrow and tissue was rigidly defined by morphology, immunophenotype and molecular genetics. Other large studies of mantle cell lymphoma have found a high proportion of cases with bone marrow involvement, ranging from 53% to 93% (Argatoff et al, 1997; Berger et al, 1994; Duggan et al, 1990; Fisher et al, 1995; Norton et al, 1995; Pittaluga et al, 1996; Swerdlow et al, 1983; Velders et al, 1996; Weisenburger et al, 1981; Zucca et al, 1995). With 83% of cases showing bone marrow aspirate involvement, and 91% of cases with a positive bone marrow biopsy, this study showed a high percentage of cases with bone marrow involvement in mantle cell lymphoma. This slightly higher than usual rate of lymphoma detection may be due to the bilateral bone marrow biopsies traditionally performed at our institution in cases of lymphoma. The inclusion of patients at later stages of disease did not appear to influence the high number of patients with bone marrow involvement, since 93% of patients with specimens available at diagnosis had either a positive aspirate or biopsy. The pattern of bone marrow biopsy involvement was quite variable in this study, with a combination of patterns present in most biopsies. The nodular pattern was most common, closely followed by interstitial, paratrabecular and diffuse involvement; a similar variety of patterns has been noted in other studies (Argatoff et al, 1997; Bartl et al, 1982; Pittaluga et al, 1996; Wasman et al, 1996). The histologic pattern of bone marrow biopsy in mantle cell lymphoma can aid in its distinction from other small B-cell lymphomas. A paratrabecular pattern in a CD5-positive B-cell lymphoma argues strongly against a diagnosis of chronic lymphocytic leukaemia, because large studies of this entity show interstitial, nodular and diffuse patterns of bone marrow involvement, and probably never a paratrabecular pattern (Rozman et al, 1984). Historically, a paratrabecular pattern is most closely associated with follicular lymphomas involving the bone marrow; clearly the differential diagnosis of a paratrabecular lymphoma must include MCL. Although a paratrabecular pattern has been described in cases of bone marrow involvement in monocytoid B-cell lymphoma, the immunophenotype of these individual cases was not described (Sheibani et al, 1988). The diverse morphologic appearance of mantle cell lymphoma has recently been recognized (Swerdlow et al, 1996; Zucca et al, 1994). Within this context, we believe that it is quite possible that most previous descriptions of CD5-positive lymphomas with a monocytoid B-cell cytology and bone marrow involvement actually represent cases of mantle cell lymphoma (Agnarsson & Kadin, 1987; Traweek & Sheibani, 1992). Since cases of mantle cell lymphoma can present with a Mantle Cell Lymphoma in Bone Marrow 307 peripheral blood lymphocytosis (28% in this study), and a far higher percentage of patients have peripheral blood involvement by mantle cell lymphoma by morphologic assessment (80% in this study), the appearance of mantle cell lymphoma in Wright-Giemsa stained smears can greatly assist in the expeditious recognition of this entity. Compared to other small lymphocytic lymphomas (B-cell CLL and follicular lymphoma), the chromatin pattern of mantle cell lymphoma is usually less condensed, and nucleoli are more readily apparent. In general, the chromatin pattern seen in the lymphoma cells was the most consistent feature separating mantle cell lymphoma from CLL. With their occasional binucleate and trinucleate cells and cytoplasmic vacuolization, mantle cell lymphoma cells are typically more polymorphous than those from the usual cases of B-cell CLL or hairy cell leukaemia. However, some cases of CLL may show a significant amount of nuclear clefting (Matutes et al, 1996), and this feature alone should not be considered to be distinctive of either entity. The Wright-Giemsa stained morphology of mantle cell lymphomas also overlaps with T-cell CLL, T-cell prolymphocytic leukaemia (PLL) and B-cell PLL. Phenotypic studies would obviously exclude the possibility of a T-cell lymphoproliferative disorder. In the absence of a tissue diagnosis in a patient with a B-cell lymphocytosis, PLL would clearly be considered in the same differential diagnosis as MCL in leukaemic phase. MCL and PLL share some cytologic features, and splenomegaly is a common feature of both disorders. In the current study, 59% of patients had varying degrees of splenomegaly. A more reticular nuclear chromatin pattern than CLL is common to both disorders. Nucleoli, the hallmark of PLL, were commonly identified in our cases of MCL, although few had the prominent, single central nucleolus of the classic prolymphocyte. The immunophenotype of PLL can be variable, and can include CD5 expression (Melo et al, 1986). In addition, the t (11;14) has been described in some cases of B-PLL (Brito-Babapulle et al, 1992). Therefore diagnostic caution must be exerted in those cases of possible PLL that have overlapping cytologic and laboratory features with MCL and exhibit a classic MCL immunophenotype. Although some studies have addressed the issue of a leukaemic phase of MCL, many of these studies are small, not all cases were immunophenotyped, and no uniform definition of a leukaemic phase prevails. Some authors have defined a leukaemic phase as those cases with morphologic involvement of the peripheral blood smear (Argatoff et al, 1997; Pittaluga et al, 1996); others have defined it as a high peripheral white blood cell count (Daniel et al, 1995; Pombo De Oliveira et al, 1989), or an absolute lymphocytosis (Norton et al, 1995; Swerdlow et al, 1983; Weisenburger et al, 1981; Zucca et al, 1995). In most earlier studies patients with an absolute lymphocytosis > /l at diagnosis appear to have a shorter survival (Duggan et al, 1990; Norton et al, 1995; Swerdlow et al, 1983; Weisenburger et al, 1981) and one study suggests that a peripheral lymphocytosis of > /l is an indicator of an especially poor prognosis (Pittaluga et al, 1996). In the current study, survival did not appear to be influenced by the presence of an absolute lymphocytosis at diagnosis, and no difference in

7 308 P. L. Cohen et al survival was noted among patients with or without peripheral blood involvement detected by morphology. Our study does suggest that circulating lymphoma cells are characteristic of most cases of mantle cell lymphoma, and their presence in the blood may be a reflection of the biology of the disease rather than a poor prognostic indicator. Finally, a few case reports have suggested that the development of a leukaemic phase heralds end-stage disease in patients with long-established mantle cell lymphoma (Vadlamudi et al, 1996). Survival did appear to be influenced by the degree of BM involvement by MCL at initial diagnosis. Median survival was only 13 months in patients when 50% of the biopsy cellularity was due to lymphoma. In contrast, patients with 50% of the biopsy involved by lymphoma had a median survival of 49 months. These data confirm another recent large study, in which patients with > 50% of the bone marrow involved by lymphoma tended to have a shorter overall survival (Argatoff et al, 1997). The high number of cases found to contain circulating neoplastic lymphocytes at the time of diagnosis in this study coupled with the fairly distinctive phenotype of mantle cell lymphoma suggests that peripheral blood immunophenotyping might obviate the need for a tissue biopsy to establish the diagnosis of mantle cell lymphoma. However, until the specificity of the characteristic mantle cell lymphoma immunophenotype is determined by studying the blood of a large number of patients with various B-cell chronic lymphoproliferative disorders, the diagnosis of mantle cell lymphoma still requires morphologic confirmation from a specimen other than blood and bone marrow. Finally, the results of our PCR studies of the MTC region of bcl-1, in which 37% of cases demonstrated gene rearrangements in this region, are in accordance with other studies % of cases of MCL have shown bcl-1 gene rearrangements by this method (Kumar et al, 1996; Luthra et al, 1995; Molot et al, 1994; Pinyol et al, 1996; Rimokh et al, 1993). Although one earlier study showed gene rearrangement by Southern analysis with probe p94ps in a high proportion of cases (6/23) (Williams et al, 1991), our results of only one of 27 rearrangements detected with this probe are similar to other investigators, who found none of 33 (Rimokh et al, 1993) and none of 21 (Pinyol et al, 1996) cases positive for bcl-1 gene rearrangements with this probe. ACKNOWLEDGMENTS We thank Dr Peter C. Wollan for performing the statistical analyses and acknowledge the expert secretarial assistance of Annette Bjorheim in preparing the manuscript. REFERENCES Agnarsson, B. & Kadin, M. (1987) An unusual B-cell lymphoma simulating hairy cell leukemia. American Journal of Clinical Pathology, 88, Argatoff, L., J.M, C., Klasa, R., Horsman, D. & Gascoyne, R. (1997) Mantle cell lymphoma: a clinicopathologic study of 80 cases. Blood, 89, Banks, P., Chan, J., Cleary, M., Delsol, G., De Wolf-Peeters, C., Gatter, K., Grogan, T., Harris, N., Isaacson, P., Jaffe, E., Mason, D., Pileri, S., Ralfkiaer, E., Stein, H. & Warnke, R. (1992) Mantle cell lymphoma: a proposal for unification of morphologic, immunologic, and molecular data. American Journal of Surgical Pathology, 16, Bartl, R., Frisch, B., Burkhardt, R., Kettner, G., Mahl, G., Fateh- Moghadam, A. & Sund, M. (1982) Assessment of bone marrow histology in the malignant lymphomas (non-hodgkin s): correlation with clinical factors for diagnosis, prognosis, classification, and staging. British Journal of Haematology, 51, Berard, C. & Dorfman, R. (1974) Histopathology of malignant lymphomas. Clinical Hematology, 3, Berger, F., Felman, P., Sonet, A., Salles, G., Bastion, Y., Bryon, P. & Coiffier, B. (1994) Nonfollicular small B-cell lymphomas: a heterogeneous group of patients with distinct clinical features and outcome. Blood, 83, Brito-Babapulle, V., Ellis, J., Matutes, E., Oscier, D., Khokhar, T., MacLennan, K. & Catovsky, D. (1992) Translocation t(11;14)(q13;q32) in chronic lymphoid disorders. Genes, Chromosomes and Cancer, 5, Cossman, J., Neckers, L., Hsu, S., Longo, D. & Jaffe, E. (1984) Low-grade lymphomas: expression of developmentally regulated B-cell antigens. American Journal of Clinical Pathology, 115, Cox, D. (1970) The Analysis of Binary Data. Chapman and Hall, New York. Daniel, M., Tigaud, I., Flexor, M., Nogueira, M., Berger, R. & Jonveaux, P. (1995) Leukaemic non-hodgkin s lymphomas with hyperdiploid cells and t(11;14)(q13;q32): a subtype of mantle cell lymphoma. British Journal of Haematology, 90, Duggan, M., Weisenburger, D., Ye, Y., Bast, M., Pierson, J., Linder, J. & Armitage, J. (1990) Mantle zone lymphoma: a clinicopathologic study of 22 cases. Cancer, 66, Fisher, R.S.D., Nathwani, B., Banks, P., Miller, T. & Grogan, T. (1995) A clinical analysis of two indolent lymphoma entities: mantle cell lymphoma and marginal zone lymphoma (including the mucosaassociated lymphoid tissue and monocytoid B-cell subcategories): a Southwest Oncology Group study. Blood, 85, Frizzera, G., Sakurai, M., Notohara, K. & Konishi, H. (1991) t(11;14)(q13q;q32) in B-cell lymphomas (intermediately differentiated lymphocytic and follicular): a report of four cases. American Journal of Clinical Pathology, 95, Harris, N., Jaffe, E., Stein, H., Banks, P., Chan, J., Cleary, M., Delsol, G., DeWolf-Peeters, C., Falini, B., Gatter, K., Grogan, T., Isaacson, P., Knowles, D., Mason, D., Muller-Hermelink, H., Pileri, S., Piris, M., Ralfkiaer, E. & Warnke, R. (1994) A revised European American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood, 84, Harris, N., Nadler, L. & Bhan, A. (1984) Immunohistologic characterization of two malignant lymphomas of germinal center type (centroblastic/centrocytic and centrocytic) with monoclonal antibodies: follicular and diffuse lymphomas of small-cleaved-cell type are related but distinct entities. American Journal of Pathology, 117, Kaplan, E. & Meier, P. (1958) Nonparametric estimation from incomplete observations. Journal of the American Statistical Association, 53, Kumar, R., Krenacs, L., Otsuki, T., Kumar, D., Harris, C., Wellmann, A., Jaffee, E. & Raffeld, M. (1996) bcl-1 rearrangement and cyclin D1 protein expression in multiple lymphomatous polyposis. American Journal of Clinical Pathology, 105, Lardelli, P., Bookman, M., Sundeen, J., Longo, D. & Jaffe, E. (1990)

8 Lymphocytic lymphoma of intermediate differentiation: morphologic and immunophenotypic spectrum and clinical correlations. American Journal of Surgical Pathology, 14, Lennert, K., Stein, H. & Kaiserling, E. (1975) Cytological and functional criteria for the classification of malignant lymphomata. British Journal of Cancer, 31, Leroux, D., Le Mark Hadour, F., Gressin, R., Jacob, M., Keddari, E., Monteil, M., Caillot, P., Jalbert, P. & Sotto, J. (1991) Non-Hodgkin s lymphomas with t(11;14)(q13;q32): a subset of mantle zone/ intermediate lymphocytic lymphoma? British Journal of Haematology, 77, Lust, J. (1996) Molecular genetics and lymphoproliferative disorders. Journal of Clinical Laboratory Analysis, 10, Luthra, R., Hai, S. & Pugh, W. (1995) Polymerase chain reaction detection of the t(11;14) translocation involving the bcl-1 major translocation cluster in mantle cell lymphoma. Diagnostics in Modern Pathology, 4, 4 7. Matutes, E., Oscier, D., Garcia-Marco, J., Ellis, J., Copplestone, A., Gillingham, R., Hamblin, T., Lens, D., Swansbury, G. & Catovsky, D. (1996) Trisomy 12 defines a group of CLL with atypical morphology: correlation between cytogenetic, clinical and laboratory features in 544 patients. British Journal of Haematology, 92, Medeiros, L., Van Krieken, J., Jaffe, E. & Raffield, M. (1990) Association of bcl-1 rearrangements with lymphocytic lymphoma of intermediate differentiation. Blood, 76, Melo, J., Catovsky, D. & Galton, D. (1986) The relationship between chronic lymphocytic leukaemia and prolymphocytic leukaemia. British Journal of Haematology, 63, Molot, R., Meeker, T., Wittwer, C., Perkins, S., Segal, G., Masih, A., Braylan, R. & Kjeldsberg, C. (1994) Antigen expression and polymerase chain reaction amplification of mantle cell lymphomas. Blood, 83, Norton, A., Matthews, J., Pappa, V., Shamash, J., Love, S., Rohatiner, A. & Lister, T. (1995) Mantle cell lymphoma: natural history defined in a serially biopsied population over a 20-year period. Annals of Oncology, 6, Pileri, S., Rivano, M., Gobbi, M., Taruscio, D. & Lennert, K. (1985) Neoplastic and reactive follicles within B-cell malignant lymphomas: a morphologic and immunological study of 30 cases. Hematological Oncology, 3, Pinyol, M., Campo, E., Nadal, A., Terol, M., Jares, D., Nayach, I., Fernandez, P., Pieris, M., Montserrat, E. & Cardesa, A. (1996) Detection of the bcl-1 rearrangement at the major translocation cluster in frozen and paraffin-embedded tissues of mantle cell lymphomas by polymerase chain reaction. American Journal of Clinical Pathology, 105, Pittaluga, S., Verhoef, G., Criel, A., Maes, A., Nuyts, J., Boogaerts, M. & De Wolf Peeters, C. (1996) Prognostic significance of bone marrow trephine and peripheral blood smears in 55 patients with mantle cell lymphoma. Leukemia and Lymphoma, 21, Pombo De Oliveira, M., Jaffe, E. & Catovsky, D. (1989) Leukaemic phase of mantle zone (intermediate) lymphoma: its characterization in 11 cases. Journal of Clinical Pathology, 42, Raffeld, M. & Jaffe, E. (1991) bcl-1, t(11;14), and mantle cell-derived lymphomas. Blood, 78, Rimokh, R., Berger, F., Delsol, G., Charrin, C., Bertheas, M., Ffrench, M., Garoscio, M., Felman, P., Coiffier, B., Bryon, P., Rochet, M., Gentihomme, O., Germain, D. & Magaud, J. (1993) Rearrangement and overexpression of the BCL-1/PRAD-1 gene in intermediate lymphocytic lymphomas and in t(11q13)-bearing leukemias. Blood, 81, Mantle Cell Lymphoma in Bone Marrow 309 Rosenberg, C., Wong, E., Petty, E., Bale, A., Tsujimoto, Y., Harris, N. & Arnold, A. (1991) PRAD1, a candidate BCL1 oncogene: mapping and expression in centrocytic lymphoma. Proceedings of the National Academy of Sciences of the United States of America, 88, Rozman, C., Montserrat, E., Rodriquez-Fernandez, J., Ayats, R., Vallespi, T., Parody, R., Rios, A., Prados, D., Morey, M., Gomis, F., Alcala, A., Gutierrez, M., Maldonado, J., Gonzalez, C., Hernandez- Nieto, L., Cabrera, A. & Fernandez-Ranada, J. (1984) Bone marrow histologic pattern the best single prognostic parameter in chronic lymphocytic leukemia: a multivariate survival analysis of 329 cases. Blood, 64, Sheibani, K., Burke, J., Swartz, W., Nademanee, A. & Winberg, C. (1988) Monocytoid B-cell lymphomas: clinicopathologic study of 21 cases of a unique type of low-grade lymphoma. Cancer, 62, Shivdasani, R., Hess, J., Skarin, A. & Pinkus, G. (1993) Intermediate lymphocytic lymphoma: clinical and pathologic features of a recently characterized subtype of non-hodgkin s lymphoma. Journal of Clinical Oncology, 11, Spier, C., Grogan, T., Fielder, K., Richter, L. & Rangel, C. (1986) Immunophenotypes in well-differentiated lymphoproliferative disorders, with emphasis on small lymphocytic lymphoma. Human Pathology, 17, Strickler, J., Medeiros, L., Copenhaver, C., Weiss, L. & Warnke, R. (1988) Intermediate lymphocytic lymphoma: an immunophenotypic study with comparison to small lymphocytic lymphoma and diffuse small cleaved cell lymphoma. Human Pathology, 19, Swerdlow, S., Habeshaw, J., Murray, L., Dhaliwali, H., Lister, T. & Stansfeld, A. (1983) Centrocytic lymphoma: a distinct clinicopathologic and immunologic entity. American Journal of Pathology, 113, Swerdlow, S., Zukerberg, L., Yang, W., Harris, N. & Williams, M. (1996) The morphologic spectrum of non-hodgkin s lymphomas with BCL1/cyclin D1 gene rearrangements. American Journal of Surgical Pathology, 20, Traweek, S. & Sheibani, K. (1992) Monocytoid B-cell lymphoma: the biologic and clinical implications of peripheral blood involvement. American Journal of Clinical Pathology, 97, Vadlamudi, G., Lionetti, K., Greenberg, S. & Mehta, K. (1996) Leukemia phase of mantle cell lymphoma: two case reports and review of the literature. Archives of Pathological Laboratory Medicine, 120, van den Oord, J., de Wolf-Peeters, C., Pulford, K., Mason, D. & Desmet, V. (1986) Mantle zone lymphoma: immuno- and enzymehistochemical studies on the cell of origin. American Journal of Surgical Pathology, 10, Vandenberghe, E., De Wolf-Peeters, C., Van Den Oord, J., Wlodarska, I., Delabie, J., Stul, M., Thomas, J., Michauz, J., Mecucci, C., Cassiman, J. & Van Den Berghe, H. (1991) Translocation (11;14): a cytogenetic anomaly associated with B-cell lymphomas of nonfollicle centre cell lineage. Journal of Pathology, 163, Vasef, M., Medeiros, J., Koo, C., McCourty, A. & Brynes, R. (1997) Cyclin D1 immunohistochemical staining is useful in distinguishing mantle cell lymphoma from other low-grade B-cell neoplasms in bone marrow. American Journal of Clinical Pathology, 198, Velders, G., Kluin, N., Elemans, J., De Boer, C., Hermans, J., Noordijk, E., Schuuring, E., Kramer, M., Van Deijk, W., Rahder, J., Kluin, P. & Van Krieken, J. (1996) Mantle-cell lymphoma: a population-based clinical study. Journal of Clinical Oncology, 14, Wasman, J., Rosenthal, N. & Farhi, D. (1996) Mantle cell lymphoma:

9 310 P. L. Cohen et al morphologic findings in bone marrow involvement. American Journal of Clinical Pathology, 106, Weisenburger, D., Nathwani, B., Diamond, L., Winberg, C. & Rappaport, H. (1981) Malignant lymphoma, intermediate lymphocytic type: a clinicopathologic study of 42 cases. Cancer, 48, Weisenburger, D., Sanger, W., Armitage, J. & Purtilo, D. (1987) Intermediate lymphocytic lymphoma: immunophenotypic and cytogenetic findings. Blood, 69, Williams, M., Meeker, T. & Swerdlow, S. (1991) Rearrangement of the chromosome 11 bcl-1 locus in centrocytic lymphoma: analysis with multiple breakpoint probes. Blood, 78, Williams, M., Swerdlow, S. & Meeker, T. (1993) Chromosome t(11;14)(q13;q32) breakpoints in centrocytic lymphoma are highly localized at the bcl-1 major translocation cluster. Leukemia, 7, Williams, M., Westermann, C. & Swerdlow, S. (1990) Genotypic characterization of centrocytic lymphoma: frequent rearrangement of the chromosome 11 bcl-1 locus. Blood, 76, Witzig, T., Banks, P., Stenson, M., Greipp, P., Katzmann, J., Habermann, T., Colgan, J. & Gonchoroff, N. (1990) Rapid immunotyping of B-cell non-hodgkin s lymphomas by flow cytometry: a comparison with the standard frozen-section method. American Journal of Clinical Pathology, 94, Zucca, E., Roggero, E., Pinotti, G., Pedrinis, E., Cappella, C., Venkco, A. & Bavalli, F. (1995) Patterns of survival in mantle cell lymphoma. Annals of Oncology, 6, Zucca, E., Stein, H. & Coiffier, B. (1994) European Lymphoma Task Force (ELTF): report on the workshop on mantle cell lymphoma (MCL). Annals of Oncology, 5, Zukerberg, L., Medeiros, J., Ferry, J. & Harris, N. (1993) Diffuse lowgrade B-cell lymphomas. American Journal of Clinical Pathology, 100,