SEARCH

SEARCH BY CITATION

Keywords:

  • intrasinusoidal infiltration;
  • bone marrow;
  • splenic marginal zone lymphoma with villous lymphocytes;
  • large granular lymphocyte leukaemia

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

Summary. We report a retrospective immunohistochemical study on bone marrow biopsies of 43 patients with different types of lymphomas showing unusual intrasinusoidal infiltration. Most of these patients presented with splenomegaly (74·4%) and peripheral lymphocytosis (83%). In 20/43 patients, lymphoid infiltrates were not detectable on haematoxylin–eosin sections. After immunohistochemistry on bone marrow biopsies and blood and bone marrow smear examinations, the following diagnoses were made: splenic marginal zone lymphoma with villous lymphocytes (SLVL) in 24 patients, large granular lymphocyte (LGL) leukaemia in 14 patients, hepatosplenic T-cell lymphoma in two patients, anaplastic large cell lymphoma in two patients and intravascular large B-cell lymphoma in one patient. In the presence of intrasinusoidal infiltrates of small lymphocytes, a B-cell phenotype (CD20+, CD76/DBA44+/–) was associated with splenic marginal zone lymphoma whereas intrasinusoidal CD3/CD45RA-positive T-cell infiltrates were strongly suggestive of LGL leukaemia. Intrasinusoidal bone marrow infiltration appears to be a common feature of distinct lymphoma subtypes. Immunohistochemical analysis is essential to detect intrasinusoidal medullary infiltrates (which may be minimal) and should be systematically performed in patients with splenomegaly and peripheral lymphocytosis.

Bone marrow biopsy findings determine the stage of disease in lymphomas and may contribute to appropriate treatment. However, conventional examination often eludes the diagnosis of lymphoid infiltrates, particularly if they are minimal or topographically unusual. We retrospectively studied bone marrow biopsies from 43 patients, in which the lymphoid infiltration, difficult to detect on routine haematoxylin–eosin (H&E) stained sections, was considered predominantly intrasinusoidal, by systemic use of immunohistochemistry with monoclonal antibodies that could react on decalcified, paraffin-embedded tissue. That intrasinusoidal bone marrow infiltration as a possible hallmark of B-cell splenic lymphoma has been reported previously (Franco et al, 1996; Labouyrie et al, 1997) but our experience after immunohistochemical study showed that different subtypes of B- and T-cell lymphomas cause intrasinusoidal involvement in the bone marrow, with a potential for further classification of conditions within the lymphoid malignancies.

Patients. The cases selected for study were from the archival data of patients treated after 1990 from the departments of Pathology at Purpan Hospital (Toulouse) and Gui de Chauliac Hospital (Montpellier). Specific coding for intrasinusoidal lymphoid infiltrates in bone marrow biopsy permitted this retrospective search. Clinical data were collected and checked by two haematologists. The age of the patients ranged from 10 to 83 years (median 58·5 years). There were 29 males and 14 females. Thirty-two patients (74·4%) presented with splenomegaly and only five with peripheral lymphadenopathy (11%) (Table I).

Table I.  Clinical characteristics of 43 patients with lymphomatous intrasinusoidal bone marrow involvement.
Lymphoma diagnosisnMedian age (years) Sex (M/F) Splenomegaly*Peripheral adenopathiesLymphocytosis > 4 × 109/l
  • *

    Palpable or diagnosed after radiological exploration.

  • One of these two patients had sarcoidosis.

  • SLVL, splenic marginal zone lymphoma with villous lymphocytes; LGL, large granular lymphocytes.

SLVL246517/724/242/2424/24
T LGL leukaemia1464·8 8/6 5/140/14 9/14
T hepatosplenic lymphoma 231 2/0 2/20/2 1/2
Haemangioendotheliomatosis 151 1/0 1/11/1 0/1
Anaplastic large cell lymphoma 222·5 1/1 0/22/2 0/1
Total: 434358·529/43 (67·4%)32/43 (74·4%)5/43 (11%)34/41 (83%)

Bone marrow biopsies. Bone marrow biopsies were fixed overnight in either Duboscq's, Bouin's or formalin fixative, decalcified in 2% nitric acid or EDTA and then paraffin embedded. For each patient, conventionally stained sections (i.e. haematoxylin-and-eosin and reticulin staining) were re-examined by two pathologists without knowledge of the final immunohistochemical results. Complete immunohistochemical analysis was performed in all cases. Tissue sections were deparaffinized in xylene and rehydrated in a series of graded alcohol. The slides were incubated in 10 mmol/l citrate buffer, pH 6·0, at 121°C for 10 min in a steam autoclave. Immunohistochemical analysis was performed with an automaton (Techmate 500; Dako, Glostrup, Denmark), using the Dako ChemMate detection kit with rabbit anti-mouse secondary antibodies and diaminobenzidine (DAB) as chromogen, according to the manufacturer's instructions. First, we applied anti-CD20 (L26) and anti-CD3. If a B-cell infiltrate was detected, the analysis was supplemented by the following antibodies: anti-CD76/DBA44, anti-CD5 and anti-CD23. In the presence of a T-cell infiltrate, we used anti-CD45RA/DBB.42, anti-CD2, anti-CD5, anti-CD4, anti-CD8, anti-CD56, anti-CD57, anti-TIA1 and antigranzyme B antibodies. For cases of anaplastic large cell lymphoma, anti-CD30, anti-EMA and anti-ALK antibodies were applied. All antibodies were from Dako, except CD5 and CD30/BerH2 (Novocastra, Newcastle, UK).

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

Histological results are summarized in Table II. Three patterns of bone marrow infiltration were observed: exclusively intrasinusoidal, intrasinusoidal and interstitial, and intrasinusoidal and interstitial with non-paratrabecular aggregates (Fig 1).

Table II.  Results initially proposed after H&E-stained sections examination in 43 bone marrow biopsies with intrasinusoidal lymphoid involvement detected by immunophenotyping.
Lymphoma diagnosisnNormal BMHyperplastic BM ± maturation anomalySuspected lymphoid infiltrate?LymphomaIntravascular lymphoid infiltrate
  1. BM, bone marrow; SLVL, splenic marginal zone lymphoma with villous lymphocytes; LGL, large granular lymphocytes.

SLVL243/24 4/246/2411/244/24
T LGL leukaemia141/14 8/140/14 5/142/14
T hepatosplenic lymphoma 20/2 1/21/2 0/20/2
B angioendotheliomatosis 10/1 1/10/1 0/10/1
Anaplastic large cell lymphoma 20/2 2/20/2 0/20/2
Total: 43434/43 (9·3%)16/43 (37·2%)7/43 (16·2%)16/43 (37·2%)6/43 (16·2%)
image

Figure 1. Patterns of lymphoid infiltrate observed in bone marrow biopsies. (A) Exclusively intrasinusoidal, (B) intrasinusoidal and interstitial, (C) intrasinusoidal and interstitial, with non-paratrabecular aggregates.

Download figure to PowerPoint

Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

In 24 patients (17 men and seven women, M:F = 2·4), the final diagnosis was splenic marginal zone lymphoma with villous lymphocytes. The mean age at the time of the biopsy was 65 years. Clinically, these patients presented with splenomegaly in all cases. Blood cell count showed a peripheral lymphocytosis (> 4 × 109/l) in all patients, but was associated with thrombocytopenia (platelet count < 150 × 109/l) in three patients and anaemia in two patients. The diagnosis was based on the presence of atypical and villous lymphoid cells in the blood and/or in the marrow smears. In peripheral blood smears, villous lymphocytes, characterized by an irregular cytoplasmic outline, short and thin villi, usually concentrated at one or both cell poles, were identified in only 18 patients. In four patients, villous lymphocytes were identified in bone marrow smears only. Two more patients were retrospectively diagnosed after re-examination of bone marrow films. The mean percentage of villous lymphocytes in bone marrow smears was 26·5% (range 8–52). The phenotype was determined in 10 patients by flow cytometry.

Histologically, the bone marrow presented a normal (5/24) or slightly hypercellular parenchyma (19/24) with moderate maturation abnormalities in megakaryocyte (13/24), erythroid (17/24) and myeloid (10/24) cell lines. The reticulin content was slightly increased in eight patients. Lymphocytic infiltrates were obvious in 11 patients, but an intravascular topography was suggested only in four of these patients. In seven of these 11 patients, small lymphoid nodules were noted. Lymphomatous involvement was suspected in six patients and undetected in seven patients. The final diagnosis was normal bone marrow in three patients, reactive hyperplasia with maturation abnormalities in four patients, possible small cell lymphoma requiring immunohistochemical confirmation in six patients and small cell lymphoma in 11 patients.

In all of these biopsies, immunohistochemical analysis revealed an exclusively intrasinusoidal B-cell infiltrate in nine patients, associated with an interstitial component in 15 patients (Fig 2). These B cells were of small to medium size, with round or slightly cleaved nuclei, finely dispersed chromatin and no evident nucleoli. They expressed CD45/LCA (24/24), CD20 (24/24) and CD76/DBA44 (14/16). They were negative for CD3, CD23 and CD43, but 2/20 patients tested showed a questionable positive reaction for CD5 without cyclin D1 expression. In seven patients with lymphoid nodules, immunohistochemistry highlighted the CD20-positive nature of aggregates in five patients. In the other two patients, they consisted of a mixture of T and B cells and were interpreted as being reactive lymphoid nodules.

image

Figure 2. B-cell splenic lymphoma with villous lymphocytes (original magnification, ×25). Upper panel: on H&E sections, the bone marrow presented hypercellular parenchyma with a slight intrasinusoidal lymphoid infiltrate. Lower panel: anti-CD20 immunostaining outlined this intrasinusoidal infiltrate and the irregular cytoplasmic outline of small lymphoid cells (insert, original magnification, ×100).

Download figure to PowerPoint

Splenectomy was performed in 16 patients with confirmation of the finding of a B-cell marginal zone lymphoma. Venous sinuses and Billroth cords were infiltrated by small cells with regular hyperchromatic and nucleolated nuclei, and an abundant cytoplasm. In two patients, however, lymphoid infiltration was composed of sheets of medium to large size cells, some of which closely resembled immunoblasts. The intravascular growth pattern was predominant and the phenotype was identical in the spleen and bone marrow. These two patients were diagnosed as splenic marginal zone lymphoma with large villous lymphocytes.

Large granular lymphocyte leukaemia: 14 patients

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

This diagnosis was retained in eight men and six women (M:F = 1·3). Their mean age was 64·8 years (range 31–84). All patients presented with asthenia which was associated with fever and recurrent bacterial infections in four patients, with splenomegaly in five patients, with autoimmune manifestations in two patients and with diabetes insipidus because of infiltration in the hypophysis in one patient.

Peripheral lymphocytosis was present in 9/14 patients but an increase in the absolute number of circulating LGLs, which had the same morphology as the LGLs noted in normal individuals (intracytoplasmic azurophilic granules), was consistently present. In 10 patients with available peripheral blood smears for retrospective examination, the mean number was 3408 × 109 LGL/l (range 920–5000). Neutropenia (< 1·5 × 109/l) was present in 10 out of 14 patients (71·4%). Anaemia was seen in six patients and thrombocytopenia (platelet count < 150 × 109/l) in five patients. LGL excess was confirmed on bone marrow smears. Flow cytometry was performed in 10 patients: nine were of T-cell phenotype (CD2+, CD3+, CD5, CD4, CD8+, CD56 and CD57+) and one of natural killer (NK)-cell phenotype (CD2+, CD3, CD4, CD8, CD56+ and CD57).

The morphological features of the bone marrow biopsy showed a moderate (n = 8) or marked (n = 5) hypercellularity. The reticulin content was slightly increased in 10 patients (71·4%). A diffuse, apparently interstitial, small lymphocyte infiltrate was obvious in five patients and associated in four patients with small poorly defined, non-trabecular, lymphoid aggregates (1–4 nodules per biopsy). The initial diagnoses were normal bone marrow in one patient, reactive/hyperplastic bone marrow in eight patients and indolent lymphoma in five patients.

In all patients, immunohistochemical analysis revealed a CD3+, CD45RA+, CD5+ and CD8+ lymphoid infiltrate, with an exclusive intrasinusoidal pattern in five patients (Fig 3), and both intrasinusoidal and interstitial pattern in the others. In one patient only, small lymphoid nodules predominantly expressed T-cell markers while in the other three patients, nodules, composed of a mixture of T and B cells, were interpreted as reactive lymphoid nodules. All these patients were negative for CD56 whereas anti-CD57 antibody showed heterogeneous, weak staining in 4/12 patients. This discrepancy with flow cytometry may be explained by antigenic degradation during the bone marrow biopsy processing.

image

Figure 3. Large granular lymphocyte leukaemia (original magnification, ×40). Upper panel: an intrasinusoidal lymphoid infiltrate was suspected in this bone marrow biopsy on H&E stain sections. Lower panel: immunochemistry revealed that this CD3+ infiltrate expressed CD45RA/DBA42 and confirmed its intrasinuoidal growth pattern.

Download figure to PowerPoint

Hepatosplenic T-cell lymphoma: two patients

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

These two patients were 21- and 43-year-old-men who presented with fever and hepatosplenomegaly. Lymphoma was diagnosed on liver biopsies, which showed sinusoidal infiltrates of pleomorphic medium size cells. These cells expressed pan-T cell markers (CD45R0, CD3) but were CD4, CD8 and T-cell receptor β (TCR-β) negative. They expressed TCR-γ with a cytotoxic profile (TIA1+, granzyme B+).

Bone marrow was hypercellular, with marked maturation abnormalities. In patient 1, the lymphoid infiltrate was suspected on H&E sections. In patient 2, histiocytes were numerous with haemophagocytic features and the initial diagnosis was hyperplastic bone marrow with haemophagocytic syndrome. Immunohistochemical analysis revealed a slight sinusoidal infiltrate, comprising medium-sized lymphoid T cells with moderately clumped chromatin and a rim of pale cytoplasm (CD2+, CD3+, CD4, CD8 and CD5). The survival time of these patients was short: 9 and 14 months respectively.

Anaplastic large cell lymphoma: two patients

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

In these two young patients (F10 and M35), bone marrow biopsy was performed at presentation. Both had a common-type anaplastic large cell lymphoma diagnosed on peripheral lymph node biopsy. H&E-stained sections showed an hypercellular marrow without maturation abnormalities or reticulin network increase. No lymphoid infiltrate was noted. The immunohistochemical analysis highlighted the presence of sparse large cells expressing the CD30, EMA and ALK proteins. The majority of these atypical cells displayed an intrasinusoidal growth pattern (Fig 4).

image

Figure 4. In this patient with an anaplastic large cell lymphoma, the bone marrow biopsy was initially considered as negative. The immunohistochemical analysis highlighted the presence of rare intravascular large atypical cells expressing the ALK protein (original magnification, ×40).

Download figure to PowerPoint

Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

We found three patients of intravascular lymphomatosis or so-called malignant angioendothelimatosis that were confirmed at autopsy. Bone marrow biopsies were negative in two patients. The positive patient was a 51-year-old man who presented with fever, persistent headaches, splenomegaly and lymphadenopathy. Bone marrow biopsy showed a highly cellular parenchyma. Immunohistochemistry revealed the B-cell nature of the lymphoma cells (CD20+, DBA42+, bcl2+, CD3, BNH9, CD30/+, LMP) and their exclusive intrasinusoidal localization (Fig 5). In spite of chemotherapy, the patient rapidly developed unrelenting neurological problems and died 2 months later.

image

Figure 5. (Original magnification, ×40) upper panel: In this patient, bone marrow was hyperplastic with marked maturation anomalies and the lymphoid infiltrate was difficult to confirm on H&E stained sections. Lower panel: Anti-CD20 immunostaining revealed an obvious intrasinusoidal exclusive large cell infiltrate compatible with an intravascular lymphomatosis.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References

An intravascular lymphoma growth pattern is well known in rare entities like ‘malignant angioendotheliomatosis’ and hepatosplenic T-cell lymphoma (Gaulard et al, 1991). Recently, two reports described intrasinusoidal bone marrow infiltration in 5/11 patients with splenic B-cell lymphoma with villous lymphocytes (Franco et al, 1996; Labouyrie et al, 1997). In the present series of 43 patients with intrasinusoidal lymphomatous infiltrates, we identified five different non-Hodgkin's lymphoma subtypes: splenic marginal zone lymphomas with villous lymphocytes (SLVL), large granular lymphocyte leukaemia (LGL), hepatosplenic T-cell lymphoma, anaplastic large cell lymphoma and intravascular B-cell lymphoma. Notably, most of these patients presented with splenomegaly and lymphocytosis without peripheral lymphadenopathy. In this clinical context, differential diagnosis was first based on peripheral blood smear evaluation. However, distinguishing neoplastic from reactive lymphocytosis by cytological criteria was difficult.

Among our 43 patients with an intrasinusoidal bone marrow lymphomatous infiltrate, splenic marginal zone lymphoma with villous lymphocytes (SLVL) was the most common aetiological group (more than half of the patients). This entity is now considered as a form of splenic marginal zone lymphoma (Isaacson et al, 1994; Gruska-Westwood et al, 1999). In our series, all patients (mean age: 65 years) presented with splenomegaly and peripheral lymphocytosis. Villous lymphocytes were initially detected and identified in only 18/24 patients on blood smears. In the large series of Troussard et al (1996), lymphocytosis was < 4 × 109/l in 24/100 patients and diagnosis required bone marrow aspiration and bone marrow biopsy. In our study, bone marrow biopsies were moderately hypercellular with slight maturation abnormalities in the three cell lineages, without any significant reticulin content increase. On H&E sections, lymphomatous involvement was obvious in 11/24, suspected in 6/24 and undetected in 7/24 patients. In all of these biopsies, immunohistochemistry revealed a small B-cell intrasinusoidal infiltrate. Cytologically, the differential diagnosis may include atypical variants of hairy cell leukaemia (HCL). However, an interstitial infiltrate with increased reticulin is in favour of HCL whereas intrasinusoidal predominant infiltrate without reticulin modification is in favour of SLVL.

The second most important aetiological group was LGL leukaemia. Large granular lymphocytes have been described as large lymphocytes with azurophilic granules in an abundant cytoplasm. They constitute 10–15% of normal peripheral blood lymphocytes. Lymphoproliferative disorders involving LGL are included in chronic leukaemias and may be of T-cell (CD3+) or NK-cell (CD3) phenotype (Harris et al, 1994). As in our cases, neutropenia is generally the most prevalent haematological consequence (85%), with recurrent bacterial infections. Anaemia and thrombocytopenia are less common (50% and 20% of patients respectively) (Loughran & Starkebaum, 1987). In our series, most bone marrow biopsies were hypercellular, with an increase of the reticulin content. Evans et al (2000) recently showed the usefulness of quantitative immunohistochemistry for differential diagnosis between LGL leukaemia and reactive bone marrow infiltrates. However, the authors discussed a diffuse interstitial pattern rather than intrasinusoidal. In our opinion, the concomitant positivity of small lymphocytes for CD3 and CD45RA/DBB42 together with a peculiar intrasinusoidal topography are strongly suggestive of LGL leukaemia. DBB42/anti-CD45RA antibody reacts with B lymphocytes of both the germinal centre and mantle zone of hyperplastic lymphoid follicles. This antigen is also expressed on a subpopulation of naive T cells and NK cells, and on a subpopulation of macrophages (Sewell et al, 1997). In haematological malignancies, DBB42 reacts with 98% of B-cell lymphomas. A weak staining in a few patients with T-cell lymphoma (Lennert's lymhoma and lymphoblastic lymphomas) has been reported (Al Saati et al, 1989). Reactive T-cell infiltrates and T-cell chronic lymphocytic leukaemia do not preferentially localize in sinusoids and remain CD45RA negative.

Two patients corresponded to the extension of a previously diagnosed hepatosplenic T-cell lymphoma. γδ T-lymphocytes represent a subset of cytotoxic T cells that exhibit preferential homing to the sinusoidal area of the spleen and the liver. Bone marrow infiltration is an almost constant feature of this lymphoma, with the same intrasinusoidal expansion (Gaulard et al, 1991; Cooke et al, 1996). One of our patients presented with the haemophagocytic syndrome.

‘Angio-endotheliomatosis’ or intravascular lymphomatosis is an extremely rare variant of aggressive large B-cell lymphoma associated mainly with neurological manifestations. Negative bone marrow biopsies are common in this disorder (Wick et al, 1986) even though immunoglobulin heavy chain gene rearrangement studies by polymerase chain reaction (PCR) were positive in 5/5 patients tested by DiGiuseppe et al, 1997). In our files, we retrieved three cases of intravascular B-cell lymphomatosis. Only one patient presented intrasinusoidal bone marrow infiltrates. Bone marrow involvement in intravascular lymphomatosis is probably not as rare as this suggests but is difficult to diagnose without immunohistochemistry (DiGiuseppe et al, 1994; Tucker et al, 1999; Parrens et al, 2000).

Regarding anaplastic large cell lymphoma (ALCL), Fraga et al (1995) highlighted the utility of immunohistochemistry with anti-CD30 and anti-EMA antibodies in bone marrow biopsy. In a series of 42 patients with ALCL and bone marrow biopsy, 17% of the patients had bone marrow involvement visible on conventional examination, but immunohistochemistry detected occult malignant cells in 23% of patients in which the bone marrow biopsy had been considered negative (Fraga et al, 1995).

Finally, with the exception of patients with ALCL, most patients in this series had a similar clinical presentation with splenomegaly and lymphocytosis. In this context, immunohistochemistry is deemed essential to detect a possible minimal infiltrate. In the presence of intrasinusoidal bone marrow infiltrates of small lymphocytes, a B-cell phenotype (CD20+, CD76/DBA44+/–) would indicate SLVL whereas intrasinusoidal CD3/CD45RA-positive T-cell infiltrates are strongly suggestive of LGL leukaemia. In both situations, blood and marrow smear examinations are critical for confirmation but in other conditions, e.g. ALCL, infiltration may be detected only on trephine biopsy.

The reasons for the confinement or predilection of lymphoid cells to the vascular system are unclear, but are probably related to the expression of specific lymphocyte–endothelium adhesion molecules. Our study suggests that this phenomenon is possible in bone marrow and is independent of the B- or T-cell lineage.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Splenic marginal zone lymphoma with villous lymphocytes (SLVL): 24 patients
  6. Large granular lymphocyte leukaemia: 14 patients
  7. Hepatosplenic T-cell lymphoma: two patients
  8. Anaplastic large cell lymphoma: two patients
  9. Intravascular lymphomatosis (so-called malignant angioendotheliomatosis): one patient
  10. Discussion
  11. Acknowledgment
  12. References
  • Al Saati, T., Caspar, S., Brousset, P. & Delsol, G. (1989) Production of anti-B monoclonal antibodies (DBB.42, DBA.44, DNA.7, and DND.53) reactive on paraffin-embedded tissues with a new B-lymphoma cell line grafted into athymic nude mice. Blood, 74, 24762485.
  • Cooke, C.B., Krenacs, L., Stetler-stevenson, M., Grebier, T.C., Raffeld, M., Kingma, D.W., Abruzzo, L., Frantz, C., Kaviani, M. & Jaffe, E.S. (1996) Hepatosplenic T-cell lymphoma: a distinct clinicopathologic entity of cytotoxic gamma delta T-cell origin. Blood, 88, 42654274.
  • DiGiuseppe, J., Nelson, W., Seifter, E., Boitnott, J. & Mann, R.B. (1994) Intravascular lymphomatosis: a clinicopathological study of 10 cases and assessment of response to chemotherapy. Journal of Clinical Oncology, 12, 25732579 .
  • DiGiuseppe, J.A., Hartman, D., Freter, C., Cossman, J. & Mann, R.B. (1997) Molecular detection of bone marrow involvement in intravascular lymphomatosis. Modern Pathology, 10, 3337.
  • Evans, H.L., Burks, E., Viswanatha, D. & Larson, R.S. (2000) Utility of immunohistochemistry in bone marrow evaluation of T-lineage large granular lymphocyte leukemia. Human Pathology, 31, 12661273.
  • Fraga, M., Brousset, P., Schlaifer, D., Payen, C., Robert, A., Rubie, H., Huguet-Rigal, F. & Delsol, G. (1995) Bone marrow involvement in anaplastic large cell lymphoma: immunohistochemical detection of minimal disease and its prognostic significance. American Journal of Clinical Pathology, 103, 8289.
  • Franco, V., Florena, A.M. & Campesi, G. (1996) Intrasinusoidal bone marrow infiltration: a possible hallmark of splenic lymphoma. Histopathology, 29, 571575.
  • Gaulard, P., Kanavaros, P., Farcet, J.P., Rocha, F.D., Haioun, C., Divine, M., Reyes, F. & Zafrani, E.S. (1991) Bone marrow histologic and immunohistochemical findings in peripheral T-cell lymphoma: a study of 38 cases. Human Pathology, 22, 331338.
  • Gruska-Westwood, A.M., Mattutes, E., Coignet, L., Wotherspoon, A. & Catocsky, D. (1999) The incidence of trisomy 3 in splenic lymphoma with villous lymphocytes: a study by FISH. British Journal of Haematology, 103, 600604.
  • Harris, N.L., Jaffe, H.S., Stein, H., Banks, P.M., Chan, J.K., Cleary, M.L., Delsol, G., De Wolf-Peeters, C., Falini, B., Gatter, K.C., Grogan, T.M., Isaacson, P.G., Knowles, D.M., Mason, D.Y., Muller-Hermelink, H.K., Pilei, S.A., Piris, M.A., Rafkiaer, E. & Warnke, R.A. (1994) A revised European–American classification of lymphoid neoplasms: a proposal from the international lymphoma study group. Blood, 84, 13611392.
  • Isaacson, P.G., Matutes, E., Burke, M. & Catovsky, D. (1994) The histopathology of splenic lymphoma with villous lymphocytes. Blood, 11, 38283834.
  • Labouyrie, E., Marit, G., Vial, J.P., Lacombe, F., Fialon, P., Bernard, P., De Mascarel, A. & Merlio, J.P. (1997) Intrasinusoidal bone marrow involvement by splenic lymphoma with villous lymphocytes: a helpful immunohistologic feature. Modern Pathology, 10, 10151020.
  • Loughran, T.P. & Starkebaum, G. (1987) Large granular lymphocyte leukemia: report of 38 cases and review of the literature. Medicine (Baltimore), 66, 397405.
  • Parrens, M., Dubus, P., Agape, P., Rizcallah, E., Marit, G., De Mascarel, A. & Merlio, J.P. (2000) Intrasinusoidal bone marrow infiltration revealing intravascular lymphomatosis. Leukemia and Lymphoma, 37, 219223.
  • Sewell, W., Cooley, M. & Hegen, M. (1997) CD45 workshop panel report. In: Leucocyte Typing VI (ed. by Kishimoto et al ), pp. 499502. Garland Publishing, Inc., New York.
  • Troussard, X., Valensi, F., Duchayne, E., Garand, R., Felman, P., Tulliez, M., Henry-Amar, M., Bryon, P. & Flandrin, G. (1996) Splenic lymphoma with villous lymphocytes: clinical presentation, biology and prognostic factors in a series of 100 patients. British Journal of Haematology, 93, 731736.
  • Tucker, T.J., Bardales, R.H. & Miranda, R.N. (1999) Intravascular lymphomatosis with bone marrow involvement. Archives of Pathology and Laboratory Medicine, 123, 952956.
  • Wick, M., Mills, S., Scheithauer, B., Cooper, P., Davitz, M. & Parkinson, K. (1986) Reassessment of malignant angioendotheliomatosis. American Journal of Surgical Pathology, 10, 112123.