Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
JCP Online First, published on December 10, 2013 as 10.1136/jclinpath-2013-201520 Original article
Bone marrow trephine biopsy involvement by lymphoma: review of histopathological features in 511 specimens and correlation with diagnostic biopsy, aspirate and peripheral blood findings Vishakha Sovani,1 Caroline Harvey,2 Andrew P Haynes,2 Andrew K McMillan,2 David M Clark,3 Simon R O’Connor1 1
Department of Histopathology, Nottingham University Hospital, Nottingham, UK 2 Department of Haematology, Nottingham University Hospital, Nottingham, UK 3 Department of Cellular Pathology, Lincoln County Hospital, Lincoln, UK Correspondence to Dr Simon O’Connor, Department of Histopathology, Nottingham University Hospital, City Campus, Hucknall Road, Nottingham, NG5 1PB, UK; [email protected] Received 1 February 2013 Revised 5 September 2013 Accepted 20 October 2013
ABSTRACT Aims This study aimed to evaluate the key features of bone marrow trephine (BMT) biopsy involvement by lymphoma. Methods 511 cases were assessed for percentage of marrow involvement, pattern of involvement (diffuse, nodular, paratrabecular, interstitial or intrasinusoidal), presence/absence of granulomas, stromal fibrosis and necrosis, presence/absence of neoplastic/reactive follicles and discordance with other biopsy sites. Correlation with aspirate and peripheral blood findings was made in a subset of 345 patients (167 aspirates, 178 blood). Results The most frequent subtype was follicular lymphoma (26.2%) followed by extranodal marginal zone (23.1%), lymphoplasmacytic (19.2%), diffuse large B cell (DLBCL) (12.5%), Hodgkin (HL) (5.7%) and mantle cell lymphomas (4.3%). The predominant pattern in follicular lymphoma was paratrabecular. Marginal zone lymphomas of all types and lymphoplasmacytic lymphoma showed a relatively even distribution between diffuse, interstitial, paratrabecular and nodular patterns. The majority of mantle cell lymphoma cases showed either diffuse or nodular patterns. A diffuse pattern was common in DLBCL and Burkitt lymphomas. An intrasinusoidal pattern was seen only in extranodal and splenic marginal zone lymphomas. Granulomas and fibrosis were uncommon in small cell B cell lymphomas but frequent in DLBCL and HL. Aspirate and trephine results concurred in 73.8% of cases overall, but this varied widely between subtypes. Peripheral blood involvement rates by lymphoma also varied, with a mean of 37.1%. Conclusions Different lymphomas often demonstrate reliably characteristic architectural patterns of marrow involvement which can help differentiate them even when cytological features do not permit this, and marrow stromal and other background changes may also be useful pointers towards a particular lymphoma subtype.
PARAGRAPH ON WHAT THE PAPER ADDS
To cite: Sovani V, Harvey C, Haynes AP, et al. J Clin Pathol Published Online First: [ please include Day Month Year] doi:10.1136/ jclinpath-2013-201520
We believe this paper adds to the currently available data on bone marrow trephine (BMT) biopsy involvement by lymphoma for the following reasons. It represents by a significant margin the largest series of trephine biopsies examined in this way to date and is also of interest as representing consecutive cases within a defined geographical region and population group in the UK ( previous series of BMT cores have mainly been from tertiary centres in the USA, where cases may often be received from outside the region and this can
Sovani V, et al. J Clin Pathol 2013;0:1–7. Copyright Article author doi:10.1136/jclinpath-2013-201520 (or their employer) 2013.
sometimes mean that population-based incidence features are unreliable). It also makes comparison of BMT, aspirate and peripheral blood findings in lymphoma over a far larger number of cases than has previously been analysed, and examines BMT findings in a significant number of Hodgkin’s lymphomas which have been neglected in most previous studies of lymphoma involving bone marrow. Although immunohistochemical, flow cytometric and molecular findings are essential in diagnostic haematopathology, BMT morphology remains an important point of reference for their interpretation, and we believe the findings of this paper contribute to the body of knowledge available on this subject.
MANUSCRIPT WHO lymphoma classification employs combined morphological, immunophenotypic and molecular genetic information to accurately define individual lymphoma subtypes as distinct clinicopathologcal entities.1 Bone marrow involvement occurs in around 40% of all cases of lymphoma,2 and BMT biopsy staging has significant implications for prognostication and therapy as well as often playing an important role in diagnosis.3 Despite the importance of immunohistochemical, flow cytometric and molecular data, BMT morphological evaluation remains for lymphoma the gold standard against which other investigations are interpreted.4 Pattern analysis forms an important part of this, complementing cytomorphological information5 and is especially useful in small to medium-sized cell lymphomas where cytological features may be similar between subtypes. This study was designed to assess patterns of BMT involvement by lymphoma in well-characterised cases classified by the 2008 WHO classification of lymphoid neoplasms, and correlate these with, where available, aspirate and peripheral blood findings. Frequency and nature of discordance between the findings in the bone marrow and other specimens was also analysed, as was concordance with aspirate and peripheral blood findings. This represents the largest series of trephines examined from this perspective to date.
MATERIALS AND METHODS Totally, 612 sequential BMT cores showing involvement by lymphoma were identified by search of the archival cases of the mid-Trent haematopathology
Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article Table 1 Percentage of marrow involved Percentage involvement Lymphoma type
90
18 (18.4) 29 (24.6) 33 (33.7) 17 (17.3) 1 (0.8) 10 (15.6) 10 (15.6) 14 (21.9) 18 (28.1) 12(18.8) 3 (13.6) 7 (31.8) 9 (40.9) 3 (13.6) 0 (0.0) 2 (10.5) 4 (21.1) 10 (52.6) 3 (15.8) 0 (0.0) 0 (0.0) 1 (8.3) 3 (25.0) 1 (8.3) 7 (58.3) 3 (50.0) 0 (0.0) 2 (33.3) 1 (16.7) 0 (0.0) 2 (6.9) 7 (49.1) 11 (37.9) 2 (6.9) 6 (20.7)
service for the 4 years 2004–2007. The service receives all BMTs from the mid-Trent region of England (Nottinghamshire and Lincolnshire), a population of 2.0 million. Diseases manifesting primarily as leukaemias were excluded, including T-cell large granular lymphocytic, prolymphocytic, hairy cell and acute lymphoblastic leukaemias and chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL). Cases from patients in whom a previous trephine had been obtained in the preceding 12 months were excluded, as were cases where immunophenotypic confirmation of the diagnosis (either by immunohistochemistry or flow cytometry) was not available; 511 of the initial 612 cases met the inclusion criteria, from 486 patients. The slides were reviewed by two haematopathologists, and the following undertaken: review of final clinical diagnosis and lymphoma subtype, assessment of percentage of marrow involvement, pattern of involvement, presence/absence of granulomas, stromal fibrosis and necrosis, and presence/absence of neoplastic/
Figure 1 Principal patterns of marrow involvement in commonest lymphoma subtypes. 2
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article
Figure 2
Distribution of primary patterns of marrow involvement by individual lymphoma subtype.
reactive follicles. Semiquantitive assessment of level of involvement was made using the following broad categories: 90% (table 1). Pattern of involvement was classified as diffuse, nodular, paratrabecular, interstitial or intrasinusoidal (figures 1 and 2). Diffuse was regarded as areas of effacement by a dense infiltrate of lymphocytes not forming nodules. Nodular was defined as circumscribed aggregates of lymphoid cells. Paratrabecular was defined as lymphoid aggregates with broad bases aligned alongside bony trabecula without intervening fat cells. Arber and George’s5 definition of interstitial infiltration was used: lymphoma cells infiltrating between fat cells without distorting the normal low-power architecture. Whether nodules were neoplastic or reactive was assessed on morphological grounds, and where necessary, with the aid of immunohistochemistry for CD23 and bcl-2. CD23 aids in localising the
dendritic cell network associated with lymphoid follicles. Where a germinal centre is present, bcl-2 negativity is highly specific for benign lymphoid aggregates, although up to 10–15% of follicular lymphomas may also be bcl-2 negative. Bcl-2 is positive within germinal centres in the vast majority of follicular lymphomas, but the majority of benign lymphoid follicles lacking germinal centres are also bcl-2 positive, so that this must be interpreted with caution.6 Discordance between marrow and other tissue biopsy by subtype was assessed by review of relevant other histological specimens. It was only possible to make correlation with aspirate/peripheral blood findings in 345 patients who underwent biopsy at Nottingham (167 aspirate, 201 blood). The pathology computer systems were used to access aspirate and peripheral blood reports.
Table 2 Lymphoma type discordances between BMT and lymph node biopsy Lymphoma in BMT Lymphoma in lymph node/ extranodal organ
DLBCL
Follicular
Lymphoplasmacytic
Extranodal marginal zone
Splenic marginal zone
Hodgkin
DLBCL Follicular CLL
– 8 –
13 – –
4 – –
2 – –
1 – –
1 – 1
CLL, chronic lymphocytic leukaemia; DLBCL, diffuse large B cell.
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
3
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article
Figure 3 Discordance of lymphoma type between bone marrow (left, showing classical Hodgkin lymphoma) and lymph node (right, involved by chronic lymphocytic leukaemia).
RESULTS AND DISCUSSION Percentage involvement and subtype discordance The BMT specimens were from 283 male and 228 female patients (1.24:1 M:F ratio) with age range 9–96 years (median 59 years). T-cell lymphomas were under-represented relative to their UK incidence rates at 9/511 (1.8%) of total cases as were HL cases (17.6%) so that detailed analysis of this group was not attempted. This may be because post-therapy biopsies, as opposed to those for staging purposes, are less frequently performed in these conditions. All cases of HL were of classical HL. There were 30 instances (5.9% of all cases) of subtype discordance between BMT and other tissue biopsies, summarised in table 2. In two cases, unusual combinations were present. One represented a case of HL involving the marrow with previous nodal involvement by CLL (figure 3) while the other showed classical HL in the marrow with diffuse large B cell (DLBCL) in a previous lymph node biopsy. Coexistence of classical HL and B-NHL is uncommon but well described, and the two diseases often show a common clonal origin in such cases.7 We did not attempt to assess for this owing to the difficulty in our experience of successfully amplifying DNA from decalcified
bone marrow specimens. There was, however, no evidence in the first case of Epstein-Barr virus positive large blast cells to suggest that it represented CLL with Reed-Sternberg cells, which may relapse as either disease. Hodgkinoid transformation of CLL or the entities representing distinct and unrelated diseases were, therefore, the most likely explanations. As expected, by far the predominant pattern in follicular lymphoma (FL) was paratrabecular, with 102/134 cases (76%) showing primarily this pattern. This was in keeping with the findings of a previous large series which showed 90% of follicular lymphomas to exhibit a paratrabecular pattern.5 It is believed that this tropism of follicular lymphoma cells towards the extracellular matrix protein-rich paratrabecular zones of the bone marrow stroma is due to the similarity between their matricellular proteins and those of the germinal centre microenvironment, including Jagged-1 Notch ligand, β-1 integrins, and type I collagen.8 9 CXC chemokine ligands such as CXCL13 are known to control organisation of B-cells within lymphoid follicles,10 and it has been suggested that paratrabecular homing may recapitulate the physiological chemotactic effect of CXCLs on haemopoietic stem cells, controlling their migration to endosteal and perivascular spaces.11 MZLs of all types showed a broad, relatively even, distribution between diffuse, interstitial, paratrabecular and nodular patterns (extranodal MZL:12/118 (10.2%), 32/118 (27.1%), 23/118 (19.5%), 49/118 (41.5%) respectively, splenic MZL:4/19 (21.1%), 3/19 (15.8%), 3/19 (15.8%), 7/19 (36.8%), nodalMZL:2/6 (33.3%), 2/6 (33.3%), 1/6 (16.7%), 1/6 (16.7%)). An intrasinusoidal pattern was present in 2/12 (18%) of cases of extranodal MZL and in 2/4 (50%) of cases of splenic MZL. This is a highly characteristic feature of MZLs11 first described in 1996.12 The small to medium-sized cells of MZL are often difficult to visualise in this location on H+E, but CD20 immunostaining readily reveals their presence in small clusters in venous sinuses (figure 4). Homing of MZL cells, particularly common in splenic marginal zone lymphoma (SMZL), to these sinusoidal niches may relate to chemoattractant/adhesive signals mediated by CXCL-12, hyaluronan, and ICAM-1.13 14 The provisional category of splenic diffuse red pulp small B-cell lymphoma, although not represented here, is described as having bone marrow features indistinguishable from splenic MZL.15 Lymphoplasmacytic lymphoma showed essentially identical patterns of infiltration to MZL except that intrasinusoidal infiltration was not seen. Most mantle cell lymphoma (MCL) cases demonstrated diffuse (12/22, 54.5%) or nodular (8/22, 36.4%) patterns. These bone marrow patterns did not correspond to those in the diagnostic lymph node biopsies, all of which demonstrated a diffuse growth pattern. MCL was the only small-medium cell lymphoma showing a diffuse pattern in >15% of cases. A diffuse pattern of involvement was common in DLBCL (33/ 64 cases, 51.6%) and Burkitt lymphomas (10/12 cases, 83.3%), probably reflecting their usually aggressive nature and rapid growth. The second commonest pattern in high-grade B-cell lymphomas was nodular (DLBCL:32.8%, Burkitt:8.3%). HL also showed a diffuse growth pattern in most cases (19/29 cases, 65.5%) followed by nodular (8/29, 27.6%). Two cases of HL (6.9%) were the only lymphomas other than small cell B-cell lymphomas to show a paratrabecular pattern, which has previously been described in HL.11
Secondary marrow changes
Figure 4 Intrasinusoidal infiltration in a case of SMZL (left, H+E; right, CD20). 4
The background changes of granuloma formation and stromal fibrosis were commonest in DLBCL and HL (table 3). Necrosis was, as expected, largely confined to DLBCL, Burkitt and HL. In this series, we applied the original diagnostic criteria for Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article Table 3 Secondary marrow changes Stromal changes/ follicles Lymphoma type
Total cases
Granulomas
Fibrosis
Necrosis
Neoplastic follicles
Reactive follicles
Follicular (%) Extranodal marginal zone (%) Lymphoplasmacytic (%) DLBCL (%) Mantle cell (%) Splenic marginal zone (%) Burkitt (%) Nodal marginal zone (%) Hodgkin (%) Total (%)
134 118 98 64 22 19 12 6 29 502
8 (6.0) 0 (0.0) 8 (8.2) 11 (17.2) 2 (9.1) 0 (0.0) 0 (0.0) 0 (0.0) 17 (58.6) 47 (9.4)
2 (1.5) 2 (1.7) 3 (3.1) 3 (4.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 6 (20.7) 16 (3.2)
0 (0.0) 0 (0.0) 2 (2.0) 15 (23.4) 0 (0.0) 0 (0.0) 3 (25.0) 0 (0.0) 2 (6.7) 22 (4.4)
11 (8.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0%) 0 (0.0) 11 (2.2)
0 (0.0) 7 (5.9) 3 (3.1) 2 (3.1) 2 (9.1) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 14 (2.8)
DLBCL, diffuse large B cell.
neoplastic follicles of being situated in an intertrabecular location, with less than 10% of the tumour being paratrabecular, and containing a meshwork of FDCs.16 Reactive follicles are a recognised feature of MZL, present in 7/118 (5.9%) in this series, and also occurring with a lesser frequency in several other lymphomas.
Aspirate and peripheral blood concordance Aspirates show reduced sensitivity compared to BMT in detecting lymphoma due to reticulin inhibiting successful aspiration and aspirate contamination by blood.17 Aspirate and trephine concurred in 71.3% of cases (table 4), ranging between 38.9% of DLBCL to 100% of Burkitt and MZLs. Previously, concordance between BMT and aspirates has been assessed at around 65% in all lymphomas18 and 80% of NHL.19 HL is often associated with fibrosis, causing ‘dry taps’. The low incidence of aspirate involvement here 1/14 (7.1%) parallels previous experience suggesting that only around 5% of aspirates are positive in HL.20 Peripheral blood involvement rates are known to vary considerably between lymphoma subtypes, and this was reflected in this series, with a range of 0–100% and mean of 40.4%. Blood film involvement is reported as broadly correlating with marrow involvement across small-medium cell B-cell NHLs with the exception of follicular and lymphoplasmacytic lymphomas which are less frequently detected in blood.21 Blood involvement is infrequent in DLBCL and T-cell lymphomas and exceptional in HL.22 Flow cytometry and molecular analysis are
increasingly used to detect far lower levels of blood involvement than is possible by blood film examination.23 The determinants of the presence of lymphoma cells in peripheral blood in NHL, other than primary leukaemias, have not been clearly defined. In a study of combined low and high-grade NHL, their detection was shown to be associated with extensive disease, but they were only a marginal independent risk factor for progression-free survival in multivariate analysis.24 The rare cases of DLBCL which actually present in leukaemic phase, however, are associated with chemoresistance and poor outcomes.25 Bone marrow touch imprint cytology is not used at our institution, but this technique has been shown to be especially valuable in assessing cytological detail in place of the aspirate in cases where there is a dry tap.26 When properly prepared they also offer some information on the topographical architecture of the marrow. A large series encompassing various haematological disorders showed 78% positive correlation between aspirate and biopsy, and 84.3% between imprint and biopsy.26
Comparison with previous studies The largest previous series of marrow involvement by lymphomas classified according to WHO classification was by Arber et al, already referred to (450 cases analysed).5 By comparison, our cohort included comparatively large numbers of extranodal marginal zone lymphoma (EMZL) and lymphoplasmacytic lymphomas (23.7% and 18.4% vs 3.8% and 8.4% of cases in Arber’s series, respectively) probably partly reflecting the fact
Table 4 Concordance between trephine, aspirate and peripheral blood (PB) involvement Aspirate
PB film
Type
Trephine
Assessed
Involved
Percent
Assessed
Involved
Percent
Follicular lymphoma Extranodal marginal Zone Lymphoplasmacytic DLBC Mantle cell Splenic marginal zone Burkitt Nodal marginal zone Hodgkin Total
134 118 98 64 22 19 12 6 29 502
43 37 29 18 5 8 5 1 14 160
34 34 25 7 4 8 5 1 1 118
79.1 91.9 86.2 38.9 80 100 100 100 7.1 73. 8
49 46 33 20 9 12 8 1 16 194
7 39 5 2 4 9 5 1 0 72
14 85 15 10 44.4 75 62.5 100 0 37.1
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
5
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article that sequential BMTs are more often used to monitor therapy in these lymphomas). Percentage involvements classified by subtype correlated fairly closely between the two series, as did the majority of the different patterns observed. A significant difference between our cohort and previously published studies was the relatively low incidence of intrasinusoidal infiltration, which we detected in 21% of cases of SMZL and 1.7% of EMZL, but not in other subtypes. Arber et al regarded sinusoidal involvement as a subset of interstitial disease, so did not formally distinguish this pattern in their analysis.5 Other series vary widely in their estimates of the incidence and distribution of this phenomenon, but most regard it as pathognomonic of MZL.12 27 28 Among the other lymphomas in our study it is a common occurrence only in intravascular large B cell lymphoma, and anaplastic large cell lymphoma.28 Costes detected it in 18% of 24 SMZL cases by CD20 immunostaining.28 Boveri et al examined 120 cases of MZL and detected it frequently in SMZL (74%), and nodal marginal zone lymphoma (NMZL) (57%), but less so in EMZL (8%) at presentation.27 They noted that the intrasinusoidal pattern tended to disappear posttreatment compared to other patterns ( particularly in cases of EMZL and NMZL relative to SMZL). On review, 11/19 (57.9%) of cases of SMZL, 1/6 (16.7%) of NMZL and 39/118 (33.1%) of EMZL in our series were follow-up trephines post-treatment. Disappearance of the intrasinusoidal pattern post-therapy might therefore account for its relatively lower incidence in our series.
CONCLUSION This study of BMT involvement patterns by individual lymphoma subtypes substantiates previous studies supporting BMT assessment as an essential morphological touchstone in the diagnostic workup of lymphoma, against which the results of ancillary techniques can be accurately interpreted. Different lymphomas often demonstrate reliably characteristic architectural patterns of marrow involvement which can help differentiate them, and marrow stromal and other background changes may also be useful pointers towards a particular lymphoma subtype. Although it has been suggested that advances in PET/ CT imaging may obviate the need for BMT in detecting marrow involvement by lymphoma,29 the unexpected detection of a lymphoma of a different subtype, as seen here in around 6% of cases, represents valuable and potentially treatment-altering information not otherwise obtainable.
bone marrow trephine data collection: VS, SO’C, DC. Review of bone marrow trephine samples: VS, SO’C, DMC. Collation and review of haematological data (aspirate and peripheral blood findings): CH, AKM, AH. Drafting, subsequent alterations and additions to manuscript and final review: All authors were involved. I (SO’C) have agreed to act as a guarantor responsible for the overall content. Competing interests None. Provenance and peer review Not commissioned; externally peer reviewed.
REFERENCES 1
2
3 4 5
6
7
8 9 10 11
12 13
14
15
16 17
Take-home messages 18
▸ Familiarity with patterns of marrow involvement by lymphomas enhances the accuracy of BMT evaluation, and complements cytomorphological and immunohistochemical findings in helping achieve a definitive diagnosis. ▸ Except in cases with associated stromal fibrosis (mostly HL and diffuse large B cell), aspirate involvement by lymphoma closely parallels that of trephines. Detection of lymphoma cells in peripheral blood is approximately half as frequent as in aspirates and has been correlated with extensive disease.
19
20
21
22
Contributors I certify that all the authors were involved in the conception and design of the study and participated in the generation, analysis and interpretation of the data and in the writing of the article. All have reviewed the final draft of the paper and given their approval to it. Please see below for individual roles: Initial 6
23
Campo E, Swerdlow SH, Harris NL, et al. The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 2011;117:5019–32. Lambertenghi-Deliliers G, Annaloro C, Soligo D, et al. Incidence and histological features of bone marrow involvement in malignant lymphomas. Ann Hematol 1992;65:61–5. Fend F, Kremer M. Diagnosis and classification of malignant lymphoma and related entities in the bone marrow trephine biopsy. Pathobiology 2007;74:133–43. Zhang Q-Y, Foucar K. Bone marrow involvement by hodgkin and non-hodgkin lymphomas. Hematol Oncol Clin North Am 2009;23:873–902. Arber DA, George TI. Bone marrow biopsy involvement by non-Hodgkin’s lymphoma: frequency of lymphoma types, patterns, blood involvement, and discordance with other sites in 450 specimens. Am J Surg Pathol 2005;29:1549–57. West RB, Warnke RA, Natkunam Y. The usefulness of immunohistochemistry in the diagnosis of follicular lymphoma in bone marrow biopsy specimens. Am J Clin Pathol 2002;117:636–43. Schmitz R, Renné C, Rosenquist R, et al. Insights into the multistep transformation process of lymphomas: IgH-associated translocations and tumor suppressor gene mutations in clonally related composite Hodgkin’s and non-Hodgkin’s lymphomas. Leukemia 2005;19:1452–8. Cacciatore M, Guarnotta C, Calvaruso M, et al. Microenvironment-Centred Dynamics in Aggressive B-Cell Lymphomas. Adv Hematol 2012;2012:1–12. Tripodo C, Sangaletti S, Piccaluga PP, et al. The bone marrow stroma in hematological neoplasms—a guilty bystander. Nat Rev Clin Oncol 2011;8:456–66. Ansel KM, Ngo VN, Hyman PL, et al. A chemokine-driven positive feedback loop organizes lymphoid follicles. Nature 2000;406:309–14. Sugiyama T, Kohara H, Noda M, et al. Maintenance of the Hematopoietic Stem Cell Pool by CXCL12-CXCR4 Chemokine Signaling in Bone Marrow Stromal Cell Niches. Immunity 2006;25:977–88. Franco V, Florena AM, Campesi G. Intrasinusoidal bone marrow infiltration: a possible hallmark of splenic lymphoma. Histopathology 1996;29:571–5. Florena AM, Tripodo C, Porcasi R, et al. Immunophenotypic profile and role of adhesion molecules in splenic marginal zone lymphoma with bone marrow involvement. Leuk Lymphoma 2006;47:49–57. Trentin L, Cabrelle A, Facco M, et al. Homeostatic chemokines drive migration of malignant B cells in patients with non-Hodgkin lymphomas. Blood 2004;104:502–8. Ponzoni M, Kanellis G, Pouliou E, et al. Bone marrow histopathology in the diagnostic evaluation of splenic marginal-zone and splenic diffuse red pulp small B-cell lymphoma: a reliable substitute for spleen histopathology? Am J Surg Pathol 2012;36:1609–18. Torlakovic E, Torlakovic G, Brunning RD. Follicular pattern of bone marrow involvement by follicular lymphoma. Am J Clin Pathol 2002;118:780–6. Sah SP, Matutes E, Wotherspoon AC, et al. A comparison of flow cytometry, bone marrow biopsy, and bone marrow aspirates in the detection of lymphoid infiltration in B cell disorders. J Clin Pathol 2003;56:129–32. El Bolkainy TMN, Abo Deif WS, Gouda HM, et al. Evaluation of bone marrow in 143 lymphomas: the relative frequency and pattern of involvement, secondary myelopathies, pitfalls and diagnostic validity. J Egypt Natl Canc Inst 2008;20:17–30. Musolino A, Guazzi A, Nizzoli R, et al. Accuracy and relative value of bone marrow aspiration in the detection of lymphoid infiltration in non-Hodgkin lymphoma. Tumori 2010;96:24–7. Moid F, DePalma L. Comparison of relative value of bone marrow aspirates and bone marrow trephine biopsies in the diagnosis of solid tumor metastasis and Hodgkin lymphoma: institutional experience and literature review. Arch Pathol Lab Med 2005;129:497–501. Hanson CA, Kurtin PJ, Katzmann JA, et al. Immunophenotypic analysis of peripheral blood and bone marrow in the staging of B-cell malignant lymphoma. Blood 1999;94:3889–96. Milosevic ́ R, Colovic ́ M, Cemerikic-́ Martinovic ́ V, et al. Hodgkin’s and Reed-Sternberg cells in the peripheral blood of a patient with advanced stage Hodgkin’s disease. Srp Arh Celok Lek 2003;131:400–2. Mancuso P, Calleri A, Antoniotti P, et al. If it is in the marrow, is it also in the blood? An analysis of 1,000 paired samples from patients with B-cell non-Hodgkin lymphoma. BMC Cancer 2010;10:644.
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article 24
25
26
Hiorns LR, Nicholls J, Sloane JP, et al. Peripheral blood involvement in non-Hodgkin’s lymphoma detected by clonal gene rearrangement as a biological prognostic marker. Br J Cancer 1994;69:347–51. Muringampurath-John D, Flowers CR, Jabbar AA, et al. Prognostic Significance of Circulating Lymphoma Cells in Newly Diagnosed Diffuse Large B-Cell Lymphoma (DLBCL). Poster presentation. 51st Annual meeting, American Society of Haematologists. 2009. https://ash.confex.com Chandra S, Chandra . Comparison of bone marrow aspirate cytology, touch imprint cytology and trephine biopsy for bone marrow evaluation. Hematol Rep 2011;3:e22.
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
27
28
29
Boveri E, Arcaini L, Merli M, et al. Bone marrow histology in marginal zone B-cell lymphomas: correlation with clinical parameters and flow cytometry in 120 patients. Ann Oncol 2009;20:129–36. Costes V, Duchayne E, Taib J, et al. Intrasinusoidal bone marrow infiltration: a common growth pattern for different lymphoma subtypes. Br J Haematol 2002;119:916–22. Khan AB, Barrington SF, Mikhaeel NG, et al. PET-CT staging of DLBCL accurately identifies and provides new insight into the clinical significance of bone marrow involvement. Blood 2013;122:61–7.
7
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Bone marrow trephine biopsy involvement by lymphoma: review of histopathological features in 511 specimens and correlation with diagnostic biopsy, aspirate and peripheral blood findings Vishakha Sovani, Caroline Harvey, Andrew P Haynes, et al. J Clin Pathol published online December 10, 2013
doi: 10.1136/jclinpath-2013-201520
Updated information and services can be found at: http://jcp.bmj.com/content/early/2013/12/10/jclinpath-2013-201520.full.html
These include:
References
This article cites 28 articles, 8 of which can be accessed free at: http://jcp.bmj.com/content/early/2013/12/10/jclinpath-2013-201520.full.html#ref-list-1
P
JCP Online First, published on December 10, 2013 as 10.1136/jclinpath-2013-201520 Original article
Bone marrow trephine biopsy involvement by lymphoma: review of histopathological features in 511 specimens and correlation with diagnostic biopsy, aspirate and peripheral blood findings Vishakha Sovani,1 Caroline Harvey,2 Andrew P Haynes,2 Andrew K McMillan,2 David M Clark,3 Simon R O’Connor1 1
Department of Histopathology, Nottingham University Hospital, Nottingham, UK 2 Department of Haematology, Nottingham University Hospital, Nottingham, UK 3 Department of Cellular Pathology, Lincoln County Hospital, Lincoln, UK Correspondence to Dr Simon O’Connor, Department of Histopathology, Nottingham University Hospital, City Campus, Hucknall Road, Nottingham, NG5 1PB, UK; [email protected] Received 1 February 2013 Revised 5 September 2013 Accepted 20 October 2013
ABSTRACT Aims This study aimed to evaluate the key features of bone marrow trephine (BMT) biopsy involvement by lymphoma. Methods 511 cases were assessed for percentage of marrow involvement, pattern of involvement (diffuse, nodular, paratrabecular, interstitial or intrasinusoidal), presence/absence of granulomas, stromal fibrosis and necrosis, presence/absence of neoplastic/reactive follicles and discordance with other biopsy sites. Correlation with aspirate and peripheral blood findings was made in a subset of 345 patients (167 aspirates, 178 blood). Results The most frequent subtype was follicular lymphoma (26.2%) followed by extranodal marginal zone (23.1%), lymphoplasmacytic (19.2%), diffuse large B cell (DLBCL) (12.5%), Hodgkin (HL) (5.7%) and mantle cell lymphomas (4.3%). The predominant pattern in follicular lymphoma was paratrabecular. Marginal zone lymphomas of all types and lymphoplasmacytic lymphoma showed a relatively even distribution between diffuse, interstitial, paratrabecular and nodular patterns. The majority of mantle cell lymphoma cases showed either diffuse or nodular patterns. A diffuse pattern was common in DLBCL and Burkitt lymphomas. An intrasinusoidal pattern was seen only in extranodal and splenic marginal zone lymphomas. Granulomas and fibrosis were uncommon in small cell B cell lymphomas but frequent in DLBCL and HL. Aspirate and trephine results concurred in 73.8% of cases overall, but this varied widely between subtypes. Peripheral blood involvement rates by lymphoma also varied, with a mean of 37.1%. Conclusions Different lymphomas often demonstrate reliably characteristic architectural patterns of marrow involvement which can help differentiate them even when cytological features do not permit this, and marrow stromal and other background changes may also be useful pointers towards a particular lymphoma subtype.
PARAGRAPH ON WHAT THE PAPER ADDS
To cite: Sovani V, Harvey C, Haynes AP, et al. J Clin Pathol Published Online First: [ please include Day Month Year] doi:10.1136/ jclinpath-2013-201520
We believe this paper adds to the currently available data on bone marrow trephine (BMT) biopsy involvement by lymphoma for the following reasons. It represents by a significant margin the largest series of trephine biopsies examined in this way to date and is also of interest as representing consecutive cases within a defined geographical region and population group in the UK ( previous series of BMT cores have mainly been from tertiary centres in the USA, where cases may often be received from outside the region and this can
Sovani V, et al. J Clin Pathol 2013;0:1–7. Copyright Article author doi:10.1136/jclinpath-2013-201520 (or their employer) 2013.
sometimes mean that population-based incidence features are unreliable). It also makes comparison of BMT, aspirate and peripheral blood findings in lymphoma over a far larger number of cases than has previously been analysed, and examines BMT findings in a significant number of Hodgkin’s lymphomas which have been neglected in most previous studies of lymphoma involving bone marrow. Although immunohistochemical, flow cytometric and molecular findings are essential in diagnostic haematopathology, BMT morphology remains an important point of reference for their interpretation, and we believe the findings of this paper contribute to the body of knowledge available on this subject.
MANUSCRIPT WHO lymphoma classification employs combined morphological, immunophenotypic and molecular genetic information to accurately define individual lymphoma subtypes as distinct clinicopathologcal entities.1 Bone marrow involvement occurs in around 40% of all cases of lymphoma,2 and BMT biopsy staging has significant implications for prognostication and therapy as well as often playing an important role in diagnosis.3 Despite the importance of immunohistochemical, flow cytometric and molecular data, BMT morphological evaluation remains for lymphoma the gold standard against which other investigations are interpreted.4 Pattern analysis forms an important part of this, complementing cytomorphological information5 and is especially useful in small to medium-sized cell lymphomas where cytological features may be similar between subtypes. This study was designed to assess patterns of BMT involvement by lymphoma in well-characterised cases classified by the 2008 WHO classification of lymphoid neoplasms, and correlate these with, where available, aspirate and peripheral blood findings. Frequency and nature of discordance between the findings in the bone marrow and other specimens was also analysed, as was concordance with aspirate and peripheral blood findings. This represents the largest series of trephines examined from this perspective to date.
MATERIALS AND METHODS Totally, 612 sequential BMT cores showing involvement by lymphoma were identified by search of the archival cases of the mid-Trent haematopathology
Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article Table 1 Percentage of marrow involved Percentage involvement Lymphoma type
90
18 (18.4) 29 (24.6) 33 (33.7) 17 (17.3) 1 (0.8) 10 (15.6) 10 (15.6) 14 (21.9) 18 (28.1) 12(18.8) 3 (13.6) 7 (31.8) 9 (40.9) 3 (13.6) 0 (0.0) 2 (10.5) 4 (21.1) 10 (52.6) 3 (15.8) 0 (0.0) 0 (0.0) 1 (8.3) 3 (25.0) 1 (8.3) 7 (58.3) 3 (50.0) 0 (0.0) 2 (33.3) 1 (16.7) 0 (0.0) 2 (6.9) 7 (49.1) 11 (37.9) 2 (6.9) 6 (20.7)
service for the 4 years 2004–2007. The service receives all BMTs from the mid-Trent region of England (Nottinghamshire and Lincolnshire), a population of 2.0 million. Diseases manifesting primarily as leukaemias were excluded, including T-cell large granular lymphocytic, prolymphocytic, hairy cell and acute lymphoblastic leukaemias and chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL). Cases from patients in whom a previous trephine had been obtained in the preceding 12 months were excluded, as were cases where immunophenotypic confirmation of the diagnosis (either by immunohistochemistry or flow cytometry) was not available; 511 of the initial 612 cases met the inclusion criteria, from 486 patients. The slides were reviewed by two haematopathologists, and the following undertaken: review of final clinical diagnosis and lymphoma subtype, assessment of percentage of marrow involvement, pattern of involvement, presence/absence of granulomas, stromal fibrosis and necrosis, and presence/absence of neoplastic/
Figure 1 Principal patterns of marrow involvement in commonest lymphoma subtypes. 2
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article
Figure 2
Distribution of primary patterns of marrow involvement by individual lymphoma subtype.
reactive follicles. Semiquantitive assessment of level of involvement was made using the following broad categories: 90% (table 1). Pattern of involvement was classified as diffuse, nodular, paratrabecular, interstitial or intrasinusoidal (figures 1 and 2). Diffuse was regarded as areas of effacement by a dense infiltrate of lymphocytes not forming nodules. Nodular was defined as circumscribed aggregates of lymphoid cells. Paratrabecular was defined as lymphoid aggregates with broad bases aligned alongside bony trabecula without intervening fat cells. Arber and George’s5 definition of interstitial infiltration was used: lymphoma cells infiltrating between fat cells without distorting the normal low-power architecture. Whether nodules were neoplastic or reactive was assessed on morphological grounds, and where necessary, with the aid of immunohistochemistry for CD23 and bcl-2. CD23 aids in localising the
dendritic cell network associated with lymphoid follicles. Where a germinal centre is present, bcl-2 negativity is highly specific for benign lymphoid aggregates, although up to 10–15% of follicular lymphomas may also be bcl-2 negative. Bcl-2 is positive within germinal centres in the vast majority of follicular lymphomas, but the majority of benign lymphoid follicles lacking germinal centres are also bcl-2 positive, so that this must be interpreted with caution.6 Discordance between marrow and other tissue biopsy by subtype was assessed by review of relevant other histological specimens. It was only possible to make correlation with aspirate/peripheral blood findings in 345 patients who underwent biopsy at Nottingham (167 aspirate, 201 blood). The pathology computer systems were used to access aspirate and peripheral blood reports.
Table 2 Lymphoma type discordances between BMT and lymph node biopsy Lymphoma in BMT Lymphoma in lymph node/ extranodal organ
DLBCL
Follicular
Lymphoplasmacytic
Extranodal marginal zone
Splenic marginal zone
Hodgkin
DLBCL Follicular CLL
– 8 –
13 – –
4 – –
2 – –
1 – –
1 – 1
CLL, chronic lymphocytic leukaemia; DLBCL, diffuse large B cell.
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
3
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article
Figure 3 Discordance of lymphoma type between bone marrow (left, showing classical Hodgkin lymphoma) and lymph node (right, involved by chronic lymphocytic leukaemia).
RESULTS AND DISCUSSION Percentage involvement and subtype discordance The BMT specimens were from 283 male and 228 female patients (1.24:1 M:F ratio) with age range 9–96 years (median 59 years). T-cell lymphomas were under-represented relative to their UK incidence rates at 9/511 (1.8%) of total cases as were HL cases (17.6%) so that detailed analysis of this group was not attempted. This may be because post-therapy biopsies, as opposed to those for staging purposes, are less frequently performed in these conditions. All cases of HL were of classical HL. There were 30 instances (5.9% of all cases) of subtype discordance between BMT and other tissue biopsies, summarised in table 2. In two cases, unusual combinations were present. One represented a case of HL involving the marrow with previous nodal involvement by CLL (figure 3) while the other showed classical HL in the marrow with diffuse large B cell (DLBCL) in a previous lymph node biopsy. Coexistence of classical HL and B-NHL is uncommon but well described, and the two diseases often show a common clonal origin in such cases.7 We did not attempt to assess for this owing to the difficulty in our experience of successfully amplifying DNA from decalcified
bone marrow specimens. There was, however, no evidence in the first case of Epstein-Barr virus positive large blast cells to suggest that it represented CLL with Reed-Sternberg cells, which may relapse as either disease. Hodgkinoid transformation of CLL or the entities representing distinct and unrelated diseases were, therefore, the most likely explanations. As expected, by far the predominant pattern in follicular lymphoma (FL) was paratrabecular, with 102/134 cases (76%) showing primarily this pattern. This was in keeping with the findings of a previous large series which showed 90% of follicular lymphomas to exhibit a paratrabecular pattern.5 It is believed that this tropism of follicular lymphoma cells towards the extracellular matrix protein-rich paratrabecular zones of the bone marrow stroma is due to the similarity between their matricellular proteins and those of the germinal centre microenvironment, including Jagged-1 Notch ligand, β-1 integrins, and type I collagen.8 9 CXC chemokine ligands such as CXCL13 are known to control organisation of B-cells within lymphoid follicles,10 and it has been suggested that paratrabecular homing may recapitulate the physiological chemotactic effect of CXCLs on haemopoietic stem cells, controlling their migration to endosteal and perivascular spaces.11 MZLs of all types showed a broad, relatively even, distribution between diffuse, interstitial, paratrabecular and nodular patterns (extranodal MZL:12/118 (10.2%), 32/118 (27.1%), 23/118 (19.5%), 49/118 (41.5%) respectively, splenic MZL:4/19 (21.1%), 3/19 (15.8%), 3/19 (15.8%), 7/19 (36.8%), nodalMZL:2/6 (33.3%), 2/6 (33.3%), 1/6 (16.7%), 1/6 (16.7%)). An intrasinusoidal pattern was present in 2/12 (18%) of cases of extranodal MZL and in 2/4 (50%) of cases of splenic MZL. This is a highly characteristic feature of MZLs11 first described in 1996.12 The small to medium-sized cells of MZL are often difficult to visualise in this location on H+E, but CD20 immunostaining readily reveals their presence in small clusters in venous sinuses (figure 4). Homing of MZL cells, particularly common in splenic marginal zone lymphoma (SMZL), to these sinusoidal niches may relate to chemoattractant/adhesive signals mediated by CXCL-12, hyaluronan, and ICAM-1.13 14 The provisional category of splenic diffuse red pulp small B-cell lymphoma, although not represented here, is described as having bone marrow features indistinguishable from splenic MZL.15 Lymphoplasmacytic lymphoma showed essentially identical patterns of infiltration to MZL except that intrasinusoidal infiltration was not seen. Most mantle cell lymphoma (MCL) cases demonstrated diffuse (12/22, 54.5%) or nodular (8/22, 36.4%) patterns. These bone marrow patterns did not correspond to those in the diagnostic lymph node biopsies, all of which demonstrated a diffuse growth pattern. MCL was the only small-medium cell lymphoma showing a diffuse pattern in >15% of cases. A diffuse pattern of involvement was common in DLBCL (33/ 64 cases, 51.6%) and Burkitt lymphomas (10/12 cases, 83.3%), probably reflecting their usually aggressive nature and rapid growth. The second commonest pattern in high-grade B-cell lymphomas was nodular (DLBCL:32.8%, Burkitt:8.3%). HL also showed a diffuse growth pattern in most cases (19/29 cases, 65.5%) followed by nodular (8/29, 27.6%). Two cases of HL (6.9%) were the only lymphomas other than small cell B-cell lymphomas to show a paratrabecular pattern, which has previously been described in HL.11
Secondary marrow changes
Figure 4 Intrasinusoidal infiltration in a case of SMZL (left, H+E; right, CD20). 4
The background changes of granuloma formation and stromal fibrosis were commonest in DLBCL and HL (table 3). Necrosis was, as expected, largely confined to DLBCL, Burkitt and HL. In this series, we applied the original diagnostic criteria for Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article Table 3 Secondary marrow changes Stromal changes/ follicles Lymphoma type
Total cases
Granulomas
Fibrosis
Necrosis
Neoplastic follicles
Reactive follicles
Follicular (%) Extranodal marginal zone (%) Lymphoplasmacytic (%) DLBCL (%) Mantle cell (%) Splenic marginal zone (%) Burkitt (%) Nodal marginal zone (%) Hodgkin (%) Total (%)
134 118 98 64 22 19 12 6 29 502
8 (6.0) 0 (0.0) 8 (8.2) 11 (17.2) 2 (9.1) 0 (0.0) 0 (0.0) 0 (0.0) 17 (58.6) 47 (9.4)
2 (1.5) 2 (1.7) 3 (3.1) 3 (4.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 6 (20.7) 16 (3.2)
0 (0.0) 0 (0.0) 2 (2.0) 15 (23.4) 0 (0.0) 0 (0.0) 3 (25.0) 0 (0.0) 2 (6.7) 22 (4.4)
11 (8.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0%) 0 (0.0) 11 (2.2)
0 (0.0) 7 (5.9) 3 (3.1) 2 (3.1) 2 (9.1) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 14 (2.8)
DLBCL, diffuse large B cell.
neoplastic follicles of being situated in an intertrabecular location, with less than 10% of the tumour being paratrabecular, and containing a meshwork of FDCs.16 Reactive follicles are a recognised feature of MZL, present in 7/118 (5.9%) in this series, and also occurring with a lesser frequency in several other lymphomas.
Aspirate and peripheral blood concordance Aspirates show reduced sensitivity compared to BMT in detecting lymphoma due to reticulin inhibiting successful aspiration and aspirate contamination by blood.17 Aspirate and trephine concurred in 71.3% of cases (table 4), ranging between 38.9% of DLBCL to 100% of Burkitt and MZLs. Previously, concordance between BMT and aspirates has been assessed at around 65% in all lymphomas18 and 80% of NHL.19 HL is often associated with fibrosis, causing ‘dry taps’. The low incidence of aspirate involvement here 1/14 (7.1%) parallels previous experience suggesting that only around 5% of aspirates are positive in HL.20 Peripheral blood involvement rates are known to vary considerably between lymphoma subtypes, and this was reflected in this series, with a range of 0–100% and mean of 40.4%. Blood film involvement is reported as broadly correlating with marrow involvement across small-medium cell B-cell NHLs with the exception of follicular and lymphoplasmacytic lymphomas which are less frequently detected in blood.21 Blood involvement is infrequent in DLBCL and T-cell lymphomas and exceptional in HL.22 Flow cytometry and molecular analysis are
increasingly used to detect far lower levels of blood involvement than is possible by blood film examination.23 The determinants of the presence of lymphoma cells in peripheral blood in NHL, other than primary leukaemias, have not been clearly defined. In a study of combined low and high-grade NHL, their detection was shown to be associated with extensive disease, but they were only a marginal independent risk factor for progression-free survival in multivariate analysis.24 The rare cases of DLBCL which actually present in leukaemic phase, however, are associated with chemoresistance and poor outcomes.25 Bone marrow touch imprint cytology is not used at our institution, but this technique has been shown to be especially valuable in assessing cytological detail in place of the aspirate in cases where there is a dry tap.26 When properly prepared they also offer some information on the topographical architecture of the marrow. A large series encompassing various haematological disorders showed 78% positive correlation between aspirate and biopsy, and 84.3% between imprint and biopsy.26
Comparison with previous studies The largest previous series of marrow involvement by lymphomas classified according to WHO classification was by Arber et al, already referred to (450 cases analysed).5 By comparison, our cohort included comparatively large numbers of extranodal marginal zone lymphoma (EMZL) and lymphoplasmacytic lymphomas (23.7% and 18.4% vs 3.8% and 8.4% of cases in Arber’s series, respectively) probably partly reflecting the fact
Table 4 Concordance between trephine, aspirate and peripheral blood (PB) involvement Aspirate
PB film
Type
Trephine
Assessed
Involved
Percent
Assessed
Involved
Percent
Follicular lymphoma Extranodal marginal Zone Lymphoplasmacytic DLBC Mantle cell Splenic marginal zone Burkitt Nodal marginal zone Hodgkin Total
134 118 98 64 22 19 12 6 29 502
43 37 29 18 5 8 5 1 14 160
34 34 25 7 4 8 5 1 1 118
79.1 91.9 86.2 38.9 80 100 100 100 7.1 73. 8
49 46 33 20 9 12 8 1 16 194
7 39 5 2 4 9 5 1 0 72
14 85 15 10 44.4 75 62.5 100 0 37.1
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
5
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article that sequential BMTs are more often used to monitor therapy in these lymphomas). Percentage involvements classified by subtype correlated fairly closely between the two series, as did the majority of the different patterns observed. A significant difference between our cohort and previously published studies was the relatively low incidence of intrasinusoidal infiltration, which we detected in 21% of cases of SMZL and 1.7% of EMZL, but not in other subtypes. Arber et al regarded sinusoidal involvement as a subset of interstitial disease, so did not formally distinguish this pattern in their analysis.5 Other series vary widely in their estimates of the incidence and distribution of this phenomenon, but most regard it as pathognomonic of MZL.12 27 28 Among the other lymphomas in our study it is a common occurrence only in intravascular large B cell lymphoma, and anaplastic large cell lymphoma.28 Costes detected it in 18% of 24 SMZL cases by CD20 immunostaining.28 Boveri et al examined 120 cases of MZL and detected it frequently in SMZL (74%), and nodal marginal zone lymphoma (NMZL) (57%), but less so in EMZL (8%) at presentation.27 They noted that the intrasinusoidal pattern tended to disappear posttreatment compared to other patterns ( particularly in cases of EMZL and NMZL relative to SMZL). On review, 11/19 (57.9%) of cases of SMZL, 1/6 (16.7%) of NMZL and 39/118 (33.1%) of EMZL in our series were follow-up trephines post-treatment. Disappearance of the intrasinusoidal pattern post-therapy might therefore account for its relatively lower incidence in our series.
CONCLUSION This study of BMT involvement patterns by individual lymphoma subtypes substantiates previous studies supporting BMT assessment as an essential morphological touchstone in the diagnostic workup of lymphoma, against which the results of ancillary techniques can be accurately interpreted. Different lymphomas often demonstrate reliably characteristic architectural patterns of marrow involvement which can help differentiate them, and marrow stromal and other background changes may also be useful pointers towards a particular lymphoma subtype. Although it has been suggested that advances in PET/ CT imaging may obviate the need for BMT in detecting marrow involvement by lymphoma,29 the unexpected detection of a lymphoma of a different subtype, as seen here in around 6% of cases, represents valuable and potentially treatment-altering information not otherwise obtainable.
bone marrow trephine data collection: VS, SO’C, DC. Review of bone marrow trephine samples: VS, SO’C, DMC. Collation and review of haematological data (aspirate and peripheral blood findings): CH, AKM, AH. Drafting, subsequent alterations and additions to manuscript and final review: All authors were involved. I (SO’C) have agreed to act as a guarantor responsible for the overall content. Competing interests None. Provenance and peer review Not commissioned; externally peer reviewed.
REFERENCES 1
2
3 4 5
6
7
8 9 10 11
12 13
14
15
16 17
Take-home messages 18
▸ Familiarity with patterns of marrow involvement by lymphomas enhances the accuracy of BMT evaluation, and complements cytomorphological and immunohistochemical findings in helping achieve a definitive diagnosis. ▸ Except in cases with associated stromal fibrosis (mostly HL and diffuse large B cell), aspirate involvement by lymphoma closely parallels that of trephines. Detection of lymphoma cells in peripheral blood is approximately half as frequent as in aspirates and has been correlated with extensive disease.
19
20
21
22
Contributors I certify that all the authors were involved in the conception and design of the study and participated in the generation, analysis and interpretation of the data and in the writing of the article. All have reviewed the final draft of the paper and given their approval to it. Please see below for individual roles: Initial 6
23
Campo E, Swerdlow SH, Harris NL, et al. The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 2011;117:5019–32. Lambertenghi-Deliliers G, Annaloro C, Soligo D, et al. Incidence and histological features of bone marrow involvement in malignant lymphomas. Ann Hematol 1992;65:61–5. Fend F, Kremer M. Diagnosis and classification of malignant lymphoma and related entities in the bone marrow trephine biopsy. Pathobiology 2007;74:133–43. Zhang Q-Y, Foucar K. Bone marrow involvement by hodgkin and non-hodgkin lymphomas. Hematol Oncol Clin North Am 2009;23:873–902. Arber DA, George TI. Bone marrow biopsy involvement by non-Hodgkin’s lymphoma: frequency of lymphoma types, patterns, blood involvement, and discordance with other sites in 450 specimens. Am J Surg Pathol 2005;29:1549–57. West RB, Warnke RA, Natkunam Y. The usefulness of immunohistochemistry in the diagnosis of follicular lymphoma in bone marrow biopsy specimens. Am J Clin Pathol 2002;117:636–43. Schmitz R, Renné C, Rosenquist R, et al. Insights into the multistep transformation process of lymphomas: IgH-associated translocations and tumor suppressor gene mutations in clonally related composite Hodgkin’s and non-Hodgkin’s lymphomas. Leukemia 2005;19:1452–8. Cacciatore M, Guarnotta C, Calvaruso M, et al. Microenvironment-Centred Dynamics in Aggressive B-Cell Lymphomas. Adv Hematol 2012;2012:1–12. Tripodo C, Sangaletti S, Piccaluga PP, et al. The bone marrow stroma in hematological neoplasms—a guilty bystander. Nat Rev Clin Oncol 2011;8:456–66. Ansel KM, Ngo VN, Hyman PL, et al. A chemokine-driven positive feedback loop organizes lymphoid follicles. Nature 2000;406:309–14. Sugiyama T, Kohara H, Noda M, et al. Maintenance of the Hematopoietic Stem Cell Pool by CXCL12-CXCR4 Chemokine Signaling in Bone Marrow Stromal Cell Niches. Immunity 2006;25:977–88. Franco V, Florena AM, Campesi G. Intrasinusoidal bone marrow infiltration: a possible hallmark of splenic lymphoma. Histopathology 1996;29:571–5. Florena AM, Tripodo C, Porcasi R, et al. Immunophenotypic profile and role of adhesion molecules in splenic marginal zone lymphoma with bone marrow involvement. Leuk Lymphoma 2006;47:49–57. Trentin L, Cabrelle A, Facco M, et al. Homeostatic chemokines drive migration of malignant B cells in patients with non-Hodgkin lymphomas. Blood 2004;104:502–8. Ponzoni M, Kanellis G, Pouliou E, et al. Bone marrow histopathology in the diagnostic evaluation of splenic marginal-zone and splenic diffuse red pulp small B-cell lymphoma: a reliable substitute for spleen histopathology? Am J Surg Pathol 2012;36:1609–18. Torlakovic E, Torlakovic G, Brunning RD. Follicular pattern of bone marrow involvement by follicular lymphoma. Am J Clin Pathol 2002;118:780–6. Sah SP, Matutes E, Wotherspoon AC, et al. A comparison of flow cytometry, bone marrow biopsy, and bone marrow aspirates in the detection of lymphoid infiltration in B cell disorders. J Clin Pathol 2003;56:129–32. El Bolkainy TMN, Abo Deif WS, Gouda HM, et al. Evaluation of bone marrow in 143 lymphomas: the relative frequency and pattern of involvement, secondary myelopathies, pitfalls and diagnostic validity. J Egypt Natl Canc Inst 2008;20:17–30. Musolino A, Guazzi A, Nizzoli R, et al. Accuracy and relative value of bone marrow aspiration in the detection of lymphoid infiltration in non-Hodgkin lymphoma. Tumori 2010;96:24–7. Moid F, DePalma L. Comparison of relative value of bone marrow aspirates and bone marrow trephine biopsies in the diagnosis of solid tumor metastasis and Hodgkin lymphoma: institutional experience and literature review. Arch Pathol Lab Med 2005;129:497–501. Hanson CA, Kurtin PJ, Katzmann JA, et al. Immunophenotypic analysis of peripheral blood and bone marrow in the staging of B-cell malignant lymphoma. Blood 1999;94:3889–96. Milosevic ́ R, Colovic ́ M, Cemerikic-́ Martinovic ́ V, et al. Hodgkin’s and Reed-Sternberg cells in the peripheral blood of a patient with advanced stage Hodgkin’s disease. Srp Arh Celok Lek 2003;131:400–2. Mancuso P, Calleri A, Antoniotti P, et al. If it is in the marrow, is it also in the blood? An analysis of 1,000 paired samples from patients with B-cell non-Hodgkin lymphoma. BMC Cancer 2010;10:644.
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Original article 24
25
26
Hiorns LR, Nicholls J, Sloane JP, et al. Peripheral blood involvement in non-Hodgkin’s lymphoma detected by clonal gene rearrangement as a biological prognostic marker. Br J Cancer 1994;69:347–51. Muringampurath-John D, Flowers CR, Jabbar AA, et al. Prognostic Significance of Circulating Lymphoma Cells in Newly Diagnosed Diffuse Large B-Cell Lymphoma (DLBCL). Poster presentation. 51st Annual meeting, American Society of Haematologists. 2009. https://ash.confex.com Chandra S, Chandra . Comparison of bone marrow aspirate cytology, touch imprint cytology and trephine biopsy for bone marrow evaluation. Hematol Rep 2011;3:e22.
Sovani V, et al. J Clin Pathol 2013;0:1–7. doi:10.1136/jclinpath-2013-201520
27
28
29
Boveri E, Arcaini L, Merli M, et al. Bone marrow histology in marginal zone B-cell lymphomas: correlation with clinical parameters and flow cytometry in 120 patients. Ann Oncol 2009;20:129–36. Costes V, Duchayne E, Taib J, et al. Intrasinusoidal bone marrow infiltration: a common growth pattern for different lymphoma subtypes. Br J Haematol 2002;119:916–22. Khan AB, Barrington SF, Mikhaeel NG, et al. PET-CT staging of DLBCL accurately identifies and provides new insight into the clinical significance of bone marrow involvement. Blood 2013;122:61–7.
7
Downloaded from jcp.bmj.com on April 10, 2014 - Published by group.bmj.com
Bone marrow trephine biopsy involvement by lymphoma: review of histopathological features in 511 specimens and correlation with diagnostic biopsy, aspirate and peripheral blood findings Vishakha Sovani, Caroline Harvey, Andrew P Haynes, et al. J Clin Pathol published online December 10, 2013
doi: 10.1136/jclinpath-2013-201520
Updated information and services can be found at: http://jcp.bmj.com/content/early/2013/12/10/jclinpath-2013-201520.full.html
These include:
References
This article cites 28 articles, 8 of which can be accessed free at: http://jcp.bmj.com/content/early/2013/12/10/jclinpath-2013-201520.full.html#ref-list-1
P
