| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Hematopathology Section, Laboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Among the biphenotypic neoplasms, composite non-Hodgkin's lymphomas, ie, tumors with two morphologically and/or phenotypically different components in the same anatomical site, are rare, and some earlier reports lack molecular studies.13,15,17-19,24,25
We present three cases of low grade B-NHL with two morphologically and immunophenotypically distinct tumor components occupying different but intimately interwoven microenvironments in the involved tissue. Molecular analysis of the two tumor components obtained by laser capture microdissection (LCM) revealed two unrelated clonal populations in all three cases despite their synchronous anatomical presentation.
| Patients |
|---|
|
|
|---|
A 58-year-old female presented with small bowel obstruction, leading to resection of a stenosed segment of the small intestine and mesenteric lymph nodes. A diagnosis of malignant non-Hodgkin's lymphoma was rendered. Clinical staging revealed no further manifestations of lymphoma and PB counts were in the normal range. Two bone marrow biopsies performed at 12 and 18 months were reported to show evidence of minimal, focal involvement by lymphoma. The patient received 22 cycles of polychemotherapy over a period of 2 years and remains in continuous complete remission 8 years after the primary manifestation.
Case 2
A 77-year-old male with a 1-year history of marked splenomegaly
developed inguinal lymphadenopathy. A lymph node biopsy was performed
and a diagnosis of lymphoma made. Flow cytometric immunophenotyping of
a bone marrow aspirate showed a population of B cells with
light
chain restriction and coexpression of CD5 and partly CD23 and FMC7.
Cytologic examination of the peripheral blood showed no involvement by
lymphoma.
Case 3
A 69-year-old female underwent laryngectomy and bilateral neck
dissection for a T3 squamous cell carcinoma of the vocal cords. The
grossly enlarged lymph nodes showed no metastases of the carcinoma, but
involvement by a malignant lymphoma. The peripheral blood showed 31,000
leukocytes/µl with 82.3% lymphocytes. Flow cytometry of the
peripheral blood revealed a large B cell population coexpressing CD5
and CD23 and showing dim
light chain expression. In addition, a
small CD5+ B cell population with
light chain restriction was
found.
| Materials and Methods |
|---|
|
|
|---|
Only paraffin-embedded tissue was available from the diagnostic
specimens of all patients. Immunophenotyping was performed with the
antibodies listed in Table 1
using an
automated immunostainer (Ventana Medical Systems, Inc., Tucson, AZ)
according to the company's protocols, with minor modifications.
Heat-induced antigen retrieval was performed with a microwave pressure
cooker as previously described.26
Incubation was performed
overnight for cyclin D1, p27, CD5, and CD10; the remaining primary
antibodies were incubated for 32 minutes. The rest of the staining
procedure was performed on the Ventana immunostainer.
|
LCM was performed on routinely immunostained slides as described previously27-29 using a PixCell laser capture microscope (Arcturus Engineering, Santa Clara, CA). In brief, the stained, dehydrated tissue section is overlaid with a thermoplastic film mounted on an optically transparent cap. The visually selected areas are bound to the membrane by short, low-energy laser pulses leading to focal melting of the polymer. Immunostains showing a positive reaction in only one of the two different tumor components were selected for LCM. Each component was microdissected separately at least three times. The cells were immersed in 50100 µl Tris buffer, pH 8.0, containing 0.5 mmol/L EDTA and 400 µg/ml proteinase K and digested for 3 hours at 55°C. After digestion, the enzyme was heat-inactivated and the extract was used directly for PCR.
Polymerase Chain Reaction (PCR), Cloning, and Sequence Analysis
The amplification of rearranged immunoglobulin genes was achieved
using previously published primers directed at common sequences of all
JH region genes (JH
,
LJH, and VLJH) and
against homologous sequences of either the framework 3
(VH) or the framework 2 (FR2A) regions of the
immunoglobulin heavy chain (IgH) genes.30,31
All PCRs were
performed with a hot start technique using the TaqStart antibody
(Clontech, Palo Alto, CA), which reversibly inactivates polymerase
activity below 70°C. Five to 10 µl of the crude extract from
microdissected tissue or 1 µg of DNA obtained from whole sections by
standard phenol/chloroform extraction32
were added as
template. For the complementarity-determining-region III (CDRIII) PCR
with primers VH and JH
,
35 cycles of amplification consisting of denaturation at 94°C for 45
seconds, annealing at 56°C for 1 minute, and elongation at 74°C for
1 minute were performed with DNA from whole tissue sections, 40 cycles
with microdissected material. Amplification with the primer FR2A was
performed as a seminested procedure in conjunction with primer LJH for
the first round of 35 cycles and primer VLJH for the 25 cycles of the
second round. The CDRIII and FR2 PCR products were analyzed on 16%
polyacrylamide or 3% Metaphor agarose (FMC Bioproducts, Rockland, ME)
gels, respectively. Primers bcl-2 and JH
were
used for the detection of t(14;18) translocations of follicular
lymphoma involving the major breakpoint region.33
For
amplification of the mantle cell lymphoma-specific t(11;14) involving
the major translocation cluster, primers bcl-1.1 and
JH
were used.34
The specificity
of the bcl-2 amplification product was confirmed by Southern blotting
and hybridization with a digoxigenin-labeled internal control
oligonucleotide. Bound probe was detected with an alkaline
phosphatase-labeled anti-DIG Fab fragment and CSPD as
chemiluminescence substrate recommended by the manufacturer (Boehringer
Mannheim, Indianapolis, IN). Results obtained from microdissected
tissue were confirmed by repeat PCR at least once. Microdissected
reactive follicles were used as polyclonal control. Negative controls
included water, DNA extraction buffer, and unused LCM caps.
Clonal amplification products were extracted from polyacrylamide gel
slices using the crush and soak technique.32
The purified
PCR products were reamplified with the same primers, ligated into the
PCR 2.1 vector (TA Cloning Kit, Invitrogen, Carlsbad, CA), and cloned
into INV
F' bacteria according to the manufacturer's
instructions. Automated fluorescent sequencing of plasmids containing
inserts was performed with the BigDye Terminator Cycle Sequencing Kit
(PE Applied Biosystems, Foster City, CA). A minimum of 4 clones was
sequenced for each amplified rearrangement. Comparison of sequences and
identification of IG heavy chain gene usage was performed with the
MacVector 5.0 software (Eastman Kodak, New Haven, CT).
| Results |
|---|
|
|
|---|
The immunohistochemical findings are summarized in Table 2
.
|
The tumor diffusely involved the mucosa and submucosa of the small
intestine with occasional flat ulcerations and spread into the muscle
and focally into the subserosal fat. On routinely stained sections, two
tumor components of different morphology could be appreciated. The
diffuse component consisted of a monotonous proliferation of small
lymphoid cells with slightly irregular nuclei and inconspicuous
nucleoli. Blasts were not identified. The second component was
surrounded by the diffuse tumor component and consisted of moderately
well defined, less densely packed follicular structures composed of
centrocytes with irregular, cleaved nuclei and occasional centroblasts
(Figure 1, A and B)
. Lymphoepithelial
lesions were absent. Regional lymph nodes showed involvement only by
the diffuse component with a mantle zone growth pattern.
Immunohistochemically, the two components could easily be distinguished
by their distinct antigen expression patterns. In addition to CD20, the
diffuse component co-expressed CD43 and CD5, whereas p27 was found only
in rare cells probably representing reactive T cells. The follicular
component stained strongly for p27 and CD10, but lacked CD5 and CD43
(Figure 1, C-E
). The cyclin-dependent kinase inhibitor p27 is not
expressed by the majority of MCL, in contrast to FL and other low grade
B-NHL.26
|
The lymph node showed a proliferation of irregular follicles with
absence of well defined mantle zones and a monotonous increase in small
lymphocytes in the interfollicular area. The follicles contained a
mixture of centrocytes and centroblasts, corresponding to a FL grade 2.
The interfollicular proliferation consisted of small round lymphocytes
without atypia and occasional paraimmunoblasts (Figure 2, A and B)
. This diffuse component
exhibited a phenotype consistent with SLL, showing co-expression of
CD20 with CD5, CD23, and bcl-2. The neoplastic follicles showed only
rare, strongly stained CD5- and CD43-positive cells, probably
representing T cells, lacked bcl-2 expression and showed strong CD10
reactivity (Figure 2, C
-E). The bone marrow showed focal involvement by
aggregates of lymphoma cells coexpressing CD5 and CD20.
|
Multiple lymph nodes showed replacement of the normal architecture
by a monotonous proliferation of small lymphocytes and interspersed but
clearly distinct nodular aggregates of medium-sized lymphoid cells with
more irregular nuclei and open chromatin (Figure 3, A and B)
. The nodular areas showed a
stronger expression of CD20 and were positive for cyclin D1, strongly
positive for IgD and
light chains, but negative for p27 and CD23
and focally negative for CD5. In addition, they contained irregular
meshworks of follicular dendritic cells stained by CD21. The diffuse
part showed strong reactivity for p27 and expressed both CD5 and CD23,
but lacked demonstrable cyclin D1 or Ig light chain expression,
consistent with CLL/SLL (Figure 3, C-F
).
|
The molecular findings and the sequences of rearranged IgH genes
are summarized in Tables 3 and 4
.
|
|
DNA extracted from whole tissue sections of the small bowel tumor
showed a single clonal band by CDRIII PCR as well as a bcl-2
rearrangement. Multiple microdissections from the diffuse areas
revealed an identical clonal band in the CDRIII PCR. In addition to a
weak band of identical size, the microdissected follicular areas
revealed a weaker but reproducible band of larger size not obtained
from whole tumor sections (Figure 4A)
.
Sequencing of both CDRIII PCR products confirmed the presence of two
unrelated clones. All microdissected follicular areas showed an
identical bcl-2 rearrangement, which was confirmed by Southern blotting
and sequencing. In contrast, only the largest of several
microdissections from the diffuse component showed a weak signal for
bcl-2, indicating the presence of contaminating FL cells (Figure 4B)
.
The findings were interpreted as composite MCL of the intestine with a
clonally unrelated FL carrying a bcl-2 rearrangement.
|
DNA obtained from whole lymph node sections as well as the
microdissected, CD5+ tumor component rendered a strong clonal band of
identical size. In addition to a weak band of the same size,
microdissections from the CD5- follicular component revealed a strong
band of different size, which was also identifiable as a minor product
in the PCRs from the CD5+ component as well as from the whole section
(Figure 5)
. Sequencing of both bands
revealed the presence of two different clonal rearrangements based on
the differences in the D and N regions, although both rearrangements
used the J5 gene segment. No bcl-2 or bcl-1 rearrangements were
identified by PCR with DNA obtained from whole tissue sections. The
case was interpreted as bcl-2-negative FL with a clonally unrelated
interfollicular neoplasm consistent with SLL.
|
All microdissected areas from the nodular component obtained from
two different lymph nodes rendered a clonal band of identical size and
sequence by CDRIII PCR. A band of identical size and sequence but
consistently lower intensity was amplified from the microdissections
from the diffuse component (data not shown). In contrast, amplification
with the FR2 primer set gave two different-sized clonal bands for the
two tumor components. (Figure 6)
.
Sequence comparisons showed that the lower FR2 band, amplified from the
cyclin D1+ nodular areas, resulted from the same clonal rearrangement
as the CDRIII product, whereas the higher FR2 band, amplified
predominantly from the CLL population, was clonally unrelated. This
sequence showed mismatches with the CDRIII VH
primer, probably accounting for its failure to amplify the CLL-specific
rearrangement. No bcl-1 or bcl-2 rearrangements were detected by PCR
with DNA obtained from whole tissue sections. In conjunction with the
results of the immunohistochemistry and the flow cytometric
immunophenotyping of the PB, the case was interpreted as
composite MCL and CLL with peripheral blood involvement, predominantly
by the CLL component.
|
| Discussion |
|---|
|
|
|---|
The current classification of low grade B-NHL is based on the concept that the different subtypes are characterized by neoplastic cells arrested at specific points during development, corresponding to various stages of normal B cell differentiation. Their distinct morphologies and immunophenotypes, the presence or absence of somatic hypermutation of the immunoglobulin genes, and characteristic genetic abnormalities define them as true disease entities.1,35 CLL/SLL, MCL, FL, and MZL are the most common and best defined categories of the low grade B-NHL. In this context, composite lymphomas showing two phenotypically distinct cell populations with features of two of these B-NHL entities in the same anatomical location are of special interest despite their rarity, since they seem to challenge the conceptual separation of the low grade B-NHL.
The definition of composite lymphoma has evolved over time. Initially based on the presence of two morphologically distinct components in the same anatomical site, refined criteria required the demonstration of distinct immunophenotypes, and ultimately, molecular evidence of distinct clonal origins of the two tumor cell populations.13,15,18,24,25,36 The most frequent occurrence of two different lymphoma morphologies presenting in the same patient is a high grade tumor arising in the setting of a concomitant or antecedent low grade neoplasm. This generally represents histological transformation of the low grade tumor, and a common clonal origin can be demonstrated in most, but not all, cases.2,3,14,16,20,21,23,37
Composite B-cell neoplasms with two phenotypically different low grade components are infrequent; only a few reports contain sufficient immunophenotypic and molecular genetic information.9,11,13,17-19,25,36
Although our three cases showed some unusual morphological features, only a detailed immunophenotypic analysis made a diagnosis of composite lymphoma possible. The differential expression of markers such as CD5, CD10, CD43, and cyclin D1 allowed us to identify the two cytological and architectural patterns as distinct tumor components and also to distinguish the FL infiltrates of Case 1 and the MCL nodules of Case 3 from residual germinal centers. Based on these findings, we chose to classify the two tumor components separately, following established criteria. Although the cases show some abnormalities of phenotype, such as the absence of bcl-2 expression in the FL of Case 2 or the lack of detectable cyclin D1 expression in the MCL component of Case 1, these variations are well described in otherwise typical examples of these entities.34,38,39 The characteristic absence of p27 expression in the MCL components of Cases 1 and 3, in contrast to both FL and SLL, confirmed previous findings and demonstrates the utility of this antibody when used in combination with the established markers mentioned above.26,40
Despite the convincing immunophenotypic evidence for the presence of composite lymphomas, biclonality could not be established by PCR amplification of DNA obtained from whole tissue sections in any of the cases. Only molecular analysis of microdissected tissue with comparison of rearranged immunoglobulin sequences allowed us to confirm the presence of two neoplasms of independent clonal origin in the same anatomical site. The lack of amplification of one of the two rearrangements in Cases 1 and 2 from DNA obtained from whole sections is most likely due to differences in amplification efficiency rather than only to the numerical predominance of one of the cell populations, because enrichments achievable with crude microdissection from hematoxylin and eosin (H&E)-stained sections failed to bring out the second clone in Case 1, which appeared only after high level enrichment attained with more precise LCM from immunostained sections. In Case 3, the CLL clone was amplifiable only with the FR2 primers. The failure to detect the CLL clone in the CDRIII PCR was probably a result of the mismatches at the VH primer binding site preventing the amplification of this rearrangement but allowing the detection of rare contaminating MCL cells present even in microdissected CLL populations.
Because biclonality and ongoing Ig gene alterations can occur in morphologically and phenotypically homogeneous B-NHL, mainly FL, it might be argued that the second rearrangement, found only in the microdissected tissues but not in the bulk tissue extracts, does not necessarily represent a distinct second neoplasm.7,8,12,41-43 However, the reproducible linkage of these second clonal bands to the areas with profoundly different phenotypes is evidence against this possibility. Furthermore, the ongoing Ig gene alterations that occur in FL, generally consist of point mutations and occasionally small inserts or deletions that will not change the VDJ gene segment usage detected by PCR. In addition, bcl-2 rearrangements, as seen in Case 2, are stable markers of the neoplastic clone and are preserved, despite ongoing Ig gene alterations.41 The failure to detect a bcl-2 rearrangement in the FL of Case 2 and of bcl-1 rearrangements in the MCL components of Cases 1 and 3 is not unusual, because only 3550% of these rearrangements can be detected with the primer sets used in this study. The few cases of composite low grade B-NHL reported with sufficient phenotypic and molecular information to allow classification according to current criteria also seem to represent two clonally unrelated neoplasms.13,17,19 Due to the apparent rarity of these cases, one can only speculate about the reasons for the simultaneous occurrence of two distinct low grade B-NHL. It should be noted, however, that none of our patients had a condition predisposing to B-cell lymphoma, such as autoimmune disease or immunosuppression.
The histological and immunophenotypic findings in our cases must be distinguished from the variable expression of antigens or changing morphology in different areas of the same tumor, which are probably due to a compartmentalization of neoplastic cells according to their functional or maturational status. Such zoning phenomena have been described in FL and especially MZL, a finding which initially led to the frequent description of composite lymphomas in this group.44 However, cytogenetics or molecular analysis with or without microdissection has usually supported a common clonal origin of the two populations.4,5,44,45
In summary, we present three cases of composite low grade B-cell lymphoma with two phenotypically distinct populations residing in separate but intimately interwoven compartments of the same anatomical site. The distinct clonal origin of the two compartments as revealed by molecular analysis of microdissected tissue confirmed the presence of two separate neoplasms and supports the current concepts of the classification of low grade B-NHL. The more frequent application of microdissection on lymphomas with unusual morphology and phenotype may reveal the existence of such cases more frequently. The ease and speed with which LCM can be performed on routinely immunostained tissue sections for standard PCR assays makes it a potentially useful tool for the analysis of molecular microheterogeneity of tumors.
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported in part by grants from the Austrian Science Funds to F. F. (Erwin-Schroedinger Stipendium J1402 MED) and L. Q.-M. (Charlotte-Buehler Stipendium H00083).
Accepted for publication February 25, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. Yakirevich, C. L. Jackson, P. A. Meitner, D. MacKenzie, R. Tavares, L. Robinson-Bostom, R. A. DeLellis, and M. B. Resnick Analysis of T-Cell Clonality Using Laser Capture Microdissection and High-Resolution Microcapillary Electrophoresis J. Mol. Diagn., September 1, 2007; 9(4): 490 - 497. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Gniadecki Neoplastic Stem Cells in Cutaneous Lymphomas: Evidence and Clinical Implications Arch Dermatol, September 1, 2004; 140(9): 1156 - 1160. [Full Text] [PDF] |
||||
![]() |
J. W. Gillespie, G. Gannot, M. A. Tangrea, M. Ahram, C. J.M. Best, V. E. Bichsel, E. F. Petricoin, M. R. Emmert-Buck, and R. F. Chuaqui Molecular Profiling of Cancer Toxicol Pathol, January 1, 2004; 32(1_suppl): 67 - 71. [Abstract] [PDF] |
||||
![]() |
O. Dereure, E. Levi, E. C. Vonderheid, and M. E. Kadin Improved Sensitivity of T-Cell Clonality Detection in Mycosis Fungoides by Hand Microdissection and Heteroduplex Analysis Arch Dermatol, December 1, 2003; 139(12): 1571 - 1575. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-L. Sanchez, J. Almeida, D. Gonzalez, M. Gonzalez, M.-A. Garcia-Marcos, A. Balanzategui, M.-C. Lopez-Berges, J. Nomdedeu, T. Vallespi, M. Barbon, et al. Incidence and clinicobiologic characteristics of leukemic B-cell chronic lymphoproliferative disorders with more than one B-cell clone Blood, October 15, 2003; 102(8): 2994 - 3002. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Cong, M. Raffeld, J. Teruya-Feldstein, L. Sorbara, S. Pittaluga, and E. S. Jaffe In situ localization of follicular lymphoma: description and analysis by laser capture microdissection Blood, May 1, 2002; 99(9): 3376 - 3382. [Abstract] [Full Text] [PDF] |
||||
![]() |
A Saito, G Fujii, Y Sato, M Gotoh, M Sakamoto, G Toda, and S Hirohashi Detection of genes expressed in primary colon cancers by in situ hybridisation: overexpression of RACK 1 Mol. Pathol., February 1, 2002; 55(1): 34 - 39. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Quintanilla-Martinez, M. Kremer, G. Keller, M. Nathrath, A. Gamboa-Dominguez, A. Meneses, L. Luna-Contreras, A. Cabras, H. Hoefler, A. Mohar, et al. p53 Mutations in Nasal Natural Killer/T-Cell Lymphoma from Mexico : Association with Large Cell Morphology and Advanced Disease Am. J. Pathol., December 1, 2001; 159(6): 2095 - 2105. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Fend and M. Raffeld Laser capture microdissection in pathology J. Clin. Pathol., September 1, 2000; 53(9): 666 - 672. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. S. J. Elenitoba-Johnson, S. D. Bohling, R. S. Mitchell, M. S. Brown, and R. S. Robetorye PCR Analysis of the Immunoglobulin Heavy Chain Gene in Polyclonal Processes Can Yield Pseudoclonal Bands as an Artifact of Low B Cell Number J. Mol. Diagn., May 1, 2000; 2(2): 92 - 96. [Abstract] [Full Text] |
||||
![]() |
S Curran, J A McKay, H L McLeod, and G I Murray Laser capture microscopy Mol. Pathol., April 1, 2000; 53(2): 64 - 68. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |