Hematopathology / LIGHT CHAINS IN GERMINAL CENTER AND MANTLE CELLS
Comparative Analysis of Light Chain Expression
in Germinal Center Cells and Mantle Cells of Reactive
Lymphoid Tissues
A Four-Color Flow Cytometric Study
Kaaren K. Reichard, MD, Robert W. McKenna, MD, and Steven H. Kroft, MD
Key Words: Flow cytometry; Immunophenotyping; Reactive tissue; Reactive lymphoid hyperplasia; Light chain ratio; Germinal center
DOI: 10.1092/9MYMD68FU8YE843D
Abstract
We studied skewing of light chain ratios (LCRs) in
germinal center cells (GCCs) relative to mantle cells
(MCs) by flow cytometry (FC) in 98 reactive lymphoid
tissues. LCRs were assessed using a 4-color
lambda/kappa/CD20/CD38 tube. GCCs and MCs were
discriminated by CD20 and CD38 density. Of 98 cases,
65 contained distinct GCCs and MCs. Light chain
expression usually was dimmer on GCCs than on MCs;
in 22 cases, the kappa and lambda clusters converged
and accurate LCRs could not be determined. Of the
remaining 43 cases, the mean GCC LCR was 1.78
(range, 1.10-3.07) vs 1.56 (range, 1.00-2.24) in the
MCs (P = .001). The overall kappa/lambda ratio in
cases containing GCCs and MCs was 1.65 (range,
1.18-2.69) compared with 1.46 (range, 1.00-1.98) in
cases containing MCs only. Of 43 cases, 19 (44%)
showed differences of 20% or more between the LCRs
of GCCs and MCs. LCRs of GCCs and MCs may differ
substantially in reactive lymphoid tissues. These subsets
may form distinct clusters and skews in their LCRs and
should not be misinterpreted as evidence of occult
lymphoma.
130
130
Am J Clin Pathol 2003;119:130-136
DOI: 10.1092/9MYMD68FU8YE843D
Multiparameter flow cytometry is a powerful tool for the
characterization of various hematolymphoid cell populations.
In particular, differential analysis of antigen expression on
immunophenotypically defined cell populations may be
performed. However, knowledge of the normal immunophenotypic variability among various subsets is essential for
accurate interpretation.
Flow cytometry is used commonly in combination with
clinical, morphologic, and immunohistochemical data in the
evaluation of hematolymphoid disorders. Mature B-lineage
lymphoid neoplasms typically are differentiated from reactive
processes by the identification of a light chain–restricted
population. The “normal” range that generally is used for
kappa/lambda ratios in reactive lymphoid tissues is between
0.8 and 2.2. It has been suggested that the kappa/lambda
ratios of some reactive lymphoid tissues may be as low as 0.3
and as high as 7.0.1 Ratios of more than 3.0 or less than 0.5
have been used in several studies to signify the presence of a
clonal B-cell process. 2,3 In our routine flow cytometric
analyses, kappa/lambda ratios outside the range of 1.0 to 2.0
are noted, and the B-lymphocyte populations are studied
thoroughly for aberrance.
Germinal center cells (GCCs) and mantle cells (MCs) are
immunophenotypically distinct subsets of B lymphocytes.
GCCs commonly are defined by their expression of CD10.
They also show increased density of CD20 and CD38
compared with MCs4 and frequently form a distinct cluster
based on these antigens. We routinely assess the light chain
expression patterns in GCCs and MCs by using a 4-color antibody combination of CD20, CD38, kappa, and lambda. During
the past few years, we have observed cases of reactive lymphoid
tissues that showed substantial differences between the
© American Society for Clinical Pathology
Hematopathology / ORIGINAL ARTICLE
kappa/lambda ratios of GCCs and MCs. On initial examination, this may suggest the possibility of a neoplastic process. In
the present study, we analyzed the light chain expression
patterns of GCCs and MCs in morphologically confirmed reactive lymphoid tissues to assess the degree of allowable skew.
Materials and Methods
Samples
All cases of reactive lymphoid tissues (from May 1997
to April 2000) with tissue sections available for review were
identified in the clinical flow cytometry database at the
University of Texas Southwestern Medical Center, Dallas.
Fine-needle aspirations were excluded. Also excluded were
any cases in which there were atypical histologic features
suggestive of B-cell lymphoma. A total of 98 cases ultimately were included in the study. The specimens included
95 lymph nodes, 1 submandibular gland, 1 tonsil, and 1
terminal ileum. A variety of morphologic diagnoses,
including florid follicular hyperplasia, granulomatous
inflammation, Castleman disease, and progressive transformation of germinal centers, were encountered. Clinical
data, including sex, age, and medical history, are given for
43 cases in ❚Table 1❚.
❚Table 1❚
Details of 43 Cases With GCC Populations and Distinct kappa and lambda Clusters
kappa/lambda Ratio
Case No./Sex/
Age (y)*
1/M/56
2/F/63
3/M/9 mo
4/F/44
5/F/69
6/F/49
7/F/70
8/M/34
9/F/30
10/M/34
11/M/22
12/F/39
13/M/41
14/F/54
15/M/42
16/M/55
17/F/33
18/M/47
19/M/10
20/M/15
21/F/39
22/M/14
23/M/43
24/F/24
25/F/33
26/F/9
27/M/43
28/M/27
29/M/39
30/F/35
31/M/5
32/M/46
33/F/19
34/M/5
35/M/19
36/F/51
37/F/38
38/F/44
39/F/17
40/M/36
41/M/47
42/M/50
43/F/46
History
Castleman disease
Left-sided lymphadenopathy, neck
Unknown
Cervical lymphadenopathy
Unknown
Lymphoma
Axillary lymphadenopathy
HIV+
Toxoplasmosis
HIV+
Neck abscess
HIV+
Inflammatory bowel disease
Tonsillar mass; lymphadenopathy
Unknown
Unknown
Axillary lymphadenopathy
Previous DLCBL
Unknown
Unknown
HIV+
Unknown
Hepatitis C
Unknown
Peripheral T-cell lymphoma
Unknown
Hepatitis C
Unknown
Groin mass
Unknown
Unknown
Unknown
Unknown
Unknown
Axillary lymphadenopathy
Unknown
Hepatitis C
Axillary lymphadenopathy
Cervical lymphadenopathy
Toxoplasmosis
HIV+, previous DLBCL
Unknown
Axillary lymphadenopathy
B Cells (% of Total
No. of Lymphocytes)
GCCs (% of
B Cells)
GCC
MC
Overall
37
50
30
45
37
25
39
45
37
21
35
29
57
36
34
38
38
41
22
39
22
75
29
21
37
34
28
33
39
44
39
36
39
40
30
20
29
55
18
35
45
38
68
13
9
12
31
22
19
18
81
12
27
19
60
10
25
8
30
12
8
22
42
55
10
19
31
7
24
16
43
5
26
30
36
19
42
19
38
9
4
22
19
15
9
55
1.32
1.21
1.22
1.26
2.23
1.18
1.96
2.84
1.39
2.30
1.58
1.80
2.03
2.04
1.25
1.22
1.90
1.67
1.57
1.73
2.74
1.14
1.60
1.46
2.29
3.07
1.56
2.82
1.46
1.77
1.63
1.68
1.52
2.15
2.18
2.14
1.27
2.05
1.35
1.60
1.10
1.50
2.80
1.77
1.18
1.42
1.22
2.03
1.66
1.96
2.07
1.22
1.35
1.37
1.00
1.51
1.92
1.30
1.56
1.30
1.63
1.56
1.47
2.24
1.39
1.22
1.15
2.19
1.93
1.08
1.99
1.63
1.50
1.60
1.56
1.24
1.82
1.57
1.80
1.31
1.93
1.38
1.55
1.47
1.40
1.64
1.71
1.18
1.40
1.23
2.07
1.57
1.96
2.69
1.24
1.60
1.41
1.48
1.56
1.95
1.30
1.46
1.38
1.63
1.56
1.58
2.52
1.37
1.29
1.25
2.20
2.20
1.16
2.35
1.62
1.57
1.61
1.60
1.29
1.96
1.69
1.93
1.31
1.94
1.37
1.56
1.41
1.41
2.28
DLBCL, diffuse large B-cell lymphoma; GCC, germinal center cell; MC, mantle cell.
* Unless otherwise noted.
Am J Clin Pathol 2003;119:130-136
© American Society for Clinical Pathology
131
DOI: 10.1092/9MYMD68FU8YE843D
131
131
Reichard et al / LIGHT CHAINS IN GERMINAL CENTER AND MANTLE CELLS
Data Analysis
Flow cytometric data were acquired using a 4-color
FACSCalibur flow cytometry instrument with CellQuest
software (Becton Dickinson) and analyzed using Paint-aGate software (Becton Dickinson). Specific cell populations
were identified using cluster analysis, various antigen
expression patterns, and forward and side angle light-scatter
properties. Nonviable cells and debris were removed based
on forward and orthogonal light-scatter properties.
B lymphocytes were identified by the expression of
CD19 after the exclusion of CD38 bright plasma cells
(CD10/CD19/CD20/CD38) or CD20 (pL/pK/CD20/CD38).
Nonlymphocyte events were removed based on forward
and side light-scatter properties ❚Image 1A❚. GCCs were
identified as a discrete population distinct from MCs on the
basis of increased CD20 and CD38 antigen intensity
(pL/pK/CD20/CD38) (Image 1A) ❚Image 1B❚. The separation of GCCs from MCs was validated by their CD10
expression in the tube containing CD10/CD19/CD20/CD38
❚Image 2❚. The percentage of total B cells (of all events)
was determined on CD10/CD19/CD20/CD38 and forward
scatter after the elimination of debris. The percentage of
GCCs equaled the number of GCC events divided by the
132
132
Am J Clin Pathol 2003;119:130-136
DOI: 10.1092/9MYMD68FU8YE843D
B
CD38
CD38
A
CD20
kappa
kappa
CD20
lambda
lambda
Side Scatter
Immunophenotypic Analysis
Fresh tissue was processed through a mesh filter (<100
µm), and the cells were resuspended in 5% newborn calf
serum in RPMI 1640 tissue culture medium (Life Technology, Rockville, MD). Cell counts were performed manually, and 500,000 cells per tube were washed with a phosphate buffered saline solution containing 0.0455% sodium
azide and 0.1% bovine serum albumin (PAB). The cells then
were incubated with a 4-color combination of antibodies.
Antibodies against CD2 (55.2), CD3 (SK7), CD4 (SK3),
CD5 (L17F12), CD7 (4H9), CD8 (SK1), CD10 (W8E7),
CD19 (SJ25C1), CD20 (L27), CD38 (HB7), CD45 (2D1),
CD45RO (UCHL-1), and monoclonal kappa (TB28-2) and
lambda (I-155-2) immunoglobulins were obtained from
Becton Dickinson (San Jose, CA). Antibodies against
FMC7, CD23 (B6), and polyclonal kappa (goat) and lambda
(goat) immunoglobulins were obtained from CoulterImmunotech (Hialeah, FL). Anti-CD30 (BerH2) was
obtained from DAKO (Carpinteria, CA). These antibodies
were conjugated with fluorescein isothiocyanate (FITC),
phycoerythrin (PE), peridinin chlorophyll protein (PerCP),
or allophycocyanin (APC). The expression of kappa and
lambda light chains was determined in a tube containing
polyclonal lambda (pL)-FITC/polyclonal kappa (pK)PE/CD20-PerCP/CD38-APC.
Specimens were incubated at 2°C to 8°C in the dark for
20 minutes, washed with PAB, and resuspended in phosphate-buffered saline containing 1% paraformaldehyde.
Forward Scatter
❚Image 1❚ A, Separation of germinal center cells (magenta)
from mantle cells (green) in a tube containing polyclonal
lambda/polyclonal kappa/CD20/CD38. Germinal center cells
are distinctly larger than mantle zone cells based on forward
light scatter properties. B, Determination of distinct
kappa/lambda ratios in germinal center cells based on
discrete kappa (dark blue, mantle cells; red, germinal center
cells) and lambda (light blue, mantle cells; yellow, germinal
center cells) clusters.
total number of B-cell events. The mean channel number of
forward light scatter was recorded for MC and GCC populations (Image 1). Kappa/lambda ratios for GCCs and MCs
in each case were determined in the pL/pK/CD20/CD38
tube (Image 1). After identification of B cells based on
CD20 and light scatter, discrete kappa and lambda clusters
were assigned different colors, and the ratio of events in the
2 clusters was calculated.
© American Society for Clinical Pathology
CD38
Statistical Analysis
The kappa/lambda light chain ratios in cases with and
without GCC populations were compared using a t test. The
size of GCCs vs MCs was compared using a paired sample t
test. The 2-tailed Fisher exact test was used to compare age,
sex, and HIV status between convergent and distinct GCC
cases. Spearman correlation coefficients were calculated for
the remaining various combinations of continuous variables.
CD19
Hematopathology / ORIGINAL ARTICLE
CD20
Specimen Characteristics
Of 98 cases of reactive lymphoid tissues, 33 (34%)
were composed solely of MCs with no discernible GCC
population ❚Image 4❚, precluding comparison of GCC and
MC kappa/lambda ratios. The other 65 cases (66%)
showed distinct GCC and MC populations based on CD20
and CD38 expression (Image 2). However, 22 of these 65
cases (34%) were excluded from the final analysis because
the GCC events showed convergence toward the midline
on the light chain expression plot (Image 3). This convergence precluded an accurate separation of kappa and
lambda clusters and the determination of a GCC
kappa/lambda ratio. Forty-three cases (66%) remained for
comparative analysis of kappa/lambda ratios in GCCs and
MCs (Table 1).
Comparison of GCCs and MCs
The average size of GCCs was larger than MCs in all
cases based on forward light scatter (P < .001) (Image
1A). The mean percentage of GCCs (of total B cells) was
29% (range, 4%-81%), and in the majority of cases, they
showed dimmer surface immunoglobulin expression than
the MCs. The mean kappa/lambda ratio in the 43 cases
containing GCCs and MCs was 1.65 (range, 1.18-2.69) vs
1.46 (range, 1.00-1.98) in the 33 cases composed of only
MCs (P = .009) ❚Table 3❚ . Of the 43 cases with GCC
populations, 7 (16%) showed overall kappa/lambda ratios
more than 2.0 vs none of 33 cases with MCs only (P =
.017). No cases in either group had a kappa/lambda ratio
less than 1.0.
For GCCs, the mean kappa/lambda ratio was 1.78
(range, 1.10-3.07) vs 1.56 (range, 1.00-2.24) for MCs (P =
.001) (Table 3). The difference between the GCC and MC
kappa/lambda ratios exceeded 20% in 19 (44%) of 43 cases.
The largest difference was 80% (1.00 in MCs vs 1.80 in
GCCs). Of the 19 cases, 14 showed higher ratios for GCCs
than for MCs; and 5 GCC ratios were lower than MC ratios.
Of the 65 cases with GCCs and MCs for evaluation, 12
were submitted with a history of HIV infection. HIV+ cases
had a higher mean percentage of GCCs (48%) than the HIV
negative cases (25%) (P = .006). In the 43 cases with distinct
GCC and MC light chain populations, the kappa/lambda
ratio was significantly higher in the HIV+ cases than in the
HIV– cases (1.37 vs 1.12; P = .04).
❚Figure 1❚ depicts the GCC kappa/lambda ratio vs the
MC kappa/lambda ratio. The ratio of the GCC kappa/lambda
ratios to that of the MCs correlated positively with the
proportion of GCCs (R = 0.47; P = .001) ❚Figure 2❚.
kappa
Patient Characteristics
The mean age of the 94 patients with 98 reactive
lymphoid tissue specimens was 32 years (range, 9 months to
78 years). There were 49 females and 45 males.
Of the 65 cases with GCC and MC populations for
comparison, 22 cases showed convergence of the GCC light
chain clusters (see next paragraph) ❚Image 3❚. Comparison of
these 22 cases with the 43 cases with distinct GCC kappa
and lambda populations showed no significant difference in
age (P = .40) or sex (P = .71) ❚Table 2❚. In 7 of 22 convergent cases and 5 of 43 distinct cases, there was a submitted
history of HIV infection (P = .087) (Table 2).
❚Image 2❚ Separation of germinal center cells (magenta)
from mantle cells (green) by increased intensity of CD38 and
CD20 (left) and validation by their expression of CD10 (right).
CD38
Results
CD10
CD20
lambda
❚Image 3❚ Separation of germinal center cells (GCCs;
magenta) from mantle cells (green) on CD38 and CD20 (left)
with convergence of the GCCs toward the midline on the
light chain plot (right).
Am J Clin Pathol 2003;119:130-136
© American Society for Clinical Pathology
133
DOI: 10.1092/9MYMD68FU8YE843D
133
133
Convergent
Clusters (n = 22)
Distinct
Clusters (n = 43)
40.0
1:1
7 (32)
36.4
1.2:1
5 (12)
Mean age (y)
Sex (male/female)
No. (%) HIV+
P
CD38
❚Table 2❚
Comparison of Selected Clinical Features Between Cases With
Convergent and Distinct kappa and lambda Germinal Center
Cell Clusters
kappa
Reichard et al / LIGHT CHAINS IN GERMINAL CENTER AND MANTLE CELLS
.40
.71
.087
CD20
Clinical Follow-up
Clinical follow-up was available for 41 of the 65
patients with distinct GCC and MC populations, including
15 of 22 cases with convergent GCC light chain clusters. All
patients lacked any evidence of lymphoma from 1 to 58
months (median, 30 months) after the reactive tissue biopsies. Of the 41 patients with follow-up, 3 had a history of
large B-cell lymphoma (LBCL). Two were patients with
HIV infection who had LBCL diagnosed in the nasopharynx
and retroperitoneum 1 month and 2 years previously, respectively. The reactive lymph nodes in both showed convergent
light chain clusters in the GCCs. These patients were alive
and free of lymphoma 38 and 26 months, respectively, after
the reactive lymph node excisions. The third patient was
HIV– and had an LBCL of the parotid gland diagnosed 7
years earlier. The reactive lymph node has distinct GCC light
chain clusters. This patient was alive and free of lymphoma
28 months after the reactive lymph node excision.
One HIV+ patient had a concurrent diagnosis of LBCL
of the spleen with highly anaplastic features diagnosed on
fine-needle aspiration. The left axillary lymph node that was
excised within several days of the splenic fine-needle aspiration showed convergent GCC light chain clusters.
lambda
❚Image 4❚ B cells in a clinical sample with no germinal center
cells (brighter CD38 and CD20) present. Dark blue, kappaexpressing B cells; light blue, lambda expressing B cells).
❚Table 3❚
Summary of kappa/lambda Ratios in 76 Cases of Reactive
Lymphoid Tissues
Type of Case
No. of Cases
No GCCs
GCCs and MCs
GCCs
MCs
33
43
43
43
Mean
SD
Range
P
1.46
1.65
1.78
1.56
0.22
0.38
0.52
0.31
1.0-2.0
1.2-2.7
1.1-3.07
1.0-2.2
.009*
.001†
GCCs, germinal center cells; MCs, mantle cells.
* No GCCs vs GCCs and MCs.
† GCCs vs MCs.
Discussion
Multiparameter flow cytometry can readily distinguish
GCCs from MCs by their expression of CD10 and increased
density of CD20 and CD38; each B-cell subset may form a
distinct cluster based on those antigens. In this retrospective
4-color flow cytometric study, we analyzed the kappa/lambda
light chain ratios of GCCs and MCs in reactive lymphoid
Germinal Center Cell Ratio/
Mantle Cell Ratio
2.0
Germinal Center Cells
3.5
3.0
2.5
2.0
1.5
1.6
1.4
1.2
1.0
0.8
0.6
1.0
1.0
1.5
2.0
Mantle Cells
2.5
❚Figure 1❚ Kappa/lambda ratios in germinal center cells vs
mantle cells. y = 1.11x + 0.056. Gray line, identity line; black
line, regression line.
134
134
1.8
Am J Clin Pathol 2003;119:130-136
DOI: 10.1092/9MYMD68FU8YE843D
3.0
0
10
20
30
40
50
60
70
80
90
Germinal Center Cells % (of Total B Cells)
❚Figure 2❚ Ratio of germinal center kappa/lambda ratio to
mantle cell kappa/lambda ratio as a function of the
percentage of germinal center cells (GCC; percentage of
total B cells). R = 0.47; P = .001.
© American Society for Clinical Pathology
Hematopathology / ORIGINAL ARTICLE
tissues and found that they may differ substantially. In addition, the kappa/lambda ratio of GCCs may exceed the ranges
commonly used to distinguish benign and neoplastic
processes. Thus, knowledge of the normal variability in light
chain ratios and the differential antigen expression of
immunophenotypically distinct groups is critical for accurate
interpretation.
Mature B-cell neoplasms are detected immunophenotypically as populations with B-lineage surface markers and
immunoglobulin light chain restriction. In many instances,
neoplastic populations form cell clusters distinct from
normal B cells on the basis of aberrant expression of 1 or
more lymphoid antigens. In such circumstances, these
discrete populations express 1 light chain exclusively and are
detectable even in the presence of a background of polyclonal B lymphocytes. However, when neoplastic populations lack overt aberrance, they typically will be detected as
increases or decreases in the kappa/lambda ratio of the entire
B-cell population. In flow cytometry–based studies Maiese
et al4 and Kaleem et al5 reported kappa/lambda ranges of
0.98 to 2.27 and 0.6 to 2.2, respectively, for reactive lymph
nodes. Values for the kappa/lambda ratio of more than 3.0 or
less than 0.5 have been used to signify a B-cell neoplastic
proliferation.2,3,5
In lymphoid tissues containing distinct GCCs and MCs,
we found a statistically significant difference between the
kappa/lambda ratios of these populations (Table 3): 44%
(19/43) showed a difference of 20% or more in these ratios.
The majority showed a higher ratio for the GCCs than MCs.
Tissues with increasing degrees of follicular hyperplasia
showed greater variability between the GCC and MC
kappa/lambda ratios. Cases composed of only MCs had
overall lower kappa/lambda ratios compared with cases with
GCCs and MCs. These findings are consistent with the
previous observation that unusually high or low
kappa/lambda ratios are most likely to be seen in florid
follicular hyperplasia.1
We routinely have considered overall kappa/lambda
ratios less than 1.0 and more than 2.0 worthy of additional
investigation to rule out occult B-cell neoplasia. In this
regard, it is notable that 7 of the 19 cases showing a difference of 20% or more between the GCC and MC
kappa/lambda ratios had kappa/lambda ratios exceeding 2.0.
This variability and skew in the kappa/lambda ratios of reactive lymphoid tissues should not be taken as evidence of a Bcell neoplasm. Importantly, none of our 98 cases demonstrated a kappa/lambda ratio less than 1.0.
The range of light chain ratios determined in the present
study is narrower than many reported in the literature. This is
likely due to methodologic differences. First, 3- or 4-color
flow cytometry permits the assessment of B-cell markers
(eg, CD19 or CD20) in the same tube as kappa and lambda.
This permits exclusion of non–B-cell populations that may
show nonspecific staining patterns from the light chain
analysis. Second, by specifically isolating kappa and lambda
cell clusters, rather than relying on the number of events
exceeding certain thresholds (as in quadrant analysis), an
additional level of noise is removed from the analysis. We
believe that this approach maximizes the precision of the
light chain ratio determination.
Investigators previously have reported occasional
abnormal kappa/lambda ratios in otherwise morphologically
benign lymphoid tissues.6-8 Palutke et al6 identified 12
lymph nodes and 1 stomach from 10 patients that showed
more than 25% lymphocytes bearing a single class
immunoglobulin. Half of the cases subsequently were diagnosed as lymphoma or were highly suggestive of lymphoma.
Levy et al7 reported 12 examples of morphologically benign
lymphoid tissues with what they considered to represent
monoclonal immunoglobulin expression; at the time of
publication, none of the patients had developed lymphoma.
Bain and Bain8 reported 12 cases of reactive lymph nodes
with abnormal kappa/lambda ratios. Follow-up information
was available for 10 of the patients. Six remained well, 1 had
persistent lymphadenopathy, 1 developed non-Hodgkin
lymphoma, 1 developed Hodgkin lymphoma, and 1 died of
lung carcinoma and pneumonia. It has been postulated that
perhaps these cases represent occasional natural instances in
which a few B-cell clones predominated and expressed a
similar class of immunoglobulin. 7 Data to support this
hypothesis come from molecular experiments by Küppers
and colleagues.9 By using micromanipulation techniques,
these authors traced B-cell development from the mantle
zone through the germinal center. They identified a progressive, antigen-driven pathway from clonally diverse mantle
cells to polytypic germinal center dark zone cells to a few
dominant B-cell clones in the light zone. These data provide
an explanation for the occasional skewed kappa/lambda
ratios seen in purely reactive states. In cases of benign reactive lymphoid tissues, particularly those showing prominent
follicular hyperplasia, the survival and predominance of a
few high-affinity clones is therefore expected. Since kappabearing B lymphocytes are naturally more abundant, one
would expect a skew more often toward the kappa than
toward the lambda light chain.
As expected, we identified a trend toward kappa in the
B cells of reactive lymphoid tissues. Figure 1 depicts the
correlation of the GCC vs MC kappa/lambda ratios with an
overall shift toward kappa. Despite the fact that this is a
physiologically plausible process, we are uncertain whether
this shift reflects a true biologic phenomenon or artifact
based on the fact that our polyclonal kappa antibody was
conjugated to a brighter fluorochrome (PE) than our lambda
reagent (FITC). In populations expressing dim light chain,
Am J Clin Pathol 2003;119:130-136
© American Society for Clinical Pathology
135
DOI: 10.1092/9MYMD68FU8YE843D
135
135
Reichard et al / LIGHT CHAINS IN GERMINAL CENTER AND MANTLE CELLS
such as GCCs, the population defined by the brighter fluorochrome might be overestimated. To address this issue, we
prospectively analyzed reactive lymphoid tissues with an
antibody combination consisting of monoclonal kappaFITC/monoclonal lambda-PE/CD20-PerCP/CD38-APC
(data not shown). However, these fluorochromes did not
adequately discriminate between the GCC clusters, and,
therefore, we cannot be certain whether the kappa skew is a
physiologic phenomenon or artifact. Nevertheless, the data
from Küppers et al,9 combined with the greater likelihood of
selecting kappa-expressing B cells to become dominant
clones, favor a true biologic phenomenon. In addition, a
positive correlation of the ratio of the GCC kappa/lambda
ratio to the MC kappa/lambda ratio with the percentage of
GCCs also was found (Figure 2). This further suggests that
as tissues become more hyperplastic, the germinal center
produces a small number of dominant B cells that more often
express kappa than lambda light chain.
An additional finding in this study was that GCCs tend
to exhibit decreased intensity of surface immunoglobulin,
which may mimic a neoplastic population with dim (or negative) surface immunoglobulin. Twenty-two of our cases
displayed such a phenomenon and were excluded from the
final analysis because the GCC light chain clusters
converged about the midline of the dot plot, precluding reliable separation of the populations (Image 3). Correlation
with other cell surface markers (CD10, CD20, CD38) will
help identify these as a normal population. Of note, cases
with convergent GCC light chain populations were more
likely to be submitted with a history of HIV infection than
those lacking this feature, although this was not found to be
statistically significant. The 12 cases reported with HIV
infection were found to have higher percentages of GCCs
than the HIV– cases. In cases with distinct GCC light chain
populations, the HIV+ cases had higher overall
kappa/lambda ratios compared with the HIV– cases. Lymph
nodes excised from HIV-infected individuals often show
florid follicular hyperplasia. These cases may be expected to
contain increased numbers of GCCs and, therefore, skews in
their ratios and/or dim surface immunoglobulin expression.
It should be noted that the present study was based on
excisional lymph node and other tissue biopsy specimens so
that reactive histologic features could be confirmed. Therefore, our results may be strictly applicable only to such specimens. It might be expected that skews in the kappa/lambda
136
136
Am J Clin Pathol 2003;119:130-136
DOI: 10.1092/9MYMD68FU8YE843D
ratio would be greater in paucicellular specimens, such as
fine-needle aspirations or core biopsies.
By using a 4-color flow cytometry technique to evaluate
reactive lymphoid tissues, we were able to distinguish GCCs
and MCs based on CD10, CD20, and CD38 expression. We
found that the GCC and MC kappa/lambda ratios may differ
substantially and occasionally exceed the usual range of 1.0
to 2.0. Either subset may form a distinct cluster in the
analysis, and GCCs frequently demonstrate decreased light
chain expression. These findings should not be misinterpreted as occult B-lineage neoplasia.
From the Department of Pathology, University of Texas
Southwestern Medical Center, Dallas.
Address reprint requests to Dr Kroft: Dept of Pathology,
University of Texas Southwestern Medical Center, 5323 Harry
Hines Blvd, Dallas, TX 75930-9072.
References
1. Knowles DM. Immunophenotypic markers useful in the
diagnosis and classification of hematopoietic neoplasms. In:
Knowles DM, ed. Neoplastic Hematopathology. 2nd ed.
Philadelphia, PA: Lippincott Williams &Wilkins; 2001:185.
2. Geary WA, Frierson HF, Innes DJ, et al. Quantitative criteria
for clonality in the diagnosis of B-cell non-Hodgkin’s
lymphoma by flow cytometry. Mod Pathol. 1993;6:155-161.
3. Taylor CR. Results of multiparameter studies of B-cell
lymphomas. Am J Clin Pathol. 1979;72:687-698.
4. Maiese RL, Segal GH, Iturraspe JA, et al. The cell surface
antigen and DNA content distribution of lymph nodes with
reactive hyperplasia. Mod Pathol. 1995;8:536-543.
5. Kaleem Z, Vollmer RT, White G, et al. Clonality evaluation
in B-cell lymphoproliferative disorders by flow cytometric
immunophenotyping [abstract]. Mod Pathol. 2001;14:152A.
6. Palutke M, Schnitzer B, Mirchandani I, et al. Increased
numbers of lymphocytes with single class surface
immunoglobulins in reactive hyperplasia of lymphoid tissue.
Am J Clin Pathol. 1982;78:316-323.
7. Levy N, Nelson J, Meyer P, et al. Reactive lymphoid
hyperplasia with single class (monoclonal) surface
immunoglobulin. Am J Clin Pathol. 1983;80:300-308.
8. Bain VG, Bain GO. Lymphocyte populations with abnormal
kappa:lambda ratios in reactive lymphoid hyperplasia. J Surg
Oncol. 1985;29:227-230.
9. Küppers R, Zhao M, Hansmann ML, et al. Tracing B cell
development in human germinal centres by molecular
analysis of single cells picked from histological sections.
EMBO J. 1993;12:4955-4967.
© American Society for Clinical Pathology