Received: 21 February 2020 Accepted: 16 September 2020

DOI: 10.1111/jvim.15913

STANDARD ARTICLE

Polyclonal B-cell lymphocytosis in English bulldogs

Emily D. Rout1 | A Russell Moore1 | Robert C. Burnett1 | Julia D. Labadie2,3 |

Kelly L. Hughes1 | Paul A. Navin4 | Janna A. Yoshimoto1 | Paul R. Avery1 |
Anne C. Avery1

1
Department of Microbiology, Immunology
and Pathology, College of Veterinary Medicine Abstract
and Biomedical Sciences, Colorado State Background: English bulldogs disproportionally develop an expansion of small B-cells,
University, Fort Collins, Colorado
2 which has been interpreted as B-cell chronic lymphocytic leukemia (BCLL). However,
Public Health Sciences Division, Fred
Hutchinson Cancer Research Center, Seattle, clonality testing in these cases has often not been supportive of neoplasia.
Washington
Hypothesis: English bulldogs have a syndrome of nonneoplastic B-cell expansion.
3
Department of Epidemiology, University of
Washington, Seattle, Washington Animals: Eighty-four English bulldogs with small-sized CD21+ B-cell lymphocytosis
4
VCA All Pets Animal Hospital, Lockport, in the blood as determined by flow cytometry.
Illinois
Methods: This is a retrospective study. We characterized this syndrome by assessing
Correspondence B-cell clonality, clinical presentation, flow cytometric features, and immunoglobulin
Anne C. Avery, 200 West Lake Street,
gammopathy patterns. We identified 84 cases with CD21+ lymphocytosis among
Colorado State University, Fort Collins, CO,
80523. 195 English bulldogs with blood samples submitted to the Colorado State University-
Email: anne.avery@colostate.edu
Clinical Immunology laboratory for immunophenotyping between 2010 and 2019.
Funding information Flow cytometry features were compared to normal B-cells and BCLL cases. PCR for
Morris Animal Foundation, Grant/Award
antigen receptor rearrangements (PARR) by multiple immunoglobulin primers was
Number: D18CA-413
performed to assess B-cell clonality. A subset of cases with gammopathy were exam-
ined by protein electrophoresis, immunofixation, and immunoglobulin subclass ELISA
quantification.
Results: Seventy percent (58/83) of cases had polyclonal or restricted polyclonal
immunoglobulin gene rearrangements, suggesting nonmalignant B-cell expansion.
The median age of all dogs in the study was 6.8 years and 74% were male. The
median (range) lymphocyte count was 22 400/μL (2000-384 400/μL) and B-cells had
low expression of class II MHC and CD25. Splenomegaly or splenic masses were
detected in 57% (26/46) of cases and lymphadenopathy in 11% (7/61). Seventy-one
percent (52/73) of cases had hyperglobulinemia and 77% (23/30) with globulin char-
acterization had IgA ± IgM polyclonal or restricted polyclonal gammopathy patterns.
Conclusions and Clinical Importance: Polyclonal B-cell lymphocytosis in English bull-
dogs is characterized by low B-cell class II MHC and CD25 expression, splenomegaly

Abbreviations: BCLL, B-cell chronic lymphocytic leukemia; CSU-CI, Colorado State University-Clinical Immunology; IG, immunoglobulin; IGH, IG heavy chain; IGHV, IGH variable; IGH-VDJ,
complete immunoglobulin heavy (IGH) chain variable (V)-diversity (D)-joining (J) gene rearrangement; IGL, IG lambda; IQR, interquartile range; Kde, kappa-deleting element; PARR, PCR for
antigen receptor rearrangements; PBLEB, polyclonal B-cell lymphocytosis in English bulldogs; PCR, polymerase chain reaction; PE/IF, protein electrophoresis/immunofixation; PPBL, persistent
polyclonal B-cell lymphocytosis; RALD, RAS-associated autoimmune lymphoproliferative disorder.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2020 The Authors. Journal of Veterinary Internal Medicine published by Wiley Periodicals LLC on behalf of American College of Veterinary Internal Medicine.

J Vet Intern Med. 2020;1–14. wileyonlinelibrary.com/journal/jvim 1

2 ROUT ET AL.

and hyperglobulinemia consisting of increased IgA ± IgM. We hypothesize that this

syndrome has a genetic basis.

KEYWORDS
clonality, canine, clinical pathology, flow cytometry, hyperglobulinemia, IgA gammopathy, PCR
for antigen receptor rearrangements

1 | I N T RO DU CT I O N substantial increases in IgA with or without IgM, with normal to

diminished IgG.
Lymphocytosis can be caused be a reactive or neoplastic process. In
dogs, reactive expansions of nonneoplastic lymphocytes in the blood
appear uncommon and are associated with only a small number of 2 | M A T E R I A L S A N D M ET H O D S
conditions, including Ehrlichia canis infection, hypoadrenocorticism
and thymoma.1-5 Neoplastic lymphocytosis comprises clonally 2.1 | Case selection
expanded lymphocytes and is a more common cause of persistent
lymphocytosis in adult dogs.6 Clonality testing by PCR for antigen The Colorado State University-Clinical Immunology (CSU-CI) labora-
receptor rearrangements (PARR) can help differentiate a monoclonal tory database was queried for English bulldog cases with blood sub-
population of neoplastic lymphocytes that have an identically sized mitted for immunophenotyping by flow cytometry between
antigen receptor rearrangement from a polyclonal population of reac- September 17, 2010 and August 31, 2019. Inclusion criteria included
tive lymphocytes, which contains diverse antigen receptor an expansion of the number of small-sized CD21+ B-cells exceeding
7,8
rearrangements. Reactive or inflammatory processes might also the upper limit of the reference interval (724 CD21+ cells/μL) for
cause increased production of polyclonal immunoglobulin proteins.9,10 canine blood samples at our institution. Small size was defined by a
Monoclonal immunoglobulin production is typically due to an size ratio of CD21+ cell median forward scatter: neutrophil median
immunoglobulin-secreting B-cell or plasma cell neoplasm, though forward scatter < 0.60.
rarely certain infectious or inflammatory conditions are associated The CSU-CI laboratory database query was also used to identify
with monoclonal gammopathy in dogs.9,11 several control groups, which are as follows: (i) blood from 30 clinically
B-cell chronic lymphocytic leukemia (BCLL) is a common hemato- healthy non-English bulldogs without lymphocytosis or evidence of
poietic neoplasm in dogs, defined by a clonal expansion of small-sized lymphoproliferative disease by flow cytometry and PARR, which were
12,13
B-cells in the blood or bone marrow. Our laboratory identifies a range of breeds and included 57% females and age ranged from 1.9
BCLL based on inclusion criteria of >5000 lymphocytes/μL on CBC to 8.9 years old; (ii) blood from 49 clinically healthy control English
with small-sized CD21+ B-cells accounting for >60% of the lympho- bulldogs with no suspicion of lymphoproliferative disease and a nor-
cyte population by flow cytometry. Small breed dogs have increased mal CD21+ B-cell count (<724/μL by flow cytometry), which included
13
risk of developing BCLL. English bulldogs have increased odds of 41% females and age ranged from 1 to 12 years old (median, 3.9 years
developing BCLL, as defined in this study, but this breed had a unique old); (iii) 53 small breed BCLL cases from BCLL-predisposed breeds13
presentation in being significantly younger at diagnosis compared to with blood submitted for flow cytometry between 1 January 2015
mixed breed dogs, having increased frequency of hyperglobulinemia, and 10 January 2019 and clonal immunoglobulin (IG) rearrangements
and their B-cells had decreased CD25 and class II MHC expression by by PARR; and (iv) PARR results from 25 small breed BCLL cases with
flow cytometry. This unique presentation raised the question of a flow cytometry diagnosis and routine PARR performed on submitted
whether English bulldogs have a different form of BCLL, or a different blood.
B-cell disease entirely. Serum immunoglobulins were evaluated in 30 English bulldog
Since detecting this unique presentation in English bulldogs, our cases, 6 control English bulldogs with no evidence of CD21+ lympho-
laboratory anecdotally found that English bulldogs with B-cell lym- cytosis by flow cytometry, 15 non-bulldog BCLL cases with hyper-
phocytosis frequently had polyclonal immunoglobulin gene globulinemia, and 4 control non-bulldogs with polyclonal gammopathy
rearrangements by PARR. These PARR findings suggested that the and no lymphocytosis.
B-cell expansions in these dogs might be nonneoplastic, and that
English bulldogs have a B-cell lymphocytosis syndrome separate
from BCLL. The goal of this study was to identify English bulldogs 2.2 | Clinical variables
with B-cell lymphocytosis, to evaluate clonality by PARR and protein
electrophoresis/immunofixation (PE/IF) modalities, and to analyze Signalment, physical exam findings, and laboratory data were provided
the clinical features of the cases. Here, we describe a syndrome of on the CSU-CI laboratory submission form. Hematology data were
polyclonal B-cell expansion in English bulldogs characterized by collected from the CBC performed at the time of flow cytometry
ROUT ET AL. 3

submission. Anemia, thrombocytopenia, and neutropenia were identi- polymorphisms or somatic hypermutation in the primer-binding
fied by the CBC laboratory's reference interval for hematocrit, platelet region. Light-chain primers target kappa-deleting element (Kde) and
count and neutrophil count, respectively. Hyperglobulinemia was IG lambda (IGL) rearrangements. The routine and expanded PARR
identified if indicated by the veterinarian on the submission form as assays were performed on all English bulldog cases with available
being present, or if the concentration of globulins was increased on a sample.
biochemistry panel or by PE. Physical exam findings, including lymph- PARR results were interpreted as clonal, polyclonal, or restricted
adenopathy, splenic abnormalities, and hepatic abnormalities, were polyclonal by the rubric provided in Table 1. These criteria were inter-
identified by palpation, ultrasound, radiographs, or a combination of nally validated and apply only to these primers and cycling conditions
these tests. When available, blood films, bone marrow cytology when analyzed by fragment analysis methods, and therefore, are not
reports, and spleen histology samples were reviewed. Histology sam- applicable for other assays and laboratories. Cases with 1 or 2 peaks
ples had immunohistochemistry performed for Pax5 expression reaching objective criteria for clonality were classified as clonal, as
(monoclonal mouse anti-human Pax5, DAK-Pax5 clone; Dako North lymphocytes can rearrange 1 or both alleles of the IG locus resulting
America Inc., Carpinteria, CA), CD3 expression (monoclonal mouse in up to 2 clonal peaks.16 Representative tracings for the electropho-
anti-human CD3, LN10 clone; Leica Biosystems Newcastle Ltd., retic patterns are provided in Figure 1. DNA quantity and quality were
Newcastle Upon Tyne, UK), and MUM1 expression (monoclonal determined to be sufficient based on PARR results, as previously
mouse anti-human MUM1, MUM1p clone; Dako North America Inc.). described.17 PARR results were independently interpreted by the
For cases with sequential samples, treatment information was authors (E.D. Rout, A.C. Avery), blinded to the signalment and flow
collected from the CSU-CI laboratory submission form. cytometry data.

2.3 | Flow cytometry 2.5 | Serum immunoglobulins

Flow cytometry was performed on blood samples at the CSU-CI labo- Agarose gel PE and IF were performed at Colorado State University to
ratory. Samples were collected, stored, and stained as previously evaluate immunoglobulins, as previously described.15,18 PE and IF
14
described by antibody panels listed in Supplemental Table 1. Expres- were performed on serum or plasma samples from a subset of English
sion of CD25, class II MHC, and CD21 on B-cells were determined in bulldog cases (n = 30), control English bulldogs (n = 6), and non-
English bulldog cases, healthy control dogs, and small breed BCLL bulldog BCLL cases (n = 10). Five additional BCLL cases had PE
cases. Because anti-CD25 is not in the same staining reaction as anti-
CD21, the percentage of B-cells expressing CD25 was determined by
gating on lymphocytes in tube 2; excluding cells expressing CD3, T A B L E 1 Immunoglobulin PCR for antigen receptor
rearrangements (PARR) assay diagnostic criteria
CD4, CD5, and CD8; and calculating %CD25 on remaining cells. Class
II MHC and CD21 expression was determined by median fluorescence Interpretation Capillary electrophoretic pattern
intensity. Six English bulldog cases sampled before May 2012 did not IGH-VDJ
have CD25 or class II MHC data available. rearrangements
Clonal 1 or 2 tall narrow peak(s) >5000 in
amplitude and >3 times the height of the
base peaks forming the polyclonal
2.4 | PCR for antigen receptor rearrangements
background
Polyclonal Multiple peaks forming a Gaussian
The CSU-CI routine PARR assay was performed as previously distribution
described.7,15 This assay detects B-cell clonality by targeting both
Restricted No peaks reach clonal criteria, but ≥1 peak is
complete and incomplete IG gene rearrangements in the IG heavy polyclonal >2000 in amplitude and >2 times the
chain (IGH) locus. Complete IGH-VDJ rearrangements consist of a V height of the base peaks forming the
(variable), D (diversity) and J (joining) gene, whereas incomplete IGH- polyclonal background

DJ rearrangements consist of a D and J gene. IGH-DJ, Kde, or IGL

rearrangements
An expanded PARR assay was developed to detect additional
complete IGH-VDJ and incomplete IGH-DJ rearrangements not Clonal Tall narrow peak >8000 in amplitude and
>3 × the height of the base peaks forming
detected with routine PARR, as well as IG light chain rearrangements.
the polyclonal background
Primers were designed and tested as described in the Supplemental
Polyclonal Multiple peaks forming a Gaussian
material. The expanded assay for complete IGH-VDJ rearrangements distribution
targets IGH variable (IGHV) subgroup genes not detected by our rou-
Abbreviations: IGH-DJ, incomplete immunoglobulin heavy chain diversity
tine PARR assay. Additionally, IGH-VDJ primers in the expanded assay
(D)-joining (J) gene rearrangement; IGH-VDJ, complete immunoglobulin
target a different region of the IGHV gene, to increase detection of heavy (IGH) chain variable (V)-diversity (D)-joining (J) gene rearrangement;
IGHV genes that did not amplify with routine PARR because of IGL, immunoglobulin lambda; Kde, kappa-deleting element.
4 ROUT ET AL.

F I G U R E 1 PCR for antigen receptor rearrangements (PARR) results for English bulldogs with B-cell lymphocytosis. GeneMarker tracings for
complete immunoglobulin heavy (IGH) chain variable (V)-diversity (D)-joining (J) (IGH-VDJ) rearrangements are presented for 4 polyclonal cases (A),
4 restricted polyclonal cases (B), and 4 clonal cases (C). The size of the PARR amplicons is plotted on the horizontal axis and the abundance of
amplicons is on the vertical axis. (A) Polyclonal cases have multiple peaks forming a Gaussian distribution. (B) Restricted polyclonal cases have 1 to
5 peaks >2000 in amplitude and 2 times the height of the polyclonal peaks forming the base. (C) Clonal cases have a peak >5000 in amplitude and
>3 times the base height

without IF. The hemolysis and lipemia indices were determined for Quantification of IgA, IgM, and IgGFC proteins was performed
all samples (Cobas c501 Roche Diagnostics, Indianapolis, IN), as with commercially available ELISA kits, as previously described.18
9,19
these factors can alter electrophoretogram morphology. Total Immunoglobulin proteins were measured in serum and plasma sam-
globulin concentration was calculated as the sum of all globulin ples from a subset of English bulldog cases (n = 12), 4 clinically healthy
fractions by PE to avoid overestimation of albumin in cases with a English bulldogs and 3 clinically healthy non-bulldogs.
20
marked gammopathy. Hyperglobulinemia was identified by PE if
the total globulin concentration exceeded our internal upper refer-
ence limit 3.5 g/dL. PE/IF results were blindly and independently 2.6 | Statistical analysis
interpreted by the authors (E.D. Rout, P.R. Avery, A.R. Moore)
using the rubric in Table 2. Representative results for the electro- For all English bulldog cases, signalment, physical exam findings,
phoretic patterns are provided in Figure 2. Immunofixation results hematologic data, and flow cytometry data were summarized, and
were interpreted to determine whether immunoglobulins were descriptive statistics were calculated. For continuous variables, nor-
predominantly of IgG (detected by the IgGFC antibody), IgA, or IgM mality was assessed visually and by a Shapiro-Wilk test. To compare
21
subclass. An IgA or IgM predominance was considered atypical, continuous variables between English bulldog cases, healthy controls
as normal dogs have predominantly IgGFC labeling with no to faint and BCLL cases, a Kruskal-Wallis test was calculated. Dunn's test for
IgA and IgM.9 multiple comparisons was subsequently used for pairwise
ROUT ET AL. 5

TABLE 2 Protein electrophoresis and immunofixation diagnostic was 6.8 years and there was a male predominance. The lymphocyte
criteria count was moderately increased in most cases, but a small number
Interpretation Electrophoretic pattern (16/84) of cases had >50 000 lymphocytes/μL (median, 22 400

Protein lymphs/μL; interquartile range [IQR], 12 000-41 100 lymphs/μL;

electrophoresis range, 2000-384 400 lymphs/μL). Anemia was present in 26/84 cases
Normal No protein increases or atypical restricted (31%) and mild to moderate in all affected cases (median hematocrit,
bands present for the sample type 30%; IQR, 27-34%; range, 21-38%). The anemia was nonregenerative
Monoclonal Single atypical restricted peak with no more in 14/17 cases with reticulocyte counts available, as defined by the
than minimal polyclonal increases laboratory's reticulocyte reference interval. Thrombocytopenia was
Biclonal Pair of closely associated peaks with no more uncommon, affecting 8/84 cases (10%), and mild to moderate in all
than minimal surrounding polyclonal
affected cases (median, 114 000 platelets/μL; IQR, 89500-152 000/
increases
μL; range, 38 000-198 000/μL). Hyperglobulinemia affected 65% of
Restricted ≥1 atypical restricted band(s) on gel amid
cases (45/69) with globulin data at diagnosis. An additional 7 cases
polyclonal moderate polyclonal increases, irrespective
of height of any peak had hyperglobulinemia detected after initial diagnosis, resulting in

Polyclonal Only diffuse/broad bands present 71% of cases (52/73) having hyperglobulinemia at some point during
the course of their syndrome. Splenomegaly or splenic masses or both
Immunofixation
were also common (26/46), but lymphadenopathy was rare (7/61).
Normal Moderate IgGFC polyclonal labeling with no to
faint polyclonal IgA and IgM Of 44 cases where lymphocyte morphology was evaluated by a
clinical pathologist, 35/44 were described as having small-sized or
Monoclonal Single restricted band of 1 IGH class with no
more than minimal IgGFC polyclonal increase small-to-intermediate-sized lymphocytes, 4 were described as having
Biclonal 2 bands (same IGH class or different classes) intermediate-sized lymphocytes, and 5 cases were described as having
with no more than minimal IgGFC polyclonal intermediate-to-large-sized lymphocytes. All cases met the flow
increase cytometric inclusion criteria for small-sized lymphocytes. Chromatin
IgA/IgM ≥1 restricted bands of IgA and/or IgM amid was described as mature, clumped, or condensed in all cases evalu-
restricted moderate IgA and/or IgM polyclonal labeling
ated, except in 1 where it was described as fine. Six cases were
polyclonal and decreased IgGFC
described as having few nucleoli. Ten cases with blood smears avail-
IgA/IgM Increased immunoglobulin labeling without
able at the Colorado State University Veterinary Diagnostic laboratory
polyclonal restricted bands in IgA and/or IgM with
decreased IgGFC were reviewed by the authors (E.D. Rout, P.R. Avery). Lymphocytes

IgGFC polyclonal Increased immunoglobulin labeling without were predominantly smaller than a neutrophil with a small round
restricted bands with predominantly IgGFC nucleus, condensed chromatin, and scant basophilic cytoplasm
labeling and no to faint IgA and IgM (Figure 3A). Rare cells were intermediate sized with slightly expanded
Abbreviation: IGH, immunoglobulin heavy chain. pale blue cytoplasm. When cases were blindly pooled with blood films
from 10 Shih Tzu dogs with BCLL, the English bulldog cases could not
be differentiated cytologically. Two English bulldog cases had bone
comparisons. Immunoglobulin quantification was compared between marrow aspirates performed. Megakaryocytes and erythroid and mye-
controls and English bulldog cases with a Mann-Whitney test. For loid series were considered within normal limits and small mature lym-
comparisons between clonal and nonclonal English bulldog cases, a phocytes accounted for 20% and 43% of nucleated cells.
Mann-Whitney test was calculated. A Fisher's exact test was used to Six cases had cytologic evaluation and 4 cases had histologic eval-
compare categorical variables across groups. Cases with missing or uation of the spleen. None of these samples were definitively diag-
unknown data on the submission form were censored for that data. nosed with lymphoma. Six cytology samples were diagnosed with
Statistical analysis was performed in R version 3.5.2 and a 2-sided lymphoid hyperplasia and 3 of those had a second diagnosis of possi-
P value <.05 was considered significant. ble lymphoma. One sample was minimally cellular, but remaining cases
were described as moderately to highly cellular with a heterogeneous
lymphoid population, consisting of predominantly small lymphocytes
3 | RESULTS with condensed chromatin, with few large lymphocytes and plasma
cells. Cytology report comments indicated that the heterogeneity and
3.1 | Clinical presentation predominance of small well-differentiated lymphocytes was consis-
tent with lymphoid hyperplasia, but given the clinical history, a small
Between September 17, 2010 and August 31, 2019, 84 of 195 English cell lymphoma or BCLL infiltration into the spleen could not be ruled
bulldogs with blood submitted to the CSU-CI laboratory for immuno- out. Histologically, there was expansion of the spleen with lymphoid
phenotyping by flow cytometry had an expanded number of small- hyperplasia of variable severity. Lymphoid hyperplasia ranged from
sized CD21+ B lymphocytes. Signalment and laboratory data from mild expansion around periarteriolar sheaths to more marked nodular
these cases are summarized in Table 3. The median age at diagnosis lymphoid hyperplasia. Nodular lymphoid hyperplasia formed small,
6 ROUT ET AL.

F I G U R E 2 Protein electrophoresis (PE) and immunofixation (IF) results in control dogs and English bulldogs with B-cell lymphocytosis. An
agarose gel protein electrophoresis gel and tracing (top) and immunofixation gel (bottom) are provided. Immunofixation gels are labeled with anti-
whole serum (WS), anti-canine IgGFC heavy chain (G), anti-canine IgA heavy chain (A), anti-canine IgM heavy chain (M), and anti-canine light chain
(L) antibodies. (A) Serum PE and IF results from a healthy control English bulldog with no CD21+ B-cell lymphocytosis reveal polyclonal
immunoglobulin proteins that predominantly label with anti-IgGFC. (B) Serum PE and IF results from a non-bulldog with polyclonal gammopathy.
Immunoglobulin proteins form a broad smear and predominantly label with anti-IgGFC. (C) Serum PE and IF results from an English bulldog B-cell
lymphocytosis case diagnosed with polyclonal gammopathy characterized by broad peaks in beta and gamma regions on PE, with no evidence of
restricted bands by PE or IF. There is heavy labeling with anti-IgA by IF relative to IgGFC, which is atypical as compared to the control (A), and a
lack of a hypergammaglobulinemia, which is atypical for an IgG-centric polyclonal gammopathy. (D) Serum PE and IF results from an English
bulldog B-cell lymphocytosis case diagnosed with restricted polyclonal gammopathy, characterized by multiple restricted peaks within a
polyclonal background. By IF, the majority of protein labels with anti-IgA. (E) Plasma PE and IF results from an English bulldog B-cell
lymphocytosis case with a clonal immunoglobulin gene rearrangement by PARR. There is a tall narrow peak in the beta 2 region on PE, which
labels with anti-IgM by IF, consistent with an IgM monoclonal gammopathy

noncoalescing follicular structures to occasionally larger rarely coa- bulldog controls (P < .001) and small breed BCLL cases (P < .001).
lescing follicles (Figure 3B). Follicles were composed of predominantly English bulldog cases had significantly higher B-cell CD21 expres-
B-cells (Figure 3D) with scattered small T-cells often radiating around sion than non-bulldog controls (P < .001), English bulldog controls
follicles and scattered plasma cells within sinusoids (not shown). The (P < .001) and small breed BCLL cases (P < .001). Comparing non-
B-cells occupying the follicular structures were heterogeneous, com- bulldog and English bulldog controls, English bulldog controls had
posed of small and intermediate-sized lymphocytes with condensed significantly lower CD25 and CD21 expression (P = .003 and P = .02,
chromatin or occasional marginal zone appearance with a prominent respectively).
central nucleolus (Figure 3C).

3.3 | Clonality
3.2 | Flow cytometry
Eighty-three of 84 English bulldog cases had sample material available
Among English bulldog cases, the median CD21+ B-cell count was for routine PARR analysis. Eighty-one cases had PARR performed on
20 606 cells/μL (IQR, 10461-42 682/μL; range, 1174-378 815/μL; the same blood sample that was submitted for flow cytometry. Two
reference interval, 0-724/μL). CD21+ B-cells accounted for 67%- cases had tissue samples (spleen and bone marrow, respectively) sub-
99% of the lymphocyte population in the blood by flow cytometry. mitted for PARR within 1 month of the blood flow cytometry analysis;
English bulldog cases were characterized by low expression of CD25 PARR results from those sources were used for analysis. Thirty-nine
and class II MHC. Sixty-eight percent of cases (53/78) had only percent of cases had polyclonal IG rearrangements, 37% of cases had
0-2% of B-cells expressing CD25, and English bulldog cases had sig- restricted polyclonal IG rearrangements, and 24% had clonal IG
nificantly lower CD25 expression than B-cells from clinically healthy rearrangements by routine PARR. Representative PARR tracings for
non-bulldogs (P < .001), clinically healthy English bulldog controls polyclonal, restricted polyclonal, and clonal cases are presented in
(P < .001) and small breed BCLL cases (P < .001) (Figure 4). B-cell Figure 1. Of the 25 cases which met inclusion criteria for small breed
class II MHC expression was also significantly lower in English bull- BCLL during an 18-month period, 100% of BCLL cases had clonal IG
dog cases compared to non-bulldog controls (P < .001), English rearrangements by routine PARR analysis, with either complete
ROUT ET AL. 7

T A B L E 3 Summary signalment data, laboratory data, and physical 2.1 g/dL (IQR, 1.8-2.5 g/dL; range, 1.4-2.7 g/dL; reference interval,
exam findings for 84 English bulldogs with B-cell lymphocytosis at the 2.8-3.7 g/dL). Approximately half of English bulldog samples had an
time of diagnosis
atypical banding pattern, which did not fit a classic polyclonal or mono-
Number affected (%) clonal pattern, so a rubric was developed to interpret these samples
Number of cases or median (IQR; (Table 2), and representative results are presented in Figure 2. Six cases
with available data range)
had an IgA or IgM monoclonal gammopathy, and all 6 cases had clonal
Signalment
IG rearrangements by PARR (Table 4). Remaining cases were classified
Male 84 62 (74%) as polyclonal (n = 7) or restricted polyclonal (n = 17) by PE alone. By IF,
Age, median (IQR; 84 6.8 (5.2-9.0; 2.5-11.0) 1 case was classified as polyclonal with IgGFC predominance; this case
range), years
had a clonal IG rearrangement by PARR and mild hyperglobulinemia
Hematologic data (3.95 g/dL). Eight cases had polyclonal gammopathy with increased IgA
Lymph count, 84 22.4 (12.0-41.1; with or without increased IgM. Fifteen cases had restricted polyclonal
median (IQR; 2.0-384.4)
gammopathy with increased IgA with or without increased IgM. All
range), ×103/μL
cases had an appropriate light chain pattern for all IG heavy chains pre-
Anemia 84 26 (31%)
sent. Of 23 cases with IgA/IgM polyclonal or restricted polyclonal
Thrombocytopenia 84 8 (10%)
gammopathy, globulins were frequently increased in the beta region on
Neutropenia 84 2 (2%)
PE and gamma globulins were often within the reference interval. 2/23
Hyperglobulinemia 69 45 (65%) (9%) cases had hypogammaglobulinemia and 15/23 (65%) cases had
Physical exam normogammaglobulinemia. In summary, 77% of cases (23/30) were
Splenomegaly/ 46 26 (57%) classified as polyclonal or restricted polyclonal with increased IgA with
splenic mass
or without increased IgM. One of the cases with IgA/IgM polyclonal or
Peripheral 61 7 (11%) restricted polyclonal IG protein had a clonal IG rearrangement by PARR,
lymphadenopathy
and the remaining cases had polyclonal or restricted polyclonal IG
Visceral 38 5 (13%)
rearrangements.
lymphadenopathy
English bulldog PE/IF results were compared to control English
Hepatomegaly/ 42 6 (14%)
bulldogs without CD21+ B-cell lymphocytosis (n = 6) and non-bulldog
hepatic mass
BCLL cases (n = 15). Three of the control English bulldogs had normal
Note: For variables besides age and lymphocyte count, the number of
PE/IF, and the other 3 dogs had mild hyperglobulinemia with mild
cases with available data, and the percentage of cases affected among
those with available data are presented. For age and lymphocyte count,
polyclonal gammopathy on PE, labeling with predominantly IgGFC
the median, interquartile range (IQR), and range of values are presented. heavy chain antibody. Two of these cases had increased acute phase
proteins, suggesting underlying inflammation in these control dogs. Of
5 BCLL cases with PE only, 4 cases had a monoclonal protein in the
IGH-VDJ primers alone (76%), incomplete IGH-DJ primers alone beta 1-beta 2 region, and the fifth case had a suspicious band in the
(12%), or both (12%). beta 2 region that could not be confirmed as M-protein without IF. Of
Eighty-five percent of English bulldog cases (n = 71) had sample 10 BCLL cases with PE and IF, 6 cases had monoclonal IgM
available for PARR with expanded IG primers. Clonal results were gammopathy, 1 case had monoclonal IgA gammopathy, 1 case had
detected with the expanded primer set in 5 cases that had been poly- biclonal IgA gammopathy, and 2 cases had polyclonal gammopathy
clonal by routine PARR. These 5 cases had polyclonal complete IGH- with IgGFC predominance.
VDJ rearrangements by routine PARR, but clonal complete IGH-VDJ Serum IgA, IgM, and IgGFC proteins were quantified by ELISA in a
rearrangements with the expanded primers. When results from rou- subset of English bulldog cases (n = 12) to confirm the increases in IgA
tine PARR and expanded PARR were combined, 37% of English bull- and IgM concentrations seen on IF (Figure 5). IgA, IgM, and IgGFC pro-
dogs were polyclonal, 33% of cases had restricted polyclonal IG teins were also measured in 3 non-bulldog healthy dogs and 4 healthy
rearrangements, and 30% were clonal (Table 4). English bulldogs without CD21+ B-cell lymphocytosis. There were no
significant differences in measurements between non-bulldog and bull-
dog control samples, so control values were combined. English bulldog
3.4 | Immunoglobulin protein analysis cases with B-cell lymphocytosis had significantly higher quantities of
IgA (P < .001) and IgM (P = .012), and significantly less IgGFC than con-
Serum or plasma globulin concentration, electrophoretic morphology, trols (P = .004). All 12 English bulldog cases had IgA values above the
and isotype location and distribution were evaluated in 30 English bull- range seen in control cases, although 1 case with marked IgM
dog cases with PE/IF. Total globulin concentrations ranged from 3.3 to gammopathy had only a mild increase in IgA. Four cases had marked
10.7 g/dL (median, 6.1 g/dL; IQR, 4.3-8.7 g/dL; reference interval, increases in IgM. All English bulldog cases had IgGFC quantities below
2.2-3.5 g/dL). Eighty percent of cases (24/30) had hypoalbuminemia. the mean concentration seen in controls and 9 cases (75%) had IgGFC
Among hypoalbuminemic cases, the median albumin concentration was quantities below the minimum value seen in controls, which correlates
8 ROUT ET AL.

F I G U R E 3 Blood smear and histopathology and immunohistochemistry of spleen from English bulldog cases with polyclonal B-cell
lymphocytosis. (A) Peripheral blood film from an English bulldog with B-cell lymphocytosis and polyclonal immunoglobulin gene rearrangements.
Lymphocytes are small with condensed chromatin and scant basophilic cytoplasm (Wright Giemsa, ×60 objective, 10 μm scale bar). (B-D)
Histopathology of spleen from a different English bulldog with B-cell lymphocytosis and polyclonal immunoglobulin gene rearrangements. There
is lymphoid hyperplasia characterized by nodules of multifocal to rarely coalescing lymphoid follicular structures (B, H&E, ×2 objective, 500 μm
scale bar). The follicular structures are composed of primarily small lymphocytes with condensed chromatin with fewer intermediate-sized
lymphocytes and scattered lymphocytes with marginal zone appearance with a single central prominent nucleolus (C, H&E, ×40 objective, 20 μm
scale bar). Lymphocytes within the follicles are predominated by B-cells with strong nuclear immunoreactivity for PAX5 (D, PAX5, Fast Red
chromogen, ×4 objective, 200 μm scale bar)

with the normogammaglobulinemia and hypogammaglobulinemia seen group included cases with polyclonal or restricted polyclonal IG
on PE/IF. The pattern of relative IG concentration by ELISA subjectively rearrangements by PARR. Clonal cases were significantly older than
matched the pattern observed by IF, confirming that English bulldog nonclonal cases (P = .002) (Figure 6A). There were significantly more
cases had increased IgA and decreased IgGFC concentrations, and a sub- males in the nonclonal group (81%) compared to clonal cases (56%)
set (n = 4) had markedly increased IgM. (P = .03) (Figure 6B). The CD21+ B-cell count was not significantly dif-
ferent between clonal cases (median, 21 700/μL; IQR, 10900-73 500/
μL; range, 1200-378 800/μL) and nonclonal cases (median, 19 700/
3.5 | Clonal versus nonclonal English bulldog cases μL; IQR, 9800-38 300/μL; range, 1400-285 000/μL) (P = .38). Class II
MHC expression was not significantly different between groups
We investigated clinical and immunophenotypic differences between (P = .12), but clonal cases had significantly lower expression of CD21
clonal cases (n = 25) and nonclonal cases (n = 58). The nonclonal (P = .04, not shown) and significantly higher expression of CD25
ROUT ET AL. 9

(P = .003) (Figure 6C). The majority of nonclonal cases (76%) had <2% T A B L E 4 Summary clonality results for immunoglobulin gene
CD25-expressing B-cells, while 48% of clonal cases had <2% rearrangements and immunoglobulin proteins in English bulldog cases
with B-cell lymphocytosis

Number of PE cases with clonal

Clonality result cases (%) PARR result
Immunoglobulin gene
rearrangementsa
Polyclonal 31/83 NA
(37%)
Restricted polyclonal 27/83 NA
(33%)
Clonal 25/83 NA
(30%)
Immunoglobulin
proteinsb
IgGFC polyclonal 1/30 1/1
(3%)
IgA polyclonal 7/30 1/7
(23%)
IgA + IgM polyclonal 1/30 0/1
(3%)
IgA restricted 3/30 0/3
polyclonal (10%)
IgA + IgM restricted 12/30 0/12
polyclonal (40%)
IgA monoclonal 2/30 2/2
(7%)
IgM monoclonal 4/30 4/4
(13%)
a
Clonality of immunoglobulin (IG) gene rearrangements was determined by
PCR for antigen receptor rearrangements (PARR) in 83 unique cases. NA,
not applicable.
b
Clonality of immunoglobulin proteins was determined by protein electro-
phoresis (PE) and immunofixation in 30 cases. The number of cases with a
clonal IG rearrangement by PARR is presented in the far-right column. All
cases with monoclonal protein had clonal IG gene rearrangements. 22/24
cases with polyclonal or restricted polyclonal proteins had polyclonal or
restricted polyclonal IG gene rearrangements.

CD25-expressing B-cells. There were no significant differences in

anemia, thrombocytopenia or splenomegaly/splenic mass between
groups. The presence of hyperglobulinemia was not significantly dif-
ferent between clonal and nonclonal cases, although the type of

F I G U R E 4 B-cell expression of CD25, class II MHC, and CD21 in

English bulldog B-cell lymphocytosis cases compared to clinically
healthy non-bulldog controls, English bulldog controls with normal B-
cell counts, and small breed B-cell chronic lymphocytic leukemia
(BCLL) cases. Expression of CD25 (A), class II MHC (B) and CD21
(C) by flow cytometry is plotted for individual cases. Lines depict the
median and interquartile range for each group. English bulldog cases
had significantly lower expression of CD25 and class II MHC and
significantly higher expression of CD21 compared to peripheral blood
B-cells from healthy non-bulldog and English bulldog controls and
small breed BCLL cases
10 ROUT ET AL.

F I G U R E 6 Signalment and B-cell CD25 expression in nonclonal

and clonal English bulldog cases with B-cell lymphocytosis. (A) The age
at diagnosis for nonclonal cases with polyclonal or restricted polyclonal
immunoglobulin PARR results and clonal cases with clonal
immunoglobulin PARR results. Clonal cases (median, 8.2 years old) were
significantly older than nonclonal cases (median, 6.3 years old)
F I G U R E 5 IgA, IgM, and IgGFC protein quantification by ELISA in
(P = .002). (B) The percentage of males and females within nonclonal
English bulldog cases with B-cell lymphocytosis. Each English bulldog
and clonal groups is presented. There were significantly more males in
case is colored consistently across the 3 graphs. Data points plotted
the nonclonal group (P = .03). (C) The percentage of B-cells expressing
at 10 g/dL were above the limits of quantification for the assay.
CD25 by flow cytometry in nonclonal and clonal cases. Nonclonal cases
Dotted lines represent the mean (black line) and range (gray lines) of
had significantly lower CD25 expression than clonal cases (P = .003)
values for 7 control dogs. English bulldog cases had significantly
greater quantities of serum IgA (P < .001), and a subset of cases had
increased IgM, compared to control dogs. English bulldogs had
significantly less IgGFC than controls (P = .004)
ROUT ET AL. 11

gammopathy was different: in the 8 clonal cases with PE/IF, 6/8 cases cases were managed with steroids alone or steroids and oral chemo-
had monoclonal IgA or IgM protein, and none of the nonclonal PARR therapy and these treated cases had reductions in the B-cell count.
cases had monoclonal PE/IF results. Peripheral lymphadenopathy was Two treated cases with sequential PE/IF analysis had reductions in
more common in clonal cases (32%) than nonclonal cases (2%) globulin concentration, though atypical PE/IF patterns and IgA or IgA
(P = .003). Of the 6 clonal cases with peripheral lymphadenopathy, and IgM increases persisted. Treatment protocols were variable in
5/6 were female, 5/6 were > 6.8 years old, and 5/6 had high CD25 these cases.
expression, in contrast to the young male signalment and low CD25
expression frequently seen in nonclonal cases.
4 | DI SCU SSION

3.6 | Sequential sample analysis Our study identified a syndrome in English bulldogs characterized by
an expanded number of nonclonal B-cells in the blood, with polyclonal
Sequential PARR samples >1 month apart were available for 18 cases or restricted polyclonal IG rearrangements by PARR. This syndrome,
with polyclonal or restricted polyclonal IG rearrangements on initial termed polyclonal B-cell lymphocytosis of English bulldogs (PBLEB),
presentation. Most cases had 1 sequential sample, but 6 cases had frequently affects young males and is commonly associated with
2 to 5 sequential samples available for analysis. The time between ini- splenomegaly or splenic masses and hyperglobulinemia. The hyper-
tial diagnosis and the most recent sample varied from 2.6 to globulinemia is characterized by a polyclonal or restricted polyclonal
64.5 months. The PARR results did not change over time for 13/18 pattern, with increased IgA with or without IgM.
cases. Cases with restricted polyclonal IG rearrangements had identi- Forty-three percent of English bulldogs with blood submitted to
cally sized peaks over time (Figure 7). In 1 case, which was followed the CSU-CI laboratory for flow cytometry over a nearly 9-year period
for 64.5 months, the restricted peaks became more pronounced over had a B-cell expansion. These cases generally have moderate lympho-
time, resulting in several tall narrow peaks with minimal polyclonal cytosis (median, 22 400 lymphocytes/μL). An expansion of a single
background (Figure 7C). Of the 5 cases where PARR results changed lymphocyte subset of this magnitude typically corresponds with lym-
over time, 2 restricted polyclonal cases became polyclonal (both were phoid neoplasia in dogs, which is consistent with the fact that 100%
receiving oral chemotherapy and prednisone), 1 polyclonal case of the small breed dogs with B-cell lymphocytosis in this study had
became restricted polyclonal, and 2 restricted polyclonal cases prog- clonal IG rearrangements. However, the majority (70%) of English bull-
ressed to having clonal IG rearrangements. In both cases, a restricted dogs with B-cell lymphocytosis did not have clonal IG rearrangements
peak initially identified increased in amplitude over time and reached by PARR, suggesting a nonneoplastic process.
clonal criteria in the sequential sample. To help rule out a false-negative PARR result, an expanded PARR
Of 14 monitored cases with treatment data, 10 cases have been assay was developed. The routine PARR assay has high sensitivity
clinically stable over time with no treatment, and the B-cell count among traditional BCLL cases, but its sensitivity in detecting neo-
remained stable or decreased over time in 7/10 cases and increased plasms with extensive somatic hypermutation is not known. We were
in 3/10 cases though the count remained less than 50 000/μL. Four concerned that the bulldogs might have a neoplasm with somatic

F I G U R E 7 Sequential PCR for antigen receptor rearrangements (PARR) results with immunoglobulin primers for 3 English bulldog cases with
B-cell lymphocytosis. For each case, immunoglobulin PARR results at initial presentation (top) and sequential PARR results (bottom)
3.4-64.5 months after diagnosis are presented. The size of the PCR amplicons is indicated along the horizontal axis. Restricted peaks persist over
time and are identical in size to those present at initial presentation. The PARR tracings maintain a similar pattern over time in 2 cases (A, B) and
become more restricted in a case that has been monitored over 5 years (C)
12 ROUT ET AL.

hypermutation, which affected PARR sensitivity. We developed an splenomegaly, lymphadenopathy, hypergammaglobulinemia, and auto-
expanded PARR assay, which assessed clonality in additional IG loci, immunity.32 This syndrome is diagnosed at a young age and generally
to test whether this assay would detect clonality in bulldog cases. has an indolent clinical course, though rare cases can undergo malig-
Detection of incomplete IGH-DJ and IG light-chain rearrangements nant transformation. Sequencing in 1 case demonstrated a restricted
increases sensitivity in detecting B-cell and plasma cell neoplasms in B-cell receptor repertoire, with expansions of many different B-cell
humans, cats and dogs, as compared to complete IGH-VDJ clones.33 There are features of human PPBL and RALD that are not
16,17,22,23
rearrangements alone. Incomplete IGH-DJ and Kde consistent with the canine PBLEB syndrome described here, including
rearrangements are less prone to somatic hypermutation, and analysis binucleate lymphocyte morphology and female predisposition in
of the IG kappa locus is useful in detecting clonality in canine tumors PPBL, and monocytosis, lymphadenopathy, hypergammaglobulinemia,
with somatic hypermutation, such as plasmacytomas.16,22 The and IgG gammopathy in RALD. However, these human syndromes do
expanded PARR assay targeting additional IGH-VDJ and IGH-DJ gene highlight the possibility for an underlying mutation or chromosomal
rearrangements and IG light-chain rearrangements identified an addi- abnormality to cause altered lymphocyte homeostasis, resulting in
tional 5 cases as clonal, but the majority of nonclonal cases remained polyclonal B-cell expansions, splenomegaly, and gammopathy.
nonclonal with the expanded PARR assay. False-negative PARR We hypothesize that this syndrome begins as a nonneoplastic
results in the English bulldogs due to nonfunctional IG condition, with a potential for malignant transformation over time. In
rearrangements, such as a J region deletion, are unlikely because cases PBLEB cases with sequential samples, we documented progression
have large quantities of immunoglobulin protein in the serum, from polyclonal IG rearrangements to restricted polyclonal IG
suggesting that gene rearrangements are functional.8 It is still possible rearrangements in 1 case, and from restricted polyclonal IG
that the expanded B-cells in these English bulldogs are rearranging an rearrangements to clonal IG rearrangements in 2 cases. Additionally,
unusual or highly variable IGHV gene not detected with these assays, clonal cases were significantly older than polyclonal cases. A subset of
but the PARR testing with this extensive pool of primers suggests that those clonal English bulldogs had features consistent with PBLEB,
most of these English bulldogs have a nonneoplastic or preneoplastic including splenomegaly without lymphadenopathy, hyperglobulinemia
syndrome. and low class II MHC and CD25 expression on B-cells, raising the pos-
One-third of the bulldog cases had an unusual restricted poly- sibility that some PBLEB cases might undergo malignant transforma-
clonal IG PARR pattern defined by IG peaks, which did not meet tion over time. In human medicine, monoclonal gammopathy of
objective criteria for clonality, within a polyclonal background. undetermined significance is considered a premalignant disease that
Approximately 50% of cases had an unusual restricted polyclonal can undergo malignant transformation and progress to Waldenstrom
PE/IF pattern, where restricted bands of IG protein were present macroglobulinemia or a plasma cell neoplasm.34 Additionally, a subset
among a polyclonal background. These patterns suggest restricted IG of apparently healthy asymptomatic people with monoclonal B-cell
diversity in the B-cell population, possibly attributed to proliferation lymphocytosis will progress to chronic lymphocytic leukemia/small
of a restricted, but not necessarily neoplastic pool of B-cells. The lymphocytic lymphoma.35 We hypothesize that bulldogs with PBLEB
nature of this restricted pool was remarkably consistent over time. also have a spectrum of disease, and some bulldogs will maintain non-
Identically sized IG rearrangements were detected in subsequent sam- neoplastic polyclonal B-cell expansions while others progress to malig-
ples, suggesting persistence of expanded B-cell populations. The nancy. Therefore, these bulldogs might require monitoring over time
unusual restricted IG protein pattern seen on PE/IF also supports the to identify progression.
hypothesis that there are restricted pools of IG-secreting B-cells. A separate subset of clonal English bulldog cases had a different
There are rare human disorders of nonmalignant polyclonal B- clinical and flow cytometry presentation than the PBLEB cases. These
cells, which can be associated with splenomegaly and hyper- cases had a female predominance, lymphadenopathy, and higher class
globulinemia. Persistent polyclonal B-cell lymphocytosis (PPBL) is a II MHC and CD25 expression. We hypothesize these bulldogs have a
benign disease characterized by expanded polyclonal B-cells in the different B-cell disease, such as conventional BCLL as seen in other
blood and increases in polyclonal serum IgM.24 Patients are generally breeds. It is unclear if these cases have a different clinical course, but
asymptomatic and a subset of cases have splenomegaly. Rare cases this information would be useful. Though these cases and the PBLEB
have massive splenomegaly, which might require splenectomy and cases that progress could both meet the clinical definition of BCLL,
mimic splenic lymphoma histologically.25-27 Familial cases suggest a we hypothesize that these are different entities with different under-
genetic predisposition and many cases have an isochromosome lying predisposition and different events leading to B-cell neoplasia.
(3q) abnormality and chromosomal instability.24,28 Most cases remain Limited data suggest that a subset of bulldogs maintain stable
polyclonal and clinically stable, but there are rare reports of cases pro- lymphocyte counts and globulin levels without a need for therapeutic
gressing to a clonal B-cell malignancy.29 PPBL is hypothesized to intervention, but a subset of cases (including polyclonal cases) require
result from hyperproliferation of B-cells with a marginal zone-like phe- treatment to control marked hyperglobulinemia and its sequela and
notype and altered CD40 signaling.25,30,31 Ras-associated autoim- massive splenomegaly (data not provided).
mune lymphoproliferative disorder (RALD) is a rare nonmalignant Hyperglobulinemia and splenomegaly were common features in
human syndrome caused by mutations in RAS genes and characterized PBLEB cases. Globulin concentrations were often moderately to
by persistent monocytosis, polyclonal B-cell lymphocytosis, massive markedly increased. All cases tested had increased IgA, a subset of
ROUT ET AL. 13

cases had concurrent increases in IgM, and IgGFC was typically normal OF F-LABEL ANTIMI CROBIAL DECLARATION
to decreased. In T-cell-dependent IgA class switching pathways, cyto- Authors declare no off-label use of antimicrobials.
kines in combination with CD40L can promote B-cell proliferation and
IgA class switching.36 In T-cell-independent pathways, soluble factors INSTITU TIONAL ANIMAL C AR E AND USE COMMITTEE
such as BAFF and APRIL induce IgA class switching.36 Several mecha- (IACUC) OR OTHER APPROVAL DECLARATION
nisms might explain the constellation of findings in these dogs. PBLEB Authors declare no IACUC or other approval was needed.
cases might have either increased factors that stimulate IgA class
switching or a decrease in inhibitors of class switch recombination. HUMAN E THICS APPROVAL DECLARATION
For example, APRIL is hypothesized to cause hyperproduction of IgA Authors declare human ethics approval was not needed for this study.
in human IgA nephropathy and BAFF overexpression in mice resulted
in a hyper-IgA syndrome with B-cell hyperplasia.37,38 Alternatively,
OR CID
certain B-cell subsets, including splenic marginal zone B-cells in mice,
Emily D. Rout https://orcid.org/0000-0003-1435-8532
class switch to IgA more effectively than other subsets, and perhaps
A Russell Moore https://orcid.org/0000-0003-4904-8788
the expanded B-cells in PBLEB are primed to readily/robustly switch
to IgA when exposed to stimuli.36,39 Many PBLEB cases had spleno-
RE FE RE NCE S
megaly/splenic masses without lymphadenopathy. A subset of cases
1. Avery AC, Avery PR. Determining the significance of persistent lym-
had massive splenomegaly and a few dogs were treated with splenec-
phocytosis. Vet Clin North Am Small Anim Pract. 2007;37(2):267-282.
tomy because of discomfort. Histology was only available in a small 2. Weiser M, Thrall M, Fulton R, Beck E, Wise L, Van Steenhouse J.
number of cases, but splenic samples were expanded by enlarged lym- Granular lymphocytosis and hyperproteinemia in dogs with chronic
phoid follicles consisting of heterogeneous B-cells. The expanded B- ehrlichiosis. J Am Anim Hosp Assoc. 1991;27:84-88.
3. Peterson ME, Kintzer PP, Kass PH. Pretreatment clinical and labora-
cells in this syndrome could originate in the spleen, there could be
tory findings in dogs with hypoadrenocorticism: 225 cases
stimulatory molecules in the spleen promoting proliferation and sur- (1979-1993). J Am Vet Med Assoc. 1996;208(1):85-91.
vival at this site, or the expanded B-cells express receptors promoting 4. Burton AG, Borjesson DL, Vernau W. Thymoma-associated lympho-
homing to the spleen rather than lymph node. A limitation of this cytosis in a dog. Vet Clin Pathol. 2014;43(4):584-588.
5. Batlivala TP, Bacon NJ, Avery AC, et al. Paraneoplastic T cell lympho-
study is that only 2 cases had bone marrow aspirates performed. In
cytosis associated with a thymoma in a dog. J Small Anim Pract. 2010;
these cases, small, mature appearing lymphocytes comprised 20% and
51(9):491-494.
43% of the bone marrow nucleated cells suggesting that this compart- 6. Yagihara H, Uematsu Y, Koike A, et al. Immunophenotyping and gene
ment might be involved. Examination of additional cases would be rearrangement analysis in dogs with lymphoproliferative disorders
necessary to confirm this suspicion. characterized by small-cell lymphocytosis. J Vet Diagn Invest. 2009;21
(2):197-202.
There appears to be a genetic predisposition in English bulldogs
7. Burnett RC, Vernau W, Modiano JF, Olver CS, Moore PF, Avery AC.
for this syndrome. English bulldog controls with normal B-cell counts Diagnosis of canine lymphoid neoplasia using clonal rearrangements
had significantly lower CD25-expressing B-cells than non-bulldog of antigen receptor genes. Vet Pathol. 2003;40(1):32-41.
controls of various breeds. Because we do not have these data for a 8. van Dongen JJ, Langerak AW, Brüggemann M, et al. Design and stan-
dardization of PCR primers and protocols for detection of clonal
population of other purebred dogs, we do not know if this finding is
immunoglobulin and T-cell receptor gene recombinations in suspect
specific to English bulldogs, nor if it is related to the syndrome being lymphoproliferations: report of the BIOMED-2 concerted action
described. Future directions are aimed at performing genome BMH4-CT98-3936. Leukemia. 2003;17:2257-2317.
sequencing to identify underlying mutations. We have documented 9. Moore AR, Avery PR. Protein characterization using electrophoresis
and immunofixation; a case-based review of dogs and cats. Vet Clin
rare cases of this syndrome in other breeds. These non-bulldog cases
Pathol. 2019;48(S1):29-44.
were identified because of their clinical presentation, including 10. Tappin SW, Taylor SS, Tasker S, Dodkin SJ, Papasouliotis K,
splenomegaly and hyperglobulinemia, and flow cytometric B-cell fea- Murphy KF. Serum protein electrophoresis in 147 dogs. Vet Rec.
tures, including low class II MHC and CD25 expression. Polyclonal IG 2011;168(17):456.
11. Giraudel JM, Pagès JP, Guelfi JF. Monoclonal gammopathies in the
gene rearrangements by PARR and polyclonal IgA gammopathies by
dog: a retrospective study of 18 cases (1986-1999) and literature
PE/IF supported a diagnosis of this syndrome in these non-English review. J Am Anim Hosp Assoc. 2002;38(2):135-147.
Bulldog cases. 12. Comazzi S, Gelain ME, Martini V, et al. Immunophenotype predicts
survival time in dogs with chronic lymphocytic leukemia. J Vet Intern
Med. 2011;25(1):100-106.
ACKNOWLEDGMENTS
13. Bromberek JL, Rout ED, Agnew MR, Yoshimoto J, Morley PS, Avery AC.
This study was made possible with support in the form of funding Breed distribution and clinical characteristics of B cell chronic lymphocytic
from the Morris Animal Foundation, D18CA-413. The authors thank leukemia in dogs. J Vet Intern Med. 2016;30(1):215-222.
the veterinarians and clients that have provided valuable information 14. Seelig DM, Avery P, Webb T, et al. Canine t-zone lymphoma: unique
and samples for these cases. immunophenotypic features, outcome, and population characteristics.
J Vet Intern Med. 2014;28(3):878-886.
15. Colopy LJ, Shiu KB, Snyder LA, Avery AC, Rout ED, Moore AR. Immu-
CONF LICT OF IN TE RE ST DEC LARAT ION noglobulin G4-related disease in a dog. J Vet Intern Med. 2019;33(6):
Authors declare no conflict of interest. 2732-2738.
14 ROUT ET AL.

16. Langerak AW, Groenen PJTA, Brüggemann M, et al. 30. Berkowska MA, Grosserichter-Wagener C, Adriaansen HJ, et al. Per-
EuroClonality/BIOMED-2 guidelines for interpretation and reporting sistent polyclonal B-cell lymphocytosis: extensively proliferated
of Ig/TCR clonality testing in suspected lymphoproliferations. Leuke- CD27+IgM+IgD+ memory B cells with a distinctive
mia. 2012;26(10):2159-2171. immunophenotype. Leukemia. 2014;28(7):1560-1564.
17. Rout ED, Burnett RC, Yoshimoto JA, Avery PR, Avery AC. Assess- 31. Voelxen N, Wehr C, Gutenberger S, et al. B-cell signaling in persistent
ment of immunoglobulin heavy chain, immunoglobulin light chain, polyclonal B lymphocytosis (PPBL). Immunol Cell Biol. 2016;94(9):
and T-cell receptor clonality testing in the diagnosis of feline lym- 830-837.
phoid neoplasia. Vet Clin Pathol. 2019;48(S1):45-58. 32. Calvo KR, Price S, Braylan RC, et al. JMML and RALD (Ras-associated
18. Harris AD, Rout E, Avery A, Bolte D, Belling-Kelly E, Moore AR. Vali- autoimmune leukoproliferative disorder): common genetic etiology
dation and method comparison of the use of densitometry to quantify yet clinically distinct entities. Blood. 2015;125(18):2753-2758.
monoclonal proteins in canine sera. Vet Clin Pathol. 2019;48(S1): 33. Levy-Mendelovich S, Lev A, Rechavi E, et al. T and B cell clonal
78-87. expansion in Ras-associated lymphoproliferative disease (RALD) as
19. Booth RA, McCudden CR, Balion CM, et al. Candidate recommenda- revealed by next-generation sequencing. Clin Exp Immunol. 2017;189
tions for protein electrophoresis reporting from the Canadian Society (3):310-317.
of Clinical Chemists Monoclonal Gammopathy Working Group. Clin 34. Kyle RA, Durie BGM, Rajkumar SV, et al. Monoclonal gammopathy of
Biochem. 2018;51:10-20. undetermined significance (MGUS) and smoldering (asymptomatic)
20. Snozek CLH, Saenger AK, Greipp PR, et al. Comparison of bromcresol multiple myeloma: IMWG consensus perspectives risk factors for pro-
green and agarose protein electrophoresis for quantitation of serum gression and guidelines for monitoring and management. Leukemia.
albumin in multiple myeloma. Clin Chem. 2007;53(6):1099-1103. 2010;24(6):1121-1127.
21. Donaghy D, Moore AR. Identification of canine IgG and its subclasses, 35. Strati P, Shanafelt TD. Monoclonal B-cell lymphocytosis and early-
IgG1, IgG2, IgG3 and IgG4, by immunofixation and commercially stage chronic lymphocytic leukemia: diagnosis, natural history, and
available antisera. Vet Immunol Immunopathol. 2020;221:110014. risk stratification. Blood. 2015;126(4):454-462.
22. Takanosu M, Nakano Y, Kagawa Y. Improved clonality analysis based 36. Cerutti A. The regulation of IgA class switching. Nat Rev Immunol.
on immunoglobulin kappa locus for canine cutaneous plasmacytoma. 2008;8(6):421-434.
Vet Immunol Immunopathol. 2019;215:109903. 37. Takahara M, Nagato T, Nozaki Y, et al. A proliferation-inducing ligand
23. Evans PAS, Pott C, Groenen P, et al. Significantly improved PCR- (APRIL) induced hyper-production of IgA from tonsillar mononuclear
based clonality testing in B-cell malignancies by use of multiple immu- cells in patients with IgA nephropathy. Cell Immunol. 2019;341:
noglobulin gene targets. Report of the BIOMED-2 Concerted Action 103925.
BHM4-CT98-3936. Leukemia. 2007;21(2):207-214. 38. McCarthy DD, Chiu S, Gao Y, Summers-deLuca LE, Gommerman JL.
24. Troussard X, Cornet E, Lesesve J-F, Kourel C, Mossafa H. Polyclonal BAFF induces a hyper-IgA syndrome in the intestinal lamina propria
B-cell lymphocytosis with binucleated lymphocytes (PPBL). Onco Tar- concomitant with IgA deposition in the kidney independent of LIGHT.
gets Ther. 2008;1:59-66. Cell Immunol. 2006;241(2):85-94.
25. Del Giudice I, Pileri SA, Rossi M, et al. Histopathological and molecu- 39. Kaminski DA, Stavnezer J. Enhanced IgA class switching in marginal
lar features of persistent polyclonal B-cell lymphocytosis (PPBL) with zone and B1 B cells relative to follicular/B2 B cells. J Immunol. 2006;
progressive splenomegaly. Br J Haematol. 2009;144(5):726-731. 177(9):6025-6029.
26. Sun P, Juskevicius R. Histological and immunohistochemical features
of the spleen in persistent polyclonal B-cell lymphocytosis closely
mimic splenic B-cell lymphoma. Diagn Pathol. 2012;7(1):1-6. SUPPORTING INF ORMATION
27. Bhagwandin SB, Weisenberg ES, Ozer H, Maker AV. Symptomatic Additional supporting information may be found online in the
massive splenomegaly in persistent polyclonal B-cell lymphocytosis
Supporting Information section at the end of this article.
requiring splenectomy. Open J Clin Med Case Rep. 2015;1(3):1-10.
28. Cornet E, Mossafa H, Courel K, Lesesve JF, Troussard X. Persistent
polyclonal binucleated B-cell lymphocytosis and MECOM gene ampli-
fication hematology. BMC Res Notes. 2016;9(1):3-13. How to cite this article: Rout ED, Moore AR, Burnett RC, et al.
29. Cornet E, Lesesve JF, Mossafa H, et al. Long-term follow-up of Polyclonal B-cell lymphocytosis in English bulldogs. J Vet
111 patients with persistent polyclonal B-cell lymphocytosis with Intern Med. 2020;1–14. https://doi.org/10.1111/jvim.15913
binucleated lymphocytes. Leukemia. 2009;23(2):419-422.