Received: 18 December 2019 Revised: 31 March 2020 Accepted: 7 April 2020

DOI: 10.1002/cyto.b.21881

ORIGINAL ARTICLE

Flow cytometric evaluation of TRBC1 expression in tissue

specimens and body fluids is a novel and specific method
for assessment of T-cell clonality and diagnosis of T-cell
neoplasms

Holly Berg | Gregory E. Otteson | Heidi Corley | Min Shi | Pedro Horna |
Dragan Jevremovic | Horatiu Olteanu

Division of Hematopathology, Department of

Pathology and Laboratory Medicine, Mayo Abstract
Clinic, Rochester, Minnesota Background: Flow cytometric detection of T-cell clonality is challenging. The current
Correspondence available methodology for T-cell receptor (TCR) Vβ repertoire evaluation is a complex
Horatiu Olteanu, Division of assay and has limited sensitivity especially for detecting low levels of disease. There-
Hematopathology, Mayo Clinic, 200 First
Street SW, Rochester, MN 55905. fore, there is an unmet need for a reliable, simple, and rapid assay to identify T-cell
Email: [email protected] clonality. The rearrangement of the TCRB gene involves the random and mutually
exclusive expression of one of two constant β chain genes (TRBC1 and TRBC2), anal-
ogous to the kappa and lambda gene utilization by B cells.
Methods: Here, we used a single TRBC1 antibody, in conjunction with other T-cell associ-
ated markers, to detect T-cell clonality in tissue biopsies and body fluids. A total of 143 tis-
sue/body fluid specimens from 46 patients with a definitive diagnosis of a T-cell neoplasm
and 97 patients with no T-cell malignancy were analyzed with a cocktail of monoclonal
antibodies including CD2/CD3/CD4/CD5/CD7/CD8/CD45/TCRγδ/TRBC1.
Results: We examined TRBC1 expression on neoplastic T-cell populations identified
based on their immunophenotypic aberrancies, and monotypic TRBC1 expression
was identified in all 46 known T-cell lymphoma cases. We applied a similar gating
strategy to the 97 cases without T-cell neoplasms, and arbitrarily dissected T-cell
populations into immunophenotypically distinct subsets; in this group, we found that
all cases revealed an expected polytypic TRBC1 expression in all subsets.
Conclusions: Single TRBC1 antibody detection of T-cell clonality by flow cytometry
is a simple, rapid, and robust assay that could be routinely utilized in flow cytometric
laboratories.

KEYWORDS

flow cytometry, T-cell clonality, T-cell neoplasms, TRBC1

1 | I N T RO DU CT I O N seen in reactive lymphoproliferations. There are several, distinct enti-

ties of T-cell neoplasms in the current 2016 WHO classification, with
T-cell neoplasms are heterogeneous tumors characterized by a broad variable clinical behavior, ranging from relatively indolent to highly
spectrum of histologic findings, which at times overlap with those aggressive malignancies (Mature & NK-cell neoplasms, 2017). The

Cytometry. 2021;100:361–369. wileyonlinelibrary.com/journal/cytob © 2020 International Clinical Cytometry Society 361

 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
362 BERG ET AL.

primary diagnostic site may be tissue-based, including lymph nodes, application also included almost exclusively peripheral blood samples
skin, or extramedullary sites, for entities such as peripheral T-cell lym- (Novikov et al., 2019), we hereby applied this strategy to a diagnostic
phoma, not otherwise specified (PTCL, NOS), mycosis fungoides/ flow cytometry panel in tissue and body fluid specimens, resulting in
Sezary syndrome (MF/SS), extranodal NK/T-cell lymphoma, nasal type rapid, accurate, and unequivocal identification of clonal T-cell
(ENKTCL), angioimmunoblastic T-cell lymphoma (AITL), T-cell lympho- populations.
blastic lymphoma (T-ALL), or anaplastic large cell lymphoma (ALCL).
Alternatively, other processes are typically found in the peripheral
blood or bone marrow (e.g., T-cell prolymphocytic leukemia [T-PLL] or 2 | P A T I E N T S A ND M E T H O D S
T-cell large granular lymphocytic leukemia). Finally, body fluids may be
the primary or (more often) the site of secondary involvement by 2.1 | Patients
some of these malignancies.
Depending on the type and site of presentation, demonstration of Tissue biopsies (lymph node, tonsil, spleen, and other types) and body
the neoplastic nature of a T-cell infiltrate has traditionally relied on a com- fluid specimens (cerebrospinal [CSF], pleural, and peritoneal fluid)
bination of histologic criteria and possibly supplemented by immunohis- received for flow cytometric analysis at Mayo Clinic in Rochester, MN,
tochemistry, genetic, and/or molecular analysis. Furthermore, there are between November 2018 and November 2019, were included in a vali-
multiple studies that have documented the utility of flow cytometric dation study aimed at incorporating TRBC1 staining into our diagnostic
immunophenotyping as a highly sensitive and specific technique in the T-cell antibody panel. All cases were reviewed, and the diagnosis was
diagnosis of T-cell neoplasms (Chen, Kesler, Karandikar, McKenna, & confirmed by one of the authors (H.O.), to identify patients with either a
Kroft, 2006; Jamal et al., 2001; Jevremovic & Olteanu, 2019; Reichard & definitive diagnosis of a T-cell neoplasm involving the specimen analyzed
Kroft, 2013). In that respect, the diagnosis of lymphoproliferative or no clinical or laboratory evidence of a current or prior T-cell malig-
disorders by flow cytometric immunophenotyping usually relies either nancy. Hematologic malignancy diagnoses were tabulated according to
on demonstration of clonality (main approach utilized for B-cell the 2016 WHO Classification of Tumors of Hematopoietic and Lym-
lymphoproliferations) or on identification of immunophenotypic aber- phoid Tissues. Clinical and laboratory information was collected from the
rancies (primarily employed for the diagnosis of T-cell malignancies). In electronic medical record, and included age at diagnosis, gender, tissue
mature B-cell lymphoproliferative disorders, demonstration of clonality is biopsy or body fluid type, anatomic site, cytogenetic/FISH, and molecu-
relatively straightforward and takes advantage of differential (restricted lar (TCR gene rearrangement) testing results. This study was approved
or absent) surface immunoglobulin expression on the neoplastic cells. by the Mayo Clinic Institutional Review Board.
Unlike this readily available approach for B-cell neoplasms, T-cell
lymphoproliferations traditionally have required the deployment of less
commonly utilized assays, such as killer immunoglobulin-like receptor and 2.2 | Flow cytometry
Vβ T-cell receptor (TCR) repertoire analysis (Morice et al., 2004; Morice
et al., 2006; Morice, Kurtin, Leibson, Tefferi, & Hanson, 2003; Olteanu, Specimen processing, staining, and acquisition were performed as previ-
Karandikar, Eshoa, & Kroft, 2010; Wu, Anderson, Othus, & Wood, 2016). ously described (Shi et al., 2019). Briefly, body fluid samples were received
However, these techniques have some limitations in being labor-inten- and kept at room temperature for up to 72 hours after collection. Lymph
sive, relatively expensive and requiring interpretive expertise that is not node and tissue biopsies were received in RPMI media, grinded through a
available in all clinical laboratories. As a consequence, the primary wire mesh to produce a 1-mL cell suspension in bovine serum albumin and
approach to the flow cytometric diagnosis of T-cell neoplasms relies on kept at 4 C for up to 96 hours after collection. Aliquots of 100 μL were
the identification of deviations from normal immunophenotypic antigen then incubated in the dark for 15 minutes at room temperature with a
patterns. In addition, analysis of TCR gene rearrangements by PCR tech- cocktail of monoclonal antibodies including CD2 PerCP-Cy5.5/CD3 PE-
nique, and/or next-generation sequencing of rearranged TCRs, consti- Cy7/CD4 Alexa Fluor 700/CD5 APC/CD7 BV421/CD8 APC-H7/CD45
tutes molecular methods that may be used for assessment of T-cell V500/TCRγδ PE/TRBC1 FITC. With the exception of the anti-TRBC1
clonality. They do not represent an ideal alternative to flow cytometry, antibody (clone JOVI-1, Ancell Corporation, Bayport, Minnesota), all
either; in addition to yielding potentially false-positive results in physio- other fluorochrome-conjugated antibodies were from BD Biosciences,
logic states, such as senescence or inflammatory conditions, they are also San Jose, California. Stained cells were resuspended in 500 μL of phos-
associated with higher costs, operational complexity, and limited availabil- phate buffer saline, and 150,000 events were acquired on an FACSLyric
ity (Bruggemann et al., 2007; Langerak et al., 2012; Schumacher, flow cytometer (BD Biosciences), equipped with FACSDiva software
Duncavage, Mosbruger, Szankasi, & Kelley, 2014; Wang & Raffeld, 2019). (BD Biosciences) for data acquisition.
We recently reported a flow cytometric strategy to assess T-cell
clonality using a single antibody (JOVI-1) against one of two mutually
exclusive T-cell receptor β constant (TRBC) regions randomly selected 2.3 | Data analysis
during TCR gene rearrangement (Maciocia et al., 2017; Shi et al., 2019).
Since our initial experience was primarily derived from peripheral blood List mode files were analyzed on Kaluza version 2.1 (Beckman Coul-
and bone marrow specimens, and the only other report of this ter, Brea, California) as previously described (Shi et al., 2019) with
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
BERG ET AL. 363

some modifications. In brief, sequential gating and real-time color cod- cell malignancy were identified. Tissue biopsies comprised lymph
ing was used to identify each major T and NK-cell population, based nodes (n = 99), spleen (n = 11), lung/mediastinum (n = 11), tonsil
on the most informative and biologically relevant parameters (exclud- (n = 2), gastrointestinal/liver (n = 3), nasopharynx (n = 2), and soft tis-
ing TRBC1). Neoplastic cells were identified as immunophenotypically sue (n = 2). The anatomic sites of the lymph node biopsies were cervi-
abnormal T-cell subsets with discretely homogenous fluorescence cal (n = 37), abdominal (n = 22), axillary (n = 21), inguinal (n = 14), and
properties. In patients without a T-cell malignancy, CD4-positive, thoracic (n = 5). Body fluids consisted of pleural fluid (n = 7), peritoneal
TCRγδ-negative/CD8-positive, CD4/CD8-double positive, fluid (n = 4), and CSF (n = 2). All specimens were obtained from
CD4/CD8-double negative T-cell subsets were identified as 69 male and 74 female patients, with a median age of 54 years at
immunophenotypically distinct main subpopulations. In addition, diagnosis (range, 18–92 years). T-cell entities consisted mostly of
TRBC1 expression was evaluated on CD4-positive/CD5-negative, PTCL, NOS (n = 15); AITL (n = 11), MF/SS (n = 7), PTCL, T-follicular
CD4-positive/CD7-negative, CD8-positive/CD5-negative, and helper type (PTCL-TFH; n = 4), T-PLL (n = 3), and T-lymphoblastic leu-
CD8-positive/CD7-negative minor subsets, for the same cohort of kemia/lymphoma (T-ALL, n = 3). Of the other diagnoses, only one
patients without a diagnosis of T-cell neoplasm. TRBC1 expression each of ALCL, ENKTCL, and HSTCL, respectively, was represented. In
was studied on a log scale histogram, and a fluorescence channel the control cohort of patients without at T-cell neoplasm, diagnoses
intermediate between discrete TRBC1-positive and TRBC1-negative included lymph nodes with a combination of benign, follicular and/or
normal T cells was visually selected as a threshold to estimate the per- paracortical lymphoid hyperplasia (n = 57), granulomatous inflamma-
centage of TRBC1-positive events. In a subset of cases with T-cell tion (n = 2), hyaline vascular Castleman disease (n = 1), B-cell lym-
malignancy, TRBC1 expression was interpreted as dim-positive, if uni- phoma, including follicular lymphoma, diffuse large B-cell lymphoma,
formly present in the aberrant cluster, and 50% of the population and MALT lymphoma (n = 14), thymoma/thymic hyperplasia (n = 9),
exceeded the visual cutoff between TRBC1-negative and benign spleen (n = 4), and body fluids with no malignant cells observed
TRBC1-positive normal T cells. Only distinct clusters of T-cell subsets (n = 10) (Table 1). One patient had a composite lymphoma (DLBCL
comprised of 150 or more events were studied. Restricted (mono- and PTCL-TFH) diagnosed in the lymph node biopsy.
typic) TRBC1 expression was arbitrarily defined when present on
>85% or <15% of a distinct T-cell cluster, based on our previous
study. All cases were analyzed retrospectively by two of the authors 3.2 | TRBC1 expression in nonneoplastic T-cell
(G.O. and H.O.) in a blinded fashion, to ensure consistency. populations

First, we examined TRBC1 expression in the 97 cases without T-cell

2.4 | TCR gene rearrangement studies neoplasms and arbitrarily dissected T-cell populations into
immunophenotypically distinct subsets. Of these, we found that all
Total cellular DNA was extracted and PCR amplification performed in cases revealed an expected polytypic TRBC1 expression in all sub-
five multiplex PCR tubes with ASR Biomed-2 primers (Invivoscribe sets (Table S1 and Figure 1). The range of TRBC1 expression in the
Technologies, San Diego, California) targeting TCR Vβ, Dβ, Jβ, Vγ, and different compartments followed a Gaussian distribution, as con-
Jγ regions. The products were separated and detected by capillary gel firmed by D'Agostino and Pearson omnibus normality testing. Posi-
electrophoresis on the ABI Prism 3130xl genetic analyzer (Applied tive and negative subsets were easily distinguishable and
Biosystems, Warrington, UK). reproducibly identified by using a uniform analytic approach, as
described in Section 2.
The median values were very similar for CD4-positive and
2.5 | Statistical analysis CD8-positive T cells, with a slight predominance for CD4-positive
helper T cells: 43.8% vs 37.9%, respectively. Further analysis for the
Statistical analyses were carried out in GraphPad Prism, version 6.0 CD4-positive compartment generated a 95th percentile of
(GraphPad Software, San Diego, California). Mean percentages of 35.8–51.1%. A similar analysis for the CD8-positive group yielded a
TRBC1-positive events between groups were studied on a two-tailed 95th percentile of 36.5–50.8%. The CD4/CD8-double positive and
t test, and the variances were compared using an F test. A statistical CD4/CD8-double negative subsets had comparable median expres-
significant p value was considered as less than .05. sion values and reference ranges. Less common (minor) subsets
included CD4-positive/CD5-negative, CD4-positive/CD7-negative,
CD8-positive/CD5-negative, and CD8-positive/CD7-negative
3 | RESULTS populations, and all showed a similar pattern of polytypic TRBC1
expression. For the CD4/CD8-double negative T-cells, the combina-
3.1 | Patient characteristics tion of T-cell antigens, including TCRγ/δ antibodies in the same tube,
allows for separation of TCRα/β and TCRγ/δ subsets (as they both
A total of 143 tissue and body fluid specimens from 46 patients with tend to show a similar, double negative expression pattern), and selec-
a definitive diagnosis of a T-cell neoplasm and 97 patients with no T- tive TRBC1 enumeration on the α/β-expressing T cells only.
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
364 BERG ET AL.

TABLE 1 Diagnoses in tissue and body fluid specimens

Lymph node and Lung/mediastinum Nasopharynx and Body fluids

Diagnosis tonsil (n = 101) Spleen (n = 11) (n = 11) GI/liver (n = 3) soft tissue (n = 4) (n = 13)
T-cell lymphoma 34 4 11 2 2 3
AITL 11 0 0 0 0 0
PTCL, NOS 8 3 1 2 0 1
PTCL-TFH 4 0 0 0 0 0
ALCL 0 0 0 0 1
ENKTCL 0 0 0 0 1 0
HSTCL 0 1 0 0 0 0
MF/SS 7 0 0 0 0 0
T-ALL 2 0 0 0 0 1
T-PLL 2 0 0 0 0 1
B-cell lymphoma 9 3 0 1 1 0
DLBCL 6 3 0 1 1 0
MALT 2 0 0 0 0 0
FL 1 0 0 0 0 0
No malignant cells 10
FH and/or PCH 55 4 1 0 1 0
Thymoma/thymic hyperplasia 0 0 9 0 0 0
Granulomatous inflammation 2 0 0 0 0 0
Castleman disease 1 0 0 0 0 0

Abbreviations: AITL, angioimmunoblastic T-cell lymphoma; ALCL, anaplastic large cell lymphoma; DLBCL, diffuse large B-cell lymphoma; ENKTCL,
extranodal NK/T-cell lymphoma, nasal type; FH, follicular lymphoid hyperplasia; FL, follicular lymphoma; HSTCL, hepatosplenic T-cell lymphoma; MALT,
extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue; MF/SS, mycosis fungoides / Sezary syndrome; PCH, paracortical lymphoid
hyperplasia; PTCL, NOS, peripheral T-cell lymphoma, not otherwise specified; PTCL-TFH, peripheral T-cell lymphoma with T follicular helper
immunophenotype; T-ALL, T lymphoblastic leukemia/lymphoma; T-PLL, T-cell prolymphocytic leukemia.

We then compared the distribution of TRBC1 expression in dif- (92–100%, median 94%), or CD4/CD8-double negative neoplastic
ferent T-cell subsets across several groups of specimens, including T-cells (96–100%, median 98%). The only case of CD4/CD8-double
lymph nodes, spleen, and body fluids (Table S1). Both major and minor positive T-cell lymphoma showed similar level of TRBC1 expression
subsets maintained a similar fraction of TRBC1-positive populations, (97%) as the other lymphomatous cases. In comparison, T-cell lympho-
both for mean/median values and 99th percentile CI. mas lacking TRBC1 expression on the neoplastic cells showed only a
minor subset of TRBC1-positive events within distinctly aberrant T-cell
populations, ranging from 0 to 5.4% (median: 1.1%; mean: 1.43%). Simi-
3.3 | TRBC1 expression in neoplastic T-cell lar to positive cases, there was no difference in the level of TRBC1 posi-
populations tivity in those clusters, regardless of the CD4 and/or CD8 expression
status (Table 2).
We then applied a similar gating strategy to cases with neoplastic T-cell Within the cohort of T-cell neoplasms, restricted TRBC1 expression
populations identified based on their immunophenotypic aberrancies. was found in 34 lymph node biopsies, 2 liver/GI biopsies, 1 mediastinal
Restricted (monotypic) TRBC1 expression was identified in all 46 mass, 1 soft tissue mass, 1 nasopharyngeal tumor, 4 splenectomy speci-
known T-cell lymphoma cases. TRBC1 expression was positive in mens, and 3 body fluids, respectively. TRBC1-positive tumors included
23/46 (50.0%) of cases. Of these, the majority were CD4-positive neo- AITL (4/11, 36.4%), ENKTCL (1/1, 100%), MF/SS (2/7, 28.6%), PTCL,
plasms (15/23; 65.2%), with fewer CD8-positive (4/23; 17.4%), NOS (11/15, 73.3%), PTCL-TFH (4/4, 100%), and T-PLL (1/3, 33.3%). Of
CD4/CD8-double positive (1/23; 4.3%), or CD4/CD8-double negative the remaining diagnoses, all cases of T-ALL (3/3, 100%), ALCL (1/1,
(3/23; 13.0%) tumors. The proportion of TRBC1-positive events within 100%), and HSTCL (1/1, 100%) were TRBC1-negative. All cases of T-ALL
distinct clusters of aberrant T cells ranged from 92 to 100%, with a were surface CD3-positive, CD4/CD8-double positive (n = 2), or
median value of 98% (mean: 97%; range: 92 to 100%). There was no CD4/CD8-double negative (n = 1), TCRα/β-positive. The ALCL case was
difference in the level of TRBC1 expression in CD4-positive T-cell surface CD3-positive, CD4-positive, TCRα/β-positive, while the single
populations (92–100%, median 99%), CD8-positive T-cell clusters case of HSTCL showed a surface CD3-positive, CD4/CD8-double
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
BERG ET AL. 365

F I G U R E 1 TRBC1 expression in a normal lymph node with reactive follicular and paracortical lymphoid hyperplasia. CD4-positive (cyan),
CD8-positive (blue), CD4/CD8-double positive, and CD4/CD8-double negative (both pink) T-cell subsets are shown. Also present are γδ T cells
(black) and B/NK cells (orange) [Color figure can be viewed at wileyonlinelibrary.com]

TABLE 2 Percent TRBC1 expression in neoplastic T-cell

expression in TRBC1-positive T-cell neoplasms was at similar levels in
populations, based on CD4 and/or CD8 immunophenotype
10/23 (43.5%) cases (Figures 2 and 3). One patient with splenic
CD4(+)/CD8(−) TRBC1(+) (n = 15) TRBC1(−) (n = 17) CD4/CD8-dual negative, TCRα/β-positive PTLC, NOS demonstrated
Mean 96.87 1.76 bright surface CD3/TRBC1 coexpression. The remainder (12/23,
Median 98.81 1.29 52.2%) showed a uniform expression that was intermediate between
Range 91.61–99.91 0–5.35 TRBC1-positive and TRBC1-negative normal T-cell subsets and was
CD8(+)/CD4(−) TRBC1(+) (n = 4) TRBC1(−) (n = 0) interpreted as dim-positive (Figure 3). The diagnoses for the 12 cases
Mean 95.22 — with dim TRBC1 expression included PTCL-THF or NOS (n = 8), AITL
Median 93.52 — (n = 3), and ENKTCL (n = 1). With the exception of two specimens

Range 92.20–99.94 — (one pleural fluid and one spleen) involved by PTCL, NOS with dim
TRBC1, but normal CD3 expression, the remaining 10 tissue biopsies
CD4(+)/CD8(+) TRBC1(+) (n = 1) TRBC1(−) (n = 2)
had concordant dim CD3 and dim TRBC1 expression.
Mean 97.25 0.18
Median — 1.18
Range — 0.14–2.21
3.4 | Comparison of TRBC1 expression and
CD4(−)/CD8(−) TRBC1(+) (n = 3) TRBC1(−) (n = 4)
molecular/genetic analysis in normal and neoplastic
Mean 97.75 0.35 T-cell populations
Median 97.98 0.18
Range 95.58–99.69 0–1.04 TCR gene rearrangement (TCR-PCR) studies were available in 38/46
(82.6%) of tissues or body fluids diagnosed with at T-cell neoplasm,
and in 8/97 (8.3%) of specimens without a T-cell malignancy. There
negative, TCRγ/δ-positive immunophenotype. The four TRBC1-negative was 100% concordance between a clonal TCR-PCR result and
PTCL, NOS lymph nodes showed surface CD3 and TCRα/β expression restricted TRBC1 expression in cases with a neoplastic T-cell clone.
and were either CD4-positive (n = 2) or CD4/CD8-double negative None of the eight specimens without T-cell malignancy showed a pos-
(n = 2), respectively. itive TCR-PCR. In addition, all eight samples without T-cell malignancy
Compared to normal, CD4-positive helper T cells, or and with a negative TCR-PCR also showed polytypic TRBC1 expres-
CD8-positive cytotoxic T cells, the fluorescence intensity of antigen sion by flow cytometry.
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
366 BERG ET AL.

F I G U R E 2 TRBC1 expression in a cervical lymph node involved by angioimmunoblastic T-cell lymphoma. The neoplastic T cells (red) are CD2
(+), surface CD3 (+), CD4 (+), CD5 (+), CD7 (−), CD8 (−), CD10 (−), and TRBC1 (−). Also shown are CD4-positive (cyan), CD8-positive (blue),
CD4/CD8-double positive, and CD4/CD8-double negative (both pink) normal T-cell subsets, γδ T cells (black), and B/NK cells (orange) [Color
figure can be viewed at wileyonlinelibrary.com]

F I G U R E 3 TRBC1 expression in an axillary lymph node involved by peripheral T-cell lymphoma with T follicular helper (TFH)
immunophenotype. The neoplastic T cells (red) are CD2 (−), surface CD3 (dim +), CD4 (dim +), CD5 (+), CD7 (minor subset +), CD8 (−), TRBC1
(dim +). Also shown are CD4-positive (cyan), CD8-positive (blue), CD4/CD8-double positive, and CD4/CD8-double negative (both pink) normal T-
cell subsets, γδ T cells (black), and B/NK cells (orange) [Color figure can be viewed at wileyonlinelibrary.com]
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
BERG ET AL. 367

4 | DISCUSSION expression, 2 specimens had dim TRBC1, but normal CD3 expression,
and the remaining 10 tissue biopsies had concordant dim CD3/dim
Previous studies have reported on the clinical utility of TRBC1 in T- TRBC1 expression. There was a single example of PTCL, NOS involv-
cell clonality assessment by flow cytometry in peripheral blood and ing the spleen, with bright CD3 and bright TRBC1 coexpresssion.
bone marrow specimens (Novikov et al., 2019; Shi et al., 2019). We Overall, 21/23 TRBC1-positive neoplasm showed concordant sCD3/
have expanded on those prior reports and demonstrated in this study TRBC1 expression intensity. These findings suggest a proportional
the value of TRBC1 immunophenotyping in the characterization of a relationship between surface CD3 and TRBC1 levels on neoplastic T
spectrum of T-cell malignancies in tissue biopsies and body fluids. All cells, at least in the majority of cases. The remaining neoplastic T-cell
46 cases with documented neoplastic T-cell populations showed cases in our cohort with absent TRBC1 expression are postulated to
restricted (monotypic) TRBC1 expression, a 100% sensitivity. We also express monotypic TRBC2. Currently, Thermo Fisher Scientific
established the expression patterns of TRBC1 in normal T-cell subsets (Waltham, Massachusetts) offers a TRBC2 antibody (Thermo Fisher
from 97 cases without a T-cell neoplasm and demonstrated that T-cell Catalog # PA5-71421). This is an unconjugated, polyclonal, rabbit-
malignancies show a narrow and distinctly restricted (positive or nega- generated antibody developed for Western blot, rather than flow
tive) expression range of TRBC1. All specimens were analyzed with a cytometry, and that recognizes the intracytoplasmic portion of the
routine diagnostic T-cell panel, including an integrated TRBC1 anti- TRBC2 epitope. As such, it is not readily applicable for routine flow
body, and support an easy-to-use, robust, and accurate approach that cytometry, without the use of a conjugated fluorochrome or anti-
may be easily implemented in the clinical workflow of a wide variety rabbit secondary antibody and also is not suitable for surface staining,
of laboratories. Further comparison with clonality assessment by as it would require a cell membrane permeabilization step, followed
molecular analysis (TCR-PCR) indicated that the flow cytometric by intracytoplasmic staining. Since our goal was to develop a robust,
TRBC1 assay has comparable sensitivity and specificity, while being rapid, and simple clinical flow cytometry assay, we did not evaluate
faster and less costly. Furthermore, TCR-PCR testing provides only a this antibody. We acknowledge that using TRBC1 in the absence of
global assessment of the submitted sample and a binary output (clonal TRBC2 is akin to assessing kappa in the absence of lambda light chain
vs non-clonal population present), whereas TRBC1 testing yields spe- expression in B cells. We and others are in the process of developing
cific clonality information on distinctly aberrant T-cell populations, as a working surface TRBC2 antibody for flow cytometric applications
defined by immunophenotypic analysis. TRBC1 assessment by flow (Onuoha et al., 2018).
cytometry may change the way other ancillary testing (such as TCR Given the high sensitivity and specificity of TRBC1 in detecting
PCR or Vbeta receptor analysis) is being used for T-cell clonality monotypic T-cell populations, it represents a potentially useful adjunct
assessment. We anticipate that the presence of a clonal TRBC1 flow in the diagnosis of T-cell neoplasms. While a diagnosis of T-cell lym-
cytometry result will obviate the need of performing a more costly phoma may be rendered in a sufficient number of cases based on his-
and time-consuming test, at least as CD4-positive aberrant T-cell tologic identification of an overtly malignant infiltrate alone, the utility
populations are concerned. of our proposed approach is particularly impactful in the subset of
When reviewing TRBC1 expression on various T-cell subsets tumors that show significant morphologic overlap with other, reactive,
from different tissue and body fluid specimens, we recorded a slightly or neoplastic processes. In those cases, inclusion of TRBC1 assess-
higher expression level in CD4-positive T cells, compared to ment by flow cytometry may allow for rapid confirmation of the neo-
CD8-positive T cells. A similar pattern has been reported in peripheral plastic cell lineage. We have encountered in our practice cases of
blood and bone marrow cases in two prior studies (Novikov borderline atypical lymph node infiltrates that required multiple biop-
et al., 2019; Shi et al., 2019). Furthermore, the CD4/CD8-dual nega- sies over the span of several months, in order to eventually substanti-
tive and CD8-positive/CD5-negative T-cell populations had a TRBC1 ate the possibility of T-cell lymphoma based on immunophenotypic
range that was lower and closer to CD8-positive T cells, compared to analysis that demonstrated restricted TRBC1 expression. Another
CD4/CD8-dual positive, CD4-positve/CD7-negative, and area where flow cytometric assessment of TRBC1 expression might
CD8-positive/CD7-negative subsets, which showed a slightly higher make substantial contributions is in the diagnosis of fine needle aspi-
TRBC1 concentration, and more similar to that of CD4-positive T rate specimens. Our cohort consisted primarily of excisional or
cells. Lymph node and body fluid T-cell subsets showed comparable incisional lymph node biopsies but also of fine needle aspirate and
TRBC1 levels in their respective T-cell subsets, while splenectomy core biopsy specimens. Since the latter category may include a higher
specimens had a slightly different pattern, particularly for the proportion of paucicellular cases, the increased sensitivity and speci-
CD8-positive/CD5-negative compartment. However, the number of ficity due to TRBC1 assessment would be useful in the evaluation and
normal cases for this specimen type was relative small (n = 5) and may classification of T-cell neoplasms in those circumstances. As
require further data accrual for confirmation. highlighted later in the discussion, surface TRBC1 expression can only
For tissues and body fluids involved by a neoplastic T-cell popula- be assessed in surface CD3-positive, TCR alpha/beta-positive T cells,
tion, we found monotypic, positive TRBC1 expression in half of the and this assessment is precluded in surface CD3-negative and/or TCR
cases. Of these, 52.2% showed dim TRBC1 positivity (compared to gamma/delta-positive T cells, as well as in B cells or NK cells.
their normal subset counterparts), and 83.3% of those cases had also Flow cytometric immunophenotyping has been shown to be
dim surface CD3 expression. Of the 12/23 neoplasm with dim TRBC1 potentially challenging when deployed for the diagnosis of T-cell
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
368 BERG ET AL.

neoplasms, owing to the complexity of the physiologic T-cell immune clusters for TRBC1 analysis. The corollary to this statement is that by
response. Nonneoplastic conditions are known to trigger expansions not following this approach, which is currently adopted by a number
of normal subsets, such as γ/δ T cells, CD4/CD8-double positive or of (but not all) laboratories, and using a more simplified analysis
double negative (α/β) subsets, and CD5 or CD7-negative T-cell instead, one may generate inaccurate TRBC1 results or staining pat-
populations. Furthermore, these (typically small) T-cell subsets may be terns due to noncontributory subsets included in the population of
physiologically more prominent in certain tissue sites, in addition to interest. Similar to that caveat, it is important to separate
pathologic stimulation (Aggarwal, Fischer, Swerdlow, & Craig, 2013; CD4/CD8-double negative populations into α/β and γ/δ-expressing
Andreu-Ballester et al., 2012; Karandikar et al., 2004; Neff, Howard, & subsets, in order to generate accurate TRBC1 expression results.
Morice, 2013; Roden, Morice, & Hanson, 2008). Following the same These points are important to emphasize because, as always, clonality
rationale, TRBC1 expression may also demonstrate highly skewed does not always equate neoplasia. This is the reason why cluster anal-
expression in what are likely reactive and/or oligoclonal T-cell ysis of flow cytometric data for identifying aberrant T-cell populations
populations, as has been shown in our previous study in peripheral with subsequent assessment of TRBC1 expression on those subsets is
blood for CD8-positive cytotoxic T cells (Shi et al., 2019). With these critical in preventing overcalling small, apparently (oligo)clonal and
considerations in mind, it was important to survey nonneoplastic T- TRBC1-restricted population as neoplastic. For example, oligoclonal
cell populations in a wide variety of specimens (both tissue and body CD8-positive cytotoxic T-cell population with apparent TRBC1
fluids) obtained from various anatomic sites, to provide reference restriction have been observed in peripheral blood of individuals with-
ranges for our patient cohort. We also need to point out that these out clinical, morphologic, or genetic evidence of T-cell lymphoma, and
are not “normal” TRBC1 expression ranges, as they were generated therefore require an interpretive qualifier (Shi et al., 2019). We did
from a series of patient specimens submitted for a spectrum of patho- not observe this type of populations in tissue biopsies, because of the
logic processes. Because our goal was to differentiate neoplastic current study design; however, there is no physiologic reason to
(monotypic) from nonneoplastic populations, rather than dis- assume that they cannot be encountered in lymph nodes or body
tinguishing normal from abnormal T-cell subsets, we have purposely fluids, as well.
chosen to include specimens involved by non-T-cell neoplasms, in In summary, we have documented the usefulness of TRBC1
addition to those with reactive changes, in order to cover the broad assessment in the diagnosis of T-cell neoplasia in tissue and body fluid
range of findings seen in nonneoplastic T-cell populations. Although specimens. We found this to be a highly sensitive assay, demonstrat-
the mean, median, and 95th percentile for TRBC1 expression were ing aberrant (restricted) TRBC1 expression in a large cohort of cases
distinctly separated and nonoverlapping between T-cell neoplasms covering the spectrum of T-cell malignancies. In addition, when inter-
and nonneoplastic cases, there were a limited number of preted in the context of detailed assessment of normal and reactive
CD4/CD8-double positive or double negative, TRBC1-positive or no T-cell compartments, as part of an analytic strategy that emphasizes
CD8-positive, TRBC1-negative examples of T-cell malignancies with accurate gating of immunophenotypically distinct T-cell subsets, it is
those particular immunophenotypes to draw more general conclu- also highly specific. To our knowledge, this is the first study to evalu-
sions regarding the level of TRBC1 expression as it relates to CD4 ate TRBC1 expression in a comprehensive series of tissue biopsies
and/or CD8 expression status. Because of that, we chose more con- and body fluid specimens. Our findings, combined with the few other
servative cutoffs (>85% or <15% TRBC1 expression) for defining studies evaluating TRBC1 assessment in peripheral blood and bone
clonality, similar to those employed in bone marrow and peripheral marrow cases, support the utility of single-antibody TRBC1 in the reli-
blood, that would be applicable across different subsets. able detection of T-cell neoplasms.
Another known limitation of this approach relates to TRBC1
assessment in cells that do not express either surface TRBC1 or CONFLIC T OF INT ER E ST
TRBC2 (Krangel, 2009). These include γ/δ T cells, benign thymocytes, The authors declare no potential conflict of interest.
and surface CD3-negative neoplasms (both mature and immature).
There may be several approaches that could alleviate these potential OR CID
shortcomings, particularly in the differential diagnosis of thymoma/ Horatiu Olteanu https://orcid.org/0000-0001-5783-4485
thymic hyperplasia and T-ALL. Development of TRBC2-specific anti-
bodies, cytoplasmic staining for TRBC1 and/or TRBC2, and RE FE RE NCE S
uncovering of TRBC1 expression patterns on maturing T cells that Aggarwal, N., Fischer, J., Swerdlow, S. H., & Craig, F. E. (2013). Splenic
lymphoid subsets with less well-recognized phenotypes mimic
may track closely other T-cell antigens are all likely to contribute to
aberrant antigen expression. American Journal of Clinical Pathology,
the refinement of TRBC1-based clonality assessment. In this context, 140, 787–794.
we also need to stress out the importance of a flexible and compre- Andreu-Ballester, J. C., Garcia-Ballesteros, C., Benet-Campos, C., Amigo, V.,
hensive analytic approach when assessing T-cell clonality by TRBC1 Almela-Quilis, A., Mayans, J., & Ballester, F. (2012). Values for alphabeta
and gammadelta T-lymphocytes and CD4+, CD8+, and CD56+ subsets
staining. Our gating strategy, predicated on the concept of cluster
in healthy adult subjects: Assessment by age and gender. Cytometry
analysis, is designed to identify immunophenotypically distinct T-cell
Part B, Clinical Cytometry, 82, 238–244.
subsets (Jamal et al., 2001; Jevremovic & Olteanu, 2019; Shi Bruggemann, M., White, H., Gaulard, P., Garcia-Sanz, R., Gameiro, P.,
et al., 2019). We would then subsequently hone in to those aberrant Oeschger, S., … Molina, T. J. (2007). Powerful strategy for polymerase
 15524957, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cyto.b.21881 by Ut Southwestern Medical Center, Wiley Online Library on [16/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
BERG ET AL. 369

chain reaction-based clonality assessment in T-cell malignancies report Novikov, N. D., Griffin, G. K., Dudley, G., Drew, M., Rojas -Rudilla, V.,
of the BIOMED-2 concerted action BHM4 CT98-3936. Leukemia, 21, Lindeman, N. I., & Dorfman, D. M. (2019). Utility of a simple and
215–221. robust flow cytometry assay for rapid Clonality testing in mature
Chen, W., Kesler, M. V., Karandikar, N. J., McKenna, R. W., & Kroft, S. H. peripheral T-cell lymphomas. American Journal of Clinical Pathology,
(2006). Flow cytometric features of angioimmunoblastic T-cell lym- 151, 494–503.
phoma. Cytometry Part B, Clinical Cytometry, 70, 142–148. Olteanu, H., Karandikar, N. J., Eshoa, C., & Kroft, S. H. (2010). Laboratory
Jamal, S., Picker, L. J., Aquino, D. B., McKenna, R. W., Dawson, D. B., & findings in CD4(+) large granular lymphocytoses. International Journal
Kroft, S. H. (2001). Immunophenotypic analysis of peripheral T-cell of Laboratory Hematology, 32, e9–e16.
neoplasms. A multiparameter flow cytometric approach. American Onuoha, S., Ferrari, M., Bulek, A., Bughda, R., Manzoor, S., Srivastava, S., …
Journal of Clinical Pathology, 116, 512–526. Pule, M. (2018). Structure guided engineering of highly specific chimeric
Jevremovic, D., & Olteanu, H. (2019). Flow cytometry applications in the antigen receptors for the treatment of T cell lymphomas. Blood, 132, 1661.
diagnosis of T/NK-cell lymphoproliferative disorders. Cytometry Reichard, K. K., & Kroft, S. H. (2013). In A. Orazi, K. Foucar, &
Part B, Clinical Cytometry, 96, 99–115. D. M. Knowles (Eds.), Flow cytometry in the assessment of hematologic
Karandikar, N. J., Kroft, S. H., Yegappan, S., Rogers, B. B., Aquino, V. M., disroders. neoplastic hematopathology (pp. 119–145). Baltimore, MD:
Lee, K. M., … McKenna, R. W. (2004). Unusual immunophenotype of Lippincott, Williams and Wilkins.
CD8+ T cells in familial hemophagocytic lymphohistiocytosis. Blood, Roden, A. C., Morice, W. G., & Hanson, C. A. (2008). Immunophenotypic
104, 2007–2009. attributes of benign peripheral blood gammadelta T cells and condi-
Krangel, M. S. (2009). Mechanics of T cell receptor gene rearrangement. tions associated with their increase. Archives of Pathology & Laboratory
Current Opinion in Immunology, 21, 133–139. Medicine, 132, 1774–1780.
Langerak, A. W., Groenen, P. J., Bruggemann, M., Beldjord, K., Bellan, C., Schumacher, J. A., Duncavage, E. J., Mosbruger, T. L., Szankasi, P. M., &
Bonello, L., … van Dongen, J. J. (2012). EuroClonality/BIOMED-2 Kelley, T. W. (2014). A comparison of deep sequencing of TCRG
guidelines for interpretation and reporting of Ig/TCR clonality testing rearrangements vs traditional capillary electrophoresis for assessment
in suspected lymphoproliferations. Leukemia, 26, 2159–2171. of clonality in T-cell lymphoproliferative disorders. American Journal of
Maciocia, P. M., Wawrzyniecka, P. A., Philip, B., Ricciardelli, I., Clinical Pathology, 141, 348–359.
Akarca, A. U., Onuoha, S. C., … Pule, M. A. (2017). Targeting the T cell Shi, M., Jevremovic, D., Otteson, G. E., Timm, M. M., Olteanu, H., &
receptor beta-chain constant region for immunotherapy of T cell Horna, P. (2019). Single antibody detection of T-cell receptor
malignancies. Nature Medicine, 23, 1416–1423. alphabeta clonality by flow cytometry rapidly identifies mature T-cell
Mature T- and NK-cell neoplasms. WHO Classification of tumours of neoplasms and monotypic small CD8-positive subsets of uncertain sig-
haematopoietic and lymphoid tissues. Edited by Swerdlow SH, nificance. Cytometry Part B, Clinical Cytometry, 98, 99–107.
Harris NL, Jaffe ES, Pilersi SA, Stein H, Thile J. Revised 4th edition ed. Wang, H. W., & Raffeld, M. (2019). Molecular assessment of clonality in
Lyon, France: IARC, 2017. pp. 345–422. lymphoid neoplasms. Seminars in Hematology, 56, 37–45.
Morice, W. G., Katzmann, J. A., Pittelkow, M. R., el-Azhary, R. A., Wu, D., Anderson, M. M., Othus, M., & Wood, B. L. (2016). Clinical experi-
Gibson, L. E., & Hanson, C. A. (2006). A comparison of morphologic ence with modified, single-tube T-cell receptor Vbeta flow cytometry
features, flow cytometry, TCR-Vbeta analysis, and TCR-PCR in qualita- analysis for T-cell Clonality. American Journal of Clinical Pathology, 145,
tive and quantitative assessment of peripheral blood involvement by 467–485.
Sezary syndrome. American Journal of Clinical Pathology, 125,
364–374.
Morice, W. G., Kimlinger, T., Katzmann, J. A., Lust, J. A., Heimgartner, P. J., SUPPORTING INF ORMATION
Halling, K. C., & Hanson, C. A. (2004). Flow cytometric assessment of Additional supporting information may be found online in the
TCR-Vbeta expression in the evaluation of peripheral blood involve- Supporting Information section at the end of this article.
ment by T-cell lymphoproliferative disorders: A comparison with con-
ventional T-cell immunophenotyping and molecular genetic
techniques. American Journal of Clinical Pathology, 121, 373–383.
How to cite this article: Berg H, Otteson GE, Corley H, et al.
Morice, W. G., Kurtin, P. J., Leibson, P. J., Tefferi, A., & Hanson, C. A.
(2003). Demonstration of aberrant T-cell and natural killer-cell antigen Flow cytometric evaluation of TRBC1 expression in tissue
expression in all cases of granular lymphocytic leukaemia. British Jour- specimens and body fluids is a novel and specific method for
nal of Haematology, 120, 1026–1036. assessment of T-cell clonality and diagnosis of T-cell
Neff, J. L., Howard, M. T., & Morice, W. G. (2013). Distinguishing T-cell
neoplasms. Cytometry. 2021;100:361–369. https://doi.org/10.
large granular lymphocytic leukemia from reactive conditions: Labora-
tory tools and challenges in their use. Surgical Pathology Clinics, 6,
1002/cyto.b.21881
631–639.