T-cell antigen expression can be observed in B-cell non-Hodgkin lymphoma (B-NHL). Although CD5 is expressed in B-cell chronic lymphocytic leukemia (B-CLL) and mantle cell lymphoma, the presence of other T-cell antigens is less common. This article reports a retrospective multicenter analysis in which flow cytometry was used to evaluate aberrant CD8 expression on the pathologic B cells of 951 bone marrow samples from patients with various types of B-NHL.
In a total of 18 patients, CD8 was coexpressed: 10 had B-CLL; 1, small lymphocytic lymphoma (SLL); 1, marginal zone lymphoma; 1, lymphoplasmacytic lymphoma; 2, diffuse large B-cell lymphoma; and 3, follicular lymphoma. There was a 1.89% overall frequency of CD8 coexpression in which B-CLL/SLL had a higher frequency (3.03%) than did the other B-cell neoplasms (1.18%). Most cases were characterized by a favorable outcome.
T-cell antigen expression in B-cell non-Hodgkin lymphoma (B-NHL) on neoplastic B cells is an interesting biologic feature. CD5 is normally expressed by B-cell chronic lymphocytic leukemia (B-CLL) and mantle cell lymphoma (MCL) and is considered a diagnostic feature of such diseases.1 However, the CD5 antigen may also be expressed in other types of B-NHL, such as follicular lymphoma (FL)2 and diffuse large B-cell lymphoma (DLBCL).3 The presence of other T-cell antigens seems to be less common, although, to date, few data are available and only 2 large studies have been published. Inaba et al4 reported 101 cases of B-NHL, classified by the Working Formulation system, with nodal and extranodal localizations, in which the authors observed that the neoplastic B cells coexpressed CD2 in 3 cases and CD7 in another 7. Their study analyzed only 4 bone marrow specimens, and no CD8 expression was detected. Kampalath et al5 studied 117 cases of B-CLL and small lymphocytic lymphoma (SLL), reporting 2 cases that coexpressed both CD2 and CD7 on the neoplastic B cells.
The presence of CD8 seems to be a peculiar feature of some B-CLL cases, and approximately 40 cases have been reported to date.6–13 Little information is available as to the frequency of CD8 on B cells in the other NHLs. Indeed, to the best of our knowledge, only 3 cases of CD8+ MCL have been reported.14,15
This article reports a very large series of B-NHL cases with bone marrow infiltration detected at diagnosis. A monoclonal antibody (MoAb) panel, which included T-cell and NK-cell antigens, was used to detect eventual CD8 coexpression on neoplastic B cells.
Materials and Methods
A total of 951 clinical files of patients with various types of B-NHL with bone marrow infiltration who were referred to 1 of our 4 institutions from 2004 to 2007 were reviewed. All cases had been examined at diagnosis by a simultaneous trephine biopsy and flow cytometry.16,17 All had polymerase chain reaction assays to detect IgH clonal rearrangement, and bcl1/JH and bcl2/JH translocations were studied,18,19 along with conventional cytogenetics. The cases were classified and stratified according to the Revised European-American classification of lymphoid neoplasms,1 as shown in Table 1.
Bone marrow samples from myeloaspirates (<1 mL) were collected into EDTA tubes. A direct 3- or 4-color method was applied, using a standard panel of monoclonal antibodies. Samples (50–100 μL) were placed into tubes and labeled with various combinations of the following MoAbs: CD45, CD3, CD4, CD8, CD5, CD10, CD19, CD20, CD22, CD23, CD11c, CD25, CD103, FMC7, CD16–56, CD38, and Zap-70. Clonality was determined by κ/λ analysis.
In the presence of an abnormal CD8 percentage, clearly exceeding the percentage of both CD3+ and NK cells, CD8/CD19/CD45, CD19/CD8/CD45, and CD5/CD8/CD19 (fluorescein isothiocyanate/phycoerythrin/peridinin chlorophyll protein, respectively) combinations were applied using different MoAb clones. Samples were evaluated by using a FACScan, FACSCalibur, or FACSCanto cytometer, using CellQuest or FACSDiva software (BD Biosciences, San Jose, CA). At least 30,000 events were acquired, and electronic gates were set on CD45/side scatter dot plots to identify lymphocytes. The electronic gate was set on the CD19+ cell population in the CD5/CD8/CD19 combination to evaluate both CD8 and CD5 coexpression.
A total of 18 of 951 cases reviewed coexpressed the CD8 molecule on pathologic B cells: 11 cases were B-CLL/SLL, 1 was marginal zone lymphoma, 1 was lymphoplasmacytic lymphoma, 2 were DLBCL, and 3 were FL Table 2, Table 3, and Image 1. A histologic diagnosis was made for all NHL cases by examination of a surgical biopsy specimen or bone marrow evaluation.
All B cells examined showed a homogeneous CD8 expression pattern, with the exception of case 10, in which the CD8 molecule was expressed by only 50% of the CD5+/CD19+ B cells at diagnosis but disappeared during a therapy-induced complete remission phase. However, it was observed again during a relapse (after first-line therapy) in which the whole lymphomatous cell population coexpressed the T-cell marker.
The frequency of CD8 coexpression was 3.03% in B-CLL/SLL and 1.18% in the other B-NHLs, with a cumulative frequency of 1.89%. The main biologic characteristic of the subjects with aberrant CD8 expression was that they were relatively young (ie, 36% of them were younger than 50 years).
There was an overall favorable outcome: complete follow-up was available in 16 of 18 cases, and 12 (75%) showed stable disease or a very good response to therapy. The same kind of response was observed in B-CLL and the other types of B-NHL.
This article reports an ample systematic investigation of the most relevant T-cell markers detected in the bone marrow samples of patients with B-CLL and other NHLs. The evaluation of T cell–associated markers is not used in consensus protocols for the flow cytometric immunophenotyping of NHL,20 nor is it systematically analyzed. All laboratories that took part in this study carried out a routine MoAb panel, including CD3, CD4, CD5, and CD8 as reference T-cell markers and CD16/CD56 as the reference NK-cell marker (at diagnosis and during follow-up). The lymphocytes were gated by the CD45high/side scatterlow gating strategy. The whole lymphocyte population was used to evaluate the percentage of T-cell antigen-binding lymphocytes, thus facilitating the detection of abnormal antigen expression.
Our data indicate that not only does CD8 coexpression on pathologic B lymphocytes in B-NHL seem to be an infrequent finding but also that it is more frequently observed in B-CLL than in the other NHLs. Literature reports an estimated frequency of only 0.5% for the aberrant expression of the CD8 molecule in about 40 previous B-CLL cases.12 The data from the retrospective study reported herein, carried out in 4 institutions covering a large patient series, reveal a higher frequency of CD8 coexpression in B-CLL, ie, around 3%.
Data as to the aberrant expression of CD8 in the other B-NHLs are scarce, and only 3 cases of MCL have been reported in literature.14,15 CD8 expression in the other NHL types (3 FLs, 2 DLBCLs, and 1 lymphoplasmacytic lymphoma) is also documented in our study and, therefore, provides novel information on this biologic feature, reporting a 1.89% frequency.
The significance of CD8 coexpression in B-CLL is controversial and has been associated with a heterogeneous clinical course. Although some authors6,11,13 are of the opinion that CD8 coexpression is not a marker of aggressive disease, other studies report that the same aberrant expression is characterized by aggressive behavior.8,12 Thus, the paucity of the cases so far described accounts for the substantial lack of information as to the clinical significance of such an aberrant phenotype. It is interesting that the 3 cases of CD8+ MCL reported14,15 were characterized by aggressive clinical behavior and involvement of the central nervous system.
Although Inaba et al4 did not observe any CD8+ case in their NHL patient series, the aberrant expression of CD2 or CD7 was associated with extranodal involvement at diagnosis. However, these results were not confirmed by Kaleem et al,21 who reported no unusually aggressive behavior in their series of 9 patients with NHL with aberrant expression of T-cell markers. These findings were in line with data reported by Jani et al22 for a patient with rare coexpression of both CD4 and CD7, for whom an indolent clinical course was documented.
Our data show that most cases (75%) of B-CLL/SLL and other B-NHLs with aberrant CD8 coexpression are characterized by a favorable outcome, with stable disease or a very good response to therapy. However, because of the limited number of patients and the dispersion of NHL subtypes, it is not possible to perform statistical analysis or to draw clinical conclusions.
Although the nature of T-cell antigens in NHL is still a question of debate, the transformation of a stem or progenitor cell or the neoplastic expansion of a normal B cell with T-cell antigen expression may be taken into consideration. Indeed, T and B cells stem from a common progenitor,23 and a small clonal CD5+/CD19+ population can even be found in healthy subjects, without any tendency toward malignant transformation.24 Noteworthy is the report by Caligaris-Cappio and Ghia,25 who, in a recent review article, discussed the origin of the B-CLL cell, arguing that the B cells from B-CLL have a phenotypic T-cell attitude. Moreover, they demonstrated that this property is documented by the fact that not only are B-CLL cells able to express some T-cell markers, but that they also make use of signal transduction mechanisms that normally operate in T cells.
Our article provides novel information as to the frequency of CD8 coexpression on pathologic B cells, not only in B-CLL but also in the other types of B-NHL. We propose that systematic flow cytometric analysis, including the investigation of several T cell–associated markers, be carried out in all laboratories involved in B-NHL diagnosis so as to enhance the understanding of the heterogeneous phenotype of B-cell lymphoid neoplasms. We are also of the opinion that the coexpression of 1 or more T-cell markers on pathologic B cells may, in selected patients, be used to enhance minimal residual disease detection by multiparametric flow cytometry. Further studies are ongoing toward this aim.
Upon completion of this activity you will be able to:
list types of B-cell lymphoma in which aberrant expression of T-cell markers may occur.
list T-cell markers that should be included in flow cytometric analysis of B-cell lymphomas.
discuss the immunophenotyping of B-cell lymphomas.
The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit™ per article. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.
The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.
Questions appear on p 306. Exam is located at www.ascp.org/ajcpcme.