Clinicopathological and Immunohistochemical Profile of Mantle Cell Lymphoma: An Institutional Experience

Introduction Mantle cell lymphoma (MCL) is a biologically aggressive B-cell non-Hodgkin lymphoma (NHL) with distinctive morphologic, immunophenotypic, and molecular characteristics. Differentiation from other chronic lymphoproliferative disorders is essential for prognostication. Aim This paper aims to study the clinicopathological features of MCL with emphasis on immunohistochemical features and disease correlation. Method To do so, clinicopathological characteristics from 21 cases of MCL (14 males, seven females, M:F=2:1) diagnosed in the last five years i.e. 2015 to 2020, were retrospectively reviewed and correlated with immunohistochemistry (IHC) data. Particularly those pertaining to cyclin D1, SRY-box transcription factor 11 (SOX11), cluster of differentiation (CD) 5, CD23, MIB E3 ubiquitin protein ligase 1 (MIB1), tumor protein 53 (TP53), c-myelocytomatosis oncogene product (c-MYC), multiple myeloma oncogene 1 (MUM1), mouse double minute 2 homolog (MDM2), and Epstein-Barr virus latent membrane protein 1 (EBV-LMP1) expression with its aberrations. Observations This study shows that MCL constituted 4.2% (21/500) of all NHLs with a mean age of 57.5 years (median 60 years, range 30 to 80 years). The disease was nodal in 19, and extranodal in the remaining two cases. 14 of 21 (67%) had generalized lymphadenopathy and 71% had bone marrow (BM) involvement. The nodal involvement was diffuse in 9/17 (53%), 8/21 (38%) had a blastoid morphology, and an in-situ MCL pattern was not seen in any of the cases selected for the study. Cyclin D1 immunoexpression correlated well with SOX11; CD5-negative in five cases; and CD23-positive in three cases. TP53 and c-MYC expression were noted in 17/19 (89.4%) and 8/17 (47%), respectively. MUM1 registered positive in six cases. None of the cases showed immunopositivity for MDM2 and EBV-LMP1. Conclusion In essence, this study indicates that morphological and immunophenotypic subclassification of mantle cell lymphoma with a wider panel of IHC markers is essential for understanding disease biology and better prognostication.


Introduction
Mantle cell lymphoma (MCL) is a relatively rare lymphoproliferative neoplasm (LPN), accounting for less than 10% of all non-Hodgkin lymphomas (NHL) [1]. Its morphology is quite homogeneous, but it varies strikingly in about 10% of the cases, making the diagnosis of MCL challenging for histopathologists. The disease is characterized by hallmark translocation t(11; 14) (q13; q32) which juxtaposes with cyclin D1 and the immunoglobulin heavy chain genes resulting in an increased expression of the cyclin D1 protein, which in turn results in cellular proliferation and increased survival [2]. Although morphology and immunohistochemical features [B-lymphocyte antigen cluster of differentiation (CD)20 or CD20-positive(+)/ CD5+/ cyclin D1+/ CD23-negative(-)/ CD10(-)] are characteristic for diagnosis, aberrant phenotypic expressions are not uncommon [3], and these are more commonly reported in blastoid phenotype [4]. Common aberrancies reported in the literature (either in isolation or in combinations) include CD5-, CD23+, CD10+, B-cell lymphoma 6 protein or BCL6+, and cyclin D1-subgroups that may pose a diagnostic dilemma for the surgical pathologist [4][5][6].
Recently, c-myelocytomatosis oncogene product (c-MYC) and tumor protein 53 (TP53) gene rearrangement were reported to be associated with aggressive biological behavior and overall inferior survival in a subgroup of MCL subjects for which, further in-depth studies are necessary [7,8]. The objective of our present study is to discern the immunophenotypic characteristics of a cohort of MCL cases on lymph node (LN) biopsies and compare them with clinicopathological features and their impact on disease biology. We also present a summary of the relevant published literature pertinent to our observations.

Materials And Methods
The archival biopsy material from the surgical pathology section of the Department of Pathology in a tertiary care center based in eastern India was searched retrospectively for cases of MCL diagnosed over the last five years (2015 to 2020). We included both nodal and extranodal cases in our series, and their morphological and immunohistochemistry (IHC) characteristics were reviewed and analyzed by two independent surgical pathologists (PNM, MNS) as per the criteria proposed by the 2015 World Health Organization (WHO) classification of hematopoietic and lymphoid neoplasm [4]. We included 21/28 MCL cases following stringent criteria where detail morphological and immunophenotypical data of lymph node and bone marrow (BM) biopsy were available for descriptive study. The remaining 7/28 cases in the leukemic phase, diagnosed purely based on peripheral smear and BM morphology with a limited panel of IHC, were excluded as LN biopsies were not performed in those, and flow cytometry immunophenotype data were not available in most of them.
Data pertaining to age, gender, lymphadenopathy, organomegaly, peripheral blood, and BM involvement were collected from medical records. Four microns thick deparaffinized and hematoxylin and eosin (H&E) stained tissue sections were subjected to routine morphological analysis describing the following features: nodal architecture, the morphological phenotype of tumor cell (classical vs. blastoid), angiocentricity, proliferation index [per 10 high power (x400) fields], presence or absence of extranodal spread, high endothelial venules (HEV), pink histiocytes, residual follicles.

Statistical analysis
All categorical variables were expressed in frequency and percentage. OS was compared between two subgroups in relation to age (≤ 60 vs. > 60 years), pattern of nodal involvement (diffuse vs. non-diffuse),

Clinicopathological features
The clinicopathological characteristics, immunohistochemical profile of all 21 cases of MCL are presented in the Tables below. These constituted 4.2% of all NHLs (21/500) diagnosed over the last five years at our center. There were 14 males and seven females (M:F=2:1) with a median age at diagnosis of 60 years (range 30 to 81 years). Generalized lymphadenopathy with associated hepatosplenomegaly was noted in 14 (66.7%) and seven (33.3%) of the cases, respectively. Two had extranodal involvement (stomach, caecum, one each). One (case no. 16) had multiple nodular skin lesions ( Figure 1) in addition to generalized lymphadenopathy, splenomegaly, and bone marrow involvement. And 15 cases (71.4%) (including positron emission tomography or PET scan finding of one case) had BM involvement at the time of initial diagnosis ( Table 1).  ≠PETCT showed F-fluorodeoxyglucose (FDG) uptake in the pelvic bones.
Histomorphological and immunohistochemical details of cases with extranodal involvement (case no.11 and case no.16) are described in Figure 1.

Histopathological findings
Excision lymph node biopsy material was available in 17 of the cases, needle core LN biopsy in two, and mucosal biopsies in two cases (as above) for histopathological examination ( Table 2).

Follow-up and overall survival in various prognostic subgroups
The median duration of follow-up post-therapy (n=17/21) was 10 months (range two to 46 months); with nine (52.9%) dead at the time of the last follow-up. The overall survival did not vary between two subgroups in regard to age (60≤ vs. >60 years, 29.6 vs. 22

Discussion
We studied 21 cases of MCL diagnosed over a period of five years that occurred in the elderly age group with advanced disease at presentation. The prevalence in our series (4.2%) was slightly more than two other Indian series where it was reported to be 2.1% and 3.4%, respectively [1,10]. Compared to a large series by Hrgovic et al. where 22 cases of cutaneous involvement by MCL was reported, we found such occurrence in one case only and the other two had focal gastrointestinal involvement [4,11]. Classical morphology dominated the blastoid phenotype (13 vs. eight, respectively). Immunophenotypic aberrancies such as CD5negative, CD23-positive, SOX11-negative were noted (9.5%, 14.3%, and 14.3% cases respectively), and TP53 and c-MYC expression were observed in a good number/higher proportion of our cases. However, none of the clinicopathological characteristics impacted the overall survival.
The pattern of nodal histomorphology in MCL varies from diffuse to nodular-diffuse or a mantle cell pattern. From a morphological point of view, the greatest challenge is to differentiate the latter from its precursor state, the so-called 'In-situ mantle cell neoplasia (ISMCN)'. ISMCN is characterized by preserved nodal architecture with reactive follicles without an expanded mantle, which characteristically harbors cyclin D1positive cells confined to the inner zone in some of the follicles. In contrast, the MCL with a mantle cell pattern could be recognized by its more numerous and crowded follicles with a decreased interfollicular area and expanded mantle zone, and the zone showing diffuse nuclear positivity for cyclin D1. The mantle cell pattern of MCL was not seen in our series. Though the 'in-situ MCL is an incidental finding, its pathological significance remains that these may represent the precursor stage of lymphomagenesis with similar molecular signature and confer a favorable prognosis [12]. We did not find any evidence of 'in-situ lesion' in our cases; a feature pointing towards an advanced disease with an increased propensity for peripheral blood spillage.
The blastoid variant exhibits large cells with nucleoli and a very brisk mitotic activity (>20-30/10 HPF) and sometimes may mimic a lymphoblastic lymphoma. Cyclin D1 positivity and absence of terminal deoxynucleotidyl transferase (TdT) helps in differentiating the two. The nodular pattern of involvement of the nodes allows other low-grade lymphomas like chronic lymphocytic leukemia (CLL) or follicular lymphoma (FL) into the differentials. The absence of nuclear indentations, presence of pro-lymphocytes, and proliferation centers negate MCL. A mixture of cleaved and non-cleaved cells in a follicular pattern favor FL. In addition, pink histiocytes, high endothelial venules, and angiocentric patterns of atypical lymphoid cells are subtle clues for the diagnosis of MCL [13]. Though they are morphological clues for aiding the diagnosis, their prognostic significance is questionable.
Classical immunophenotype of MCL is CD20-positive/ CD5-positive/ cyclin D1-positive/ CD23-negative/ CD10-negative. Though cyclin D1 is one of the most persistent nuclear markers for MCL, it can be absent in a subset of lymphoma known as cyclin D1-negative MCL. This subgroup was clinically aggressive with advanced stage at diagnosis with frequent peripheral blood and extranodal involvement; thus a poor response to therapy [15]. Hence, it is important to diagnose this entity, so that other treatment options could be offered in these groups of MCL. In our series, cyclin D1 was universally expressed by MCL cells; though blastoid phenotype had heterogeneous nuclear positivity compared to stronger intensity among classical subgroup [16].
The largest series by Miao et al. has shown 6% absence of CD5 expression [17] and in the present study, CD5 was not seen in 9% of cases (2/21) [Blastoid (1), Classical (1)], though weak expression was observed in 14% of cases (3/21). Shih et al. showed that all CD5-negative MCL have a classical morphology [18]. A plausible explanation for these differences could possibly be attributed to the lesser number of cases included in the study groups. CD23 is frequently negative in MCL. A study by Saksena et al. has shown that 13% of MCL cases showed CD23 expression and these cases are associated with leucocytosis, a leukemic presentation, bone marrow involvement, CD200 expression, and a lower frequency of SOX11 positivity [19]. In the present study, 14% (3/21) of cases showed membrane positivity for CD23 (one strong expression and two weak expressions) and 66.6% of patients had an elevated leukocyte count and all of them showed bone marrow involvement.
SOX11 immuno-expression was absent in three of our cases (15%) (two blastoid phenotypes). Xu et al. described a large series (n=75) of SOX11-negative MCL and reported that such subgroups have classic morphology, increased CD23 coexpression, lower proliferation rate, increased extranodal involvement, and more propensity for leukemic transformation compared to the SOX11-positive subgroup [20]. Similar observations were also made by Nygren et al. who found that the SOX11-negative group has significantly lower OS compared with SOX11-positive cases (median OS: 1.5 vs. 3.2 years, respectively; P=0.014) [21]. In contrast, the SOX11-negative phenotype offered a significantly better survival advantage over the positive subgroup (five-year OS: 78% vs. 38%, respectively, P=0.001) in another study by Fernandez et al. [22].