Coexpression of MYC and BCL2 oncoproteins in primary nodal versus primary extranodal diffuse large B-cell lymphoma

INTRODUCTION

Diffuse large B-cell lymphoma (DLBCL) is a clonal neoplasm distinguished by the heterogeneity of large B lymphoid cells, presenting distinct morphological subgroups and molecular variants based on gene expression profiling.^[1] Cytogenetic studies have classified DLBCL based on the presence of MYC (8q24), BCL2 (18q21), and BCL6 (3q27) rearrangements, indicating a more aggressive clinical course.^[1] The 5^th edition of the World Health Organization (WHO) classification of hematolymphoid tumors recognizes a prognostic subset, high-grade B-cell lymphoma with MYC and BCL2 and BCL6 rearrangements, also known as double-hit/triple-hit lymphomas (DHL/THL).^[2] While not a separate diagnostic entity in the WHO Blue Book, lymphomas coexpressing MYC and BCL2 oncoproteins by immunohistochemistry (IHC) are designated as double-expressor phenotype lymphomas (DEPL), which have consistently portrayed poorer outcomes than non-double-expressor phenotype DLBCL in various studies.^[3,4] Notably, the clinical behavior of DEPL is relatively better than that of DHL/THL.^[5,6]

To date, the frequency and clinicopathological parameters of DEPL patients remain poorly characterized. ^[7,8] This study aims to elucidate the distribution and clinicopathological associations of DEPL in primary nodal (PN) versus primary extranodal (PEN) sites, contributing to a deeper understanding of disease trajectory in DEPL patients depending on the primary sites.

MATERIALS AND METHODS

RESULTS

Clinicopathologic features

The clinicopathologic profiles of DEPL, PN-DEPL, and PEN-DEPL patients are summarized in Table 2. The median age of DEPL patients was 56 (range, 41−86) years, with a male predominance (69%). The majority presented with B symptoms (65%), higher eastern cooperative oncology group performance status (ECOG-PS) (67%), elevated lactate dehydrogenase (LDH) levels (56%), higher-stage disease (71%), bone marrow (BM) involvement (56%), and maximum revised international prognostic index (R-IPI) score (50%). Of the 48 cases, the majority were of the pathological activated B-cell (ABC) type (46%) with a Ki67 index >90% (60%).

Table 2: Clinicopathologic profile of patients of DEPL, PN-DEPL, and PEN-DEPL.

Parameters	Subcategories	DEPL n(%)	PN-DEPL n(%)	PEN-DEPL n(%)	P-value
Total cases		48	33 (69)	15 (31)
Age	≤60	20 (42)	17 (51)	3 (20)	0.0592
	>60	28 (58)	16 (49)	12 (80)
Sex	Male	33 (69)	25 (76)	8 (53)	0.1799
	Female	15 (31)	8 (24)	7 (47)
B symptoms	absent	17 (35)	6 (18)	11 (73)	0.0007
	present	31 (65)	27 (82)	4 (27)
ECOG	0-1	16 (33)	8 (24)	8 (53)	0.0963
	≥2	32 (67)	25 (76)	7 (47)
LDH	normal	21 (44)	9 (27)	12 (80)	0.0013
	elevated	27 (56)	24 (73)	3 (20)
Stage	I /II	14 (29)	4 (12)	10 (67)	0.0003
	III/IV	34 (71)	29 (88)	5 (33)
BM involvement	No	21 (44)	19 (58)	2 (13)	0.0051
	Yes	27 (56)	14 (42)	13 (87)
Subtype	GCB	17 (35)	15 (45)	2 (13)	0.005916
	ABC	22 (46)	10 (30)	12 (80)
	unclassifiable	9 (19)	8 (25)	1 (7)
Ki67 labeling index	≤90	19 (40)	18 (55)	1 (7)	0.0016
	>90	29 (60)	15 (45)	14 (93)
R-IPI score	≤3	5 (10)	3 (9)	2 (13)	0.017005
	4	19 (40)	9 (27)	10 (67)
	5	24 (50)	21 (64)	3 (20)

The figures in bold indicate significant associations. DEPL: Double-expressor phenotype lymphoma, PN-DEPL: Primary nodal double-expressor phenotype lymphoma, PEN-DEPL: Primary extranodal double-expressor phenotype lymphoma, N: Number, ECOG: Eastern cooperative oncology group, LDH: Lactate dehydrogenase, GCB: Germinal center B-cell, ABC: Activated B-cell, BM: Bone marrow, R-IPI: Revised international prognostic index

The incidence of PN-DEPL was higher compared to PENDEPL. The median ages of patients with PN-DEPL and PEN-DEPL were 56 and 61 years, respectively. Males were affected almost equally in both groups. B symptoms, LDH level, disease stage, BM involvement, pathological subtype, Ki67 index, and R-IPI score were compared between PNDEPL and PEN-DEPL patients. PN-DEPL patients had a higher incidence of B symptoms (82%), elevated LDH levels (73%), III/IV stage disease (71%), and maximum R-IPI score (64%). In contrast, BM involvement (87%), ABC-type phenotype (80%), disease stage I/II (67%), and Ki67 index >90% (93%) were more common in PEN-DEPL patients. The immunohistochemical features of PN-DEPL and Germinal center B-cell (GCB) type are illustrated in Figure 1, where large monomorphic atypical cells [Figure 1a] are showing positive expression for CD20 [Figure 1b], CD10 [Figure 1c], and BCL2 [Figure 1d]. Figure 2 displays the morphological and IHC findings of PEN-DEPL (skin), ABC type, where sheets of atypical lymphoid infiltrates are seen in the dermis [Figure 2a], showing immunonegativity for CD3 [Figure 2b], strong immunoexpression for CD20 [Figure 2c], and BCL2 [Figure 2d]. A higher power view of the same is depicted in Figure 3, where large tumor cells with non-cleaved nuclei are seen [Figure 3a] with immunopositivity for CD20 [Figure 3b], MYC [Figure 3c], and BCL2 [Figure 3d].

Close

Figure 1:

Immunohistochemical findings of primary nodal-double-expressor phenotype of lymphoma, germinal center B-cell type: (a) Diffuse sheet of monomorphic large cells (Hematoxylin and eosin, ×40), (b) the tumor cells strongly express CD20, (c) CD10, and (d) BCL2 (Diaminobenzidine, ×40).

Export to PPT

Close

Figure 2:

Immunohistochemical findings of primary extranodal-double-expressor phenotype of lymphoma (skin), activated B-cell phenotype (low power): (a) A diffuse sheet of atypical lymphoid cells seen in the dermis (Hematoxylin and eosin, ×20). (b) CD3 highlights reactive T lymphocytes (Diaminobenzidine, ×20). (c) CD20 stains atypical B-cells (Diaminobenzidine, ×20). (d) The tumor cells express BCL2 (Diaminobenzidine, ×20).

Export to PPT

Close

Figure 3:

Immunohistochemical findings of primary extranodal-double-expressor phenotype of lymphoma (skin), activated B-cell phenotype (high power): (a) The tumor cells are large in size with non-cleaved nuclei (Hematoxylin and eosin, ×40). (b) The tumor cells show immunopositivity for CD20, (c) MYC, (d), and BCL2 (Diaminobenzidine, ×40).

Export to PPT

DISCUSSION

DLBCL is the most prevalent non-Hodgkin lymphoma, exhibiting variable clinical outcomes.^[1,2] DEPL, characterized by concurrent expression of MYC and BCL2 proteins, represents a subtype with a poorer prognosis necessitating aggressive interventions.^[9,10]

In our study, 28.7% of DLBCL cases were classified as DEPL, a frequency comparable to previous studies.^[6,11] Among 167 cases of DLBCL, Johnson et al. showed 21% DEPL.^[3] In a study by Hu et al., 34% of the cases displayed dual expression.^[5] Ananthamurthy found 27.5% to be DEPL out of a total of 40 cases.^[11] Overall, 20–30% of patients with DLBCL demonstrated coexpression except for the study by Mehta et al., which stated a lower frequency of 11.6%, and Hashmi et al., which documented a higher frequency of 35.8%.^[9,12] Discrepancies in reported frequencies may be attributed to variations in MYC/BCL2 cutoffs in different studies.

We found that DEPL was more common in males, consistent with many previous reports, except for the findings by Hashmi et al. who reported females as a predominant population.^[9,11,12] The comparable age at diagnosis (56 years) aligns with existing literature.^[3,5,7] Most DEPL patients presented with adverse features such as B symptoms, elevated LDH levels, poor ECOG-PS, and higher-stage disease (III/IV) with BM involvement, in agreement with findings by Johnson et al. and Green et al.^[3,4] ABC-type phenotype was more common in DEPL in our cohort. Hu et al. and Hashmi et al. reported similar findings and postulated that ABC phenotype in DEPL may pose an inferior prognosis.^[5,12] However, the study by Ananthamurthy found almost equal representation of both GCB and ABC phenotypes in DEPL.^[11] The association of coexpression of MYC/BCL2 with a higher Ki67 labeling index, as observed in our study, corroborates findings by Hashmi et al.^[12] However, the role of the Ki67 labeling index in boosting DLBCL cases for dual expression or genomic studies remains debated, as highlighted by Mehta et al.^[9]

A meticulous review of published literature revealed few studies documenting the occurrence and clinicopathological parameters of DEPL.^[13,14] Only a handful of studies explored the clinicobiologic profiles of PN versus PEN-DLBCL patients, depicting significant clinical and genetic differences between them.^[15-17] To the best of our knowledge, this study is the first attempt to explore the clinicopathological features of DEPL based on their primary sites.

The observed predominance of PN-DEPL over PENDEPL in our study, reflecting the higher frequency of PNDLBCL, aligns with existing literature.^[18,19] This finding is in agreement with the study by Hashmi et al. but differs from Mehta et al., who reported an equal distribution of DEPL in nodal versus extranodal sites in the cohort of 20 cases.^[9,12] A slight male predominance was noted in both PN-DEPL and PEN-DEPL groups in our study, consistent with studies by Shi et al., Candelaria et al., and Khan et al. concerning the PN/PEN-DLBCL cohort.^[20-22]

Significant associations between PN-DEPL and PEN-DEPL were observed in B symptoms, LDH level, disease stage at presentation, BM involvement, pathological subtype, Ki67 index, and R-IPI. Despite the relative rarity of DEPL, we attempted to draw parallels between DLBCL and DEPL based on their primary sites. Møller et al. reported that patients with PEN-DLBCL were older and had higher ECOG-PS.^[19] However, in our study, no significant association was found regarding age and ECOG-PS between PN-DEPL and PENDEPL patients.

PN-DEPL patients were more likely to present with B symptoms (82%), elevated LDH levels (73%), stage III/IV disease (71%), and maximum R-IPI score (64%) compared to PEN-DEPL patients, indicating a higher tumor burden. These findings align with studies by Shi et al.^[20] and Khan et al.^[22] Consistent with previous literature, we noticed that ABC-type phenotype (80%), Ann Arbor stage I/II (67%), Ki67 index >90% (93%), and BM involvement (87%) were more common in PEN-DEPL.^[20] In contrast, Khan et al.^[22] reported a lower rate of BM involvement in PEN-DLBCL, and Kim et al. found no differences in the frequencies of GCB and ABC subtypes in the compared groups.^[18] Our observation, however, complies with the study by Candelaria et al.,^[21] which also showcased the gastrointestinal tract as the most common site of extranodal lymphoma. The variation in findings among different studies emphasizes the complexity and heterogeneity of DEPL in different primary sites.

Numerous studies have aimed to assess the relative prognostic importance of genetic alterations and phenotypic and microenvironmental biomarkers in DLBCL. Although high-throughput proteomic profiling has not yet been applied to DLBCL subtyping, the concurrent expression of MYC and BCL2 proteins is a well-established prognostic marker in this lymphoma type.^[23] So far, only a handful of studies have reported on the coexpression of MYC/BCL2 in different primary sites, and therefore, the prognostic role of double expression status in this clinical outline remains unpredictable. Future studies are recommended to evaluate the reasons and factors contributing to these unknown associations.

The study’s strength lies in the availability of a complete IHC workup, allowing an unambiguous diagnosis of DEPL and an examination of its characteristic profile based on primary sites. This study provides an estimate of the burden of this aggressive entity, encouraging further prognostic studies and therapeutic trials.

The drawback in our study involved the fact that DEPL cases subjected to molecular testing might have shown gene rearrangement (MYC/BCL2). Depending on the results, patients of DEPL would have stratified to a smaller cohort of DHL cases, thus altering the true sample size. It is also worth mentioning here that, unlike the reproducibility of the molecular cytogenetic technique, the IHC analysis has more variability associated with it. Another limitation can be the cutoff threshold assigned for consideration of positivity for MYC and BCL2 oncoprotein expression might have compromised the sample size and subsequent associations. Furthermore, we could not compare the outcomes of DEPL, PN-DEPL, and PEN-DEPL patients, which need to be urgently validated in a larger prospective cohort.

CONCLUSIONS

The retrospective nature of the study, resource constraints, and limitations related to the reproducibility of IHC analysis, molecular testing, and follow-up data should be acknowledged. Despite these limitations, the study comprehensively characterizes the clinical and pathological framework of DEPL, highlighting significant differences between PN-DEPL and PEN-DEPL. Testing for MYC and BCL2 rearrangements remains imperative in this subset to survey if they delineate to DHL. Further, research with larger prospective cohorts and long-term follow-up is recommended to assess molecular differences and understand the biological behavior of DEPL in nodal and extranodal primary sites for optimal treatment selection.