Introduction: Lymphomas encompass over 80 malignancies defined by morphology, Immunohistochemistry (IHC), molecular, and genetic features in the 2022 WHO classification. IHC remains central for lineage determination, proliferative assessment, and therapeutic targeting. However, immunophenotypic data from Malaysia are limited.

Methods: We conducted a retrospective study of 398 mature lymphomas diagnosed in a tertiary hospital (2015–2022) with adequate IHC reports. Sixty immunomarkers were analysed.

Results: Pan-B cell markers (CD20, CD79a) were strongly expressed across B-cell lymphomas but absent in T-cell and NK/T-cell neoplasms. BCL2 was frequent in indolent B-cell variants but absent in Burkitt lymphoma, while BCL6 and CD10 highlighted germinal centre–derived subtypes. MUM1 was enriched in activated B-cell type DLBCL, high-grade B-cell lymphomas, and Hodgkin lymphoma. CD30 and PAX5 reliably identified classical Hodgkin lymphoma, and Cyclin D1 confirmed mantle cell lymphoma. Among 177 DLBCL cases, immunophenotypic variation was observed between germinal centre, activated B-cell, and non-GCB subtypes, with co-expression of BCL2, BCL6, and C-MYC in high-grade B-cell lymphoma suggestive of double/triple-hit biology. Ki-67 indices were ≥70% in 72% of assessed cases, emphasising the high proliferative burden of aggressive subtypes. Expression patterns also differed between nodal and extranodal presentations, reflecting clinical heterogeneity.

Conclusions: This first large-scale audit of lymphoma immunophenotypes in Malaysia demonstrates broad concordance with international data while providing novel local insights. The findings reinforce the diagnostic and prognostic value of IHC in refining classification and guiding precision-based lymphoma management.

• Lymphoma Immunomarkers; Patients; Immunohistochemistry; Malaysia

Ghee LK, Selvamani S, Burud IAS, Sood S, Baba AA (2026) Analysis of Lymphoma Immunomarkers from Patients in a Tertiary Hospital in Malaysia. J Cancer Biol Res 13(1): 1154.

The WHO classification of haematopoietic and lymphoid tumours, introduced in 2001 and most recently updated in 2022, recognizes over 80 lymphoma subtypes based on morphology, Immunohistochemistry (IHC), molecular, and genetic features [1]. Mature lymphomas, which typically present as discrete tissue masses, were in the past broadly divided into Hodgkin Lymphoma (HL) and Non-Hodgkin Lymphoma (NHL). HL has distinct clinical features and treatment outcomes, while NHL now comprises multiple subtypes with diverse biological behaviour. Morphology and IHC form the backbone of routine lymphoma diagnosis. IHC, through antigen–antibody interactions, enables detection of lineage-specific proteins, identification of maturation stage, recognition of genetic alterations, assessment of proliferative activity, and guidance for targeted therapy [2,3]. Increasingly, IHC plays a pivotal role in precision medicine [4]. Although each type of lymphoma has a characteristic immunohistochemical signature, each immunomarker is not always positive or negative within any type of lymphoma. The pattern of immunomarkers within a type of lymphoma may also vary in a local population, indicating variations in biology that can be studied to understand how lymphomas differ and to guide targeted therapy. Experts advise against the use of overly-broad panels of immunostains [5], upon analysing the rates of positivity of standard panels of immunomarkers in their setting. The analysis of a cohort in Malaysia has not previously been reported. The aim of this study is to characterise the immunomarkers seen across lymphomas diagnosed in a Malaysian population.

We carried out a retrospective study at Hospital Tuanku Jaafar, Seremban involving patients diagnosed with lymphoma between 2015 to 2022 and have published the demographic characteristics of this cohort of patients [6]. The study was confined to mature lymphomas in order that solid tissue masses were present at diagnosis for biopsy. Diagnosis was made based on histology and immunohistochemistry. From the wide list from the WHO classification of lymphoma diagnoses were grouped into 15 subtypes, with large B cell (including Primary Mediastinal (PMBL) and plasmablastic), marginal zone (including nodal marginal zone and Mucosa-Associated Lymphoid Tissue (MALT) in order not to have too many groups with small numbers. DLBCL the largest group was subtyped further. DLBCL/ABC and DLBCL/non GCB were combined. Patient sample materials were cut into sections of 3-micron thickness and stained with haematoxylin-eosin for histological evaluation. Cases were diagnosed based on morphology with the assistance of ancillary stains such as immunohistochemistry stain denote by expression of related antigen to fulfil the diagnostic criteria. Cases were considered positive for an antigen according to diagnostic criteria for each stain and subtype. All specimens were reviewed by a pathologist for confirmation and classification. A second opinion of a lymphoreticular histopathologist was routinely sought for difficult cases. We excluded cases if immunohistochemistry was inadequate. Ethics approval was obtained from the Ministry of Health, Malaysia [NMRR ID-23-02984-GWV (IIR)]. Patients who had other concurrent malignancies and patients with leukaemia and immature lymphomas were excluded.

A total of 398 histologically diagnosed lymphoma cases had adequate immunomarker reports. Sixty different immunomarkers were used. CD 20 was the most commonly encountered positive marker, and was present in 308 of the 377 cases tested for this marker. The next most frequently identified marker was BCL6. The most commonly tested immunomarkers are listed in Table 1. Another set of eight less frequently tested markers are listed in Table 2, and include CD 45 and CD 30 among others. 

Features by Immunomarkers 

The immunohistochemical profiles from our cohort reveal distinct and diagnostically significant patterns across various lymphoma subtypes. Pan-B cell markers such as CD20 and CD79a showed high positivity rates (99% and 100%, respectively) in Diffuse Large B-Cell Lymphoma (DLBCL) and were strongly expressed across all B-cell lymphomas, while their presence was notably lower in Hodgkin lymphoma (29% for CD20) and largely absent in T-cell and NK/T-cell lymphomas, reaffirming their specificity to the B-cell lineage. BCL2 expression was frequent among most B-cell lymphomas, particularly the small or low-grade variants, where it was consistently present except for Burkitt where it was not found in any of 5 tested. Conversely, BCL6 was typically absent in Small Lymphocytic Lymphoma (SLL), marginal zone, and mantle cell lymphomas but strongly expressed in follicular, low-grade B-cell lymphomas, and high-grade B-cell subtypes—supporting its role in germinal center derivation. CD10 followed a similar trend, prominently expressed in follicular lymphoma (90%) and BL (100%), reinforcing its germinal centre association [7]. MUM1 showed a variable pattern, appearing in a majority of DLBCL (72%), high-grade B-cell lymphomas (71%), and Hodgkin lymphoma (90%), but less frequently in BL and low-grade lymphomas. CD3 expression was seen across T-cell lymphomas (86%) and all NK/T-cell, ALCL, and ATL cases, confirming its reliability as a T-lineage marker. CD5 showed high positivity in mantle cell lymphoma and SSL (both 100%), and in most T-cell lymphomas, but remained uncommon in other B-cell types. CD30 was a hallmark of Hodgkin lymphoma, showing 96% positivity, and was also frequently positive in NK/T-cell lymphomas, ALCL, and ATL. After exclusion of the few cases where the site of origin is unknown the rate immunomarkers were positive or negative in nodal or extranodal sites (which included lymphoma in Waldeyers ring) were noted [Table 1 and 2]. Immunomarkers such as CD20, CD79a, and CD45 were more frequently positive in extranodal lymphomas. This may reflect better antigen preservation or a higher prevalence of certain subtypes in extranodal sites. In contrast, BCL2, MUM1, and CD30 were more frequently expressed in nodal disease, aligning with the nodal predominance of Hodgkin and T-cell lymphomas that commonly express these markers. Although less frequently tested, several immunomarkers in this group provided some important diagnostic clues when applied selectively. Cyclin D1 stood out as a highly specific marker for mantle cell lymphoma, being positive in 88% of cases and rarely detected in other subtypes. C-MYC expression was most evident in Burkitt and high grade B-cell lymphomas, reflecting their proliferative drive [8]. Meanwhile, CD45 was notably under-expressed in Hodgkin lymphoma (4%), a known feature distinguishing it from other lymphomas. PAX5 emerged as a consistently reliable marker for B-cell lineage, showing high positivity in most B-cell lymphomas but particularly strong expression in Hodgkin lymphoma (98%). Its use was targeted toward confirming B-cell origin in ambiguous or mixed-lineage presentations, especially in classical Hodgkin cases where CD20 may be absent or weak. CD15, typically associated with classical HL, was positive in 73% of Hodgkin cases in this series. CD23 showed a distinctive pattern—highly expressed in SSL (100%) and to a lesser extent in large B-cell variants (83%), supporting its utility in distinguishing SLL from other low-grade B-cell neoplasms.

Table 1: Positive rate of most frequently tested immunomarkers of lymphoma cases in Seremban 2015-2022

Interestingly, while nodal and extranodal lymphomas both showed strong PAX5 positivity, CD15 and CD23 were far more common in nodal cases. Conversely, CD56 was only detected in extranodal presentations, likely reflecting the predominance of NK/T-cell lymphomas in extranodal sites such as the nasopharynx or skin.

The Ki-67(%) proliferation index, available in 155 patients, was grouped into positive categorical ranges, Table 3. Reticulin staining grade was recorded in <10% of cases and not analysed. In 37% of cases, the score was above 90%, and 72% were above 70%. High grade B cell lymphoma showed the highest proportion of high Ki-67

Table 2: Positive rate of Less Frequently Tested Immunomarkers of lymphoma cases in Seremban 2015-2022

index with 90% positive above 90%. The small/low grade variants as expected had lower Ki67 scores.

Other Immunomarkers

There were 45 other immunomarkers that were less frequently used, usually in extranodal lymphoma. EMA was used 33 times; the rest were used on less than 20 occasions. A number of markers were negative in all occasions they were used, including CD34, S100, Somatophysin, ChromograninA, CKMNF116, CK20, GFAP (all in brain lymphoma), CK7, CD117, Desmin, Myogenin, Actin, DOG1 AFP, ER, PR, WT1, CD31, Kappa and P53. They were used mainly to exclude non-lymphoma malignancies. Of note CD4 was always positive in 7 cases and CD8 was positive 5/7 of occasions (71%) in T cell lymphoma. CD68 was negative in all 5 cases tested for HL. SOX11 was positive in all 3 cases of Mantle cell lymphoma but negative in 4 other occasions in other lymphomas. TIA1 was highly positive in the 5 of the 6 occasions it was used in NK/T cell, T-cell, ALC and AT lymphomas. It was negative in HL used once. CD7 was positive in 7/10 cases of T- cell lymphoma tested and once in NK/T cell lymphoma

OCT2 and BOB1 were tested together in 8 patients with similar results in 7 cases. Both were positive together in one case each of B cell NOS, DLBCL, Large B cell and HL. Similarly negative in 3 HL. One HL patient was positive for BOB1 but negative for OCT2.

Table 3: Ki67 proliferation index (%) distribution in 155 cases of Lymphoma subtypes in Seremban

Features by Lymphoma subtypes

Large B cell variants: The large B-cell variants show a similar high rate of having CD20, CD79a, CD7, CD45, CD4 BCL2, BCL6 and PAX5, except BL is negative for BCL2. BL was also unlike the other large B-cell variants in being highly positive for CD10. It was also consistently positive for C-MYC but there were only 3 cases. Notably, BL demonstrated a highly distinct profile: strong expression of CD10 and C-MYC, absence of BCL2, CD3, CD5, and Cyclin D1, and minimal expression of MUM1—fitting its aggressive biology and classic immunophenotype.

We subtyped 190 cases of DLBCL into DLBCL Germinal Centre B cell (GCB) (n=47 24.7%) and DLBCL nonGCB/Aactivated B cell (ABC) (n=73 38.4%) while 70 cases (36.8%) were not assigned or not otherwise specified (DLBCL/NOS) and of these 177 had adequate immunomarker reports. There are variations of the immunomarker profiles among these subtypes. The DLBCL/NOS group reflects DLBCL in total. However, MUM1, CD5 and C-MYC are higher in DLBCL nonGCB/ ABC than DLBCL/GCB. On the other hand CD10, being associated with the germinal centre is higher in DLBCL/ GCB [7]. DLBCL NOS have very high Ki67% scores like high grade B-cell lymphoma (Table 3).

Small/low grade B cell variants: Indolent B lymphomas, including marginal zone, follicular, small lymphocytic and mantle cell lymphomas, were universally positive for CD20, BCL2 and CD79a. The few other low grade B-cell lymphoma are also similar except not always positive for CD20. CD45 was also universally positive, except that it was not tested in SSL and mantle cell lymphoma. CD5 differentiated these small cell variants, being always positive for small lymphocytic and mantle cell lymphoma but usually negative for marginal zone, follicular and other low-grade B cell lymphoma.

Hodgkin: The hallmark of HL were PAX5 (98%), CD30 (96%) MUM1 (90%) and CD15 (73%). CD5 was negative in all 9 tested. HL also stands in contrast to other B-cell lymphoma in not being highly positive for CD4 and CD45 (15%) and also CD7, CD20 and CD79a (less than 30%).

T-cell: T-cell lymphoma was most often positive for CD3, CD5, CD45, BCL2 and BCL6.

The pattern of positive and negative immunomarkers among different subtypes of lymphoma helps affirm the diagnosis based on histological features seen on normal staining and microscopy.

Diagnostic Consistency across Populations

The positivity rates of the wide range of various markers across the range of B cell lymphomas in our series was similar to rates found in a similar study in the United States [5]. The positivity rates markers in T-cell lymphomas, [9] and in HL were similar to other studies [10]. This alignment reinforces the validity of established immunohistochemical panels in local use bridging geographic and demographic variability..

Immunophenotypic Distinctions by Lymphoma Subtype

BL stands in contrast to other large B cell lymphomas in being strongly positive for CD10, a zinc-dependent metalloprotease found on the cell surface and a germinal centre marker [11,12], and C-MYC, which is also characteristic, but not specific for BL [13]. We found a complete absence of BCL2. BCL-2, a protein that inhibits apoptosis at the mitochondria [14,15], that is less expressed in BL [16,17].

Among indolent lymphomas, markers like BCL2 and CD79a were frequently expressed, while CD5 positivity helped differentiate mantle cell lymphoma and SLL from other low-grade subtypes. SLL is characteristically positive for CD5, CD19, CD23, CD43, and CD200, with dim expression of CD20, CD22, CD79b, and kappa but the most important characteristic of SLL is CD5 and CD23 positivity [18]. These two markers were consistently positive in all our cases, but it is in CD23 that SLL stands out in this series.

Mantle cell lymphoma as expected, also expressed Cyclin D1 [19]. The cases in this series tested for SOX11 were all positive, which is an indication of being aggressive malignancies [19].

In HL, the hallmark profile of CD30+, CD15+, and PAX5+ with weak or absent CD45 and CD20 expression was also well preserved, emphasising the ability of these markers in differentiating HL from other entities [3]. CD15 notably was rarely expressed in other lymphoma types, reinforcing its role as a characteristic, though not exclusive marker in HL pathology. CD68, is associated with poorer outcomes in HL, and its absence is good news for our patients [20]. It has been noted that FISH study of BCL2 and C-MYC showed a significantly higher aberrations in nodal compared with extranodal sites in B cell lymphoma, excluding HL [21]. That is also observed in our series for BCL2 where HL accounts for very few cases tested. However, nodal B cell lymphomas in our series do not show higher C-MYC expression.

Site-Specific Trends in Marker Expression

Nodal and extranodal lymphomas demonstrated differing patterns of antigen expression. For instance, markers such as CD20 and CD79a were more frequently positive in extranodal presentations, which may be attributed to the higher prevalence of B-cell lymphomas in these sites or superior antigen preservation in fixed tissues. Conversely, nodal lymphomas showed higher expression of CD3, CD30, and CD5, aligning with the predominance of T-cell and Hodgkin lymphomas in nodal sites.

Prognostic Implications of Immunohistochemistry

Beyond diagnostics, our findings contribute valuable prognostic insights. The co-expression of CMYC, BCL2, and BCL6 was observed in 88.6% of high-grade B-cell lymphomas in our series, suggestive of potential “double- hit” or “triple-hit” molecular phenotypes, which are known to correlate with aggressive clinical behaviour and inferior outcomes. This co-expression pattern aligns with the elevated proliferation indices documented in our cohort.

We noted CD5 more common in DLBCL/ABC which is associated with a poorer prognosis compared to DLBCL/ GCB. The expression of CD5 has been is generally associated with a poorer prognosis in DLBCL [22].

Ki-67 is a marker of cell proliferation, and a higher Ki-67 proliferation index in many lymphoma subtypes, including some extranodal presentations, has a more aggressive course and poorer outcome [23]. The Ki-67 index is high is our patients.

Limitations: Inherently, immunohistochemistry detects mutations that drive cancer, and is highly sensitive and specific for some mutations but not in all. The rate they are noted to be present or absent may also be affected by the specimen collected. This is a retrospective study and the selection of immunomarkers are done by convention and not in planned predetermined study design.

Strengths: This restrospective study reflects real practice in pathological services in Malaysia, recording the percentage positive in a cohort for each immunomarkers. It shows how common outliers occur. It is the largest collection of lymphoma cases published to date in Malaysia and the first to audit and analyse the prevalence of presence or absence of immunomarkers commonly used. The pathologists routinely consulted experts for difficult cases.

Our findings provide a robust local immunophenotypic landscape of lymphomas in Malaysia, showing close concordancewithglobaldataandhighlightingthediagnostic and prognostic power of immunohistochemistry. CD20 and BCL6 are the most commonly encountered markers in lymphomas. CD15 and CD23 tend to be present much more often in extranodal lymphomas. Our results show the importance of continued integration of IHC profiling in lymphoma workups, including proliferation indices and co-expression markers, to refine classification, guide therapy, and anticipate clinical behaviour. 

1. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 2024; 30.
2. Cho J. Basic immunohistochemistry for lymphoma diagnosis. Blood Res. 2022; 57: 55-61.
3. de Leval L, Jaffe ES. Lymphoma Classification. Cancer J. 2020; 26: 176-185.
4. Sun R, Medeiros LJ, Young KH. Diagnostic and predictive biomarkers for lymphoma diagnosis and treatment in the era of precision medicine. Mod Pathol. 2016; 29: 1118-1142.
5. Boyd SD, Natkunam Y, Allen JR, Warnke RA. Selective immunophenotyping for diagnosis of B-cell neoplasms: Immunohistochemistry and flow cytometry strategies and results. Appl Immunohistochem Mol Morphol. 2013; 21: 116-131.
6. Lim KG, Nasarudin MN, Khaw HYC, Lim JY, Nordin NS, Hameed AAI, et al. Demographic and Pathological Features of 430 Cases of Lymphoma from a Tertiary Hospital in Seremban, Malaysia (2015–2022). Malays J Med Sci. 2025; 32: 82-89
7. Basso K, Dalla-Favera R. Chapter 7 - BCL6: Master regulator of the germinal center reaction and key oncogene in B Cell Lymphomagenesis. In: Alt FW, editor. Advances in Immunology. Academic Press. 2010; 105: 193-210.
8. Ott G, Rosenwald A, Campo E. Understanding MYC-driven aggressive B-cell lymphomas: Pathogenesis and classification. Blood. 2013; 122: 3884-3891.
9. Elghawy O, Cao M, Xu J, Landsburg DJ, Svoboda J, Nasta SD, et al. Prevalence and Prognostication of CD5+ Mature T-Cell Lymphomas. Cancers. 2024; 16: 3430.
10. O’Malley DP, Dogan A, Fedoriw Y, Medeiros LJ, Ok CY, Salama ME. American registry of pathology expert opinions: Immunohistochemical evaluation of classic Hodgkin lymphoma. Ann Diagn Pathol. 2019; 39: 105-110.
11. Gaillard F. Radiopaedia. WHO classification of haematolymphoid tumors. 2024.
12. Wang S, Xiao Y, An X, Luo L, Gong K, Yu D. A comprehensive reviewof the literature on CD10: Its function, clinical application, and prospects. Front Pharmacol. 2024; 8; 15.
13. Said J, Lones M, Yea S. Burkitt lymphoma and MYC: What else is new? Adv Anat Pathol. 2014; 21: 160-165.
14. Klanova M, Kazantsev D, Pokorna E, Zikmund T, Karolova J, Behounek M, et al. Anti-apoptotic MCL1 Protein represents critical survival molecule for most Burkitt lymphomas and BCL2-negative diffuse large b-cell lymphomas. Mol Cancer Ther. 2022; 21: 89-99.
15. Opferman JT, Kothari A. Anti-apoptotic BCL-2 family members in development. Cell Death Differ. 2018; 25: 37-45.
16. Masqué-Soler N, Szczepanowski M, Kohler CW, Aukema SM, Nagel I, Richter J, et al. Clinical and pathological features of Burkitt lymphoma showing expression of BCL2--an analysis including gene expression in formalin-fixed paraffin-embedded tissue. Br J Haematol. 2015; 171: 501-508.
17. Yoshida M, Ichikawa A, Miyoshi H, Kiyasu J, Kimura Y, Niino D, et al. Low incidence of MYC/BCL2 double-hit in Burkitt lymphoma. Pathol Int. 2015; 65: 486-489.
18. Yoshino T, Tanaka T, Sato Y. Differential diagnosis of chronic lymphocytic leukemia/small lymphocytic lymphoma and other indolent lymphomas, including mantle cell lymphoma. J Clin Exp Hematop. 2020; 60: 124-129.
19. Navarro A, Beà S, Jares P, Campo E. Molecular Pathogenesis of Mantle Cell Lymphoma. Hematol Oncol Clin North Am. 2020; 34: 795-807.
20. Yoon DH, Koh YW, Kang HJ, Kim S, Park CS, Lee SW, et al. CD68 and CD163 as prognostic factors for Korean patients with Hodgkin lymphoma. Eur J Haematol. 2012; 88: 292-305.
21. Salam DSDA, Thit EE, Teoh SH, Tan SY, Peh SC, Cheah SC. C-MYC, BCL2 and BCL6 Translocation in B-cell Non-Hodgkin Lymphoma Cases. J Cancer. 2020; 11: 190-198.
22. Miyazaki K, Yamaguchi M, Imai H, Tamaru S, Kobayashi T, Shiku H, et al. Gene Expression Profiling of Diffuse Large B-Cell Lymphomas Supervised by CD5 Expression. Blood. 2010; 116: 4164.
23. He X, Chen Z, Fu T, Jin X, Yu T, Liang Y, et al. Ki-67 is a valuable prognostic predictor of lymphoma but its utility varies in lymphoma subtypes: Evidence from a systematic meta-analysis. BMC Cancer. 2014; 14: 153.