Joint Assessment of CD38 , CD138 , and Immunoglobulin Light Chains Supports the Differential Diagnosis Between Classical Hodgkin Lymphoma and Plasmablastic Lymphoma

Classical Hodgkin lymphoma (cHL) is a Bcell-derived malignancy with unique morphological and phenotypic features. Histologically, cHL is characterized by a rich inflammatory background with scattered, atypical neoplastic cells (i.e., Hodgkin and Reed/Sternberg [HRS] cells). These are positive for CD30 with weak expression of PAX5, frequent positivity for CD15 and substantial lack of B-cell program transcriptional factors (OCT2, BOB1). However, the post-germinal center B-cell marker IRF4/MUM1 is almost universally expressed, while EBER in situ hybridization (ISH) is positive in a subset of cases [1, 2]. The differential diagnosis of cHL includes a variety of B and T-cell lymphomas with HRS-like cells (e.g., diffuse large B-cell lymphomas with anaplastic features; ALK-negative or ALK-positive anaplastic large cell lymphoma), as well as subsets of immunodeficiency-related lymphoproliferative disorders (LPDs). In such cases, the correct diagnosis relies on clinical-pathological correlations and on extensive phenotyping [3]. In our practice, we also faced a challenging case of plasmablastic lymphoma (PBL) of the parotid gland, which further expands the differential diagnosis of cHL. This lymphoma disclosed a cHL-like fibro-inflammatory background with scattered HRS-like cells, occasionally forming clusters. The neoplastic cells were strongly positive for CD30, IRF4/MUM1, CD38 and EBER ISH, with faint positivity for PAX5, partial/weak positivity for CD79, and negativity for other pan B-cell markers, CD15 and CD138 (Figure 1). Although a diagnosis of PBL was favored by CD38 and focal immunoglobulin light chain expression, to date no studies have investigated CD38 expression in cHL nor has the role of plasma cell markers been addressed in the differential diagnosis between cHL and PBL. To fill this gap, we conducted a multicentric study on plasma cell and terminal B-cell differentiation markers in cHL, specifically considering their impact in the differential diagnosis with PBL. We considered a retrospective, consecutive series of 87 cHL and 14 PBL, retrieved from the Pathology Units of Padua University Hospital (Padua—Italy) and ASST Spedali Civili of Brescia (Brescia—Italy) over a 7-year period (Jan 2018 to Jan 2025). Inclusion criteria included: (i) histologically confirmed diagnosis of cHL and PBL; (ii) availability of originally stained slides for review; (iii) availability of adequate formalin-fixed paraffin-embedded material for further phenotypic studies. All cases were reviewed by 2 hematopathologists blind to the original diagnosis. After selection of representative tissue sections, immunohistochemistry (IHC) was performed in the BOND-MAX automated immunostainer (Leica Biosystems, Milan—Italy), using primary antibodies against CD38, CD138, IRF4/MUM1, and kappa and lambda light chains (Table S1). All cases also underwent light chain transcript evaluation by RNAscope technology (Probe-Hs-IGK and Probe-Hs-IGL; Advanced Cell Diagnostic, Newark—CA, USA; cutoff for positivity: 5% of neoplastic cells). These analyses were run in parallel in the two centers participating in the study. For immunoglobulin light chains, the cutoff for positivity was set at 5% to minimize false positive results, due to the possible occurrence of minor light chain-restricted B-cell populations in the inflammatory milieu of Hodgkin lymphoma [4]. For CD38, CD138 and/or IRF4/MUM1, the cutoff for positivity was set at 30%, following prior studies on B-cell lymphomas with plasmablastic differentiation [5]. Digital imaging analysis was performed using the Aperio ImageScope software (Leica Microsystem). Comparison between cHL and PBL cases was run by nonparametric tests for quantitative (Wilcoxon-Mann-Whitney test) and qualitative (Fisher's exact test) variables. Fisher's exact test was also used to test correlations between a newly developed immunohistochemical score and the underlying diagnosis, after stratification of cases based on 1-point threshold (details reported below). Differences between groups were considered statistically significant for p-values < 0.05. The cohort of cHL patients included 46 males and 41 females, with a median age at diagnosis of 14.5 years (range: 3–74 years). Histologically, most cases had features of nodular sclerosis cHL (79/87 [90.8%]) with positivity for CD30 (87/87 [100%]), CD15 (73/82 [89.0%]) and PAX5 (82/87 [94.3%]), negativity for CD20 (65/87 [74.7%]) and variable positivity for EBER ISH (12/78 [15.4%]). Among the plasma cell/terminal B-cell differentiation markers, IRF4/MUM1 was expressed in all the cases (87/87 [100%]), while CD38 and CD138 were universally negative. Immunohistochemistry for immunoglobulin light chains was negative in 86/87 (98.9%) cases, with only one case showing weak kappa expression in a minority of HRS cells. RNAscope for kappa or lambda chains was also positive in HRS cells in 7/87 (8.1%) cases, displaying kappa chains in 4/87 (4.6%) and lambda chains in 3/87 (3.5%) cases (Table 1; Figure 2). In all such cases, positive cells constituted only a subset of all HRS cells (range: 5% to 50%). 14.5 (range: 3–74) 63.5 (range: 21–84) The cohort of PBL patients included 7 males and 7 females, with median age at diagnosis of 63.5 (range: 21–84 years). Most cases presented at extra-nodal sites (oral or nasal cavity: 5/14 [35.7%]; skin: 2/14 [14.3%]; thoracic wall, abdominal cavity, urethra and large bowel: 1/14 [7.1%] case, each), while 3/14 (21.4%) cases were diagnosed in lymph nodes. All PBL showed strong IRF4/MUM1 expression, with consistent negativity for CD20 and PAX5 and variable positivity for EBER (8/14 [57.1%]). CD138 was expressed in 9/14 (64.3%) and CD38 in 11/14 (78.6%) cases. Light chain expression was documented in 11/14 (78.6%) cases by IHC (monotypic kappa: 8/14 [57.1%]; monotypic lambda: 2/14 [14.3%]; double positivity for kappa and lambda chains: 1/14 [7.1%]). RNAscope results overlapped with IHC analysis (Table 1; Figure 2). Overall, CD138, CD38 and immunoglobulin light chains (both IHC and RNAscope) were more frequently expressed in PBL than cHL (p < 0.01). Despite this, no single marker could effectively distinguish cHL from PBL. To increase the diagnostic yield of IHC characterization, we developed an integrated score based on CD138, CD38 and immunoglobulin light chain immunohistochemical expression, by attributing 1 point to each positive marker. A total score ≥ ≥ 1 significantly correlated with the diagnosis of PBL (p < 0.01), with a sensitivity of 100% and specificity of 98.8%. Our data confirm the results of prior studies on the negativity of CD138 and immunoglobulin light chains in cHL [6, 7] and originally document negativity for CD38 in all tested cHL. In keeping with prior studies [8], we also confirm negativity for kappa and lambda chains by RNAscope in the vast majority of cHL, albeit with higher expression rates than IHC analysis. The reasons for such discrepancy are unknown, yet possible explanations include (i) a higher sensitivity of RNAscope over IHC analysis, (ii) immunoglobulin gene mutations impairing protein translation, and/or (iii) post-transcriptional regulatory mechanisms preventing light chain translation in subsets of cHL. Further studies are needed to clarify the biological underpinnings of our finding. From a clinical perspective, our data identify a new tool for the differential diagnosis between cHL and PBL, which adds to already established clinical-pathological and genetic criteria. In fact, distinction between cHL and PBL is usually straightforward and relies on clinical, histological and (when available) molecular findings. Clinically, cHL occurs at younger age and presents primarily as nodal disease, while PBL affects adult to elderly patients, arises in extra-nodal sites, and is strongly associated with immunodeficiency [1, 2]. Histologically, cHL and PBL usually differ in the number, distribution and phenotype of the neoplastic cells, although some features may be disguising (e.g., positivity for CD30, down-modulation of the B-cell program, strong expression of IRF4/MUM1, variable positivity for EBER). In such a context, an integrated score that jointly considers CD138, CD38, and immunoglobulin light chains may support the differential diagnosis, outperforming the diagnostic yield of any of these markers alone. In cases with enough neoplastic cells, genetic studies may further support the differential diagnosis, since PBL bears more frequent MYC rearrangements, as well as CARD11, TET2, EP300 and MAPK pathway (NRAS, KRAS, HRAS, MAP2K1, BRAF) mutations. By contrast, cHL is more frequently associated with XPO1, ARID1A, NF-kB pathway (NFKBIA, NFKBIE, IKBKB, TNFAIP3) and immunescape-related (B2M, CIITA, HLA-B) mutations [9-12]. In conclusion, HRS cells of cHL do not express CD38 and CD138 and are generally negative for immunoglobulin light chains. The frequent positivity of PBL for ≥ ≥ 1 of these markers supports the differential diagnosis with cHL, integrating other clinical-pathological and genetic parameters. Further studies on larger cohort of patients are needed to confirm our findings. The study was approved by the local ethical committee (CET-ACEV; study ID: 519n/AO/24; copy of study protocol filed at CET-ACEV). Written consent for all reachable patients. The authors declare no conflicts of interest. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Table S1: Antibodies used for IHC analysis. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.