Lymphomas can originate in almost all tissues and organs, and may subsequently infiltrate peripheral tissues, distant tissues and bone marrow (BM) (1,2). Primary BM lymphoma (PBM) is a rare entity defined by lymphoma confined to the BM without evidence of nodal or extranodal involvement at diagnosis. Diffuse large B-cell lymphoma (DLBCL) is the most common histological subtype of PBM, yet it accounts for <1% of all DLBCL cases and is typically associated with an aggressive clinical course and poor prognosis (3). Patients often present with constitutional symptoms (fever, night sweats and weight loss) and peripheral blood cytopenias, which can mimic a wide spectrum of benign and malignant hematological diseases, frequently resulting in diagnostic delay or error (4). Diagnosis requires strict exclusion of systemic involvement (no lymph node/organ enlargement confirmed by physical examination and imaging) and is easily misdiagnosed as lymphoma of secondary BM involvement. Peripheral blood lymphocyte counts and morphology are usually unremarkable (5).

Myelofibrosis (MF) is a clonal myeloproliferative neoplasm (MPN) of the BM caused by an abnormal clone of hematopoietic stem cells (6). MF can be classified into primary MF (PMF), a Philadelphia chromosome (Ph)-negative MPN, and secondary MF, which refers to reactive MF induced by a variety of underlying diseases or conditions. Accurate differentiation between the two is critical for determining appropriate therapeutic strategies and requires an integrated assessment of clinical presentation, laboratory findings, BM histopathology and molecular genetics (7). The pathophysiological association between DLBCL and MF is considered to involve cytokine-mediated activation of fibrotic pathways (for example, JAK/STAT and SMAD). Profibrotic factors such as transforming growth factor-β (TGF-β) and platelet-derived growth factor (PDGF), secreted by lymphoma cells, lead to collagen deposition and subsequent extramedullary hematopoiesis. Clinically, cytopenias, hepatosplenomegaly and systemic symptoms predominate, with a median survival time of ~6 years (8).

The present study reports a rare and diagnostically challenging case of primary BM-CLBCL (PBM-DLBCL) that was initially misdiagnosed as a T-cell lymphoma with associated MF based on initial BM histopathology and clinical presentation. The diagnosis of PBM-DLBCL was established following two BM aspirates and biopsies. The diagnosis and therapeutic timeline over the course of the patient's two hospitalizations is summarized in Table I. The present case illustrates the potential diagnostic pitfalls and emphasizes the necessity of repeated and comprehensive BM histopathological assessment in patients with persistent cytopenias.

The present case report illustrated the substantial diagnostic challenges posed by PBM-DLBCL. The initial presentation characterized by pancytopenia and BM fibrosis and the presence of a clonal T-cell population, initially suggested a T-cell lymphoproliferative disease. The second biopsy revealed a proliferation of medium-to-large lymphocytes that had not been detected in the initial fibrotic sample. Crucially, IHC demonstrated that these large cells were strongly positive for CD20 and CD79a (B-cell markers), with an elevated Ki-67 index, confirming a dominant monoclonal B-cell process. T-cell populations were present but merely constituted a background component. The markedly different finding on the second BM biopsy underscores the pivotal role of repeat marrow evaluation in patients with refractory cytopenia or a progressive clinical course.

The initial clonal T-cell population represented an enigmatic finding. Its absence on the second BM evaluation suggests it may have reflected an incidental clonal T-cell expansion of uncertain clinical significance or a reactive process that was subsequently overtaken by the more aggressive DLBCL clone. Although the presence of a true composite lymphoma is conceivable, it appears less likely. Comparative TCR-γ gene rearrangement analysis or more comprehensive molecular profiling on the second sample were not performed due to patient refusal, leaving this question unresolved.

The patient received supportive care and regular monitoring in an outpatient setting. Following the initial biopsy, the working diagnosis was a T-cell lymphoproliferative disorder with associated MF. However, the patient failed to respond to first-line immunosuppressive therapy (cyclosporine) and subsequent glucocorticoid treatment, a finding that fundamentally challenged the initial diagnostic impression. Glucocorticoid resistance is highly unusual in classical T-cell lymphomas and effectively ruled out steroid-responsive conditions such as autoimmune cytopenia or certain forms of aplastic anemia. Treatment failure thus became the key catalyst prompting re-evaluation and a definitive second BM biopsy. In effect, the lack of treatment response served as the most critical ‘diagnostic test’ at this stage, directing investigation toward a more aggressive underlying pathology.

Lack of response to glucocorticoids is characteristic of high-grade lymphomas, including DLBCL. Although glucocorticoids exhibit lympholytic effects and are commonly used in combination chemotherapy (for example, R-CHOP), they are largely ineffective as long-term monotherapy in DLBCL. Their action primarily induces apoptosis in mature lymphocytes, but aggressive malignant clones such as DLBCL rapidly develop resistance (11). Thus, the persistent clinical and hematological deterioration despite glucocorticoid therapy was inconsistent with an indolent T-cell disorder or immune-mediated cytopenia and instead aligned with the profile of an aggressive lymphoma such as DLBCL. This clinical course provided critical corroborative evidence for the ultimate diagnosis and reinforced the histopathological findings of the second biopsy.

Secondary MF is frequently a reactive process to an underlying malignancy, analogous to the MF observed in metastatic cancer (12). However, its occurrence in the rare context of PBM-DLBCL generates a complex clinical scenario that can obscure the underlying diagnosis. In the present case, the secondary MF was hypothesized to be cytokine-mediated. The DLBCL cells likely secreted profibrotic cytokines such as TGF-β and PDGF, which subsequently activated the JAK/STAT and SMAD pathways in BM stromal cells, ultimately leading to collagen deposition and marrow fibrosis, as confirmed by the reticulin staining. This cytokine-driven remodeling of the BM microenvironment represents a plausible central mechanism (13,14). According to the 2024 CSCO guidelines for PMF, diagnosis requires the presence of all three major criteria, including megakaryocyte hyperplasia with atypia, grade ≥2 reticulin fibrosis and JAK2/CALR/MPL mutation, along with at least one minor criterion (for example, splenomegaly, anemia or increased serum LDH level). Βy contrast, establishing a diagnosis of secondary MF requires the exclusion of PMF or other MPNs with differentiation by BM pathology and mutation analysis for JAK2/CALR/MPL (15). In the last decade, two targeted therapies have been approved for the treatment of MF, both JAK2 inhibitors, namely ruxolitinib and fedratinib (16). While the majority of patients with DLBCL present with advanced-stage disease at diagnosis, >60% can be cured with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone) immunochemotherapy (11). In the present case report, the clear evidence of a B-cell lymphomatous process supports the classification as secondary MF. Although molecular testing for JAK2, CALR and MPL mutations would have been invaluable to formally exclude an underlying MPN, the patient's refusal precluded this analysis, presenting a significant limitation.

The review by Zamò et al (17) demonstrated that molecular characterization of rare hematological malignancies necessitates multicenter collaboration to accrue sufficient sample sizes. For PBM-DLBCL with secondary MF, further studies can be conducted in the following aspects: i) Molecular mechanisms: Delineating the microenvironmental differences between patients with PBM-DLBCL with versus without MF by single-cell sequencing or spatial transcriptomics to identify fibrosis-driving factors. ii) Prognostic impact: Evaluating whether the presence of MF exacerbates BM failure or contributes to treatment resistance in PBM-DLBCL requires long-term follow-up data support. iii) Clinical management: Assessing the feasibility and efficacy of combined therapeutic approaches, including immunochemotherapy (for example, R-CHOP), targeted agents (JAK inhibitors) and antifibrotic interventions (for example, TGF-β inhibitors).

Currently, the efficacy of conventional DLBCL regimens (for example, R-CHOP) in PBM- DLBCL with secondary MF remains uncertain. Although using rituximab markedly improves the prognosis of patients with CD20-positive B-cell lymphomas, its efficacy has not been rigorously validated in PBML due to the lack of randomized controlled clinical trials (18,19). The potential utility of incorporating a JAK inhibitor (for example, ruxolitinib) to target the fibrotic morrow microenvironment in this setting remains unexplored and warrants future investigation. In conclusion, the extreme rarity of PBM-DLBCL with MF creates substantial gaps about its pathophysiology and optimal therapeutic strategies. In the future, cases could be collected through international registry systems (for example, NIH or ESMO rare tumor databases) and combined with multi-omics analyses to facilitate personalized treatment approaches. The patient's refusal of treatment precluded any assessment of therapeutic response and resulted in a loss of valuable prognostic data.

PBM-DLBCL represents a diagnostically challenging malignancy, underscoring the importance of heightened clinical awareness. PBM-DLBCL should be considered in patients presenting with unexplained pancytopenia. The present case report has several limitations, primarily due to the patient's refusal of further testing. The lack of PET/CT imaging precludes definitive exclusion of occult extramedullary disease, though it was not detected on available CT scans. Additionally, the absence of molecular data (JAK2/CALR/MPL and MYC/BCL2 analysis by FISH) and an extended IHC panel (for example, MUM-1, BCL-2, BCL-6 and C-MYC) hampers accurate prognostic stratification for both the MF and the DLBCL. The 11-month interval between the initial steroid-refractory presentation and the second BM biopsy, although ultimately leading to a definitive diagnosis, represents a potential delay; however, it was the progression of symptoms and cytopenias that necessitated the repeat procedure. In conclusion, accurate diagnosis of PBM-DLBCL requires a comprehensive clinical and laboratory evaluation, the utilization of appropriate diagnostic modalities in cases of unexplained cytopenia, and a meticulous clinical approach to ensure optimal patient safety and outcomes. PBM-DLBCL is a diagnostic chameleon capable of mimicking T-cell malignancies, aplastic anemia and other causes of marrow failure. The present case report highlights that the presence of a clonal T-cell population does not necessarily confirm a T-cell lymphoma diagnosis and may represent a misleading secondary finding. Unexplained and persistent pancytopenia, especially when accompanied with systemic symptoms and elevated LDH, warrants a low threshold for repeat BM evaluation. Moreover, the complex interplay between lymphoma and the marrow microenvironment, leading to fibrosis, represents a critical avenue for further research to elucidate pathogenesis and identify potential therapeutic targets. Overall, this case underscores the diagnostic challenges and the imperative for thorough and at times, repeated diagnostic assessment.