Flow Cytometry‐Guided Diagnosis of Cortical T‐Lymphoblastic Lymphoma Presenting With a Massive Mediastinal Mass and Borderline Bone Marrow Involvement

ABSTRACT

Background

T‐cell lymphoblastic lymphoma (T‐LBL) and T‐cell acute lymphoblastic leukemia (T‐ALL) are considered a single biological entity, distinguished primarily by the extent of bone marrow infiltration. This distinction becomes diagnostically challenging in patients with borderline marrow blast counts and predominantly extramedullary disease, risking overclassification as leukemia.

Methods

We describe a 57‐year‐old man who presented with dyspnea and a large mediastinal mass causing superior vena cava syndrome, with pleural and pericardial effusions. Diagnostic workup included complete blood count and biochemistry, imaging, fine‐needle aspiration, core biopsy, pleural fluid cytology, multiparameter flow cytometry of pleural fluid and bone marrow, immunohistochemistry, and cerebrospinal fluid analysis.

Results

CBC and biochemistry were largely unremarkable aside from elevated LDH and CRP. Cytology and biopsy showed a diffuse blastoid lymphoid infiltrate. Pleural fluid flow cytometry identified an aberrant immature T‐cell population (cytoplasmic CD3+, CD1a+, CD2+, CD4+, CD5(dim)+, CD7+, CD8+, CD10+, TdT+; surface CD3−, CD34−, MPO−), confirming cortical T‐ALL/LBL and excluding B‐ and myeloid‐lineage differentiation. Bone marrow morphology suggested ~10% blasts, but flow cytometry detected only 3.8% CD34+ cells with a myeloid‐skewed, non‐T‐lymphoblastic phenotype, arguing against significant marrow involvement. The patient was classified as T‐LBL and treated per the GMALL protocol, achieving complete remission.

Conclusion

Flow cytometry resolved a discrepancy between morphologic and immunophenotypic assessment of bone marrow involvement, preventing overclassification as T‐ALL. Multiparameter flow cytometry is essential for accurate lineage assignment and staging of precursor T‐lymphoid neoplasms in cytologic specimens and borderline marrow samples.

1. Introduction

T‐cell lymphoblastic lymphoma (T‐LBL) is a highly aggressive neoplasm derived from immature thymic T‐cell precursors and represents the tissue‐based counterpart of T‐cell acute lymphoblastic leukemia (T‐ALL) [1, 2]. The distinction between these entities is largely determined by the extent of bone marrow involvement, with significant marrow infiltration supporting T‐ALL [1, 2]. T‐LBL commonly presents with a bulky anterior mediastinal mass due to thymic origin and may result in severe compressive complications such as superior vena cava (SVC) syndrome, airway compromise, and pericardial fluid [3, 4]. Cytologic specimens, including fine‐needle aspiration (FNA) and body fluids, often provide the first diagnostic material in these emergent presentations. Multiparameter flow cytometry plays a central role in lineage assignment, assessment of immaturity, and staging of precursor lymphoid neoplasms [5, 6, 7]. This case highlights the critical role of multiparameter flow cytometry in avoiding overdiagnosis of T‐ALL in patients with borderline bone marrow blast counts and predominantly extramedullary disease.

1.1. Patient Information and Clinical Findings

A 57‐year‐old patient presented to the emergency room with progressive dyspnea, fatigue, and systemic symptoms. Imaging revealed a large mediastinal mass with associated pleural and pericardial effusions, raising concern for a lymphoproliferative disorder. Clinical findings suggested cardiorespiratory compromise, including features consistent with tamponade‐like physiology. He denied fever, night sweats, or weight loss.

Laboratory finding at the emergency room: complete blood count (CBC) was largely within normal limits, with a white blood cell count of 8.5 × 10³/μL, hemoglobin 13.4 g/dL, and platelet count 297 × 10³/μL. Differential count showed a balanced distribution without overt circulating blasts. Biochemical analysis did not demonstrate hyperuricemia in the available data. However, lactate dehydrogenase (LDH) was elevated (263 U/L), consistent with increased cellular turnover. Inflammatory markers were also elevated, with CRP of 11.3 mg/L. Liver function tests were within normal limits, including ALT (15 U/L), alkaline phosphatase (61.8 U/L), and total bilirubin (0.6 mg/dL). Notably, globulin levels were mildly elevated (34.3 g/L), while albumin remained within the normal range. Patient was promptly started on intravenous fluids and allopurinol as prophylaxis for tumor lysis syndrome, along with dexamethasone according to protocol. Given the presence of significant SVC compression, urgent steroid therapy was initiated; however, no radiotherapy or endovascular stenting were performed at that stage, as the patient showed early clinical stabilization under medical management.

1.2. Radiologic Findings

Chest CT revealed a large lobulated soft tissue mass involving the superior and middle mediastinum with circumferential pericardial involvement. Although exact dimensions were not provided in the original radiology report, the lesion was associated with severe mass effect, including SVC stenosis with contrast reflux, critical narrowing of the brachiocephalic vein, and bilateral airway and vascular compression. A small pericardial effusion was present without radiologic evidence of overt cardiac tamponade, although the patient exhibited tamponade‐like clinical physiology. Due to progressive pleural effusion and cardiorespiratory compromise, a left pigtail catheter was inserted, draining more than 2 L of yellow, slightly turbid fluid and resulting in marked clinical improvement.

The imaging findings were highly suggestive of an aggressive lymphoproliferative malignancy. The differential diagnosis included T‐lymphoblastic lymphoma, primary mediastinal large B‐cell lymphoma, and Hodgkin lymphoma, whereas thymic and germ cell tumors were considered less likely given the diffuse infiltrative pattern and extensive nodal involvement.

1.3. Cytological and Histological Findings

1.3.1. FNA

Imprint smears showed numerous small‐ to medium‐sized blastoid lymphoid cells present singly and in clusters. Cells demonstrated high nuclear‐to‐cytoplasmic ratio, finely dispersed chromatin, inconspicuous nucleoli, frequent mitoses, and apoptotic debris. These features raised suspicion for a precursor lymphoid neoplasm, although morphology alone was not lineage‐specific [6, 8].

1.3.2. Core Biopsy

The biopsy showed complete effacement of lymph node architecture by a diffuse infiltrate of monotonous blastoid cells with brisk mitotic activity and focal single‐file soft tissue infiltration (more details in Supporting Information under Fine‐Needle Aspiration and Core biopsy sections).

1.3.3. Pleural Fluid Cytology and Flow Cytometry

Pleural fluid analysis demonstrated an exudate with 56,000 leukocytes per microliter, with 94% mononuclear cells. Cytologic examination showed numerous atypical blastoid lymphoid cells (Figure 1). Multiparameter flow cytometry identified an aberrant immature T‐cell population. Acute Leukemia screening tube revealed blasts (red) expressing cytoplasmic CD3, CD7, and were negative for surface CD3, CD19, and cytoplasmic CD79a, confirming T‐lineage and excluding B‐cell differentiation. CD34 was negative. Cytoplasmic MPO was negative, excluding myeloid differentiation (Figure 2).

FIGURE 1.

Cytomorphology of T‐cell lymphoblastic lymphoma in pleural fluid. Pleural fluid cytology shows small‐ to medium‐sized blastoid lymphoid cells arranged singly and in loose clusters (Left, 100 μm). High‐power view (right, 25 μm) demonstrates a high nuclear‐to‐cytoplasmic ratio, finely dispersed chromatin, inconspicuous nucleoli, scant basophilic cytoplasm, and frequent mitotic/apoptotic figures (insert). Findings are consistent with a precursor T‐lymphoblastic neoplasm.

FIGURE 2.

Flow cytometric immunophenotyping of pleural fluid using the Acute Leukemia screening tube. Multiparameter flow cytometry performed on pleural fluid demonstrating an aberrant immature T‐cell population. Doublets were excluded by comparing FSC‐H versus FSC‐A (and/or SSC‐H versus SSC‐A), allowing selection of single cells and elimination of aggregates. The blast population (red) is identified on the CD45 vs. SSC plot by dim CD45 expression and low side scatter, distinct from residual lymphocytes (purple). Initial gating of the blasts revealed cytoplasmic CD3, CD7 positivity. Blasts were negative for surface CD3, CD19, and cytoplasmic CD79a, confirming T‐lineage and excluding B‐cell differentiation. CD34 is negative. Cytoplasmic MPO is negative, excluding myeloid differentiation. Overall, this gating strategy identifies a population of immature T‐lineage blasts (cytoCD3⁺, CD7⁺, CD34−/dim, MPO−), consistent with T‐lymphoblastic neoplasia, with limited bone marrow involvement.

CD1a, CD2, Cytoplasmic CD3, CD4, CD5 (dim), CD7, CD8, CD10, nuclear TdT were positive. The cells were negative for surface CD3, cytoplasmic MPO, CD13, CD11C, CD34, CD117, HLADR, and CD56 antigens (Figure 3).

FIGURE 3.

Multiparameter flow cytometry of blast population. Doublets were excluded by comparing FSC‐H versus FSC‐A (and/or SSC‐H versus SSC‐A), allowing selection of single cells and elimination of aggregates. Blasts (red) identified on CD45 versus SSC by dim CD45 and low side scatter, distinct from residual lymphocytes (purple). Following blast gating, cells show cytoplasmic CD3, bright CD7, CD2, CD1a, nuclear TdT, CD10 (aberrant) expression, with absent surface CD3 and HLADR expression. The aberrant S.CD3‐CytoCD3 + S.CD7+ blasts are negative for CD117, CD13, CD11c, and CD56. The immunophenotype is consistent with precursor T‐cell lymphoblastic lymphoma and excludes myeloid or B‐cell lineage differentiation.

In immunohistochemical (IHC) staining, the neoplastic cells were diffusely positive for CD1a, CD2, CD3, CD4, CD5 (weak), CD7, CD8, CD10, TdT, and BCL2, and negative for CD20, CD19, PAX5, CD79a, MPO, CD34, and CD117, supporting an immature T‐cell phenotype and excluding B‐cell and myeloid differentiation.

Importantly, the Ki‐67 proliferation index was extremely high (95%–100%), highlighting the highly proliferative and aggressive nature of the tumor and supporting a diagnosis of lymphoblastic lymphoma in the appropriate morphologic and immunophenotypic context. These findings supported a diagnosis of precursor T‐cell neoplasm. Flow cytometry is considered the gold standard for immunophenotypic characterization of precursor lymphoid neoplasms, particularly in limited cytologic specimens and body fluids [5, 6, 7, 9]. In this case, it confirmed lineage, immaturity, and excluded myeloid differentiation (more details in Supporting Information under Immune Stains section).

1.4. Bone Marrow Evaluation

Bone marrow examination demonstrated approximately 10% blasts by morphological assessment. Multiparameter flow cytometry identified only 3.8% CD34‐positive blast cells and did not detect an aberrant T‐lymphoblastic population (Figure 4). Blast percentage determined by morphology and the proportion of CD34‐positive cells identified by flow cytometry reflect different biological and diagnostic parameters.

FIGURE 4.

Bone marrow flow cytometry demonstrating absence of T lymphoblastic leukemia involvement. (A–B) Bone marrow aspirate smears showing normocellular marrow with preserved trilineage hematopoiesis and orderly maturation of myeloid, erythroid, and megakaryocytic lineages. No morphologic evidence of T‐lymphoblastic infiltration is identified. Findings are consistent with bone marrow remission without overt involvement by T‐cell acute lymphoblastic leukemia. Scale bars: 100 μm (A) and 50 μm (B). (C) Doublets were excluded by comparing FSC‐H versus FSC‐A (and/or SSC‐H versus SSC‐A), allowing selection of single cells and elimination of aggregates. Multiparameter flow cytometry of the bone marrow aspirate shows no expanded blast population on CD45 versus SSC analysis. Gating the surface CD7 versus cytoplasmic CD3 positivity cell population along with their surface CD3 expression revealed a normal immunophenotype of T cells. CD34‐positive cells (red) are within normal limits and lack aberrant immunophenotypic features, with no cytoplasmic CD3 expression. The T‐cell compartment demonstrates a normal immunophenotypic pattern without an immature T‐lymphoblastic population. No pathologic population corresponding to the mediastinal blasts is identified, excluding bone marrow involvement. Residual lymphocytes (purple).

Bone marrow examination showed approximately 10% blasts by morphologic assessment. Flow cytometric analysis identified a 3.8% population of CD34‐positive immature cells. This population showed strong CD34 expression (100%) and predominantly expressed myeloid‐associated markers CD13 (86%) and CD33 (75%), with partial CD117 expression (26%). Most cells demonstrated a CD34 + CD38− phenotype (84%), consistent with an immature hematopoietic progenitor population. A small subset showed CD19 (12%) and CD7 (19%) expression. 10% blasts estimated morphologically and the 3.8% CD34‐positive population detected by flow cytometry likely reflects differences in the populations assessed by the two methodologies, including the contribution of immature hematopoietic cells that cannot be reliably classified as blasts by flow cytometry based on CD34 expression alone. Importantly, flow cytometry did not demonstrate a significant population with the aberrant T‐lymphoblastic phenotype identified in the mediastinal lesion. These findings support the interpretation of borderline bone marrow involvement without definitive immunophenotypic evidence of a significant T‐lymphoblastic population. Taken together, the bone marrow morphology, flow cytometric findings, and clinical presentation support a diagnosis of T‐LBL with limited bone marrow involvement below the diagnostic threshold for T‐ALL. Cerebrospinal fluid examination revealed no evidence of malignant cells.

These procedures were performed before steroid administration to avoid potential treatment‐related alterations in cellular morphology and antigen expression. Molecular characterization was not available, which limits further biologic subclassification.

1.5. Clinical Course and Follow‐Up

Six months from diagnosis, the patient achieved complete remission and is currently in the re‐induction phase corresponding to Week 19 of the GMALL protocol. Clinically, he reports feeling well without significant symptoms. Laboratory evaluation demonstrates hemoglobin of 10.4 g/dL, platelet count of 187 × 10³/μL, and leukocyte count of 7.5 × 10³/μL with an absolute neutrophil count of 6.1 × 10³/μL. Renal function and bilirubin levels are within normal limits. Overall, the patient is tolerating therapy well and remains in complete remission at this stage of treatment. Follow‐up bone marrow aspiration demonstrated active trilineage hematopoiesis with orderly myeloid maturation, normoblastic erythropoiesis, and no cytological evidence of residual T‐cell acute lymphoblastic leukemia. No increase in lymphoblasts was identified morphologically, while flow cytometry detected a small population of 1.86% T‐lymphoblasts, consistent with T‐ALL in remission. Current PET‐CT imaging showed no pathological FDG uptake, with no evidence of metabolically active lymphoproliferative disease above or below the diaphragm.

2. Discussion

T‐LBL commonly presents with a mediastinal mass and may be associated with life‐threatening complications, including SVC syndrome, airway compression, and large pleural or pericardial effusions [3, 4]. In such settings, rapid establishment of the diagnosis is critical to enable timely initiation of therapy. However, the cytomorphologic features of lymphoblasts may overlap with those of other high‐grade lymphoid neoplasms, making ancillary studies essential for accurate classification [6, 9].

The immunophenotypic profile in our patient was consistent with cortical T‐cell acute lymphoblastic leukemia/lymphoma (cT‐ALL/LBL), as demonstrated by strong CD1a expression. According to the European Group for the Immunological Characterization of Leukemias (EGIL) classification, CD1a is the defining marker of the cortical thymocyte stage of T‐cell differentiation and distinguishes cortical T‐ALL from pro‐T, pre‐T, and mature T‐ALL subtypes [10, 11]. Given that CD1a is physiologically expressed on cortical thymocytes, its presence reflects the thymic origin of the neoplastic clone. Consistent with this biology, cortical T‐ALL/LBL is frequently associated with thymic involvement and anterior mediastinal masses. The extensive mediastinal mass observed in our patient, including circumferential pericardial involvement and compression of major mediastinal structures, is therefore in keeping with the cortical phenotype. This association is observed across both T‐ALL and T‐LBL, supporting the concept that these entities represent different clinical manifestations of the same thymic precursor neoplasm [10, 11]. In the present case, flow cytometry provided rapid and definitive immunophenotypic characterization of the pleural fluid, demonstrating an immature T‐cell population expressing TdT and T‐lineage markers. This allowed prompt confirmation of pleural involvement by T‐LBL and facilitated timely treatment initiation. The patient was subsequently treated according to a pediatric‐inspired adult acute lymphoblastic leukemia protocol (GMALL) and is currently in remission.

An additional diagnostic challenge arose from the bone marrow findings. Morphologic examination identified approximately 10% blast‐like cells, raising concern for possible leukemic involvement. However, multiparameter flow cytometry did not detect an aberrant T‐lymphoblastic population and identified only a small population of CD34‐positive precursor cells without abnormal T‐lineage antigen expression. This phenomenon highlights a recognized limitation in the evaluation of T‐LBL, in which bone marrow involvement may be focal, patchy, or below the threshold of detection in aspirate‐based studies. Morphologic blasts are defined based on cytologic features, whereas flow cytometric quantification of CD34‐positive cells identifies a specific immunophenotypic subset that may not encompass the entire blast population.

The WHO classifies T‐LBL and T‐ALL as a single entity (T‐lymphoblastic leukemia/lymphoma), recognizing their shared biological features [12, 13]. When a distinction is made clinically, T‐ALL is defined by ≥ 25% blasts in the bone marrow [12, 14]. The International Consensus Classification (ICC) similarly recognizes these as related entities but does not mandate a specific bone marrow threshold for all diagnostic purposes [15]. In clinical practice and research, a ≥ 20% bone marrow blast threshold is also commonly used to define T‐ALL, particularly when following general acute leukemia diagnostic criteria [16]. One recent study defined “T‐ALL was diagnosed using a bone marrow biopsy, which revealed ≥ 20% lymphoblasts,” while “T‐LBL was defined by dominant nodal/extranodal disease with < 20% marrow blasts” [16]. The distinction is often based on the dominant clinical presentation rather than a strict bone marrow cutoff. T‐LBL typically presents with a mediastinal mass and limited bone marrow involvement (approximately 17%), whereas T‐ALL is characterized by significant bone marrow infiltration and leukemic manifestations [16]. In borderline cases, diagnostic classification may follow institutional practice, with T‐ALL defined by the presence of significant marrow involvement regardless of extramedullary disease [8, 16, 17]. Accurate classification requires integration of clinical presentation, morphology, immunophenotyping, and imaging findings rather than reliance on a single diagnostic modality. In the present case, the predominant mediastinal and pleural disease, together with the absence of a detectable aberrant T‐lymphoblastic population in the bone marrow, supported classification as T‐LBL with limited marrow involvement below the diagnostic threshold for T‐ALL.

This case illustrates the complementary roles of cytomorphology and flow cytometry in the evaluation of lymphoblastic neoplasms. Most importantly, it highlights a potential diagnostic pitfall in which morphologic suspicion of marrow involvement may lead to overclassification as leukemic disease. Integration of flow cytometric findings with clinical and pathologic features was essential for accurate disease classification and staging, ultimately guiding appropriate patient management.

Author Contributions

Chen Glait‐Santar conceived, designed, collected, and compiled the clinical data. Yakir Moshe assisted in clinical data analysis and manuscript revision. Itzhaki‐Alfia Ayelet added important insights and reviewed the methodology. Chen Glait‐Santar collection and analysis of data, writing and revision of the manuscript. Ben‐Zion Katz critical revision of the manuscript for important intellectual content. All authors reviewed the draft and approved the final version for submission.

Funding

The authors have nothing to report.

Consent

Exempt by Tel Aviv Medical Center IRB committee.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.