ALCL is a rare disease in children, accounting for 10–15% of all childhood NHL (11
). The majority of ALCLs in children and adolescents are ALK-positive. In our cohort, the male to female ratio was 2:1. Previous studies demonstrated that there may be a male predominance, particularly in ALK(+) cases, in which the male to female ratio is ~3:1 (12
Histologically, several ALCL variants have been described. Morphological variants of ALCL include the following types: common, lymphohistiocytic, small cell and rare subforms, including neutrophil-rich types (13
). Of these variants, the common, lymphohistiocytic and small cell types are the most frequently encountered. The ‘horseshoe’ or ‘wreath’ cell is considered the cytologic hallmark of this disease. Histological characteristics may have a high potential for future risk stratification and treatment (14
Since ALCL demonstrates a broad morphological spectrum, definitive diagnosis and differential diagnosis of ALCL from other forms of lymphoma and reactive lymphadenopathy are difficult. Unlike other lymphomas, the tumor masses comprise normal reactive T cells and neoplastic cells in ALCL and in a number of cases only a careful search is likely to reveal the presence of neoplastic cells. Diagnosis of common type ALK(+) ALCL has become straightforward owing to the widespread availability of reliable anti-ALK antibodies. However, ALK(+) ALCL includes a morphological spectrum with small cell and lymphohistiocytic variants that represent ~10–20% of cases and is easily confused with reactive lymphadenopathy (15
). In our study, the paraffin sections of two cases of lymphohistiocytic ALCL demonstrated varying degrees of necrosis, which causes the misdiagnosis of inflammation. Scattered neoplastic cells were observed with a careful search and recognized by an immunohistochemical stain of CD30- and ALK-positive cells. In a neutrophil-rich type patient who was misdiagnosed with an inflammatory disease in a local hospital, the polymorphous neoplastic cells with irregular nuclei dispersed and were intermixed with a large number of mature neutrophils. CD30- and ALK-positive results suggest a diagnosis of ALCL in this case. The neutrophil-rich variant of anaplastic large cell lymphoma (NR-ALCL) is a rare type of NHL. Diagnosis by lymph node biopsy is difficult owing to the rarity of this tumor, its resemblance to HD and other NHL, its similarity to an infectious process and its occasional confusion with metastatic carcinoma and melanoma (16
). Numerous ALCL cases present lymph node sclerosis and may be misdiagnosed as HD in which CD30 is also positive (17
). As ALCL is a peripheral T-cell-derived malignancy, an immunohistochemical stain for T cell-specific markers, CD15, EMA and ALK protein may be useful in the differential diagnosis between ALCL and HD. In addition, the nodal ALCL should be differentiated from metastatic undifferentiated carcinoma, malignant melanoma, Langerhans cell histiocytosis and soft tissue sarcoma. Therefore, lymphadenopathy in children with abnormal cells in a background of inflammation should be considered as ALCL. Immunohistochemical staining of ALK and CD30 is a useful approach to confirm the diagnosis of ALCL.
Flow cytometry provides rapid analysis of multiple characteristics of separate cell populations based on their sizes, cytoplasmic characteristics and antigens expressed. Lymph node specimens are now routinely submitted for flow cytometric analysis in patients with suspected lymphoma and are considered to be standard practice in a number of centers. By comparison, relatively few studies have addressed the use of flow cytometry in the evaluation of ALCL. In the present study, we analysed 15 samples from patients with ALK(+) ALCL with FCI, 13 of which were consistent with pathological results. The sensitivity and specificity of FCI for diagnosing ALCL were 86.7 and 100%, respectively. The samples were CD30-positive and immunophenotyically aberrant with respect to T-cell antigen expression (CD2, CD3, CD4, CD5 and CD7), which is in concordance with the previous study (9
). CD3 antigen was often expressed at a dim intensity compared with background normal T cells and such a marked contrast in the same FCI plot contributes to the diagnosis of ALCL. The frequent expression of CD25 in ALCL suggests that this antigen is a potentially useful marker in the immunophenotypic diagnosis of ALCL and a potential therapeutic target (18
ALCL should also be differentiated from HD when diagnosed using FCI. Unlike ALCL in which the neoplastic cells dispersed in lymph node tissue, the Reed-Sternberg (RS) cells of HD usually scatter, where they constitute <1% and frequently <0.01% of the total cells (19
). Therefore, routine analysis of 10,000 events by FCI is not able to bring a population that represents 0.01% of the total cells well within range of sensitivity of clinical cytometry, thus FCI is not able to establish a diagnosis in routine practice. By comparison, neoplastic cells have a relatively higher proportion of total cells in lymph node tissue with ALCL and are rapidly detected with FCI. We suggest that when CD45-, CD30- and T-cell marker-positive neoplastic cells constitute at least 5% of the total cells, a diagnosis of ALCL may be made using FCI.
Unlike other PTCL-NOS in which the neoplastic cells are usually small- to medium-sized, the majority of ALCL cells are large and show increased FSC and SSC. Therefore they are concentrated largely or entirely outside the lymphocyte gate, similar to monocytes or granulocytes. Conventional gating for lymphoma cells with low FSC and SSC may lead to false-negativity and careful analysis is required since the proportion of the aberrant population with a diagnostic value may be extremely small. Flexible gating strategies are important in the diagnosis of ALCL. In this study, 2 specimens (lymphohistiocytic type) with extensive necrosis were not confirmed as ALCL by flow cytometry due to the lack of analyzable cells present in the specimens. Similarly, Muzzafar et al
) were unable to identify neoplastic cells by flow cytometry in 4 of 23 (17.4%) adults with ALCL. There are a number of technical factors and potential pitfalls that make ALCL diagnosis by flow cytometry particularly challenging. False-negative FCI results in ALCL may be due to the necrosis and apoptosis commonly associated with such tumors and to the fragility of the large atypical neoplastic cells, which may be easily disrupted during sample processing (20
). Sampling issues may also be involved due to the neoplastic cells focally distributed in the lymph node (21
). Clinical application of FCI for ALCL may be expanded, with the exception of diagnosing ALCL in lymph node biopsy specimens. Our previous study revealed that in pleural effusion from a case with ALCL, which was considered negative for ALCL by morphological examination, FCI detected a minor proportion (9.3%) of aberrant T-cell population with high FSC/SSC, a positive expression of CD4, CD7, CD2, CD45RO and CD30 and a negative expression of CD5 and CD69 (22
). Moreover, Damm-Welk et al
considered that FCI using antibodies against ALK and CD30, sensitively and specifically detects circulating ALCL cells in bone marrow or blood (23
). Therefore, FCI holds a clear advantage over morphological examination in such circumstances.
In summary, ALK(+) ALCL is a distinct subset of NHL morphologically and immunophenotypically. FCI may be used as an adjunct to histopathological examination for reliable diagnosis of pediatric ALCL with high specificity and sensitivity. It is rapid and suitable for emergency situations, allowing for therapeutic decisions to be made promptly. However, flexible gating strategies and careful analysis are required to identify neoplastic cells with FCI.