ALK-positive anaplastic large cell lymphoma with PAX-5 expression: report of a case

ALK-positive anaplastic large cell lymphoma (ALCL) is a well-defined entity included in the 2008 WHO classification, comprising about 5% of non-Hodgkin lymphomas and 10 to 30% of childhood lymphomas. Although lacking several T-cell antigens, it is generally considered a T-cell derived neoplasm. Since both ALCL and Hodgkin lymphoma (HL) strongly express CD30, and they can show overlapping morphological features, the B-cell specific marker PAX5 is widely used for distinguishing the two entities. PAX5 is considered the most specific B-cell marker, and is defined as "the guardian of B-cell identity". We report an exceptional case of ALK-positive ALCL with PAX5 expression. A 38-years old patient presented with fever and generalized lymphadenopathy lasting for two months at the time of biopsy. A basic cytofluorimetric analysis was performed, showing the presence of a large cell population with dim CD5 and CD4 expression. Histopathological evaluation revealed an effaced lymph node architecture, due to the presence of a neoplastic infiltration by medium-sized cells in wide aggregates; occasionally larger cells with "hallmark" morphology were present; intrasinusoidal involvement was evident. These atypical cells showed a strong staining for CD30, as well as perforin, while granzyme B was only focally expressed. Other T-cell markers (CD2, CD3, CD5, CD7, ZAP70) and B-cell markers (CD20, CD79a) were negative, with the notable exception of PAX5, which showed a dim but specific staining, similar to that observed in HL. ALK was diffusely expressed in neoplastic cells, with a predominantly cytoplasmic granular staining. Double chromogenic stains and immunofluorescence were also performed to confirm the findings. ALK-positive diffuse large B-cell lymphoma was ruled out because this case didn't show the typical immunoblastic/plasmablastic morphology, didn't express plasma cell markers and was diffusely CD30-positive. The patient was unfortunately lost at follow-up. Four cases of PAX5-positive ALCL were recently described, of which only one was ALK-positive, diagnosed on a vertebral lesion. A previous study had identified three PAX5-positive ALK-negative ALCL, but no ALK-positive cases. To our knowledge, this case is the first description of lymph node involvement by PAX5-positive and ALK-positive ALCL. Our data show that the morphological presentation is totally overlapping with classical ALCL; we also confirm that PAX5 can be rarely expressed in ALCL and should not be taken as a final proof of the B-cell origin of a neoplastic population.