A comprehensive immunophenotypic marker analysis of hairy cell leukemia in paraffin-embedded bone marrow trephine biopsies--a tissue microarray study.

Hairy cell leukemia (HCL) is an uncommon B cell lymphoproliferation characterized by a unique immunophenotype. Due to low number of circulating neoplastic cells and 'dry tap' aspiration, the diagnosis is often based on BM trephine biopsy. We have performed a consecutive immunohistochemical analysis to evaluate diagnostic usefulness of various HCL markers (CD11c, CD25, CD68, CD103, CD123, CD200, annexin A1, cyclin D1, DBA.44, HBME-1, phospho-ERK1/2, TRAP, and T-bet) currently available against fixation resistant epitopes. We analyzed tissue microarrays consisting of samples gained from 73 small B-cell lymphoma cases, including hairy cell leukemia (HCL) (n = 32), HCL variant (HCL-v) (n = 4), B-cell chronic lymphocytic leukemia (B-CLL) (n = 11), lymphoplasmacytic lymphoma (LPL) (n = 3), mantle cell lymphoma (MCL) (n = 10), splenic diffuse red pulp small B cell lymphoma (SDRPL) (n = 2), splenic B cell marginal zone lymphoma (SMZL) (n = 8), and splenic B cell lymphoma/leukemia, unclassifiable (SBCL) (n = 3) cases. The HCL cases were 100% positive for all but 2 (DBA.44 and CD123) of these markers. Annexin A1 showed 100% specificity and accuracy, which was followed by CD123, pERK, CD103, HBME-1, CD11c, CD25, CD68, cyclin D1, CD200, T-bet, DBA.44, and TRAP, in decreasing order. In conclusion, our results reassured the high specificity of annexin A1 and pERK, as well as the diagnostic value of standard HCL markers of CD11c, CD25, CD103, and CD123 also in paraffin-embedded BM samples. Additional markers, including HBME-1, cyclin D1, CD200, and T-bet also represent valuable tools in the differential diagnosis of HCL and its mimics.

Monoclonal antibody HBME-1 reacts with a minor subset of B cells with villous surface and can be useful in the diagnosis of hairy cell leukemia and other indolent lymphoproliferations of villous B lymphocytes.

The Hector Battifora mesothelial epitope-1 (HBME-1) monoclonal antibody has been generated against human mesothelioma cells and recognizes a biochemically unknown membrane epitope. We have accidentally found that the HBME-1 reacts with scattered lymphocytes showing villous surface in hyperplastic lymphoid tissue. To evaluate its reactivity pattern, we have performed a consecutive immunohistochemical study in nonneoplastic bone marrow and lymphoid samples (n = 40), as well as in malignant lymphoproliferations (n = 427), including hairy cell leukemia (HCL) (n = 72), HCL variant (HCL-v) (n = 13), splenic diffuse red pulp small B cell lymphoma (SDRPL) (n = 8), splenic B cell marginal zone lymphoma (SMZL) (n = 59), and splenic B cell lymphoma/leukemia, not further classifiable on bone marrow morphology (SBCL) (n = 37) cases. The staining pattern of HBME-1 was compared to DBA.44. HBME-1(+) villous lymphocytes were constantly detected in low number in nonneoplastic lymphoid tissues. With multicolor immunofluorescence staining, HBME-1(+) lymphocytes showed a CD20(+)/CD79a(+)/IgM(+) B cell phenotype. In B cell lymphoproliferations of villous lymphocytes, HBME-1 reactivity was demonstrated in 96 % of HCL, 39 % of HCL-v, 50 % of SDRPL, 12 % of SMZL, and 19 % of SBCL cases. Nodal and extranodal marginal zone lymphoma cases were positive in 12 % of the cases. A small minority (4 %) of the other B cell lymphomas and no T cell lymphoma revealed tumor cell reactivity with HBME-1. In conclusion, our study has established that HBME-1 reacts with a minor subset of B lymphocytes and a small proportion of B cell lymphomas, which has not been described previously. We suggest that HBME-1 can be a useful marker in the diagnosis of HCL and other indolent lymphoproliferations of villous B lymphocytes.