DBA.44 Positivity Is Predictive of Hairy Cell Leukemia Variant over Splenic Marginal Zone Lymphoma When Classic Hairy Cell Leukemia Is Excluded
Devon S Chabot-Richards, Micaiah H Evans, Orrin Myers, Qian-Yun Zhang, Kathryn Foucar. University of New Mexico, Albuquerque, NM
Background: Hairy cell leukemia (HCL), hairy cell leukemia variant (HCL-v), and splenic marginal zone lymphoma (SMZL) have distinct clinical presentations, treatment options, and outcomes, however, there can be significant morphologic and immunophenotypic overlap between these entities. We sought to discover the relative importance of various morphologic, immunohistochemical, and flow cytometric diagnostic criteria in the diagnosis of these entities.
Design: Archive files from our database were searched for cases diagnosed as B-cell chronic lymphoproliferative disorder, not otherwise specified, HCL, HCL-v, SMZL, MCL, CLL or LPL. The diagnostic slides and IHC for CD20, DBA.44, TRAP, TIA-1, annexin A1, CD123, cyclin D1, TCL-1, CD25, and CD5 and flow cytometric data were reviewed independently by four blinded reviewers. A random subset of cases was reviewed twice to assess intra-reviewer consistency. The reviewers' interpretation was compiled. Log-binomial regression with a generalized estimating equations approach to estimating robust standard errors was used to assess associations between consensus diagnoses and IHC and flow cytometry results. IHC intensity and flow cytometry results were given ordinal coding with larger values for more intense or stronger positive, respectively.
Results: We found 7 cases of HCLV, 8 cases of SMZL, and 23 cases of HCL. The diagnoses of HCLv and SMZL were confirmed on 2 and 3 splenectomy cases, respectively. The other diagnoses are based on blood, bone marrow and immunophenotype features similar to spleen-confirmed cases. HCL-v was associated with leukocytosis, bone marrow sinusoidal and interstitial infiltrates similar to those seen in HCL, and positive staining for DBA.44 (5/7 cases) and TRAP and negative CD25, cyclin D1, annexin A1, and TCL-1. SMZL was associated with interstitial and nodular bone marrow infiltrate and positive TRAP and negative DBA.44 (1/7 cases), CD25, cyclin D1, annexin A1, and TCL-1. HCL is strongly associated with positive annexin A1, CD123, Cyclin-D1, and DBA.44. There were no significant differences among the reviewers in the final diagnoses.
Conclusions: Although there is considerable overlap in the WHO diagnostic criteria for these entities, careful application of diagnostic criteria with a panel of immunohistochemical stains and flow markers can allow consistent diagnosis of them. DBA.44 positivity in particular is predictive of HCL-v over SMZL when HCL is excluded.
Monday, March 19, 2012 1:00 PM
Poster Session II # 226, Monday Afternoon