A suggested immunohistochemical algorithm for the classification of T-cell lymphomas involving lymph nodes

Abstract

T-cell lymphomas are a heterogeneous group of neoplasms derived from mature T lymphocytes. These neoplasms are uncommon and usually diagnostically challenging. The focus of this article is to suggest an immunohistochemistry-based, practical approach to assist in the diagnosis of nodal T-cell lymphomas. These neoplasms fall into two major groups: those with many CD30+ tumor cells (group A) and neoplasms that are negative or show only partial expression of CD30 (group B). The differential diagnosis of group A neoplasms mainly includes ALK+ anaplastic large-cell lymphoma (ALCL), ALK-negative ALCL, mycosis fungoides with CD30+ large-cell transformation, adult T-cell leukemia/lymphoma, extranodal T-cell lymphomas involving lymph nodes (usually regional), and peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS). Group B neoplasms also include two groups based on the presence or absence of T follicular helper (TFH) markers. Those neoplasms expressing at least 2 TFH markers include angioimmunoblastic T-cell lymphoma, nodal PTCL with a TFH phenotype, and follicular T-cell lymphoma. Neoplasms expressing ≤1 TFH marker can be further subdivided based on the expression of CD8 and cytotoxic markers and mainly include PTCL-NOS and a series of unusual subsets including primary Epstein-Barr virus–positive nodal natural killer/T-cell lymphoma, PTCL-NOS with a cytotoxic immunophenotype, and γ/δ T-cell lymphomas. Using this algorithmic approach, we suggest that the pathologist can establish a diagnosis for most nodal T-cell lymphomas encountered in daily practice.