An algorithmic approach to the diagnosis of NK and T cell lymphomas

Lymphomas of natural killer (NK) and T cell lineages are uncommon disorders, although as a group they are more usually encountered in Asia compared to Western populations. In part due to their rarity, diagnosis and classification of T cell lymphomas often pose a challenge to clinicians and pathologists. Although there are morphological features that are characteristic of certain subtypes, correct classification of NK and T cell neoplasms relies heavily on the immunophenotype. With few exceptions, non-random genetic alterations such as translocations are less often seen in T cell neoplasms, adding to the diagnostic difficulty. Given these limitations, pathological diagnosis and classification of NK and T cell lymphomas are anything but straightforward. This paper attempts to present a practical algorithmic approach for the general pathologist who is confronted with these neoplasms.