Extracavitary KSHV-associated large B-Cell lymphoma: a distinct entity or a subtype of primary effusion lymphoma? Study of 9 cases and review of an additional 43 cases.

Primary effusion lymphoma (PEL) is a rare form of aggressive B-cell lymphoma in HIV patients, which typically presents with lymphomatous effusions in the body cavities without forming mass lesions. PEL is associated with Kaposi sarcoma-associated herpesvirus (KSHV) (also called human herpesvirus-8) with distinct clinical and pathologic features. Rare cases of KSHV-associated large B-cell lymphoma (KSHV-LBL) have been observed in the lymph nodes or extranodal sites without lymphomatous effusions during the course of disease. KSHV-LBL is generally similar to classic PEL on the basis of the clinical presentation (HIV(+) male), morphology (immunoblastic, plasmablastic, or anaplastic), immunophenotype (CD45(+), CD20(-), CD79a(-), CD30(+), CD138(+), and EMA(+)), presence of Epstein-Barr virus infection, and clonal immunoglobulin gene rearrangements. However, it is not clear whether KSHV-LBL is a distinct entity or represents part of the spectrum of classic PEL; in particular, there is no consensus diagnostic term for KSHV-LBL. In this study, we investigated the clinicopathologic features of 9 cases of KSHV-LBL from our files. An additional 43 such cases and 84 cases of classic PEL from the English literature were reviewed and compared with each other. In contrast to the classic PEL, KSHV-LBL had a very significant lower expression of CD45 (74% vs. 94%, P=0.004) but significant higher expression of CD20 (17% vs. 5%, P=0.04) and CD138 (70% vs. 38%, P=0.05). KSHV-LBL also had slightly higher positivity of CD79a (23% vs. 5%, P=0.13) and immunoglobulin light chain expression, although the difference was not statistically significant [κ chain (12% vs. 0%) and λ chain (31% vs. 21%)]. The expressions of EMA and CD30 were slightly lower in KSHV-LBL compared with those observed in PEL (57% vs. 75% and 63% vs. 76%, respectively). Interestingly, 29% (10/34) of cases of KSHV-LBL revealed aberrant CD3 expression, which may mislead to a diagnosis of T-cell lymphoma, particularly anaplastic large cell lymphoma in combination with the anaplastic morphology and expression of CD30 and EMA. Although KSHV-LBL shows different clinical presentations and some variations in immunophenotype from classic PEL, it is still uncertain, on the basis of our findings, whether it is justifiable to separate them as 2 distinct entities. Nevertheless, we feel it is necessary to have a consensus diagnostic term, and we recommend a tentative one as "KSHV-associated large B-cell lymphoma (KSHV-LBL)" to replace many different names previously used.