Cytotoxic Cell Antigen Expression in Anaplastic Large Cell Lymphomas of T- and Null-Cell Type and Hodgkin's Disease: Evidence for Distinct Cellular Origin

Abstract

Anaplastic large cell lymphoma (ALCL) is composed of large, frequently bizarre, cells of T- or null-cell phenotype that show a preferential sinusoidal growth pattern and consistent CD30 positivity. Whether these tumors represent a single entity or several, and what the exact cell origin, is controversial. Recently, granzyme B, a cytotoxic granule component, was reported in a small percentage of ALCL, suggesting that some cases may originate from cytotoxic lymphocytes. To further investigate this possibility, we performed an immunohistochemical study of 33 ALCLs of T- and null-cell type, using monoclonal antibodies to cytotoxic cell-associated antigens, including CD8, CD56, CD57, and the cytotoxic granular proteins perforin and TIA-1. In addition, CD4 expression was also evaluated. ALCL cases included 27 classical systemic forms and variants, 3 primary cutaneous (PC) forms, and 3 acquired immunodeficiency syndrome-associated forms. Cytotoxic antigen expression was also studied in 51 cases of Hodgkin's disease (HD) and 17 large B-cell lymphomas (LBCLs) with anaplastic cytomorphology and/or CD30 positivity. We found that 76% of ALCLs, representing all subtypes except the PC forms, expressed either TIA-1, perforin, or both proteins. Expression of TIA-1 and perforin were highly correlated (P < .001). On the basis of their immunophenotypic profiles, several subtypes of cytotoxic antigen positive and negative ALCL could be recognized. Fifty-five percent of ALCLs (18 of 33) displayed an immunophenotypic profile consistent with cytotoxic T cells. Six cases expressed cytotoxic granular proteins in the absence of lineage specific markers, and one case expressed both T-cell – and natural killer cell–like markers. These 7 cases (21%) were placed into a phenotypic category of cytotoxic lymphocytes of unspecified subtype. Twenty-four percent (8 cases) of ALCLs were cytotoxic granule protein negative. All but one of these displayed a T-cell phenotype. Cytotoxic granule protein expression did not correlate with the presence of the NPM-ALK fusion transcript. Only 10% of the 51 HD cases were found to be TIA-1+, and none expressed perforin. Cytotoxic antigen expression was absent in LBCL. The expression of cytotoxic granule proteins in the majority of ALCL implies a cytotoxic lymphocyte phenotype and suggests that most cases originate from lymphocytes with cytotoxic potential. Furthermore, the demonstration of cytotoxic cell related proteins may be a useful addition to the current panel of antibodies used to distinguish ALCL, HD, and anaplastic LBCL.

Primary CD30‐positive cutaneous T‐cell lymphomas and lymphomatoid papulosis frequently express cytotoxic proteins

“…Although recent studies agree on the expression of cytotoxic proteins in systemic ALCL, [18][19][20] the results concerning primary cutaneous ALCL are still controversial. [20][21][22] Indeed, in keeping with our data, Foss et al found perforin transcripts in 3/3 cases of cutaneous CD30þ ALCL 21 and Kummer et al found granzyme B protein expression in 10/14 CD30þ cutaneous ALCL. 22 In contrast, Krenacs et al reported no expression of TIA-1 and perforin proteins in three cases of cutaneous CD30þ ALCL.…”

“…Expression of perforin was also studied in all but one case, using ascites fluid of a rat anti-mouse IgG2 monoclonal antibody denoted P1-8 (kindly provided by Hideo Yagita, Juntendo University School of Medicine, Tokyo, Japan), which had been shown to cross react with human perforin. 20,27…”

“…Some cases, with only very rare scattered positive, apparently neoplastic cells (< 5%) were considered as negative, according to previous report. 20 Cytocentrifuged cells from YT2C2 and YTIndi NK cell lines as well as one case of NK cell lymphoma expressing TIA-1, granzyme B and perforin were used as positive controls.…”

“…18,20,21 Concerning cutaneous CD30þ lymphoproliferations, controversial results have been reported. 20,21 However, there is a recent systematic study on CTCL showing that CD30þ ALCL and LP express a cytotoxic phenotype. 22 In that study, 22 the TIA-1 and granzyme B expression was analysed whereas the perforin expression was not investigated.…”

A case of systemic anaplastic large cell lymphoma with ‘Hodgkin-like’ appearance and skin involvement mimicking lymphomatoid papulosis

“…T cell receptor phenotype of systemic ALCL are various; CD4 50-80% rather than CD8 20-45%. 1 In contrast to systemic ALCL, dominant phenotype of CD30+ lymphoproliferative disorders of skin is determined with CD4 2 . CD8+ case of primary cutaneous ALCL is rare, sporadically reported.…”