Apoptosis and cell cycle-related genes and proteins in classical Hodgkin lymphoma: application of tissue microarray technique.

The etiology and pathogenesis of Hodgkin lymphoma (HL) are not yet known. There are implications of genes involved in programmed cell death (apoptosis), and there have been repeated suggestions of an association with Epstein-Barr virus (EBV). The aim of this study was to investigate the protein expression patterns of key cell cycle-related genes, together with evidence of apoptosis and EBV status, in relation to clinical stage in HLs. A double immunohistochemical and in situ hybridization technique was used to detect the expression of bcl-2, p53, retinoblastoma (Rb), p21, Ki67 (MIB 1), and topoisomerase IIalpha (TopoIIalpha), together with latent membrane protein-1 and EBER for EBV status and TdT-mediated dUTP-FITC nick end-labeling (TUNEL) as a measure of apoptosis, on tissue microarray sections of 62 cases of classic HL (35 NS, 17 MC, 8 LR, and 2 LD). A panel of phenotypic markers was used to facilitate recognition of Hodgkin and Reed-Sternberg (H-RS) cells: CD3, CD20, CD30, CD15, and EMA. The H-RS cells of 62 classic Hodgkin lymphomas were bcl-2-positive in 35 cases (56.45%), p53-positive in 14 (22.58%), and positive for both EBV latent membrane protein-1 and EBER in 37 (59.68%); there was complete concordance of results for EBV by both procedures. No correlation was found between expression of bcl-2, p53, or EBV markers in H-RS cells and clinical stage (P > 0.05). Expression of Rb, Ki67, p21, and TopoIIalpha did, however, show significant differences with clinical stage. Expression of Rb and p21 in CD30-positive H-RS cells decreased with more advanced stage (P < 0.001). In contrast, Ki67 and ToPoIIalpha expression increased with later stage (P < 0.01). No correlation was found between expression of any of these markers in H-RS cells and the subtypes of nodular sclerosis HL, mixed cellularity HL, and LRHL (P > 0.05). TUNEL was found in the nonneoplastic cellular background in all cases and in H-RS cells in only 10 of 62 cases (16.12%) (8 nodular sclerosis HL, 1 mixed cellularity HL, and 1 LRHL). There was a significant correlation between high expression of bcl-2 and a low score by TUNEL (P < 0.05). These data are consistent with the notion that overexpression of bcl-2 may be linked to blockage of apoptosis-mediated death of H-RS cells in classic HL. Abnormal expression of p53-related protein may not play a major role in HL, because it is present in H-RS cells in only a minority of cases. Increased expression of Ki67 and TopoIIalpha by H-RS cells is significantly associated with advanced stage and may indicate aggressive disease. Adverse clinical outcome in HL also is associated with loss of Rb and p21 protein expression, consistent with the possible roles of Rb and p21 in inhibition of the growth of H-RS cells. Within the limitations of the methods used, almost two thirds of cases of HL provide evidence of an association with EBV. The tissue microarray technique is valuable not only for examination of large numbers of cases of a disease by a complex panel of markers but also potentially as a control for staining quality in immunohistochemistry and in situ hybridization.