Immunocompromised individuals are at high risk for life-threatening diseases, especially those caused by cytomegalovirus (CMV), Epstein-Barr virus (EBV) and adenovirus. Conventional therapeutics are primarily active only against CMV, and resistance is frequent. Adoptive transfer of polyclonal cytotoxic T lymphocytes (CTLs) specific for CMV or EBV seems promising, but it is unclear whether this strategy can be extended to adenovirus, which comprises many serotypes. In addition, the preparation of a specific CTL line for each virus in every eligible individual would be impractical. Here we describe genetic modification of antigen-presenting cell lines to facilitate the production of CD4(+) and CD8(+) T lymphocytes specific for CMV, EBV and several serotypes of adenovirus from a single cell culture. When administered to immunocompromised individuals, the single T lymphocyte line expands into multiple discrete virus-specific populations that supply clinically measurable antiviral activity. Monoculture-derived multispecific CTL infusion could provide a safe and efficient means to restore virus-specific immunity in the immunocompromised host.

Plasmablastic lymphoma is an aggressive neoplasm that shares many cytomorphologic and immunophenotypic features with plasmablastic plasma cell myeloma. However, plasmablastic lymphoma is listed in the World Health Organization (WHO) classification as a variant of diffuse large B-cell lymphoma. To characterize the relationship between plasmablastic lymphoma and plasmablastic plasma cell myeloma, we performed immunohistochemistry using a large panel of B-cell and plasma cell markers on nine cases of plasmablastic lymphoma and seven cases of plasmablastic plasma cell myeloma with and without HIV/AIDS. The expression profiles of the tumor suppressor genes p53, p16, and p27, and the presence of Epstein-Barr virus (EBV) and human herpes virus type 8 (HHV-8) were also analyzed. All cases of plasmablastic lymphoma and plasmablastic plasma cell myeloma were positive for MUM1/IRF4, CD138, and CD38, and negative for CD20, corresponding to a plasma cell immunophenotype. PAX-5 and BCL-6 were weakly positive in 2/9 and 1/5 plasmablastic lymphomas, and negative in all plasmablastic plasma cell myelomas. Three markers that are often aberrantly expressed in cases of plasma cell myelomas, CD56, CD4 and CD10, were positive in 5/9, 2/5, and 6/9 plasmablastic lymphomas, and in 3/7, 1/5, and 2/7 plasmablastic plasma cell myelomas. A high Ki-67 proliferation index, overexpression of p53, and loss of expression of p16 and p27 were present in both tumors. No evidence of HHV-8 infection was detected in either neoplasm. The only significant difference between plasmablastic lymphoma and plasma cell myeloma was the presence of EBV-encoded RNA, which was positive in all plasmablastic lymphoma cases tested and negative in all plasma cell myelomas. In conclusion, most cases of AIDS-related plasmablastic lymphoma have an immunophenotype and tumor suppressor gene expression profile virtually identical to plasmablastic plasma cell myeloma, and unlike diffuse large B-cell lymphoma. These results do not support the suggestion in the WHO classification that plasmablastic lymphoma is a variant of diffuse large B-cell lymphoma.

Recent studies with cDNA microarrays showed that diffuse large B-cell lymphoma (DLBCL) cases with gene expression profiles similar to germinal center (GC) B cells had much better prognosis than DLBCL cases with gene expression profiles resembling activated B cells. The goal of the current study is to evaluate if using a panel of GC B-cell (CD10 and Bcl-6) and activation (MUM1/IRF4 and CD138) markers by immunohistochemistry defines prognosis in patients with de novo DLBCL. Immunohistochemical stains for the above markers were performed on paraffin-embedded tissues from 42 de novo DLBCL patients. Median follow-up in all patients was 41 months (range, 1-103 months) and in surviving patients was 65 months (range, 14-103 months). These cases could be classified into three expression patterns: GC B-cell pattern (pattern A) expressing CD10 and/or Bcl-6 but not activation markers; activated GC B-cell pattern (pattern B) expressing at least one of GC B-cell markers and one of activation markers; and activated non-GC B-cell pattern (pattern C) expressing MUM1/IRF4 and/or CD138 but not GC B-cell markers. Patients with pattern A had much better overall survival than those with the other two patterns (Kaplan-Meier survival analysis, P < 0.008, log rank test). Using multivariate Cox proportional hazards regression analysis, the international prognostic index scores and the expression pattern of these markers were independent prognostic indicators. Our results suggest that expression patterns of this panel of GC B-cell and activation markers by immunohistochemistry correlate with the prognosis of patients with DLBCL. Immunohistochemical analysis on paraffin-embedded tissues is more readily available than gene expression profiling by cDNA microarray and may provide similar prognostic information.