DISCONTINUED The work of the Lymphoma Biology Section is ultimately directed towards the development of improved therapy for the treatment of lymphoid malignancies in children and young adults, with special emphasis on Burkitts lymphoma (BL). A major goal is to better elucidate the nature and consequences of the genetic changes that lead to lymphoid neoplasia, based on the hypothesis that such information will lead to improved diagnostic tools, the identification of prognostic factors of value in patient stratification for therapy, and the identification of molecular targets for novel therapeutic endeavors. We have also attempted to identify molecular differences in different age groups, different geographic regions, and in tumors arising in normal versus immunosuppressed hosts. Our clinical protocols have sought to take advantage of unique opportunities at the NCI to explore, in pilot studies, the hypotheses that 1) adults with BL or BL-like lymphomas require therapy similar to the effective protocols used in children, and 2) the lives of children with HIV-associated lymphomas can be significantly prolonged by judiciously designed chemotherapy. The first hypothesis appears to be valid in the completed pilot protocol (in all, 117 patients were accrued on study), in which the event free survival (EFS) rate in 102 patients with B cell lymphomas was 88% at 5 years, with no survival difference between children and adults. Even patients with the most extensive disease (bone marrow and CNS involvement) achieved an EFS rate of close to 80%. This protocol is now being used in adults by the UK NLI Cooperative Group, which has reported a slightly inferior result, but one still significantly improved compared to previous results in the UK. Late cardiac failure, a significant problem in an earlier protocol (77-04), but has yet to be seen in protocol 89-C-41, in which the total dose of adriamycin is much less. The protocol has also been used in Israel, without adriamycin, with apparently identical results in a small group of children. The small study for immunosuppressed patients has been successful in preventing patients from dying from lymphoma or its treatment. Our studies of pathogenetic mechanisms were focused, in the past, on the c-myc gene, (deregulated by the 8;14 translocation in BL), and on Epstein Barr virus but have, more recently, been expanded. We have demonstrated that the functional loss of the cdk inhibitor, p16 is almost universal in BL. We have also studied the mechanisms whereby BL cells escape apoptosis and shown that in most BL cell lines the Fas apoptotic pathway is inactivated by at least two different mechanisms. One of these, surprisingly, involves inactivation of the pro-apoptotic protein, Bax, and we have gone on to demonstrate that Bax interacts with an early component of the Fas pathway. We have also shown loss of Bax expression in 7 of 20 Burkitt lymphoma biopsies by immunohistochemistry. To date, we have not found Bax mutations in HIV associated B cell lymphomas. Our studies with EBV have been primarily directed towards EBNA-1, the only EBV gene expressed in BL. We have identified sequence changes in EBNA-1 that demonstrate body compartment and tumor specificity. Interestingly, the commonest EBNA-1 subtype in normal lymphocytes (P-ala) appears to be incompatible with lymphomagenesis (including HIV associated lymphomas), while the spectrum of EBNA-1 subtypes differs in BL and nasopharyngeal carcinoma. Another subtype, V-pro, appears to be unable to transform, since it is absent from lymphoblastoid cell lines, and rarely found in tumors. We have characterized Peruvian nasal NK/T lymphomas, identified a high frequency of p53 mutations (86%), and demonstrated the presence of mixtures of EBNA-1 subtypes with incremental accumulation of mutations, strongly suggesting that EBNA-1 mutations arise in vivo. We have detected similar atypical variants of EBNA-1 in CNS lymphomas. We have constructed EBNA-1 transgenic mice, with several transgenic lines from each of three EBNA-1 subtypes (V-pro, V-leu and P-ala) to determine whether there is a difference in oncogenicity. We are also performing complementation studies by crossing these mice with lymphoma-susceptible mice transgenic for myc. We have identified a cellular protein, p32/TAP, that binds EBNA-1. Finally, we have shown specific killing of BL cells in vitro, using EBNA-1 to drive a pro-drug activating enzyme, or the Zta gene of EBV. Zta activates the viral lytic cycle, even in the presence of inhibitors of viral DNA replication, thereby causing cell death. HIV associated work, 10%.