Plasmablastic lymphoma (PBL) was originally described in 1997 as an AIDS associated tumour although cases have been described in individuals not infected with HIV. Due to the high number of people living with HIV in South Africa, a substantial number of cases are diagnosed annually and 45 cases were included in this study. This represented the largest cohort of PBL affecting the oral mucosa published to date. Three main aspects of PBL were investigated: pathological features, viral status and certain genetic characteristics. The results from the genetic studies were the most important and interesting. These included rearrangements of the IGH gene in 63% and MYC- rearrangements in 62% of PBL’s. Seven of 43 cases (16%) showed rearrangement of both the IGH gene alleles, a finding never described before. New genetic findings also included increased CCND1 gene copy numbers in 17/41 (42%) and increased IGH gene copy numbers in 6/41 (15%) of cases. The exact role of MYC-rearrangements in the development of PBL is unclear. Many factors may be responsible for MYC deregulation but in the case of PBL of the oral cavity the possible role of Epstein Barr Virus (EBV) infection was considered. All but one of the patients with known HIV-status (32/45) was HIV positive and I supported the proposal that the diagnosis of PBL should serve as a sign of immunodeficiency, either as diagnostic thereof or as a predictor of a progressive state of immunodeficiency in patients with known HIV/AIDS status. The HIV-negative patient in this study was the only one that presented with an EBV-negative PBL on in situ hybridisation. The clinico-pathological features of the current study therefore strongly suggested an association between EBV, PBL and HIV/AIDS although the exact nature thereof remains uncertain. Routine genetic evaluation of tumours diagnosed as PBL should be introduced, as this may have prognostic and eventually treatment implications in the future. The exact panel of genes to be evaluated with a possible diagnosis of PBL should still be determined but examination of IGH and MYC for rearrangements should be included. This study proved the histomorphological features including the degree of plasmacytic differentiation not to have any diagnostic role although its prognostic value should be determined. The results of the immunohistochemical investigations performed in this study confirmed PBL always to be negative for CD20 but proved PBL not to be a morphological or immunohistochemical diagnosis by any means. In conclusion, it became clear that PBL should never be diagnosed without thorough clinical, systemic, pathological and genetic investigations, especially in the backdrop of HIV/AIDS. No pathologist should make the diagnosis of PBL and no clinician should accept such a diagnosis or decide on the treatment modality for the patient involved unless all other possibilities of systemic plasma cell disease have been excluded. / Thesis (PhD)--University of Pretoria, 2011. / Oral Pathology and Oral Biology / unrestricted