Mahlangu, Johnny Ndoni2008-10-202008-10-202008-10-20http://hdl.handle.net/10539/5766Background: The t(8;14) translocation in Burkitt’s lymphoma (BL) was the first non-random cytogenetic lesion to be described in lymphoproliferative disorders. This lesion occurs in 75-85% of all BL cases. However, the breakpoints in this cytogenetic lesion are very variable and far apart such that the t(8;14) translocation is not always amenable to standard polymerase chain reaction analysis. This is mainly due to the inability of the Thermus aquaticus (Taq) polymerase enzyme to synthesize long DNA products. Long range polymerase chain reaction (LD-PCR) with a high fidelity polymerase enzyme mix capable of longer PCR product synthesis has recently become available. In early studies, LD-PCR appeared to be capable of amplifying the t(8;14) translocation in the majority of published sporadic Burkitt’s lymphoma analyses. The utility of t(8;14) translocation LD-PCR for routine use in the diagnosis of BL in our setting has not yet been studied. The aim of this study was to establish and optimize the t(8;14) LD-PCR technique and to apply it in the retrospective analysis of all BL diagnosed in the University of the Witwatersrand teaching hospitals in a ten year period from January 1994 to December 2003. Materials and methods: High molecular weight non-degraded DNA was extracted from control cell lines as well as stored, unstained bone marrow slides remaining after routine diagnostic workup of previously identified Burkitt’s lymphoma patients. Three hundred nanograms of patient and control DNA were amplified with the LD-PCR high fidelity polymerase enzyme mix under reaction conditions which were optimized using the tissue plasminogen activator (tPA) gene as well as known Burkitt’s lymphoma cell lines as controls. Each control and patient DNA sample was amplified with tPA primers as well as four pairs of MYC/IgH primer sets. The resulting amplicons were size fractionated on an agarose gel and visualized with ethidium bromide under ultraviolet (UV) light. The fractionated DNA fragment sizes were compared to those of the t(8;14) translocation positive controls, tPA controls and known DNA molecular weight markers. Results: One hundred and ten Burkitt’s lymphoma diagnoses were made in the three teaching hospitals of the University of the Witwatersrand from January 1994 to December 2003. Bone marrow involvement by BL was present in 84 of these cases. Archival bone marrow slides were available in 74 of the 84 BL patients. Intact high molecular DNA on which the t(8;14) LD-PCR analysis could be performed was present in 41 of the 74 BL patients. In the presence of appropriate controls, an t(8;14) translocation specific product was demonstrable by t(8;14) LD-PCR analysis in only 6 of 41 BL patients. Conclusion: In this t(8;14) LD-PCR retrospective analysis of a large number known Burkitt’s lymphomas, the diagnostic yield in carefully selected patients was extremely poor. With five primer pairs required per BL sample analysis, this technique was found too labour intensive and costly in our hands making it unsuitable for routine diagnostic use. The reasons for the poor diagnostic yield remains unclear and may need to be explored in future studies. Emerging alternative techniques for the diagnosis of BL such as fluorescence in situ hybridization and microarray gene expression analyses may prove to be better diagnostic tools than LD-PCR in its current form.enBurkitt's lymphomamolecular analysisThesis