LIQUID BIOPSY FOR EARLY, NON‐INVASIVE DIAGNOSIS OF EBV‐POSITIVE BURKITT LYMPHOMA IN RESOURCE LIMITED SETTINGS

Introduction: Burkitt Lymphoma (BL) is an aggressive B- cell malignancy that is highly prevalent in Sub-Saharan Africa (SSA), contributing to 50% of paediatric cancer. In this region where Epstein Barr virus (EBV) is endemic, more than 90% of cases of BL are associated with EBV (EBL). Survival rate for BL in SSA is less than 50%, largely due to late presentation of advanced disease coupled with mis- or delayed diagnosis all of which delays initiation of effective chemo-immunotherapy. An accurate diagnosis of BL via immunohistochemistry is dependent upon obtaining a tissue biopsy. The limited number of histopathologists, infrastructural challenges and lack of reagents for immunohistochemistry are largely implicated in the diagnostic delays resulting from this diagnostic modality. We developed a liquid biopsy test for the detection of EBL that complements tissue histopathology while circumventing these challenges. Methods: We designed a custom sequencing panel of approximately 140 kb targeting genes commonly mutated in BL, the full MYC gene and its common translocation partners and three EBV genes (EBER1, EBER2 and EBNA2). Our sample set of 150 samples was split into a training cohort and a test cohort. The training cohort was used to develop the test, which was then validated on the test cohort. Three diagnostic models were tested using logistic regression; a clinical model (with clinical parameters predictive of EBL), a liquid biopsy model (using MYC translocation and the presence/absence of EBV DNA) and a combined model (combining the clinical and the liquid biopsy parameters). The model with the best performance was then validated on the test cohort and performance assessed by means of AUC, sensitivity and specificity. Histopathology diagnosis with a limited IHC panel was used as the gold standard. Results: In the training cohort, the combined model performed best (AUC: 0.95, sensitivity 87% and specificity 83%), followed by the liquid biopsy model (AUC: 0.90, sensitivity 93% and specificity of 74%). Upon validation using the test cohort, the combined model had an AUC of 0.96 (CI: 0.4, 1.0) with sensitivity of 92% and specificity of 82%. Conclusion: The presence of clinical features characteristic of EBL in children and young adults in SSA, combined with results from a liquid biopsy test using the presence of EBV DNA and MYC translocation, is diagnostic of EBL with appreciable sensitivity and specificity. This technology has the potential of revolutionizing the management of EBL by providing early non-invasive diagnosis to thousands of patients in need. Keywords: Aggressive B-cell non-Hodgkin lymphoma, Diagnostic and Prognostic Biomarkers, Non-Hodgkin (Pediatric, Adolescent, and Young Adult) No conflicts of interests pertinent to the abstract.