RESUMO
BACKGROUND: Adult sporadic Burkitt lymphoma (BL) is a rare but highly aggressive subtype of lymphoma which lacks its own unique prognostic model. Systemic inflammatory biomarkers have been confirmed as prognostic markers in several types of malignancy. Our objective was to explore the predictive value of pretreatment inflammatory biomarkers and establish a novel, clinically applicable prognostic index for adult patients with sporadic BL. METHODS: We surveyed retrospectively 336 adult patients with newly diagnosed sporadic BL at 8 Chinese medical centers and divided into training cohort (n = 229) and validation cohort (n = 107). The pretreatment inflammatory biomarkers were calculated for optimal cut-off value. The association between serum biomarkers and overall survival (OS) was analyzed by Kaplan-Meier curves and Cox proportional models. The risk stratification was defined based on normal LDH level, Ann Arbor stage of I and completely resected abdominal lesion or single extra-abdominal mass < 10 cm. RESULTS AND CONCLUSIONS: Univariate and multivariate analyses revealed that platelets< 254 × 109/L, albumin< 40 g/L, lactate dehydrogenase≥334 U/L independently predicted unfavorable OS. We used these data as the basis for the prognostic index, in which patients were stratified into Group 1 (no or one risk factor), Group 2 (two risk factors), or Group 3 (three risk factors), which were associated with 5-year OS rates of 88.1, 72.4, and 45%, respectively. In the subgroup analysis for high-risk patients, our prognostic model results showed that high-risk patients with no more than one adverse factor presented a 5-year survival rate of 85.9%, but patients with three adverse factors had a 5-year survival rate of 43.0%. Harrell's concordance index (C-index) of the risk group score was 0.768. Therefore, the new prognostic model could be used to develop risk-adapted treatment approaches for adult sporadic BL.