Clinical Predictors and Prognostic Model for Pediatric Lymphoblastic Lymphoma Treated With Uniform BFM90 Protocol: A Single-Center Experience of 65 Patients From Asia

Abstract

BackgroundThere is a lack of clinical predictors for prognosticating lymphoblastic lymphoma (LBL). In view of this lacuna, we evaluated outcomes and prognostic factors for LBL treated with a uniform protocol at our center. Patients and MethodsThis study included consecutive patients of pediatric LBL aged ≤18 years from January 2003 to January 2017. Patients were staged using the St Jude staging system. All patients were treated with acute lymphoblastic leukemia like BFM90 protocol. The Kaplan–Meier method was used for survival analysis. A statistical model was made using stepwise regression and forward selection of the factors predicting event-free survival (EFS) and overall survival (OS). ResultsSixty-five patients were evaluated with a median age of 12 years (range, 1-18 years) and male:female ratio of 2.25:1. Fifty-four patients presented with mediastinal disease. Median follow-up was 54.57 months (range, 0.6-140.5 months). EFS at 10 years was 62 ± 6% (95% confidence interval [CI], 0.49-0.73) and OS 71 ± 5% (95% CI, 0.57-0.81). In multivariate analysis, symptom duration ≤30 days, white blood cell (WBC) count >12000/µL and serum albumin ≤3.5 g/dL predicted inferior EFS and OS. A prognostic model with these 3 factors suggested that those without any of these risk factors had an OS of 92 ± 5% whereas those with 2 or 3 factors had an OS of 37 ± 14%. ConclusionOur outcomes are 15% to 20% lower than in the published literature. Low albumin level, high WBC count at baseline, and symptom duration <30 days emerged as adverse predictors for EFS and OS. These clinical predictors and prognostic model for pediatric LBL should be validated in prospective cohorts.