LONG-TERM OUTCOMES OF CHILDHOOD ACUTE LYMPHOCYTIC LEUKEMIA WITH MODIFIED BERLIN-FRANKFURT-MÜNSTER PROTOCOL IN RIO DE JANEIRO, BRAZIL: A MULTICENTRIC HOSPITAL-BASED SURVIVAL ANALYSIS STUDY

Abstract

Introduction/objectives: Acute Lymphocytic Leukemia (ALL) is the most common pediatric cancer and it seems to be more frequent in Hispanics. Berlin-Frankfurt-Münster (BFM) protocol for ALL was preceded by a clinical trial carried out in the 1970s and, in the 1980s, hospitals from Rio de Janeiro began to use this protocol seeking for a more effective treatment. Our aim is to determine the survival probabilities and the main prognostic factors for children and adolescents with ALL in Rio de Janeiro and to compare these results with other BFM studies. Methods: We conducted a survival analysis study with a retrospective cohort of 695 patients from 0‒19 years treated with a modified BFM protocol at four public hospitals in the city of Rio de Janeiro, Brazil, between 1998 and 2018. Prognostic factors included age, sex, leukocyte count, cytogenetic risks according to National Cancer Institute (NCI) and BFM risk classifications, early response to treatment, among others. Cohort was divided into two 11-year periods for analysis purposes. We calculated event-free survival and overall survival rates at 5 and 10 years. Finally, we performed a multivariate analysis using Cox proportional hazards models. Results: At 5 years, in the 1st decade, the event-free survival was 50.7% (±3.0) and the overall survival was 61.5% (±2.9). This rates for low-risk patients (n = 26) were 84% and 87.8% respectively, for medium-risk (n = 111), 66.5% and 74.8% and for high-risk (n = 144), 33.4% and 47.7%. In the 2nd decade, the event-free survival was 61.6% (±2.8) and the overall survival was 70.5% (±10.6). This rates for low-risk patients (n = 32) were 84.5% and 92.6%, for medium-risk (n = 166), 63.1% and 75.2% and for high-risk (n = 208), 57.9% and 63.8%. Immunophenotypic classification showed 81.15% B and 18.85% T cells cases. The percentages of high-risk patients based on BFM risk (51.2%), high-risk NCI (54.7%), leukocyte count greater than 50,000 (27.1%) and age less than one or greater than 10 years (28.6%) were higher than the ones found in other ALL studies from developed countries. When it comes to quality of treatment across different periods (before 2008 vs. after 2008), the number of deaths before complete remission (inducted death) was 10 vs. 11 (3.5% vs. 2.7%, p = 0.545), during the first complete remission, 33 vs. 24 (11.6% vs. 5.9%, p = 0.007) and abandonment cases were 12 vs. 5 (4% vs. 1.2%, p = 0.017). Discussion: Our cohort presented a higher leukocyte count at the time of diagnosis and an increased prevalence of high-risk classification compared to cohorts of other BFM studies, mainly composed by Caucasians. Those results lead us to consider that this higher high-risk proportion may be caused by 3 factors: delays in treatment, the use of an overestimate risk classification on the 8th day of treatment and/or unique genetic features of our population, which has Indigenous, Black and European ancestries. In the latest period, our results were similar to those of AllIC-BFM 2009. Conclusions: Over the past 2 decades, the overall and event-free survivals of children and adolescents with ALL have improved significantly in Rio de Janeiro, Brazil. We also observed a significant reduction in deaths in the first complete remission. Further investigations are needed to define the factors associated with the elevated proportion of high-risk cases. This study marks the beginning of Rio de Janeiro-based multicenter survival analysis, focusing on pediatric ALL using BFM-modified protocols.