Pediatric acute lymphoblastic leukemia: database analysis of patients and treatments in a national public healthcare system

Abstract

Introduction: Despite the achievement of progressively higher cure rates, acute lymphoblastic leukemia (ALL) remains the main cause of cancer-related deaths in children, adolescents, and young adults, particularly in low- and middle-income countries. According to SEER data, the 3-year survival for ALL ages <20 in 2015 in the US was 90.4%-92.5%. Since new immunotherapy strategies are approved up to the age of 25 years, these young adults should also be included in outcomes analyses. It is important to determine outcomes and the gap between the overall survival observed in developing and in high-income countries to guide policies and priorities regarding specific healthcare needs and, ultimately, to improve the outcomes of ALL treatments worldwide.

Objectives: To describe the treatment lines, protocols and fatality rates in children, adolescents and young adults up to the age of 25 years diagnosed with ALL (PedALL) in Brazil, based on a public healthcare national database.

Methods: The "DataSUS" is an anonymized open-access database of all patients treated in Brazilian public healthcare centers, divided in three datasets: outpatient, inpatient, and death reports. Citizens have a unique number and all diagnoses, diagnostic procedures and medical interventions are included. This national registry is used to track public expenses. DataSUS has information regarding ALL (ICD-10, C91.0) including protocols, treatment lines, and survival. We performed a descriptive cross-sectional study of PedALL included in the database between Jan/2013 and Dec/2018 to understand the burden of the disease in the public healthcare system. Then, we selected a cohort of patients with PedALL included in DataSUS as "First-line chemotherapy for pediatric leukemia" between Jan/2014 and Dec/2015, and followed these patients until Dec/2018, to describe the management and outcomes of PedALL in our country.

Results: Between 2013 and 2018, 17,658 patients ages 0-25 years (86% < 18 years) had the underlying diagnosis of ALL, a mean of 6,045 patients per year: 2,725 newly diagnosed, and the remaining, continuation therapy. The male/female ratio was 1.40. The patients underwent a total of 86,332 procedures in this timeframe including diagnostic procedures, chemotherapy, multiple treatment lines, hematopoietic stem cell transplant, transplant-related exams, and the management of disease and treatment-related complications. Treatment protocols most frequently reported were BFM and two national protocols, GBTLI-LLA and RE-LLA. A total of 1,266 patients died between 2013-2018, a mean of 211 patients each year, 79.3% of them younger than 18 years of age. In the cohort diagnosed between Jan/2014 and Dec/2015, 2,368 patients 0-18 years of age and 91 ages 19-25 years were first included in DataSUS system as "First-line chemotherapy for pediatric leukemia". The male:female ratio was 1.44 and 1.84 for the 0-18 and 19-25 years of age, respectively. The median time from diagnosis to treatment initiation was 2.3 months. Within the three year-follow-up, a 2 nd line chemotherapy was registered in 142 patients (6%), 3 rd line in 11 patients (0.46%) and a 4 th line in a single patient (0.04%). The mean duration of each treatment line was 16.4, 8.3, and 4 months, respectively, for the 1 st, 2 nd, and 3 rd line. Patients were hospitalized for a mean total of 65.3 days: 9.7 admissions per patient, for a mean of 6.7 days each. The absolute death count was 303 (0-18) and 13 (19-25) patients, which would represent a 3-year fatality rate of 12.8% (0-18) to 14.3% (19-25); 61% of the deaths occurred in the first year of treatment.

Conclusion: The burden of ALL for the public health care system is large, with over 80 thousand procedures every year. This has a high emotional and socio-economic impact to the families and to the society. More than 200 children and young adults with ALL die every year in Brazil, but fatality rate is very similar to SEER data. The two-month delay between diagnosis and treatment initiation can be an explanation for the early mortality observed in our database and must certainly be appointed as a great opportunity for improvement. Standardization of the national PedALL protocol and new treatment modalities may greatly improve this scenario.