Prevalence of Induction Failure in Children With Acute Lymphoblastic Leukemia in Kenyatta National Hospital

Abstract

Background: Acute lymphoblastic leukaemia (ALL) is the most prevalent malignancy of childhood worldwide accounting for 25-30 per cent of all cancers occurring in this age group. Induction is the first phase of treatment aimed at reducing the tumour burden to less than 5% blast cells in the bone marrow at the end of induction (EOI). Induction failure is the persistence of greater than 5% blast cells in the bone marrow at the end of induction and is an independent prognostic factor for overall survival that portends a poor outcome. The rate of induction failure is high in low- and middle- income countries (15-30%) compared to high-income countries (5- 10%). The rate and factors contributing to induction failure among children with acute lymphoblastic leukemia in low-middle income countries like Kenya is poorly established. This study sought to establish the prevalence and factors associated with induction failure in childhood acute lymphoblastic leukemia in Kenyatta National Hospital and thereby provide valuable information for improvement of careof children with ALL for better outcomes. Broad objective: The study objectives were to determine the prevalence and identify patient, disease and treatment factors associated with induction failure in children with acute lymphoblastic leukaemia treated in Kenyatta National Hospital (KNH). Study design: A retrospective cross-sectional study that included all children aged 0-15 years diagnosed with acute lymphoblastic leukaemia and treated in Kenyatta National Hospital from January 2015 to December 2019. The study design was informed by a desired large sample size which could not have been achieved prospectively due to constraints of time. Study site: The study was conducted in the Paediatric medical and oncology wards in KNH. Material and Methods: Case records of 114 children aged 0-15 years with ALL treated in KNH between January 2015 and December 2019 were identified from the medical records department. All that met the inclusion criteria were abstracted to collect data regarding demographics, clinical presentation and examination findings, initial complete blood count (CBC) features, initial bone marrow aspirate features, treatment information and remission induction status at the end of induction as detailed in appendix 1. Data were entered using

IBM SPSS V.25. Descriptive statistics were generated for continuous and categorical variables as appropriate. Morphologic status and the independent variables were cross tabulated and Chi Square was used to assess for association between the dependent and in- dependent variables and odds ratios calculated. P-values were obtained for levels of statistical significance at 95% confidence interval. Results: Induction failure was seen in 33 of 102 children who completed induction (32.4%). Case fatality rate during induction was 10% (n=12). There was an association between central nervous system (CNS) disease and the dependent variable OR 3.43(1.33-8.86) (p=0.009). Central nervous system disease increased fourfold the risk of induction failure. This study did not find any association between induction failure and demographic characteristics of patients such as gender OR 0.752 (1.75-3.23) (p=0.506), age OR 0.39 (0.14- 1.06) (p=0.142) and BMI OR 1.10 (0.48-2.53) (P=0.247) at 95% CI. Conclusion: The prevalence of induction failure was high in children diagnosed and treated for ALL in KNH. This is comparable to findings in other low-and middle-income countries. CNS disease was the only factor associated with induction failure.