Seasonal variation exists in B-Cell Precursor Childhood Acute Lymphoblastic Leukemia diagnosis, but not in Acute Myeloid Leukemia, Brain Tumors or Solid Tumors – a Swedish population-based study

Abstract

Background Recent molecular studies of B-cell precursor acute lymphoblastic leukemia (BCP-ALL) have started to delineate the nature and timing of genetic variants and those responsible for subsequent progression to overt leukemia. However, the etiology behind both initiation and progression remains largely unknown. Nonetheless, theories, but also epidemiological evidence, of how exposure to common infections and other microbes in our environment modulates the risk of developing childhood BCP-ALL, have emerged. In light of these theories and the well-known phenomena of seasonality in infectious disease spread, childhood ALL has been analyzed for signs of seasonal variation, with differing results. Methods In this study we applied the Bayesian Generalized Auto Regressive Integrated Moving Average with external variables (GARIMAX) model, adapted for count data via a negative binomial distribution, to study seasonal variation of incidence in a Swedish population-based cohort of 1601 BCP-ALL cases. The studied cases were aged 0-18 years at diagnosis and identified from the Swedish Childhood Cancer Registry (SCCR). Also, two subgroups of BCP-ALL represented by the most abundant genetic subtypes, ETV6/RUNX1 and HeH respectively, were analyzed accordingly. All analyses were performed in two stages. The first stage identified the presence of the repeatable pattern using harmonic functions, and the second stage consisted of the identification of the peak months in the series. Results An informative seasonal variation in BCP-ALL incidence numbers, displaying a peak in August and September, was detected in the total cohort of 1601 individuals. No seasonality was detected analyzing the subtype groups HeH and ETV6/RUNX1 positive BCP-ALL, respectively. Conclusion The manifested seasonality in BCP-ALL with a peak in August-September may suggest that the prolonged period of minimal viral spread during Swedish summer vacation causes a temporary halt in the last step of progression to overt disease and consequently an accumulated number of cases presenting in August-September.