Seasonal variation in month of diagnosis in children and adolescents
with acute lymphoblastic leukemia: 15-year trends at a single center
Abstract
Background: The date of acute lymphoblastic leukemia (ALL) diagnosis has
been studied regarding potential etiologic roles with contrasting
results and the issue remains controversial. We analyzed seasonality of
diagnosis in a homogenous Hispanic cohort. Design/Method: Clinical files
and electronic databases of consecutive children and adolescents with
ALL in Northeast Mexico over 2004-2018 were scrutinized. Data included
date of diagnosis, age, sex, and risk group. Patients were divided into
children <16 and adolescents 16 to 20 years. Statistical
analysis included Chi-square tests for heterogeneity and multiple
Poisson regression using parametric harmonic modelling to detect monthly
variation. Results: During the fifteen-year study period 394 consecutive
patients with ALL were included. There were 323 (81.9%) children, and
71 (18.1%) adolescents. 176 (54.5%) children had high-risk disease.
Heterogeneity across months of diagnosis was confirmed (P=0.038). A
statistically significant monthly variation in the date of diagnosis
documented by Poisson regression was found for the whole group, with a
peak on March (P < 0.001). When analyzing by age group, cases
of children with ALL predominated in March. Adolescents had a separate
peak in July, with a 4.5 : 1 male to female ratio. Conclusions:
Seasonality at month of diagnosis in children and adolescents with ALL
was identified. Male adolescents had a high July peak suggesting a
hormonal influence in ALL development in this age group.