B116
Seasonality of birth and diagnosis for cancer in children and young people from northern England, 1968-2005
Nermine Basta1, Peter James1, Alan Craft2, Richard McNally1
1Institute of Health and Soceity, Newcastle University, UK, 2Northern Institute of Cancer, Newcastle University, UK
Background
Previously seasonal trends have been reported in certain childhood cancers, including acute lymphocytic leukaemia, Hodgkin lymphoma and brain tumours. Seasonal variation suggests that environmental factors may be involved in aetiology. The aim of this study was to investigate seasonal variation in the incidence of cancer in children and young people.
Method
All cases diagnosed during the period 19682005 were extracted from the Northern Region Young Persons Malignant Disease Registry. The following main diagnostic groups were analysed: leukaemia, lymphoma, central nervous system tumours, sympathetic nervous system tumours, retinoblastoma, renal tumours, hepatoblastoma, bone tumours, soft tissue sarcoma, germ cell tumours and carcinomas. The chi-squared test for non-uniform heterogeneity was used. Sinusoidal (harmonic) models were fitted to the data using Poisson regression analysis. Month of birth and month of diagnosis were used separately as covariates in these models. Analyses were also performed by gender and age group (014, 1524 years).
Results
There were a total of 5909 cases; 2959 aged 014 years (1659 males, 1300 females) and 2950 aged 1524 years (1590 males, 1358 females). For 014 year old boys, there was statistically significant seasonal variation in month of birth for acute non-lymphocytic leukaemia (P= 0.04; September peak) and astrocytoma (P=0.03; October peak). Based on month of diagnosis, there was significant seasonal variation in girls for lymphomas (P=0.05; March peak) and Hodgkin lymphoma (P=0.005; January peak), and in boys for osteosarcoma (P=0.05; October peak). For 1524 year olds, significant seasonal variation in month of birth for malignant melanoma in females (P=0.03; March peak) and cervical carcinoma (P=0.03; October peak) was observed.
Conclusion
These findings suggest that seasonal environmental factors especially around the time of birth may be involved in the aetiology of certain diagnostic groups. Putative agents include sunlight, pesticides, diet and infections.
