Seasonal Variation in Incidence of Pediatric Medulloblastoma in the United States, 1995–2001

Abstract

Purpose Brain tumors are the second most common pediatric malignancy, second only to leukemia and lymphoma in incidence. There is a suggestion in the literature that one of the most common subtypes of malignant childhood brain tumor, medulloblastoma, has some seasonal variation in incidence by month of birth. This descriptive study aims to confirm the potential seasonal pattern of medulloblastoma incidence, thus providing direction for further etiologic studies. Methods Data from the Central Brain Tumor Registry of the United States included primary benign and malignant brain tumor cases diagnosed in children (0–19 years) between years 1995–2001 from thirteen contributing state cancer registries. Seasonal analyses were performed using Edwards' test for sinusoidal variation, which uses case frequencies per month, and tests whether frequencies follow a sine function over 12 months. Results Seasonal variation in incidence by month of birth was highly statistically significant for medulloblastoma, not otherwise specified (NOS) ( p = 0.016), with the peak occurring in the month of October. Medulloblastoma, NOS also demonstrated seasonal variation in incidence by the month of birth in children aged 5–19 years ( p = 0.041), especially female children aged 5–19 years ( p = 0.034), again with the peak occurring in the month of October. There was no significant seasonal variation in incidence by the month of birth for all brain tumors or for the other most common pediatric tumor subtypes of pilocytic astrocytoma, all other astrocytoma, and ependymoma. Conclusion These preliminary results indicate seasonal variation in medulloblastoma incidence by the month of birth and may provide an evidence for an environmental exposure etiology, though further studies are needed to explore the gender and age group significance.