This Article  • PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citing articles on HighWire  • Citing articles on Web of Science (14)  • Contact me when this article is cited  Related Content  • Similar articles in JAMA  Social Bookmarking   What's this?

To the Editor: Prior studies have indicated a possible role for in utero or early postnatal causes of acute lymphoblastic leukemia in children.^1-4 If such causes include seasonally prevalent infectious agents, then temporal patterns of birth month of affected children or the month of onset of leukemia should be expected to vary accordingly. However, studies of this issue have been generally too small to determine trends.We report the seasonal pattern in birth month and month of diagnosis for children with leukemia in the population of Denmark.

Methods
We obtained data from the Danish Cancer Registry, which was started in 1943 and which has almost complete reporting of incident cancers in Denmark. The Cancer Registry receives notifications of malignant and related diseases from hospital departments at the time of diagnosis and is notified if changes in the initial diagnosis occur. We studied the pattern of annual cyclic variation in birth month and month of first diagnosis for all 458 children younger than 4 years with acute lymphoblastic leukemia who were diagnosed in Denmark from 1950 to 1994. We adjusted the analyses for variation in the number of all births by month. To study cyclic variation we used periodic regression, fitting a sine curve to the observed data.

Results
The ratio of case births at the peak month to case births at the trough month was 1.4 (95% confidence interval [CI], 1.0-2.0), with the peak of the fitted curve (Figure 1) in April. For date of diagnosis, the peak month of occurrence was October, and the peak-to-trough ratio was 1.6 (95% CI, 1.2-2.0). Both functions had an acceptable goodness-of-fit evaluation based on the ² distribution.

View larger version (15K): [in this window] [in a new window] Figure. Trends in Month of Birth and Diagnosis of Children With Acute Lymphoblastic Leukemia Periodic regression functions are indicated by solid lines. Dashed line represent the 95% confidence intervals of the regression function and closed circles represent observed frequencies.

Comment
The timing of the peak in diagnosis date that we observed is close to that observed for date of first symptom in the study by Westerbeek et al.⁵ In contrast, Ross et al³ found a summer peak in diagnosis of acute lymphoblastic leukemia, although they included subjects as old as 20 years in their analysis.⁶ However, our study is the first to examine variation in birth month.

Two hypotheses about the role of infection in the etiology of leukemia have been proposed.^{2, 4} Greaves² suggested that an initial mutational event occurs in utero when immature B cells are rapidly dividing; subsequent mutations occur after birth and are strongly influenced by the timing of exposure to infectious agents in infancy. Kinlan⁴ proposed that childhood leukemia could be a rare response to in utero or postnatal viral exposure. We expect that year-to-year variations in prevalence of seasonal infectious agents, sensitivity of the fetus to the agent, and subsequent promotional events would all obscure any observable seasonality. Nonetheless, our data provide some evidence that the occurrence of acute lymphoblastic leukemia in childhood may result, in part, from exposure to 1 or more in utero or postnatal infectious agents.

AUTHOR INFORMATION
Funding/Support: This study was funded by research grant 9510030 from the Danish Cancer Society.

Henrik Toft Sørensen, PhD; Lars Pedersen, MSc
Department of Clinical Epidemiology
Aarhus University Hospital and Aalborg Hospital
Aarhus, Denmark

Jargen H. Olsen, PhD
Danish Cancer Society
Institute of Cancer Epidemiology
Copenhagen, Denmark

Kenneth J.M Rothman, DrPH
Department of Epidemiology
Boston University School of Public Health
Boston, Mass

1. Wiemels JL, Cazzniga G, Daniotti M, et al. Prenatal origin of acute lymphoblastc leukemia in children. Lancet. 1999;354:1499-1503. FULL TEXT | WEB OF SCIENCE | PUBMED 2. Greaves MF. Aetiology of acute leukemia. Lancet. 1997;349:344-349. FULL TEXT | WEB OF SCIENCE | PUBMED 3. Ross JA, Coppes MJ, Robison LL. Population density and risk of childhood acute lymphoblastic leukemia [comment]. Lancet. 1999;354:532. PUBMED 4. Kinlen LJ. High-contact paternal occupations, infection and childhood leukaemia: five studies of unusual population-mixing of adult. Br J Cancer. 1997;76:1539-1545. WEB OF SCIENCE | PUBMED 5. Westerbeek RM, Blair V, Eden OB, et al. Seasonal variation in the onset of childhood leukemia and lymphoma. Br J Cancer. 1998;78:119-124. WEB OF SCIENCE | PUBMED 6. Ross JA, Severson RK, Swensen AR, Pollock BH, Gurney JG, Robison LL. Seasonal variations in the diagnosis of childhood cancer in the United States. Br J Cancer. 1999;81:549-553. FULL TEXT | WEB OF SCIENCE | PUBMED

Letters Section Editor: Stephen J. Lurie, MD, PhD, Senior Editor.

JAMA. 2001;285:168-169.

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?

THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES

Do sunlight and vitamin D reduce the likelihood of colon cancer? Time for a paradigm shift?
JONGBLOET
Int J Epidemiol 2006;35:1359-1360.
FULL TEXT  

Effect of month of birth on the risk of suicide
SALIB and CORTINA-BORJA
Br. J. Psychiatry 2006;188:416-422.
ABSTRACT | FULL TEXT  

Season of birth and risk of brain tumors in adults
Brenner et al.
Neurology 2004;63:276-281.
ABSTRACT | FULL TEXT