© The Author 2006. Published by Oxford University Press.
CORRESPONDENCE |
RESPONSE: Re: Childhood Leukemia Incidence in Britain, 1974–2000: Time Trends and Possible Relation to Influenza Epidemics
Affiliation of authors: Childhood Cancer Research Group, Oxford University, UK
Correspondence to: Mary E. Kroll, MSc, Childhood Cancer Research Group, Oxford University, 57 Woodstock Rd., Oxford, OX2 6HJ, UK (e-mail: mary.kroll@ccrg.oxford.ac.uk).
We thank Dr McNally and Professor Eden for comments on our recent paper, which reported time trends in subtypes of childhood leukemia in Britain during 1974–2000 and noted the coincidence of small peaks in incidence of acute lymphoblastic leukemia (ALL) with epidemics of influenza (1). We agree that the national time trend for B-precursor ALL is consistent with one of two previous regional studies (2,3) covering about 13% and 9% of the national population, respectively. We are grateful for this opportunity to amplify some points concerning influenza.
We chose to focus on influenza because it is caused by an unstable virus that mutates frequently: an epidemic represents the emergence of a new strain to which there is little existing herd immunity. Most adults in Britain have at least partial immunity to other common infections, such as measles, chicken pox and adenovirus. Lack of herd immunity is part of the mechanism proposed in Kinlen's hypothesis (4) and may be related to the mechanism of Greaves' hypothesis (5).
We agree, and said, that the association may be due to chance. It is impossible to conduct a valid significance test because the observation was made after the leukemia data had been analyzed, without a well-defined prior hypothesis.
The association is consistent with either a nonspecific effect of infection or a specific effect of influenza (or particular strains of influenza or some unknown secondary infection), combined with lack of immunity in the adults around the child and perhaps other factors. The proportionate increase in ALL was less than the proportionate increase in influenza. McNally and Eden say that this means that influenza can be involved for "only a limited number of individuals." Even a limited association could be important. However, it is not clear that either Kinlen's or Greaves' hypothesis would predict a simple dose–response effect because both involve immunity as well as infection. The weekly all-ages consultation rate for "flu and flu-like illnesses" in selected general practices is designed to identify changes in activity, not to measure infection accurately. (Many "flu-like illnesses" are not influenza. Many respiratory illnesses are not reported to a doctor. Various strains of influenza circulate each winter. Young children have frequent respiratory infections, some of which may be undiagnosed influenza.) It certainly does not measure individual variations in immune function or local variations in prevalent strains of influenza and herd immunity.
The observed "epidemic pattern" was based on the time of diagnosis of ALL, which is consistent with both Greaves' and Kinlen's hypotheses—not with Smith's hypothesis (6), according to which an infection occurring in utero or around the time of birth would be predicted to generate an epidemic pattern based on time of birth.
There is considerable evidence that infection and immunity are involved in the etiology of childhood leukemia, particularly ALL. If not due to chance, this observation appears to implicate a particular infection in at least some cases and identify the time of exposure as being shortly before the diagnosis of ALL.
REFERENCES
(1) Kroll ME, Draper GJ, Stiller CA, Murphy MFG. Childhood leukemia incidence in Britain, 1974–2000: time trends and possible relation to influenza epidemics. J Natl Cancer Inst 2006;98:417–20.
(2) McNally RJQ, Birch JM, Taylor GM, Eden OB. Incidence of childhood precursor B-cell acute lymphoblastic leukaemia in north-west England. Lancet 2000;356:485–6.[CrossRef][ISI][Medline]
(3) Feltbower RG, Moorman AV, Dovey G, Kinsey SE, McKinney PA. Incidence of childhood acute lymphoblastic leukaemia in Yorkshire, UK [research letter]. Lancet 2001;358:385–7.[CrossRef][ISI][Medline]
(4) Kinlen L, Doll R. Population mixing and childhood leukaemia: Fallon and other US clusters. Br J Cancer 2004;91:1–3.[CrossRef][ISI][Medline]
(5) Greaves MF. Aetiology of acute leukaemia. Lancet 1997;349:344–9.[CrossRef][ISI][Medline]
(6) Smith M. Considerations on a possible viral etiology for B-precursor acute lymphoblastic leukemia of childhood. J Immunother 1997;20:89–100.
Related Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2006 98: 1746.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||