Skip Navigation


American Journal of Epidemiology Advance Access originally published online on December 20, 2006
American Journal of Epidemiology 2007 165(5):496-504; doi:10.1093/aje/kwk039
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
165/5/496    most recent
kwk039v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Roman, E
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roman, E
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

ORIGINAL CONTRIBUTIONS

Childhood Acute Lymphoblastic Leukemia and Infections in the First Year of Life: A Report from the United Kingdom Childhood Cancer Study

E Roman1, J Simpson1, P Ansell1, S Kinsey2, CD Mitchell3, PA McKinney4,5, JM Birch6, M Greaves7, T Eden8 and on behalf of the United Kingdom Childhood Cancer Study Investigators

1 Epidemiology and Genetics Unit, Department of Health Sciences, University of York, York, United Kingdom
2 Department of Paediatric and Adolescent Oncology and Haematology, St. James University Hospital, Leeds, United Kingdom
3 Paediatric Haematology/Oncology, John Radcliffe Hospital, Oxford, United Kingdom
4 Information and Services Division, National Health Service in Scotland, Edinburgh, United Kingdom
5 Paediatric Epidemiology Group, University of Leeds, Leeds, United Kingdom
6 CRUK Paediatric and Familial Cancer Research Group, Royal Manchester Children's Hospital, Manchester, United Kingdom
7 Institute of Cancer Research, Chester Beatty Laboratories, London, United Kingdom
8 Young Adult Cancer Trust, Christie Hospital NHS Trust, Manchester, United Kingdom

Correspondence to Prof. Eve Roman, Epidemiology and Genetics Unit, Department of Health Sciences, University of York, YO10 5DD York, United Kingdom (e-mail: eve.roman{at}egu.york.ac.uk).

Received for publication March 28, 2006. Accepted for publication July 26, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The United Kingdom Childhood Cancer Study was designed to examine the relation between childhood cancer and preceding exposure to infectious diseases. The authors analyzed the relation between diagnosis (1991–1996) of acute lymphoblastic leukemia (ALL) at ages 2–5 years and clinically diagnosed infections in infancy. Almost all study children (96% of both cases and controls) were taken to a general practitioner for a non-immunization-associated visit at least once before their first birthday. Children diagnosed with ALL had significantly more clinically diagnosed infectious episodes in infancy than did controls; the average number of episodes was 3.6 (95% confidence interval (CI): 3.3, 3.9) versus 3.1 (95% CI: 2.9, 3.2). This case-control difference was most apparent in the neonatal period (≤1 month); 18% of controls and 24% of ALL cases were diagnosed with at least one infection (odds ratio = 1.4, 95% CI: 1.1, 1.9; p < 0.05). Cases who had more than one neonatal infectious episode tended to be diagnosed with ALL at a comparatively young age; the mean age at ALL diagnosis was 37.7 months for cases with two or more episodes versus 45.3 months for cases with only one episode or none (p < 0.01). These findings support the hypothesis that a dysregulated immune response to infection in the first few months of life promotes transition to overt ALL later in childhood.

child; infection; leukemia, lymphocytic, acute

Abbreviations: ALL, acute lymphoblastic leukemia; CI, confidence interval; OR, odds ratio; UKCCS, United Kingdom Childhood Cancer Study


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Infections are estimated to account for 15–20 percent of cancers worldwide, and the proposition that the timing and pattern of infectious disease exposure could be a determinant of the type of leukemia commonly seen in children is not new (1–6). At present, however, no specific agents have been identified, and exactly how in-utero and postnatal infectious exposures might affect the likelihood of subsequent leukemia development in children remains a much debated topic (7–9).

The United Kingdom Childhood Cancer Study (UKCCS) was specifically designed to investigate the potential etiologic role of infections as one of its objectives (10). The main hypothesis was that acute lymphoblastic leukemia (ALL) could arise as a rare abnormal response to a common infection (10). The two postulated mechanisms as to how this might occur have become known as the "Kinlen" and "Greaves" hypotheses, after their respective architects. The former speculates that acute leukemia, and possibly non-Burkitt's lymphoma, could arise as a consequence of exposure to a specific infection, particularly evident at times of unusual population movement and mixing (8, 11). The latter, which forms the focus of the present report, proposes that a deficit of exposure to infectious agents in infancy and subsequent "delayed" infectious challenge may be causal in the development of B-cell precursor common ALL, which, with its characteristic peak at ages 2–5 years, accounts for approximately 75 percent of ALLs diagnosed in children (4, 12).

Unraveling the relation between any individual person's cancer risk and his or her history of exposure to infectious agents and response to infectious challenge is fraught with difficulties. For children, a bewildering array of proxy measures aimed at quantifying their likely exposure to infectious agents at various time points have been used, including measures of family socioeconomic status and urban/rural location (13, 14); indicators of parental social contact outside the home, such as occupation, migration, and distance traveled to work (15–18); and markers of the child's own social activity, including birth order and maternally reported child-care attendance outside the home, together with illness history and immunization (13, 19–28).

We examined the relation between clinically diagnosed infections in infancy and subsequent leukemia development among children in the UKCCS. These data were systematically abstracted from primary-care records compiled before diagnosis/interview. The ability to access medical records in this way is a unique facet of the organization of the United Kingdom health care system. In this paper, we provide important background information on the conduct of this component of the study. We describe the main findings relating to acute leukemia diagnosed in the peak years (ages 2–5 years) and its relation with clinically diagnosed infection in the first year of life.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
The UKCCS is a national population-based case-control study. Its conduct and ethical approval are described in detail elsewhere (10). Briefly, children aged 0–14 years diagnosed with ALL between 1991 and 1996 in Great Britain were eligible. For each case, two controls matched by sex, month and year of birth, and region of residence at diagnosis were randomly recruited from primary-care population registers. At interview, parents were asked to consent to our access of their child's primary-care records. For each child, all information contained within these routinely compiled health records from birth to diagnosis (using a pseudo-diagnosis date for control children) was subsequently abstracted onto specially designed forms by centrally trained research staff (29).

The data collected included all diagnoses recorded contemporaneously by the child's general practitioner (i.e., primary-care physician), as well as all signs and symptoms recorded at the same time; referrals to hospital consultants and other specialists; results of all investigations; and detailed information on all medicines and other treatments prescribed. Data collection and entry were structured around dated "events" (e.g., general practitioner consultation, hospital admission, blood tests, screening procedures), with all data being entered centrally under the supervision of experienced primary-care research nurses. Disease and drug coding are key issues in handling and analysis of such data—adopting a unified approach facilitates coding, and this has been achieved using a sophisticated system of computerized "pick-lists" embedded in the data-entry program. Illnesses and symptoms were centrally coded according to the International Classification of Diseases, Tenth Revision (30), and drugs were coded according to a schema based on the British National Formulary (31). Strict quality control procedures, including duplicate data entry of approximately one in four randomly selected records, were used throughout by one of the authors (P. A.).

For the purposes of study management, 10 UKCCS administrative areas were created; the conduct of the study within each area was the responsibility of a single epidemiologic center. Six of the 10 UKCCS centers systematically collected primary-care records from within their area, but each in a different way (see footnotes to table 1). Overall, the primary-care (general practitioner) records of 1,567 (92 percent) targeted case children with cancer and 2,393 (70 percent) control children without cancer were traced and abstracted (table 1). All cases had information on at least one of their matched controls abstracted, and more than half had information on both. The lower overall control proportion (70 percent of controls vs. 92 percent of cases) reflects regional policy variation with respect to the abstraction of data on the number of controls sought. Only one region (Oxford) routinely abstracted data on both controls; others opted for one control per case at varying times as the study progressed—Leeds, Manchester, and Cambridge doing so almost from the outset and Edinburgh and Southampton changing the procedure part-way through. Importantly, as with other components of the UKCCS that only collected additional data on one of the two individually matched controls, this control—the lowest numbered (first randomly selected) interviewed control in the series—was identified in advance of abstraction (32).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Numbers of children whose parents were interviewed and whose general practice records were abstracted, by study region, United Kingdom Childhood Cancer Study, 1991–1996

 
Infections occurring in the first year of life among children who subsequently developed leukemia between 2 and 5 years of age form the focus of the present report; these data are presented in the bottom half of table 1. Sixteen of the 471 case children also had Down's syndrome and were excluded from further analyses because of the known relations both between Down's syndrome and leukemia and between Down's syndrome and childhood infection (33).

We calculated enumeration district "deprivation" indices for the UKCCS administrative areas contributing to these analyses using the same methods as described for the United Kingdom as a whole (10). As in previous UKCCS publications (23, 24, 34–36), in order to increase precision and statistical power we used all available controls as the comparison group in the main analyses. Odds ratios, 95 percent confidence intervals, and two-sided p values were estimated using unconditional logistic regression, with adjustment for UKCCS administrative area of residence at diagnosis, sex, and age (in single years) at diagnosis (37). In order to check the veracity of our findings with respect to misclassification and/or selection bias, we repeated the analyses using conditional logistic regression and data from the Oxford region (100 percent of leukemias diagnosed at ages 2–5 years and 98.3 percent of matched controls). All analyses were conducted using Stata software (38).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Of the 455 children with leukemia diagnosed between 2 and 5 years of age, 425 (93.4 percent) had ALL; 342 (80.5 percent) of these ALLs were B-cell precursor common ALL. Findings for these three overlapping groups are presented in the tables and figures, and results for ALL are highlighted in the text.

Information on sociodemographic characteristics and primary-care consultations in the first 12 months of life is given in table 2. There were no statistically significant differences between cases and controls for any of the factors presented. The use of all available controls resulted in small differences between cases and controls with regard to the original matching factors—age and sex—and adjustment for both of these factors, along with UKCCS administrative area, was included a priori in all risk ratio estimate calculations. The socioeconomic area-based deprivation distributions of case and control families included in the analysis were similar, and no subsequent adjustments were required.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Characteristics of children with leukemia and general practitioner consultations* in the first year of life, United Kingdom Childhood Cancer Study, 1991–1996

 
All children in the study were taken to their general practitioner at least once during the first year of life, but a few only visited for routine immunization. Overall, 95.7 percent of control children and 95.8 percent of case children made at least one non-immunization-associated visit to their general practitioner before their first birthday. No systematic variations between cases and controls were observed when the data were distributed by leukemia subtype, deprivation category, and UKCCS administrative area (table 2).

Children who subsequently developed leukemia between ages 2 and 5 years visited their general practitioner marginally more frequently during the first year of their life than children who did not; the average numbers of visits per child were 7.8 (95 percent confidence interval (CI): 7.2, 8.4) and 6.9 (95 percent CI: 6.6, 7.2) for ALL cases and controls, respectively (table 3). This difference reflects the fact that children who developed leukemia were significantly more likely (p < 0.05) to visit the general practitioner with an infection and/or symptoms consistent with an infection (fever or pyrexia, vomiting, cough, and rash); the average numbers of visits per child were 4.5 (95 percent CI: 4.2, 4.9) and 3.9 (95 percent CI: 3.7, 4.1) for ALL cases and controls, respectively. Findings were similar when consultations involving potentially infectious symptoms but with no definitive diagnoses were excluded: The average number of visits with a clinically diagnosed infection was 3.6 (95 percent CI: 3.3, 3.9) for children with ALL and 3.1 (95 percent CI: 2.9, 3.2) for controls. For other types of consultation (e.g., routine health checks and follow-up visits, feeding problems, general irritability, diaper (nappy) rash, etc.), no significant differences were observed; the average number of visits totaled approximately three in controls and just over three in all leukemia groups.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Frequency of general practitioner consultation{dagger} in the first year of life for children with leukemia, United Kingdom Childhood Cancer Study, 1991–1996

 
The excess of clinically diagnosed infection in children who developed leukemia between ages 2 and 5 years is apparent from the first month of life (the neonatal period), as can be seen from table 4, where data on total infections are accompanied by data on the six most commonly diagnosed infant infections. Overall, 18 percent of controls and 24 percent of ALL cases were diagnosed with at least one infection during their first month of life (odds ratio (OR) = 1.4, 95 percent CI: 1.1, 1.9; p = 0.016); and by the end of their first year, this figure had risen to 85 percent among controls and 88 percent among ALL cases (OR = 1.3, 95 percent CI: 0.9, 1.8).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Numbers of child leukemia patients whose parents consulted a general practitioner for an infectious illness at least once during the child's first month and/or year of life and related odds ratios, United Kingdom Childhood Cancer Study, 1991–1996

 
The increase in risk associated with early infection appeared to be fairly generalized, and with the exception of infections of the lower respiratory tract, all of the odds ratios shown in table 4 are raised. However, the type of infection diagnosed in infancy varied with age. Fungal conditions tended to occur early (the median age at first diagnosis was 2.5 months in controls), and the odds ratios for ALL were 1.9 (95 percent CI: 1.1, 3.2; p = 0.025) and 1.4 (95 percent CI: 1.0, 1.9; p = 0.035) for the first month and the first year of life, respectively. In our data, all case children diagnosed with a fungal infection in the first 12 months of life had Candida albicans of the mouth and/or skin, as did all but two of the controls (both of whom had tinea or "ringworm"). Only four (7 percent) of the neonates (two controls and two cases with ALL) had received an antibiotic prescription preceding their fungal diagnosis.

While fungal conditions are comparatively easy to diagnose, separating conditions of viral origin from those of bacterial origin is not straightforward at any age and is particularly difficult in young babies. For example, while it seems likely that most of the upper respiratory tract infections were viral (for ALL diagnosed at age ≤12 months, OR = 1.3, 95 percent CI: 1.0, 1.7; p = 0.024) and most of the ear infections (otitis media) were bacterial (for ALL diagnosed at age ≤12 months, OR = 1.2, 95 percent CI: 0.9, 1.6), some will have had other causes and those of the gastrointestinal tract will have had mixed etiologies. Accordingly, further separation of the data by agent was not attempted.

As can be seen from figure 1, children who visited their general practitioner more than once with a neonatal infection tended to be diagnosed with ALL approximately 7 months earlier than children with only one infection or no infections (p < 0.01). With respect to cytogenetic subtype, data were available for 365 (86 percent) children diagnosed with ALL between 2 and 5 years of age. Within this group, no obvious differences between the types of leukemia among children who had infections and those who did not were apparent (p > 0.05). Furthermore, the findings did not vary by UKCCS administrative region, number of controls per case (two or one), or method of analysis (unconditional or conditional logistic regression). Reassuringly, with respect to the latter two factors, the odds ratios for the Oxford region (100 percent of cases and 98.3 percent of controls) for neonatal infections were 2.1 (95 percent CI: 1.1, 4.1; p = 0.022) and 1.9 (95 percent CI: 1.1, 3.5; p = 0.031) as calculated by conditional and unconditional (adjusted for matching variables) logistic regression, respectively.


Figure 1
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIGURE 1. Box-and-whisker plots showing the median (horizontal line), interquartile range (edges of box), and full range (bars) of age at diagnosis (months) of acute lymphoblastic leukemia among children with zero, one, and two or more clinically diagnosed neonatal infections, United Kingdom Childhood Cancer Study, 1991–1996. Outliers are indicated by open circles. Numbers of children and mean ages (months) at diagnosis for cases with zero, one, and two or more neonatal infections were 324 and 45.2 (95% confidence interval (CI): 43.8, 46.6), 78 and 45.8 (95% CI: 42.8, 48.7), and 23 and 37.7 (95% CI: 32.5, 42.8), respectively.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In the UKCCS, we found that children who developed ALL at the diagnostic peak, between the ages of 2 and 5 years, had more clinically diagnosed infections in the first year of life than unaffected children. This excess was apparent from the first month of life, with children with more than one neonatal episode of infectious illness developing their leukemia at a younger age than children with only one episode or none. In this context, it is important to remember that our analyses were restricted to leukemias diagnosed between the ages of 2 and 5 years—hence, the interval between neonatal infection and subsequent leukemia diagnosis was at least 23 months.

To our knowledge, the relation between clinically diagnosed infection in the first year of life and subsequent leukemia development has not been previously explored. In the absence of definitive serologic evidence, this measure is probably the most specific indicator of infection, since infectious illness requires infectious exposure. Clearly, however, not all exposures lead to symptoms, and not all symptoms result in a clinical diagnosis of illness. Indeed, even antibody detection does not yield information on timing, dose, or exposure to uncharacterized agents; and in this sense, it is impossible to directly test the delayed-infection hypothesis (9).

The lack of obvious bias is critical to the interpretation of the analyses presented here. The data came from medical records compiled contemporaneously before diagnosis/interview, and our findings were not affected by recall or reporting biases—problems that bedevil studies that rely on parental self-reporting (21, 25–27, 39–42). It also seems unlikely that our results could have been biased in any other way, since ascertainment was high (94 percent) for leukemias diagnosed between the ages of 2 and 5 years, with all cases having information from at least one of their UKCCS administrative area-matched controls. Furthermore, the deprivation distributions of cases and controls were similar (43), and no evidence of any systematic socioeconomic variation in primary-care attendance patterns in the first year of life was found, either across the UKCCS as a whole or within any specific geographic area. The homogeneity of infant consultation patterns reported here and elsewhere (44, 45) contrasts with the pattern seen for older children, where sociodemographic circumstances influence some (46, 47), but by no means all (48), health-care-seeking behaviors. Hence, while our findings for infant consultations are unlikely to have been biased by socioeconomic differences between cases and controls, this is less certain for other potential proxies of infectious disease exposure, such as parental occupation (15, 17, 18), day-care attendance outside the home (24, 26, 27), and infant feeding practices (49).

Taken at face value, our results do not support our original hypothesis that a deficit of exposure to infectious agents—as measured here by clinically diagnosed illness early in life—is associated with an increased risk of ALL development (12). Likewise, the generality of the increased risk observed across a range of infections seems counter to the notion of a specific leukemogenic agent (8). However, the UKCCS broader hypothesis (10) on infection predicts that genetic susceptibility may underpin an abnormal immune response, and we have previously reported a positive association between B-cell precursor common ALL and certain human leukocyte antigen DP alleles (50). Other immune-response genes are currently being explored.

Although the suggestion that pediatric ALL originates prenatally or within the first few weeks of life is not new (51), there is now compelling molecular evidence that childhood ALL may be initiated in utero (52–54). In this context, an abnormal or dysregulated immune response to infection after birth may promote subsequent genetic events required for transition to overt ALL (9). Our previous analysis using formal day-care attendance as a proxy for early infections indicated that early exposure may, as predicted by the "delayed infection" hypothesis, provide some degree of protection against ALL (24). Our current data argue somewhat differently and indicate that in at least some cases, early infection is positively associated with early-onset ALL. One possibility that merits further exploration is that an abnormal immune response to infection, promoting childhood ALL, can arise either as a result of a lack of exposure in infancy, with a consequent failure of immune network modulation, or as a result of a genetically biased immune response that may be independent of the timing of infectious disease exposure.


   ACKNOWLEDGMENTS
 
The United Kingdom Childhood Cancer Study (UKCCS) was sponsored and administered by the Leukaemia Research Fund. The study was conducted by 12 teams of investigators (10 clinical and epidemiologic and two biologic) based in university departments, research institutes, and the National Health Service in Scotland. The work was coordinated by a management committee. It was supported by the United Kingdom Children's Cancer Study Group (a group of pediatric oncologists) and by the National Radiological Protection Board. Financial support was provided by the Cancer Research Campaign, the Imperial Cancer Research Fund, the Leukaemia Research Fund, and the Medical Research Council through grants to their units; by the Leukaemia Research Fund, the Department of Health, the Electricity Association, the Irish Electricity Supply Board, the National Grid Company plc (Warwick, United Kingdom), and Westlakes Research (Trading) Ltd. (Cumbria, United Kingdom) through grants for the general expenses of the study; by the Kay Kendall Leukaemia Fund for the associated laboratory studies; and by the Foundation for Children with Leukaemia for the study of electric fields.

The authors thank the members of the United Kingdom Childhood Cancer Study Group for their support. They also thank the staffs of the local hospitals, the general practitioners and their staffs, and the UKCCS interviewers and technicians.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Ward G. (1917) Infective theory of acute leukaemia. Br J Child Dis 14:10–20.
  2. Cooke J. (1942) The incidence of acute leukemia in children. JAMA 119:547–50.[Abstract/Free Full Text]
  3. Stewart A, Webb J, Hewitt D. (1958) A survey of childhood malignancies. Br Med J 30:1495–507.
  4. Greaves MF. (1988) Speculations on the cause of childhood acute lymphoblastic leukemia. Leukemia 2:120–5.[Web of Science][Medline]
  5. Smith M. (1997) Considerations on a possible viral etiology for B-precursor acute lymphoblastic leukemia of childhood. J Immunother 20:89–100.
  6. Kinlen L. (1988) Evidence for an infective cause of childhood leukaemia: comparison of a Scottish new town with nuclear reprocessing sites in Britain. Lancet 2:1323–7.[CrossRef][Web of Science][Medline]
  7. Smith MA, Simon R, Strickler HD, et al. (1998) Evidence that childhood acute lymphoblastic leukemia is associated with an infectious agent linked to hygiene conditions. Cancer Causes Control 9:285–98.[CrossRef][Web of Science][Medline]
  8. Kinlen LJ. (1995) Epidemiological evidence for an infective basis in childhood leukaemia. Br J Cancer 71:1–5.[Web of Science][Medline]
  9. Greaves M. (2006) Infection, immune responses and the aetiology of childhood leukaemia. Nat Rev Cancer 6:193–203.[CrossRef][Web of Science][Medline]
  10. United Kingdom Childhood Cancer Study Investigators. (2000) The United Kingdom Childhood Cancer Study: objectives, materials and methods. United Kingdom Childhood Cancer Study Investigators. Br J Cancer 82:1073–102.[CrossRef][Web of Science][Medline]
  11. Kinlen L. (2004) Infections and immune factors in cancer: the role of epidemiology. Oncogene 23:6341–8.[CrossRef][Web of Science][Medline]
  12. Greaves MF. (1997) Aetiology of acute leukaemia. Lancet 349:344–9.[CrossRef][Web of Science][Medline]
  13. Dockerty JD, Draper G, Vincent T, et al. (2001) Case-control study of parental age, parity and socioeconomic level in relation to childhood cancers. Int J Epidemiol 30:1428–37.[Abstract/Free Full Text]
  14. Adelman AS, McLaughlin CC, Wu XC, et al. (2005) Urbanisation and incidence of acute lymphocytic leukaemia among United States children aged 0–4. Br J Cancer 92:2084–8.[CrossRef][Web of Science][Medline]
  15. Kinlen LJ, Hudson CM, Stiller CA. (1991) Contacts between adults as evidence for an infective origin of childhood leukaemia: an explanation for the excess near nuclear establishments in west Berkshire? Br J Cancer 64:549–54.[Web of Science][Medline]
  16. Roman E, Watson A, Bull D, et al. (1994) Leukaemia risk and social contact in children aged 0–4 years in southern England. J Epidemiol Community Health 48:601–2.[Free Full Text]
  17. Kinlen L, Jiang J, Hemminki K. (2002) A case-control study of childhood leukaemia and paternal occupational contact level in rural Sweden. Br J Cancer 86:732–7.[CrossRef][Web of Science][Medline]
  18. Fear NT, Simpson J, Roman E. (2005) Childhood cancer and social contact: the role of paternal occupation (United Kingdom). Cancer Causes Control 16:1091–7.[CrossRef][Web of Science][Medline]
  19. Ma X, Does MB, Metayer C, et al. (2005) Vaccination history and risk of childhood leukaemia. Int J Epidemiol 34:1100–9.[Abstract/Free Full Text]
  20. Kneale GW, Stewart AM, Wilson LM. (1986) Immunizations against infectious diseases and childhood cancers. Cancer Immunol Immunother 21:129–32.[Web of Science][Medline]
  21. Dockerty JD, Skegg DCG, Elwood JM, et al. (1999) Infections, vaccinations, and the risk of childhood leukaemia. Br J Cancer 80:1483–9.[CrossRef][Web of Science][Medline]
  22. Groves FD, Gridley G, Wacholder S, et al. (1999) Infant vaccinations and risk of childhood acute lymphoblastic leukaemia in the USA. Br J Cancer 81:175–8.[CrossRef][Web of Science][Medline]
  23. Roman E, Simpson J, Ansell P, et al. (2005) Perinatal and reproductive factors: a report on haematological malignancies from the UKCCS. Eur J Cancer 41:749–59.[CrossRef][Web of Science][Medline]
  24. Gilham C, Peto J, Simpson J, et al. (2005) Day care in infancy and risk of childhood acute lymphoblastic leukaemia: findings from UK case-control study. BMJ 330:1294–9.[Abstract/Free Full Text]
  25. Schuz J, Kaletsch U, Meinert R, et al. (1999) Association of childhood leukaemia with factors related to the immune system. Br J Cancer 80:585–90.[CrossRef][Web of Science][Medline]
  26. Neglia JP, Linet MS, Shu XO, et al. (2000) Patterns of infection and day care utilization and risk of childhood acute lymphoblastic leukaemia. Br J Cancer 82:234–40.[CrossRef][Web of Science][Medline]
  27. Jourdan-Da Silva N, Perel Y, Mechinaud F, et al. (2004) Infectious diseases in the first year of life, perinatal characteristics and childhood acute leukaemia. Br J Cancer 90:139–45.[CrossRef][Web of Science][Medline]
  28. Westergaard T, Andersen PK, Pedersen JB, et al. (1997) Birth characteristics, sibling patterns, and acute leukemia risk in childhood: a population-based cohort study. J Natl Cancer Inst 89:939–47.[Abstract/Free Full Text]
  29. Ansell P, Mitchell CD, Roman E, et al. (2005) Relationships between perinatal and maternal characteristics and hepatoblastoma: a report from the UKCCS. Eur J Cancer 41:741–8.[CrossRef][Web of Science][Medline]
  30. World Health Organization. (1992) ICD-10: international statistical classification of diseases and related health problems. (World Health Organization, Tenth Revision. Geneva, Switzerland).
  31. British Medical Association, Royal Pharmaceutical Society of Great Britain. (1992) British National Formulary—number 24 (September 1992). (British Medical Association and the Pharmaceutical Press, London, United Kingdom).
  32. Day N, Skinner J, Roman E, et al. (1999) Exposure to power-frequency magnetic fields and the risk of childhood cancer. United Kingdom Childhood Cancer Study Investigators. Lancet 354:1925–31.[CrossRef][Web of Science][Medline]
  33. Roizen NJ and Patterson D. (2003) Down's syndrome. Lancet 361:1281–9.[CrossRef][Web of Science][Medline]
  34. Fear NT, Roman E, Ansell P, et al. (2003) Vitamin K and childhood cancer: a report from the United Kingdom Childhood Cancer Study. Br J Cancer 89:1228–31.[CrossRef][Web of Science][Medline]
  35. Skinner J, Maslanyj M, Mee TJ, et al. (2000) Childhood cancer and residential proximity to power lines. United Kingdom Childhood Cancer Study Investigators. Br J Cancer 83:1573–80.[CrossRef][Web of Science][Medline]
  36. McKinney PA, Fear NT, Stockton D, et al. (2003) Parental occupation at periconception: findings from the United Kingdom Childhood Cancer Study. Occup Environ Med 60:901–9.[Abstract/Free Full Text]
  37. Breslow NE and Day NE. (1980) Statistical methods in cancer research. Vol 1. The analysis of case-control studies. (IARC Scientific Publication no. 32). (International Agency for Research on Cancer, Lyon, France).
  38. Corporation Stata. (2003) Stata statistical software: release 8.2. (Stata Corporation, College Station, TX).
  39. Rosenbaum PF, Buck GM, Brecher ML. (2005) Allergy and infectious disease histories and the risk of childhood acute lymphoblastic leukaemia. Paediatr Perinat Epidemiol 19:152–64.[Web of Science][Medline]
  40. Infante-Rivard C, Fortier I, Olson E. (2000) Markers of infection, breast-feeding and childhood acute lymphoblastic leukaemia. Br J Cancer 83:1559–64.[CrossRef][Web of Science][Medline]
  41. Perrillat F, Clavel J, Auclerc MF, et al. (2002) Day-care, early common infections and childhood acute leukaemia: a multicentre French case-control study. Br J Cancer 86:1064–9.[CrossRef][Web of Science][Medline]
  42. Chan LC, Lam TH, Li CK, et al. (2002) Is the timing of exposure to infection a major determinant of acute lymphoblastic leukaemia in Hong Kong? Paediatr Perinat Epidemiol 16:154–65.[CrossRef][Web of Science][Medline]
  43. Smith A, Roman E, Simpson J, et al. (2006) Childhood leukaemia and socioeconomic status: fact or artefact? A report from the United Kingdom Childhood Cancer Study (UKCCS). Int J Epidemiol Aug 31(e-pub ahead of print). (DOI: 10.1093/ije/dyl193).
  44. McConnachie A, Wilson P, Thomson H, et al. (2004) Modelling consultation rates in infancy: influence of maternal and infant characteristics, feeding type and consultation history. Br J Gen Pract 54:598–603.[Web of Science][Medline]
  45. Goldfeld SR, Wright M, Oberklaid F. (2003) Parents, infants and health care: utilization of health services in the first 12 months of life. J Paediatr Child Health 39:249–53.[CrossRef][Web of Science][Medline]
  46. Arber S. (1987) Social class, non-employment, and chronic illness: continuing the inequalities in health debate. Br Med J (Clin Res Ed) 294:1069–73.
  47. Saxena S, Eliahoo J, Majeed A. (2002) Socioeconomic and ethnic group differences in self reported health status and use of health services by children and young people in England: cross sectional study. BMJ 325:520–5.[Abstract/Free Full Text]
  48. Cooper H, Smaje C, Arber S. (1998) Use of health services by children and young people according to ethnicity and social class: secondary analysis of a national survey. BMJ 317:1047–51.[Abstract/Free Full Text]
  49. Beral V, Fear NT, Alexander F, et al. (2001) Breastfeeding and childhood cancer. Br J Cancer 85:1685–94.[CrossRef][Web of Science][Medline]
  50. Taylor GM, Dearden S, Ravetto P, et al. (2002) Genetic susceptibility to childhood common acute lymphoblastic leukaemia is associated with polymorphic peptide-binding pocket profiles in HLA-DPB1*0201. Hum Mol Genet 11:1585–97.[Abstract/Free Full Text]
  51. Stewart A. (1961) Aetiology of childhood malignancies: II. Congenitally determined leukaemias. Br Med J 1:452–60.[Free Full Text]
  52. Greaves MF and Wiemels J. (2003) Origins of chromosome translocations in childhood leukaemia. Nat Rev Cancer 3:639–49.[CrossRef][Web of Science][Medline]
  53. Greaves MF, Maia AT, Wiemels JL, et al. (2003) Leukemia in twins: lessons in natural history. Blood 102:2321–33.[Abstract/Free Full Text]
  54. Roman E, Doyle D, Lightfoot T, et al. (2006) Molar pregnancy, childhood cancer and genomic imprinting—is there a link? Hum Fertil 9:171–4.[CrossRef]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
A. M. Hughes, S. Crouch, T. Lightfoot, P. Ansell, J. Simpson, and E. Roman
Eczema, Birth Order, and Infection
Am. J. Epidemiol., May 15, 2008; 167(10): 1182 - 1187.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
M. T. Dorak, R. J. Q. McNally, and L. Parker
RE: "CHILDHOOD ACUTE LYMPHOBLASTIC LEUKEMIA AND INFECTIONS IN THE FIRST YEAR OF LIFE: A REPORT FROM THE UNITED KINGDOM CHILDHOOD CANCER STUDY"
Am. J. Epidemiol., August 1, 2007; 166(3): 364 - 365.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
165/5/496    most recent
kwk039v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Roman, E
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Roman, E
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?