Population mixing and childhood leukaemia and non-Hodgkin's lymphoma in census wards in England and Wales, 1966-87

The Association Between Childhood Leukemia and Population Mixing: An Artifact of Focusing on Clusters?

Supplemental Digital Content is available in the text.

Background:

Studies investigating the population-mixing hypothesis in childhood leukemia principally use two analytical approaches: (1) nonrandom selection of areas according to specific characteristics, followed by comparisons of their incidence of childhood leukemia with that expected based on the national average; and (2) regression analyses of region-wide data to identify characteristics associated with the incidence of childhood leukemia. These approaches have generated contradictory results. We compare these approaches using observed and simulated data.

Methods:

We generated 10,000 simulated regions using the correlation structure and distributions from a United Kingdom dataset. We simulated cases using a Poisson distribution with the incidence rate set to the national average assuming the null hypothesis that only population size drives the number of cases. Selection of areas within each simulated region was based on characteristics considered responsible for elevated infection rates (population density and inward migration) and/or elevated leukemia rates. We calculated effect estimates for 10,000 simulations and compared results to corresponding observed data analyses.

Results:

When the selection of areas for analysis is based on apparent clusters of childhood leukemia, biased assessments occur; the estimated 5-year incidence of childhood leukemia ranged between zero and eight per 10,000 children in contrast to the simulated two cases per 10,000 children, similar to the observed data. Performing analyses on region-wide data avoids these biases.

Conclusions:

Studies using nonrandom selection to investigate the association between childhood leukemia and population mixing are likely to have generated biased findings. Future studies can avoid such bias using a region-wide analytical strategy. See video abstract at, http://links.lww.com/EDE/B431.

Childhood cancer research in oxford III: The work of CCRG on ionising radiation

BACKGROUND

High doses of ionising radiation are a known cause of childhood cancer and great public and professional interest attaches to possible links between childhood cancer and lower doses, particularly of man-made radiation. This paper describes work done by the Childhood Cancer Research Group (CCRG) on this topic METHODS: Most UK investigations have made use of the National Registry of Childhood Tumours and associated controls. Epidemiological investigations have included national incidence and mortality analyses, geographical investigations, record linkage and case-control studies. Dosimetric studies use biokinetic and dosimetric modelling.

RESULTS

This paper reviews the work of the CCRG on the association between exposure to ionising radiation and childhood cancer, 1975-2014.

CONCLUSION

The work of CCRG has been influential in developing understanding of the causes of 'clusters' of childhood cancer and the risks arising from exposure to ionising radiation both natural and man-made. Some clusters around nuclear installations have certainly been observed, but ionising radiation does not seem to be a plausible cause. The group's work has also been instrumental in discounting the hypothesis that paternal preconception irradiation was a cause of childhood cancers and has demonstrated an increased leukaemia risk for children exposed to higher levels of natural gamma-ray radiation.

Childhood cancer research in Oxford II: The Childhood Cancer Research Group

BACKGROUND

We summarise the work of the Childhood Cancer Research Group, particularly in relation to the UK National Registry of Childhood Tumours (NRCT).

METHODS

The Group was responsible for setting up and maintaining the NRCT. This registry was based on notifications from regional cancer registries, specialist children's tumour registries, paediatric oncologists and clinical trials organisers. For a large sample of cases, data on controls matched by date and place of birth were also collected.

RESULTS

Significant achievements of the Group include: studies of aetiology and of genetic epidemiology; proposals for, and participation in, international comparative studies of these diseases and on a classification system specifically for childhood cancer; the initial development of, and major contributions to, follow-up studies of the health of long-term survivors; the enhancement of cancer registration records by the addition of clinical data and of birth records. The Group made substantial contributions to the UK government's Committee on Medical Aspects of Radiation in the Environment.

CONCLUSION

An important part of the ethos of the Group was to work in collaboration with many other organisations and individuals, both nationally and internationally: many of the Group's achievements described here were the result of such collaborations.

A causal mechanism for childhood acute lymphoblastic leukaemia

In this Review, I present evidence supporting a multifactorial causation of childhood acute lymphoblastic leukaemia (ALL), a major subtype of paediatric cancer. ALL evolves in two discrete steps. First, in utero initiation by fusion gene formation or hyperdiploidy generates a covert, pre-leukaemic clone. Second, in a small fraction of these cases, the postnatal acquisition of secondary genetic changes (primarily V(D)J recombination-activating protein (RAG) and activation-induced cytidine deaminase (AID)-driven copy number alterations in the case of ETS translocation variant 6 (ETV6)-runt-related transcription factor 1 (RUNX1)+ ALL) drives conversion to overt leukaemia. Epidemiological and modelling studies endorse a dual role for common infections. Microbial exposures earlier in life are protective but, in their absence, later infections trigger the critical secondary mutations. Risk is further modified by inherited genetics, chance and, probably, diet. Childhood ALL can be viewed as a paradoxical consequence of progress in modern societies, where behavioural changes have restrained early microbial exposure. This engenders an evolutionary mismatch between historical adaptations of the immune system and contemporary lifestyles. Childhood ALL may be a preventable cancer.

Can changes in population mixing and socio-economic deprivation in Cumbria, England explain changes in cancer incidence around Sellafield?

Previously excesses in incident cases of leukaemia and non-Hodgkin lymphoma have been observed amongst young people born or resident in Seascale, Cumbria. These excesses have not been seen more recently. It is postulated that the former apparent increased risk was related to 'unusual population mixing', which is not present in recent years. This study investigated changes in measures of population mixing from 1951-2001. Comparisons were made between three specified areas. Area-based measures were calculated (migration, commuting, deprivation, population density). All areas have become more affluent, although Seascale was consistently the most affluent. Seascale has become less densely populated, with less migration into the ward and less diversity with respect to migrants' origin. There have been marked changes in patterns of population mixing throughout Cumbria. Lesser population mixing has been observed in Seascale in recent decades. Changes in pattern and nature of population mixing may explain the lack of recent excesses.

Population mixing for leukaemia, lymphoma and CNS tumours in teenagers and young adults in England, 1996-2005

Background

Little aetiological epidemiological research has been undertaken for major cancers occurring in teenagers and young adults (TYA). Population mixing, as a possible proxy for infectious exposure, has been well researched for childhood malignancies. We aimed to investigate effects of population mixing in this older age group using an English national cancer dataset.

Methods

Cases of leukaemia, lymphoma and central nervous system (CNS) tumours amongst 15–24 year olds in England (diagnosed 1996–2005) were included in the study. Data were obtained by ward of diagnosis and linked to 1991 census variables including population mixing (Shannon index); data on person-weighted population density and deprivation (Townsend score) were also used and considered as explanatory variables. Associations between TYA cancer incidence and census variables were investigated using negative binomial regression, and results presented as incidence rate ratios (IRR) with 95% confidence intervals (CI).

Results

A total of 6251 cases of leukaemia (21%), lymphoma (49%) and CNS tumours (30%) were analysed. Higher levels of population mixing were associated with a significant decrease in the incidence of CNS tumours (IRR = 0.83, 95% CI = 0.75-0.91), accounted for by astrocytomas and ‘other CNS tumours’; however, there was no association with leukaemia or lymphoma. Incidence of CNS tumours and lymphoma was 3% lower in more deprived areas (IRR = 0.97, 95% CI = 0.96-0.99 and IRR = 0.97, 95% CI =0.96-0.98 respectively). Population density was not associated with the incidence of leukaemia, lymphoma or CNS tumours.

Conclusions

Our results suggest a possible role for environmental risk factors with population correlates in the aetiology of CNS tumours amongst TYAs. Unlike studies of childhood cancer, associations between population mixing and the incidence of leukaemia and lymphoma were not observed.

An examination, with a meta-analysis, of studies of childhood leukaemia in relation to population mixing

Background:

Marked influxes of people into rural areas, termed rural population mixing (PM), have been associated with excesses of childhood leukaemia (CL), consistent with mini-epidemics of a mainly immunising, subclinical infection to which CL is a rare response. For such situations of rural PM would promote contacts between infected and susceptible individuals, the latter tending to have a higher than average prevalence in rural or isolated areas. Confusion has arisen from some workers applying the term PM to non-rural situations lacking known recent change.

Methods:

Available PM studies using the original definition of influxes were examined, a meta-analysis carried out of studies of CL in relation to exposure to high levels of rural PM, and also a detailed analysis by age group.

Results:

The meta-analysis of 17 studies shows a significant CL excess in association with rural PM: overall relative risk (RR) at ages 0–14: 1.57; 95% confidence interval 1.44–1.72; at 0–4 years 1.72 (1.54–1.91). This contrasts with the absence of an excess of CL in similarly exposed urban areas (RR 1.00; 0.93–1.07), pointing to a high level of immunity there. The mixed results of studies using other definitions of PM were summarised. The excess associated with rural PM below age 2 years (RR 1.51; 1.17, 1.92) was not appreciably different from that at later childhood ages.

Conclusion:

Much of the inconsistency among studies ostensibly about CL and PM reflects the use of definitions other than that originally proposed. The broad similarity of the CL excess below age 2 with that at older childhood ages is inconsistent with the Greaves’ delayed infection hypothesis, since any infection underlying the former is difficult to consider as delayed.

Environmental factors and childhood acute leukemias and lymphomas

This review considers recent studies regarding the role of environmental factors in the etiology of childhood leukemia and lymphoma. Potential environmental risk factors identified for childhood leukemia include exposure to magnetic fields of more than 0.4 micro Tessla, exposure to pesticides, solvents, benzene and other hydrocarbons, maternal alcohol consumption (but only for certain genotypes), contaminated drinking water, infections, and high birth weight. The finding of space-time clustering and seasonal variation also supports a role for infections. There is little evidence linking childhood leukemia with lifetime exposure to ionizing radiation although fetal exposures to X-rays are associated with increased risk. Breast-feeding, consumption of fresh fruit and vegetables and having allergies all appear to be protective. Burkitt lymphoma (BL) is confined to areas of the world where malaria is endemic, with the additional involvement of the Epstein-Barr virus (EBV) as a co-factor. Environmental risk factors suggested for other types of non-Hodgkin lymphoma (NHL) include exposure to ionizing radiation (both lifetime and antenatal), pesticides, and, in utero exposure to cigarette smoke, benzene and nitrogen dioxide (via the mother). Hodgkin lymphoma (HL) is especially associated with higher levels of socioeconomic deprivation, but breast-feeding seems to confer lower risk. This is consistent with an infection or immune-response mediated etiology for HL.

Does population mixing measure infectious exposure in children at the community level?

Epidemiological studies focusing on the etiology of childhood chronic diseases have used population mixing as a proxy for the level of infection circulating in a community. We compared different measures of population mixing (based on residential migration and commuting) and other demographic variables, derived from the United Kingdom Census, with hospital inpatient data on infections from two Government Office Regions in England (Eastern and the West Midlands) to inform the development of an infectious disease proxy for future epidemiological studies. The association between rates of infection and the population mixing measures was assessed, using incidence rate ratios across census areas, from negative binomial regression. Commuting distance demonstrated the most consistent association with admissions for infections across the two regions; areas with a higher median distance travelled by commuters leaving the area having a lower rate of hospital admissions for infections. Deprived areas and densely populated areas had a raised rate of admissions for infections. Assuming hospital admissions are a reliable indicator of common infection rates, the results from this study suggest that commuting distance is a consistent measure of population mixing in relation to infectious disease and deprivation and population density are reliable demographic proxies for infectious exposure. Areas that exhibit high levels of population mixing do not necessarily possess raised rates of hospital admissions for infectious disease.

What do epidemiologists mean by 'population mixing'?

There is growing evidence that some chronic diseases are caused, or promoted, by infectious disease. 'Population mixing' has been used as a proxy for the range and dose of infectious agents circulating in a community. Given the speculation over the role of population mixing in many chronic diseases, we review the various methods used for measuring population mixing, and provide a classification of these. We recommend that authors fulfill two criteria in publications: measures are demonstrably associated with the putative risk factors for which population-mixing is acting as a proxy and fundamental characteristics of the chosen measures are clearly defined.

Population mixing, socioeconomic status and incidence of childhood acute lymphoblastic leukaemia in England and Wales: analysis by census ward

In this population-based study of acute lymphoblastic leukaemia (ALL) diagnosed among children aged under 15 years in England and Wales during 1986-1995, we analysed incidence at census ward level in relation to a range of variables from the 1991 census, which could be relevant to theories of infectious aetiology. 'Population-mixing' measures, used as surrogates for quantity and diversity of infections entering the community, were calculated from census data on the origins and destinations of migrants in the year before the census. Incidence at ages 1-4 years tended independently to be higher in rural wards, to increase with the diversity of origin wards from which in-migrants had moved during the year before the census, and to be lower in the most deprived areas as categorised by the Carstairs index. This last association was much weaker when urban/rural status and in-migrants' diversity were allowed for. There was no evidence of association with population mixing or deprivation for ALL diagnosed at ages 0 or 5-14 years. The apparent specificity to the young childhood age group suggests that these associations are particularly marked for precursor B-cell ALL, with the disease more likely to occur when delayed exposure to infection leads to increased immunological stress, as predicted by Greaves. The association with diversity of incomers, especially in rural areas, is also consistent with the higher incidence of leukaemia predicted by Kinlen, where population mixing results in below average herd immunity to an infectious agent.

Childhood leukaemia and population movements in France, 1990-2003

In a national study, we investigated the incidence of childhood leukaemia (CL) over a 14-year period in France in relation to several measures based on the proportion of individuals who changed address between the last two national censuses. A positive association was found with the proportion of migrants who came from a distant place. The further the migrants came, the higher was the incidence of leukaemia, particularly among children aged 0-4 years in 'isolated' communes at the time of diagnosis (RR=1.4, 95% CI: 1.1,1.8 in the highest category of migration distance). Although the role of the population density was less obvious, a more marked association was found above a certain threshold. No association with the proportion of commuters was observed.

Childhood acute lymphoblastic leukaemia in relation to population mixing around the time of birth in South Hungary

In a retrospective epidemiological study of 481,984 live births in South Hungary, we investigated whether higher levels of population mixing around the time of birth is a risk factor for acute lymphoblastic leukemia (ALL) under age 5 years. Poisson regression was used to investigate the relationship between risk of ALL and the population-mixing index based on the number of incomers in each county district for each year, standardized to have a range of 0-1. Among all children, the risk of ALL increased significantly with increasing population mixing around the time of birth (trend across the range of 0-1 RR = 2.1 95% CI: 1.02-4.44). This effect was more marked for boys (RR = 3.1 95% CI: 1.13-8.51), which supports a sex-specific effect of exposures on risk of ALL.

Population-mixing at the place of residence at the time of birth and incidence of childhood leukaemia in France

The association between the risk of childhood leukaemia before age 7 years and population-mixing at the place of residence at birth was investigated by retrospectively considering all the children born in mainland French communes between 1st January 1990 and 31st December 1998. An increased risk of acute lymphoblastic leukaemia was found with higher levels of migration for children residing at birth in isolated communes with a population density > or =50 people per km2 (SIRR = 2.59, 95% CI: 1.48-4.49). No association was observed with lower population densities. For children residing in non-isolated communes at birth, the results were similar but less marked. The risk tended to increase only for population densities > or =5000 people per km2 (SIRR = 1.57, 95% CI: 0.99-2.52). The findings are consistent with epidemic models and support the hypothesis of an infectious aetiology relating to population-mixing. Population density may be seen as an indicator of the opportunity of contacts between inhabitants and should therefore be taken into account when investigating an infectious hypothesis. This is the first systematic study of population-mixing at the place of residence at the time of birth to be conducted on a national scale.

Detecting small-area similarities in the epidemiology of childhood acute lymphoblastic leukemia and diabetes mellitus, type 1: a Bayesian approach

Childhood acute lymphoblastic leukemia and diabetes mellitus, type 1, have common epidemiologic and etiologic features, including correlated international incidence and associations with infections. The authors examined whether the diseases' similar large-scale distributions are reflected in small geographic areas while also examining the influence of sociodemographic characteristics. Details of 299 children (0-14 years) with acute lymphoblastic leukemia and 1,551 children with diabetes diagnosed between 1986 and 1998 were extracted from two registers in Yorkshire, United Kingdom. Standardized incidence ratios across 532 electoral wards were compared using Poisson regression, confirming significant associations between population mixing and the geographic heterogeneity of both conditions. Bayesian methods analysis of spatial correlation between diseases by modeling a bivariate outcome based on their standardized incidence ratios was applied; spatial and heterogeneity components were included within a hierarchical random effects model. A positive correlation between diseases of 0.33 (95% credible interval: -0.20, 0.74) was observed, and this was reduced after control for population mixing (r = 0.18), population density (r = 0.14), and deprivation (r = 0.06). The Bayesian approach showed a modest but nonsignificant joint spatial correlation between diseases, only partially suggesting that the risk of both was associated within some electoral wards. With Bayesian methodology, population mixing remained significantly associated with both diseases. The links between diabetes and acute lymphoblastic leukemia observed for large regions are weaker for small areas. More powerful replications are needed for confirmation of these findings.

An infectious aetiology for childhood acute leukaemia: a review of the evidence

There are three current hypotheses concerning infectious mechanisms in the aetiology of childhood leukaemia: exposure in utero or around the time of birth, delayed exposure beyond the first year of life to common infections and unusual population mixing. No specific virus has been definitively linked with childhood leukaemia and there is no evidence to date of viral genomic inclusions within leukaemic cells. The case-control and cohort studies have revealed equivocal results. Maternal infection during pregnancy has been linked with increased risk whilst breast feeding and day care attendance in the first year of life appear to be protective. There is inconclusive evidence from studies on early childhood infectious exposures, vaccination and social mixing. Some supportive evidence for an infectious aetiology is provided by the findings of space-time clustering and seasonal variation. Spatial clustering suggests that higher incidence is confined to specific areas with increased levels of population mixing, particularly in previously isolated populations. Ecological studies have also shown excess incidence with higher population mixing. The marked childhood peak in resource-rich countries and an increased incidence of the childhood peak in acute lymphoblastic leukaemia (ALL) (occurring at ages 2-6 years predominantly with precursor B-cell ALL) is supportive of the concept that reduced early infection may play a role. Genetically determined individual response to infection may be critical in the proliferation of preleukaemic clones as evidenced by the human leucocyte antigen class II polymorphic variant association with precursor B-cell and T-cell ALL.

Childhood leukaemia incidence and the population mixing hypothesis in US SEER data

We evaluated the infectious aetiology hypothesis of childhood leukaemia that rapid population influx into rural areas is associated with increased risk. Using data from the US SEER program, we found that in changes in rural county population sizes from 1980 to 1989 were associated with incidence rates for childhood acute lymphocytic leukaemia (ALL). The observed associations were strongest among children 0-4 years of age, born in the same state as diagnosis, in extremely rural counties, and when counties adjacent to nonrural counties were excluded. Similar analyses for brain and central nervous system (CNS) cancer in children, a disease less linked to this infectious hypothesis, provide evidence against methodologic bias. Similar evaluations for other decades were not meaningful due to limited sample sizes and, perhaps, increased population mobility.

Childhood cancer in relation to infections in the community during pregnancy and around the time of birth

In a retrospective cohort study of 404,106 live births in the northern region of England, 1975-1986, we investigated whether higher levels of community infections during the mother's pregnancy and in early life were risk factors for cancer, by diagnostic group (leukaemia and non-Hodgkin's lymphoma, Hodgkin's disease, brain/spinal tumours, neuroblastoma, other tumours), diagnosed 1975-2001 under age 15 years. Logistic regression was used to relate risk to measures of community infections (measles, respiratory and other infections) in 3 prenatal and 2 postnatal quarters. There was an increased risk of Hodgkin's disease among children exposed around birth to higher levels of measles (odds ratio for trend = 2.3, 95% confidence interval 1.3-4.2, p = 0.01). For other diagnostic groups, there was no consistent evidence of an association between risk and exposure to infections. Although the significant association observed for Hodgkin's disease may be a chance finding, consequent to multiple hypothesis testing or the ecologic nature of the study, it is consistent with other recent epidemiologic results suggesting that the risk of Hodgkin's disease may be associated with exposure to infections.

Childhood solid tumours in relation to population mixing around the time of birth

In a retrospective cohort study of 673 787 live births in the Northern Region of England, 1975-1994, we investigated whether a higher level of population mixing around birth was a risk factor for solid tumours, by diagnostic group (Hodgkin's disease, brain and spinal tumours, neuroblastoma, other solid tumours), diagnosed during 1975-2001 under age 15 years. Logistic regression was used to relate risk to population mixing, based on (i) all movers and (ii) incomers from outside the region. Both ward and county district level analyses were performed. There was a decreased risk of brain and spinal tumours with increasing population mixing based on incomers from outside the region (OR for trend across three categories=0.79, 95% CI: 0.66-0.95, P=0.01 in the ward level analysis). Although this may be because of chance, it is consistent with a role of exposure to infection and immunological response in the aetiology of these tumours. For other tumour groups, there was no consistent evidence of an association between risk and population mixing.