Inherited genetic susceptibility to acute lymphoblastic leukemia in Down syndrome.

Children with Down syndrome (DS) have a 20-fold increased risk of acute lymphoblastic leukemia (ALL) and distinct somatic features, including CRLF2 rearrangement in ∼50% of cases; however, the role of inherited genetic variation in DS-ALL susceptibility is unknown. We report the first genome-wide association study of DS-ALL, comprising a meta-analysis of 4 independent studies, with 542 DS-ALL cases and 1192 DS controls. We identified 4 susceptibility loci at genome-wide significance: rs58923657 near IKZF1 (odds ratio [OR], 2.02; Pmeta = 5.32 × 10-15), rs3731249 in CDKN2A (OR, 3.63; Pmeta = 3.91 × 10-10), rs7090445 in ARID5B (OR, 1.60; Pmeta = 8.44 × 10-9), and rs3781093 in GATA3 (OR, 1.73; Pmeta = 2.89 × 10-8). We performed DS-ALL vs non-DS ALL case-case analyses, comparing risk allele frequencies at these and other established susceptibility loci (BMI1, PIP4K2A, and CEBPE) and found significant association with DS status for CDKN2A (OR, 1.58; Pmeta = 4.1 × 10-4). This association was maintained in separate regression models, both adjusting for and stratifying on CRLF2 overexpression and other molecular subgroups, indicating an increased penetrance of CDKN2A risk alleles in children with DS. Finally, we investigated functional significance of the IKZF1 risk locus, and demonstrated mapping to a B-cell super-enhancer, and risk allele association with decreased enhancer activity and differential protein binding. IKZF1 knockdown resulted in significantly higher proliferation in DS than non-DS lymphoblastoid cell lines. Our findings demonstrate a higher penetrance of the CDKN2A risk locus in DS and serve as a basis for further biological insights into DS-ALL etiology.

Risk of solid subsequent malignant neoplasms after childhood Hodgkin lymphoma-Identification of high-risk populations to guide surveillance: A report from the Late Effects Study Group.

BACKGROUND: Survivors of Hodgkin lymphoma (HL) in childhood have an increased risk of subsequent malignant neoplasms (SMNs). Herein, the authors extended the follow-up of a previously reported Late Effects Study Group cohort and identified patients at highest risk for SMNs to create evidence for risk-based screening recommendations. METHODS: The standardized incidence ratio was calculated using rates from the Surveillance, Epidemiology, and End Results program as a reference. The risk of SMN was estimated using proportional subdistribution hazards regression. The cohort included 1136 patients who were diagnosed with HL before age 17 years between 1955 and 1986. The median length of follow-up was 26.6 years. RESULTS: In 162 patients, a total of 196 solid SMNs (sSMNs) were identified. Compared with the general population, the cohort was found to be at a 14-fold increased risk of developing an sSMN (95% confidence interval, 12.0-fold to 16.3-fold). The cumulative incidence of any sSMN was 26.4% at 40 years after a diagnosis of HL. Risk factors for breast cancer among females were an HL diagnosis between ages 10 years and 16 years and receipt of chest radiotherapy. Males treated with chest radiotherapy at age <10 years were found to be at highest risk of developing lung cancer. Survivors of HL who were treated with abdominal/pelvic radiotherapy and high-dose alkylating agents were found to be at highest risk of developing colorectal cancer and females exposed to neck radiotherapy at age <10 years were at highest risk of thyroid cancer. By age 50 years, the cumulative incidence of breast, lung, colorectal, and thyroid cancer was 45.3%, 4.2%, 9.5%, and 17.3%, respectively, among those at highest risk. CONCLUSIONS: Survivors of childhood HL remain at an increased risk of developing sSMNs. In the current study, subgroups of survivors of HL at highest risk of specific sSMNs were identified, and evidence for screening provided.

Fanconi anaemia, BRCA2 mutations and childhood cancer: a developmental perspective from clinical and epidemiological observations with implications for genetic counselling.

Fanconi anaemia (FA) is an inherited condition characterised by congenital and developmental abnormalities and a strong cancer predisposition. In around 3-5% of cases FA is caused by biallelic mutations in the BRCA2 gene. Individuals heterozygous for BRCA2 mutations have an increased risk of inherited breast and ovarian cancer. We reviewed the mutation spectrum in BRCA2-associated FA, and the spectrum and frequency of BRCA2 mutations in distinct populations. The rarity of FA due to biallelic BRCA2 mutations supports a fundamental role of BRCA2 for prevention of malignant transformation during development. The spectrum of malignancies seen associated with FA support the concept of a tissue selectivity of BRCA2 mutations for development of FA-associated cancers. This specificity is illustrated by the distinct FA-associated BRCA2 mutations that appear to predispose to specific brain or haematological malignancies. For some populations, the number of FA-patients with biallelic BRCA2 disruption is smaller than that expected from the carrier frequency, and this implies that some pregnancies with biallelic BRCA2 mutations do not go to term. The apparent discrepancy between expected and observed incidence of BRCA2 mutation-associated FA in high-frequency carrier populations has important implications for the genetic counselling of couples with recurrent miscarriages from high-risk populations.

The contrasting age-incidence patterns of bone tumours in teenagers and young adults: Implications for aetiology.

Bone tumours comprise 0.2% of cancers overall but 5.7% in 15-24 year olds. To explore the relationship with adolescence we have analysed age-incidence patterns of bone tumours in a large national dataset. Data on incident cases of bone tumours in 0-84 year olds in England, 1979-2003, were extracted from national cancer registration data. Incidence rates per million person-years by 5-year age-group, sex, morphology and primary site were calculated and adjusted to the world standard population. Nine thousand one hundred forty-six cases were identified giving an overall age-standardized rate of 7.19 per million person-years. The distribution by morphology was: osteosarcoma, 34.2%; chondrosarcoma, 27.2%; Ewing sarcoma, 19.3%; other, 19.4%. The distribution varied by age. Ewing sarcoma was most common in 0-9 year olds, osteosarcoma in 10-29 year olds and chondrosarcoma in 30-84 year olds. 29.2% of all tumours occurred in 0-24 year olds. Highest incidence of osteosarcoma and Ewing sarcoma in females was in 10-14 year olds. In males, peak incidence occurred at 15-19 years and exceeded that in females. Chondrosarcoma incidence steadily increased with age. The proportions of Ewing sarcomas occurring in respective bones were consistent with those of the adult skeleton by weight. In osteosarcoma tumours of long bones of lower limb were markedly over-represented in the adolescent peak, being six times more than at any other site. Variation in incidence patterns with age and site suggests pubertal bone growth to be a key factor in osteosarcoma while different biological pathways could be relevant for Ewing sarcoma.

Comparative incidence patterns and trends of gonadal and extragonadal germ cell tumors in England, 1979 to 2003.

BACKGROUND: It is believed that gonadal and extragonadal germ cell tumors (GCTs) arise from primordial germ cells and may have similar etiopathogenesis. Unlike testicular GCTs, there has been limited comprehensive population-based analysis of ovarian and extragonadal GCTs. METHODS: All malignant GCTs and all central nervous system (CNS) GCTs with benign and uncertain behavior that were registered in England in the age group 0 to 84 years from 1979 to 2003 were included in the current study. Incidence rates were calculated and adjusted to the world standard population. RESULTS: There were 33,364 GCTs (92.5% testes, 3.9% ovary, 3.2% extragonadal) in individuals aged 0 to 84 years. The CNS was the most common extragonadal site. An initial peak in incidence at ages 0 to 4 years of nongerminomas was observed at all sites except ovary. Second incidence peaks between ages 10 to 39 years that were more marked among males also were observed at all sites. The ages at these incidence peaks varied by site and were 10 to 14 years (CNS), 15 to 19 years (ovary), 25 to 29 years (other extragonadal sites), and 30 to 34 years (testes). A statistically significant increase in incidence over time was observed in germinomas (testes, CNS) and nongerminomas (testes, ovary). CONCLUSIONS: The age-incidence patterns observed suggested a common initiation of GCTs in embryonic/fetal life with variable rates of tumor progression as a result of subsequent events that may be site specific. The authors concluded that future genetic studies should consider GCTs from all sites to enable a better understanding of their etiology.

Cancer at ages 15-29 years: the contrasting incidence in India and England.

BACKGROUND: There has been a steady increase in published research from Europe and North America on the epidemiology of cancers in young people. There are limited data from the developing world. We contrast the incidence of cancer at ages 15-29 years in India and England. PROCEDURE: Malignant neoplasms in those aged 15-29 years registered during 2001-2003 in five urban population-based cancer registries (PBCRs) of India and in eight PBCRs in England were included. Site-based classification was used. Age-standardized incidence rates were expressed per 100,000 person years. RESULTS: In India, 4,864 (5.8%) of 84,450 cases and in England, 8,137 (1.2%) of 65,6752 cancer cases occurred in those aged 15-29 years. For this age group, the incidence rate for males and females in India were 12.91 and 14.19, and in England were 27.75 and 28.88, respectively. In males aged 15-29 years, the three most common cancers in India were leukemia, lymphoma, and central nervous system tumors and in England were cancers of male genital organs, lymphoma, and leukemia. Cancers of female genital organs, breast, and leukemia were most common in females in India and cancers of female genital organs, lymphoma, and melanoma in England. For cancers of mouth, stomach, and gall bladder, the incidence was higher in India. CONCLUSION: Incidence of cancer at ages 15-29 years in England is higher at most sites than in India. Variation in environmental exposures between the two countries might be an explanation. Under-ascertainment of cases and gender bias in seeking healthcare may also influence reported incidence rates in India.

Relationship between height at diagnosis and bone tumours in young people: a meta-analysis.

OBJECTIVE: Some evidence exists that patients with osteosarcoma and Ewing sarcoma are taller than the general population. However, previous studies are under-powered, lack comprehensive data and show inconsistencies. METHODS: Relevant studies linking osteosarcoma and Ewing sarcoma with height at diagnosis were identified in two major online databases, Medline (1950 to 2009) and Embase (1980 to 2009). Outcomes in individual studies were reported as standard deviation (SD) scores or percentages of study population with height at diagnosis above the median of the reference population. We performed separate random-effects meta-analyses for each outcome and tumour type. RESULTS: 14 studies examined the height of patients with osteosarcoma or Ewing sarcoma. Meta-analyses on SD scores found patients with osteosarcoma were 0.260 SD (95% CI: 0.088-0.432) taller than the reference population (five studies). A meta-analysis on percentages found 62% (95% CI: 57%-67%) of patients were estimated to have a height above the median (six studies). Patients with Ewing sarcoma were 0.096 SD (95% CI 0.004-0.188) taller (four studies). Only one study reported the percentage of Ewing sarcoma patients with height above the median. CONCLUSION: The average height of patients with osteosarcoma, but not Ewing sarcoma, was significantly above the average height of the reference population by 2-3 centimetres. The observed differences indicate the involvement of pubertal longitudinal bone growth in osteosarcoma development while different biological pathways could be relevant for Ewing sarcoma.

Age-incidence patterns of primary CNS tumors in children, adolescents, and adults in England.

Around 25% of all tumors in those 0-14 years of age and 9% in those 15-24 years of age involve the CNS. They are the most common cause of cancer-related deaths in both age groups. In adults 25-84 years of age, the proportion of CNS tumors is 2%; 5-year overall survival is 10%-15%; and survivors have considerable morbidity. Comprehensive up-to-date population-based incidence data on these tumors are lacking. We present incidence rates for primary CNS tumors based on data derived from the high-quality national cancer registration system in England. A total of 54,336 CNS tumors of malignant, benign, and uncertain behavior were registered across the whole of England from 1995 through 2003. The age-standardized rates for all ages (0-84 years) was 9.21 per 100,000 person-years. This is higher than previously reported for England because it includes nonmalignant CNS tumors and hence gives a more accurate picture of burden of disease. The age-standardized rates for those 0-14 years of age, 15-24 years of age, and 25-84 years of age were 3.56, 3.26, and 14.57 per 100,000 person-years, respectively. In this article, we describe the changing patterns in the epidemiology of primary CNS tumors in these three age groups with respect to sex, tumor behavior, and histology using the current WHO classification. This information will provide a reference for future studies nationally and internationally and make comparisons relevant and meaningful.

Breast cancer in patients with germline TP53 pathogenic variants have typical tumour characteristics: the Cohort study of TP53 carrier early onset breast cancer (COPE study).

Germline TP53 pathogenic variants are rare but associated with a high risk of cancer; they are often identified in the context of clinically diagnosed Li-Fraumeni syndrome predisposing to a range of young onset cancers including sarcomas and breast cancer. The study aim was to conduct a detailed morphological review and immuno-phenotyping of breast cancer arising in carriers of a germline TP53 pathogenic variant. We compared breast cancers from five defined groups: (1) TP53 carriers with breast cancer (n = 59), (2) early onset HER2-amplified breast cancer, no germline pathogenic variant in BRCA1/2 or TP53 (n = 55), (3) BRCA1 pathogenic variant carriers (n = 60); (4) BRCA2 pathogenic variant carriers (n = 61) and (5) young onset breast cancer with no known germline pathogenic variant (n = 98). Pathologists assessed a pre-agreed set of morphological characteristics using light microscopy. Immunohistochemistry (IHC) for HER2, ER, PR, p53, integrin alpha v beta 6 (αvß6) integrin, α-smooth muscle actin (α-SMA) and pSMAD2/3 was performed on tissue microarrays of invasive carcinoma. We confirmed a previously reported high prevalence of HER2-amplified, ductal no special type invasive breast carcinoma amongst known TP53 germline pathogenic variant carriers 20 of 36 (56%). Furthermore we observed a high frequency of densely sclerotic tumour stroma in cancers from TP53 carriers (29/36, 80.6%) when compared with non-carriers, 50.9% (28/55), 34.7% (50/144), 41.4% (65/157), 43.8% (95/217) in groups 2-5 respectively. The majority of germline TP53 gene carrier breast tumours had a high intensity of integrin αvß6, α-SMA and pSMAD2/3 expression in the majority of cancer cells. In conclusion, aggressive HER2 positive breast cancers with densely sclerotic stroma are common in germline TP53 carriers. High levels of αvß6 integrin, α-SMA and pSMAD2/3 expression suggest that the dense stromal phenotype may be driven by upregulated transforming growth factor beta signalling.

Cancer risks among relatives of children with Hodgkin and non-Hodgkin lymphoma.

A role for genetic susceptibility in the aetiology of childhood lymphomas was investigated in 454 families of children with histologically confirmed Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL) from Northwest England. Cancers in parents were obtained from the UK National Health Service Central Register and in other close relatives by interview with the parents. The cancer incidence among relatives was compared with expected incidence based on cancer registry data for England. There were 197 cancers in relatives (SIR 1.0 95% CI 0.8-1.1). In families of children with HL, there was an excess of HL in the first degree relatives (SIR 5.8 95% CI 1.2-16.9). Excesses of HL diagnosed under population median age (SIR 4.1 95% CI 1.1-10.6) were seen among all relatives and relatives of children who were below the median age at diagnosis (SIR 5.5 95% CI 1.1-16.0). In families of children with NHL, there were non-significant excesses of central nervous system (CNS) tumours in the first degree relatives (SIR 2.9 95% CI 0.8-7.4) and in the second and third degree relatives (SIR 1.5). There were significant excesses of CNS tumours diagnosed under the population median age (SIR 2.8 95% CI 1.1-5.8) in all relatives. Excess CNS tumours were also seen among relatives of children below the median age at diagnosis (SIR 3.2 95% CI 1.1-7.6). In conclusion, genetic susceptibility in some families of children with lymphoma might be operating, but aetiologies in HL and NHL appear to be different. Possible interpretations of our findings, in the context of putative genetic and infectious aetiologies, are discussed.

Registration and classification of adolescent and young adult cancer cases.

Cancer registries are an important research resource that facilitate the study of etiology, tumor biology, patterns of delayed diagnosis and health planning needs. When outcome data are included, registries can track secular changes in survival related to improvements in early detection or treatment. The surveillance, epidemiology, and end results (SEER) registry has been used to identify major gaps in survival for older adolescent and young adult (AYA) patients compared with younger children and older adults. In order to determine the reasons for this gap, the complete registration and accurate classification of AYA malignancies is necessary. There are inconsistencies in defining the age limits for AYAs although the Adolescent and Young Adult Oncology Progress Review Group proposed a definition of ages 15 through 39 years. The central registration and classification issues for AYAs are case-finding, defining common data elements (CDE) collected across different registries and the diagnostic classification of these malignancies. Goals to achieve by 2010 include extending and validating current diagnostic classification schemes and expanding the CDE to support AYA oncology research, including the collection of tracking information to assess long-term outcomes. These efforts will advance preventive, etiologic, therapeutic, and health services-related research for this understudied age group.

High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin's disease: report from the Late Effects Study Group.

PURPOSE: We present an update of a previously reported Late Effects Study Group cohort of 1,380 children with Hodgkin's disease (HD) diagnosed between 1955 and 1986 in patients aged 16 years or younger. We describe the pattern and incidence of subsequent neoplasms (SNs) occurring with extended follow-up. PATIENTS AND METHODS: Median age at diagnosis of HD was 11.7 years (range, 0.3 to 16.9 years) and at last follow-up was 27.8 years. Median length of follow-up was 17.0 years. RESULTS: An additional 103 SNs were ascertained (total SNs = 212). The cohort was at an 18.5-fold increased risk of developing SNs compared with the general population (standardized incidence ratio [SIR], 18.5, 95% CI, 15.6 to 21.7). The cumulative incidence of any second malignancy was 10.6% at 20 years, increasing to 26.3% at 30 years; and of solid malignancies was 7.3% at 20 years, increasing to 23.5% at 30 years. Breast cancer was the most common solid malignancy (SIR, 56.7). Other commonly occurring solid malignancies included thyroid cancer (SIR, 36.4), bone tumors (SIR, 37.1), and colorectal (SIR, 36.4), lung (SIR, 27.3), and gastric cancers (SIR, 63.9). Risk factors for solid tumors included young age at HD and radiation-based therapy. Thirty-two patients developed third neoplasms, with the cumulative incidence approaching 21% at 10 years from diagnosis of second malignancy. CONCLUSION: Additional follow-up of this large cohort of HD survivors documents an increasing occurrence of known radiation-associated solid tumors, (breast and thyroid cancers), as well as emergence of epithelial neoplasms common in adults, (colon and lung cancers) at a younger age than expected in the general population, necessitating ongoing surveillance of this high risk population.

Is there a common aetiology for certain childhood malignancies? Results of cross-space-time clustering analyses.

We previously demonstrated significant space-time clustering amongst cases of childhood leukaemia (in particular acute lymphoblastic leukaemia (ALL)), central nervous system (CNS) tumour (especially astrocytoma), soft tissue sarcoma and Wilms' tumour. We hypothesised that there may be common aetiological mechanisms between some of these diagnostic groups. To test this hypothesis we analysed for cross-space-time clustering between these diagnostic groups, using population-based data from north-west England. Data were examined by a second-order procedure based on K-functions. Reference points in time and space were dates and addresses at birth and diagnosis. The results showed statistically significant (P < 0.05) cross-clustering between cases of leukaemia and CNS tumour and between cases of ALL and astrocytoma. There was no statistically significant cross-clustering of Wilms' tumours and soft tissue sarcomas with any other malignancy. In conclusion, these findings are consistent with common, possibly infectious, aetiological mechanisms for childhood leukaemia (particularly ALL) and CNS tumours (particularly astrocytoma).

Cancer incidence and mortality among the parents of a population-based series of 2604 children with cancer.

The purpose is to determine cancer incidence and mortality from all causes among parents of children with solid tumors and to formulate hypotheses about heritable risks and associations with parental environmental exposures. A total of 2604 children ages < 15 years included in the Manchester Children's Tumor Registry, 1954-1996, was eligible. Parents of index children were traced and followed up to December 31, 2000, through the United Kingdom National Health Service Central Register. Data on cancer diagnosis and all causes of deaths were obtained. Standardized incidence (SIR) and standardized mortality (SMR) ratios, Ps, and 95% confidence intervals were calculated from serial age and sex-specific cancer incidence and mortality data for England and Wales. There was a significant excess of cancers in parents overall (observed number = 481, expected number = 428.9, SIR = 1.1; P < 0.05) because of central nervous system tumors (SIR = 1.7; P < 0.05), carcinoma breast (SIR = 1.3; P < 0.05), bone and soft tissue sarcoma (SIR = 2.9; P < 0.01), and retinoblastoma (SIR = 62.1; P < 0.001). Diagnoses in the index children associated with excess risk of cancer in their parents were retinoblastoma (SIR = 1.8; P < 0.01) and gonadal germ cell tumors (SIR = 2.3; P < 0.01). There was no excess risk of death from all causes considered together (observed number = 791, expected number = 814.2) nor from any specific cause. These results indicate that there is a small excess of incident cancers among parents of children with solid tumors, probably mainly because of known genes, but mortality is similar to the general population. The study is very reassuring for families of children with cancer.

Further clues concerning the aetiology of childhood central nervous system tumours.

Previously, we reported space-time clustering and seasonal variation in childhood central nervous system (CNS) tumours for the period 1954-1998. These previous studies provided evidence that infections may be involved in aetiology. To determine whether there were also localised spatial factors involved in aetiology we analysed the geographical distribution of CNS tumours in children aged 0-14 years using Manchester Children's Tumour Registry (MCTR) data for the period 1976-2000. Specifically, the Potthoff-Whittinghill test for spatial clustering was applied and Poisson regression was used to analyse the relationship between incidence rates and small-area population density, ethnic composition and deprivation index. No relationships were seen for all CNS tumours together and only a few for the subgroups. The previous findings of space-time clustering and seasonal variation, involving astrocytoma and ependymoma, together with the lack of spatial clustering and ecological relationships for these tumours provide evidence that astrocytoma and ependymoma may be associated with a highly mobile transient aetiological agent. An example of such an agent is an infection that occurs in mini-epidemics.

Geographical and ecological analyses of childhood Wilms' tumours and soft-tissue sarcomas in North West England.

The aim of this paper was to study the geographical distribution of Wilms' tumours (WT) and soft-tissue sarcomas (STS) for 0-14 year olds included in a population-based registry from North West England during 1976-2000. Standardised morbidity ratios (SMRs) were calculated. Relationships between incidence rates and small area (ward) population density, ethnic composition, deprivation index and urban-rural status were examined using Poisson regression. There was a non-linear relationship between WT incidence and population density (P=0.008), with a higher incidence associated with wards with low deprivation scores (P=0.02); and which included a greater proportion of whites (P=0.01). For STS, a higher incidence was associated with wards with low deprivation scores (P=0.04); and which were 'more rural/less urban' (P=0.03). These results are consistent with a role for localised environmental exposures, in combination with lifestyle factors, in the aetiology of WT. For STS, there is some evidence for the involvement of environmental and/or lifestyle factors.

Identification of a rare polymorphism in the human TP53 promoter.

The majority of families with classic Li-Fraumeni Syndrome (LFS) and a significant proportion of Li-Fraumeni-like (LFL) families have a germline mutation in the TP53 tumor suppressor gene. However around 20% of LFS and 60% of LFL families have no identifiable genetic defect in the coding region or splice junctions of TP53, and the genetic basis for cancer susceptibility in these families remains largely uncharacterized. To determine whether promoter mutations could be responsible for the Li-Fraumeni phenotype, we sequenced the TP53 promoter in index cases from members of classic LFS and LFL families without detectable TP53 mutations. We identified an identical single nucleotide deletion within the C/EBP- like site of the promoter in two out of eighteen such families (11%), compared to only one of a total of 366 control samples (0.3%). Although this result is highly significant (P=0.006, Fischer's exact test), the mutation did not affect the expression of TP53 in our hands. We provide evidence that this site is not utilized in the wild type TP53 promoter and further, that mutation of this site in LFS/LFL does not have a functional effect. We conclude that the sequence variant is a rare polymorphism arising within the TP53 promoter. However, the significantly increased frequency of this variant in LFS/LFL remains intriguing.

UK case control study of brain tumours in children, teenagers and young adults: a pilot study.

BACKGROUND: Tumours of the central nervous system are the second most common group of childhood cancers in 0-14 year olds (24% of total cancers) and represent a major diagnostic group in 15-24 year olds. The pilot case-control study aimed to establish methodologies for a future comprehensive aetiological investigation among children and young adults. METHODS: Eligible cases were newly diagnosed with an intracranial tumour of neuroepithelial tissue aged 0-24 years. The pilot recruited patients through Leeds and Manchester Principal Treatment Centres. Controls were drawn from general practice lists. Controls were frequency matched by age and gender. RESULTS: We interviewed 49 cases and 78 controls comprising 85% of the target sample size. Response rates were 52% for cases and 32% for controls. Completion of the questionnaire was successful, with a very small proportion of missing data being reported (5-10%). The age distribution of cases and controls was similar with around three-quarters of interviewed subjects aged 0-14. Half of cases and almost two-thirds of controls reported using a mobile phone with the majority starting between 10-14 years of age. Prevalence of breastfeeding was lower in cases than controls (Odds Ratio 0.4; 95% CI 0.2-1.2), whilst cases were more likely to be delivered by caesarean section (OR 1.6; 95% CI 0.6-4.4). Cases were significantly more likely to have a birthweight > 3.5 kg compared to controls. Cases were also more likely to come from a family with 3 or more siblings than controls (OR 3.0; 95% CI 0.7-13.6). The majority of participants (>80%) were in favour of taking either blood or saliva to aid molecular epidemiological research. CONCLUSIONS: Successful methods were established for identifying and recruiting a high proportion of case subjects, exploiting strong links with the clinical teams at the treatment centres. Control procedures proved more difficult to implement. However, working closely with national clinical and professional research networks will enable improved control identification and recruitment, with good prospects for collecting biological samples in the future.

Median percent change: a robust alternative for assessing temporal trends.

A typical summary statistic for temporal trends is the average percent change (APC). The APC is estimated by using a generalized linear model, usually under the assumption of linearity on the logarithmic scale. A serious limitation of least-squares type estimators is their sensitivity to outliers. The goal of this study is twofold: firstly, we propose a robust and easy-to-compute measure of the temporal trend based on the median of the rates (median percent change - MPC), rather than their mean, under the hypothesis of constant relative change; secondly, we investigate the performance of several models for estimating the rate of change when some of the most common model assumptions are violated. We provide some guidance on the practices of the estimation of temporal trends when using different models under different circumstances. The robustness property of the median is assessed in a simulation study, which shows that the MPC provides strong reductions in estimation bias and variance in presence of outliers. We also demonstrate how a mathematical property of the median helps addressing the issue of zero counts when estimating trends on the log-scale. Finally, we analyzed an English cancer registration dataset to illustrate the proposed method. We believe that, as a good practice, both APC and MPC should be presented when sensitivity issues arise.