Malignant Disease Registry Research Articles

Temporal clustering of neuroblastic tumours in children and young adults from Northern England.

BackgroundThe aetiology of neuroblastic tumours is unclear with both genetic and environmental factors implicated. The possibility that an infectious agent may be involved has been suggested. ‘Temporal clustering’ occurs if cases display an irregular temporal distribution and may indicate the involvement of an agent that exhibits epidemicity. We tested for the presence and nature of temporal clustering using population-based data from northern England.MethodsWe extracted all cases of neuroblastic tumours diagnosed in children and young adults aged 0–24 years during 1968–2011 from the Northern Region Young Persons’ Malignant Disease Registry. This is a population-based registry, covering a population of approximately 900,000 young persons, and includes all cases resident in northern England at the time of diagnosis. Tests for temporal clustering were applied using a modified version of the Potthoff-Whittinghill method. Estimates of extra-Poisson variation ( widehat{upbeta} ) and standard errors (SEs) were obtained.Results227 cases of neuroblastic tumours were diagnosed during the study period. All the analyses between fortnights and between months found significant extra-Poisson variation, with widehat{upbeta} =0.846 (SE = 0.310, P = 0.004) for the analysis between fortnights within months. Restricting the analyses to the 76 cases diagnosed at ages less than 18 months showed significant extra-Poisson variation between fortnights within months ( widehat{upbeta} =1.532, SE = 0.866, P = 0.038), but not between months. In contrast, analyses of cases aged 18 months to 24 years showed significant extra-Poisson variation between quarters within years, as well as over shorter timescales.ConclusionsTransient environmental agents may be involved in the aetiology of neuroblastic tumours. The initiating factor might be a geographically-widespread agent that occurs in ‘mini-epidemics’.Electronic supplementary materialThe online version of this article (doi:10.1186/s12940-015-0058-z) contains supplementary material, which is available to authorized users.

Socioeconomic patterning in the incidence and survival of teenage and young adult men aged between 15 and 24 years diagnosed with non–seminoma testicular cancer in northern england

PurposePrevious research from developed countries has shown a marked increase in the incidence of testicular cancer in the past 50 years. This has also been demonstrated in northern England, along with improving 5-year survival. The present study aims to determine if socioeconomic factors may play a role in both etiology and survival from non–seminoma testicular cancer. Materials and methodsWe extracted all 214 cases of non–seminoma testicular cancer diagnosed in teenage and young adult men aged between 15 and 24 years during 1968 to 2006 from the Northern Region Young Persons’ Malignant Disease Registry, which is a population-based specialist regional registry. Negative binomial regression was used to examine the relationship between incidence and both the Townsend deprivation score (and component variables) and small-area population density. Cox regression was used to analyze the relationship between survival and both deprivation and population density. ResultsDecreased incidence was associated with living in areas of higher household overcrowding for young adults aged between 20 and 24 years (relative risk per 1% increase in household overcrowding = 0.79; 95% CI: 0.66–0.94) but no association was detected for young people aged between 15 and 19 years. Community-level household unemployment was associated with worse survival (hazard ratio per 1% increase in household unemployment = 1.04; 95% CI: 1.00–1.08). ConclusionsThis study has shown that increased risk of non–seminoma testicular cancer in teenage and young adult men may be associated with some aspect of more advantaged living. In contrast, greater deprivation is linked with worse survival prospects. The study was ecological by design and so these area-based results may not necessarily apply to individuals.

Seasonality in the Incidence of Cervical Carcinoma in Teenagers and Young Adults in Northern England, 1968–2005

Infection with human papillomavirus is an established risk factor for cervical carcinoma. However, the role of other environmental factors is less well established. To further investigate whether other agents may be involved, the authors have analyzed seasonal variation in cervical cancer with respect to month of birth and separately month of diagnosis. All 85 cases diagnosed in 15–24-yr-olds during the period 1968–2005 were extracted from the specialist population-based Northern Region Young Persons' Malignant Disease Registry. The chi-square heterogeneity test was used to assess overall nonuniform variation in month of birth and separately month of diagnosis. Poisson regression analysis was used to fit sinusoidal (harmonic) models to the data using month of birth and month of diagnosis in separate models. Based on month of birth, there was statistically significant heterogeneity (p = .03) and a significant sinusoidal pattern, with an incidence peak involving births in the autumn months (p = .03). Based on month of diagnosis, there was marginally significant heterogeneity (p = .06). The evidence of seasonal variation around time of birth for cervical carcinoma is highly novel and suggests possible early etiological involvement of environmental factors. (Author correspondence: Richard.McNally@ncl.ac.uk)

038 Survival from childhood and young adult cancer in the north of England, 1968–2005

ObjectiveTo investigate survival from cancer in children and young adults resident in the North of England.MethodsCases aged 0–24 years diagnosed with a primary malignancy during the period 1968–2005 were obtained...

Season of birth and diagnosis for childhood cancer in Northern England, 1968–2005

The aim of this study was to investigate seasonal variation in the incidence of cancer in children aged 0-14 years. Details of 2959 primary malignant cases (1659 males, 1300 females), diagnosed during the period 1968-2005, were extracted from a specialist registry (the Northern Region Young Persons' Malignant Disease Registry). Seasonal variation was analysed with respect to month of birth and diagnosis. The chi-squared heterogeneity test was used to test for non-uniform variation. Poisson regression analysis was used to fit sinusoidal (harmonic) models to the data, using month of birth and month of diagnosis, respectively, as covariates in separate models. There was significant sinusoidal variation based on month of birth for acute lymphoblastic leukaemia (ALL) aged 1-6 years (P = 0.04; peak in March). For 0- to 14-year-old boys, there was statistically significant sinusoidal variation in month of birth for acute non-lymphocytic leukaemia (P = 0.04; peak in September) and astrocytoma (P = 0.03; peak in October). Based on month of diagnosis, there was statistically significant sinusoidal variation in girls for all lymphomas (P = 0.048; peak in March) and Hodgkin lymphoma (HL) (P = 0.005; peak in January), and in boys for osteosarcoma (P = 0.049; peak in October). This study confirms previous findings of seasonal variation around the month of birth for childhood ALL (at the peak ages) and provides further evidence of seasonal variation around month of birth for astrocytoma and around month of diagnosis for HL. The results are consistent with a role for environmental factors in the aetiology of these diagnostic groups. Further studies are needed to examine putative candidate agents.

Seasonal Variation in Birth and Diagnosis of Cancer in Children and Young People in Northern England, 1968-2005

ISEE-0062 Background and Objective: Seasonal variation is consistent with an environmental aetiology. The aim of this study was to investigate seasonal variation in cancer incidence in children and young people, using population-based data. Methods: Data on all cases diagnosed during the period 1968-2005 were extracted from The Northern Region Young Person's Malignant Disease Registry. The following diagnostic groups were analysed: leukaemia, lymphoma, central nervous system tumours, sympathetic nervous system tumours, retinoblastoma, renal tumours, hepatoblastoma, bone tumours, soft tissue sarcoma, germ cell tumours and carcinomas. The chi-squared heterogeneity test was used to test for departure from uniform distribution. Poisson regression analysis was used to fit sinusoidal (harmonic) models to the data, using month of birth and month of diagnosis, respectively, as covariates in separate models. The sinusoidal model assumption was assessed using the chi-square goodness of fit test. Analyses were carried out separately by gender and age group (0-14, 15-24 years). Results: There were a total of 5909 childhood cancer cases; 2959 aged 0-14 years (1659 males, 1300 females) and 2950 aged 15-24 years (1590 males, 1358 females). For 0-14 year old boys, there was statistically significant seasonal variation in month of birth for acute non lymphocytic leukaemia (P = 0.04; peak in September) and astrocytoma (P = 0.03; peak in October). Based on month of diagnosis, there was statistically significant seasonal variation in girls for lymphomas (P = 0.05; peak in March) and Hodgkin lymphoma (P < 0.01; peak in January), and in boys for osteosarcoma (P = 0.05; peak in October). For 15-24 year olds, significant seasonal variation in month of birth for malignant melanoma in females (P = 0.03; peak in March) and cervical carcinoma (P = 0.03; peak in October) was observed. Conclusion: These findings suggest that seasonal environmental factors, particularly around the time of birth, such as infections or exposure to sunlight may be involved in the aetiology of specific diagnostic groups.

Seasonal variation in birth and diagnosis of cancer in children and young people in Northern England, 1968–2005

To investigate seasonal variation in the incidence of cancer in children and young people, using population-based data. Northern England, UK. Data on all cases were extracted from the Northern Region...

Geographical analysis of thyroid cancer in young people from northern England: Evidence for a sustained excess in females in Cumbria

A previous study found a thyroid cancer excess in Cumbria following the Chernobyl explosion, but did not analyse sex-specific effects. This study examines sex differences in the incidence of thyroid cancer. Ninety-five primary thyroid carcinomas (69 females, 26 males) diagnosed in those aged 0–24 during 1968–2005 were identified from the Northern Region Young Persons’ Malignant Disease Registry. Age-standardised incidence rates (ASRs), rate ratios (RRs) and 95% confidence intervals (CIs) were calculated. For males, the ASR was 0.6 per million person-years during the pre-Chernobyl period (1968–1986), and was 1.8 per million person-years during the post-Chernobyl period (1987–2005). For females, the ASR was 2.4 pre-Chernobyl and was 3.9 post-Chernobyl. The previously noted excess in Cumbria was entirely confined to females (Cumbrian females: RR for post-Chernobyl compared with pre-Chernobyl = 10.8; 95% CI: 1.4–85.3). These findings may be consistent with sex-specific differences in susceptibility to an environmental exposure, such as fallout from the Chernobyl nuclear accident.

Congenital anomaly and childhood cancer: A population‐based, record linkage study

The cause of the majority of childhood malignancies is unknown. Association with the presence of congenital anomalies has been noted in some studies. In this study, we describe and quantify the association between congenital anomalies and childhood cancer. Cases from the Northern Congenital Abnormality Survey (NorCAS) and the Northern Region Young Persons Malignant Disease Registry (NRYPMDR) were matched using four key variables: surname at birth, date of birth, postcode at birth, and infant gender. All potential matches were checked manually. A second match was performed for children with matched birth dates, postcodes and gender but different surnames, to identify children whose surname at diagnosis was not the same as at birth. Thirty-nine children with a congenital anomaly and a diagnosis of cancer were identified from a cohort of 599,290 children born during 1985-2001. There was an almost threefold overall increased risk of malignancy among these children (RR = 2.9; 95% CI 2.1, 3.9), particularly of acute lymphoblastic leukaemia (RR = 2.7; 95% CI 1.5, 5.0), acute myeloblastic leukaemia (RR = 22.0; 95% CI 12.1, 40.0), other leukaemia (RR = 7.5; 95% CI 1.3, 43.6) and lymphomas (RR = 5.3; 95% CI 2.4, 12.0). This increased rate remained when children with Down syndrome were excluded (RR = 1.8, 95% CI 1.2, 2.7). Our findings suggest that children born with congenital anomalies are at increased risk of certain types of cancer. Further linkage studies between high quality registers, may provide insights into the origins of such malignancies.

Incidence and survival for testicular germ cell tumor in young males: A report from the Northern Region Young Person’s Malignant Disease Registry, United Kingdom

The incidence of testicular cancer has increased markedly in most developed countries, although the reasons for this are unclear. In this study, 253 patients with testicular cancer diagnosed younger than 25 years from 1968–1999 were identified from the Northern Region Young Persons’ Malignant Disease Registry. The age-standardized incidence rate increased from 0.93 in 1968–1978 to 1.60 per 100,000 in 1990–1999. The increase in incidence was confined to those patients >15 years old, with the rate in younger children remaining very low. The 5-year survival increased significantly from 46% (95% confidence interval [CI] 33% to 58%) to 92% (95% CI 85% to 96%) during 1968–1999 and was significantly higher for seminoma than nonseminoma, 91% (95% CI 78% to 96%) and 77% (95% CI 70% to 82%), respectively. Although increased surveillance and public health campaigns designed to raise awareness of the disease and the advantages of self-examination should continue, further research is required into the etiology of this relatively common cancer among young males.

Paternal occupational exposure to electro‐magnetic fields as a risk factor for cancer in children and young adults: A case‐control study from the North of England

Numerous studies have implied that paternal occupational exposures, in particular electromagnetic fields (EMF) and ionizing radiation, may be involved in the etiology of childhood cancers. We investigated whether an association exists between paternal occupations at birth involving such exposures and cancer risk in offspring, using data from the Northern Region Young Persons' Malignant Disease Registry (NRYPMDR). Cases (n=4,723) were matched, on sex and year of birth, to controls from two independent sources: (i) all other patients from the NRYPMDR with a different cancer, (ii) 100 cancer-free individuals per case from the Cumbrian Births Database. An occupational exposure matrix was used to assign individuals to exposure groups. There was an increased risk of leukemia among the offspring of men employed in occupations likely to be associated with EMF or radiation exposures (OR 1.31, 95% CI 1.02-1.69), particularly in males aged less than 6 years (OR 1.81, 95% 1.19-2.75). No significant association was seen in females. Increased risks were also seen for chondrosarcoma (OR 8.7, 95% CI 1.55-49.4) and renal carcinoma (OR 6.75, 95% CI 1.73-26.0). These associations were consistent between control groups and remained after adjustment for socio-economic status. This large case-control study identified a significantly increased risk of leukemia among the offspring of men likely to have been occupationally exposed to EMF, with differing associations between males and females. Increased risks of chondrosarcoma and renal carcinoma were also seen, although based on smaller numbers. Further detailed investigations in this area are required to understand this association.

Associations of Birthweight and Maternal Fertility Problems with Childhood Cancers.

Higher birthweight, maternal history of miscarriage and low birth order have been associated with increased risk of childhood leukaemias and some solid tumours. No study has investigated these factors together and differences in disease etiology between girls and boys have been generally overlooked. In a retrospective case-control study, 732 childhood (≤15 yr) cancer cases from the population-based Northern Region Young Persons' Malignant Disease Registry (NRYPMDR) whose hospital birth records could be accessed and 3723 controls matched for date and hospital of birth were compared. We examined maternal reproductive history and birthweight for gender-specific associations using conditional logistic regression. In univariate analysis, maternal history of miscarriage showed an association with all cancers (OR = 1.29; 95% CI = 1.05 to 1.62, P = 0.02). In individual cancer groups, this association was significant for acute lymphoblastic leukaemia (ALL) (n=225, OR = 1.56; 95% CI = 1.07 to 2.27, P = 0.02), and marginally significant in osteosarcoma and neuroblastic tumours (neuroblastoma and ganglioneuroblastoma). There was no significant association with birth order. Being first born was a weak risk factor for ALL in boys only (OR=1.3, 95% CI = 0.8 to1.8). In boys but not in girls, the risk of ALL increased with birthweight (OR = 1.06 per 100 gr increase; 95% CI = 1.01 to 1.11, P = 0.01). When birthweights were normalized using UK standards for gestational age and gender, the associations were not more marked. A multivariate model for ALL confirmed the independence of associations with miscarriage history and birthweight. Gestational age was not a risk marker and did not explain the associations with birth weight and miscarriage history. Consideration of gender unravelled significant associations of maternal reproductive history and size at birth with childhood cancer markedly different between girls and boys. Most notably, associations with birth weight and miscarriage and the weak association with being first-born in childhood ALL were all stronger in males. The findings for birthweights normalized for gestational age suggested that size at birth rather than in utero growth trajectory is of etiologic importance in childhood ALL.

Cancer in adolescents and young adults aged 15–24 years: A report from the North of England young person's malignant disease registry, UK

Descriptions of population-based data have rarely been published specifically for adolescents and young adults with cancer. Data on young adults (15-24 years) diagnosed with cancer in the North of England from 1968 to 1997 were obtained from the Northern Region Young Person's Malignant Disease Registry. Temporal changes in incidence and survival rates were investigated. There were 2,329 first cancers diagnosed over the study period (M:F 1.22:1). Overall age standardized incidence was 174 cases per million 15-24 years old, per year, 190 for males and 157 for females. The most common cancers in young adults were Hodgkin disease (19%), carcinomas (15%), central nervous system tumors (14%), germ cell tumors (13%), and leukemia (11%). Comparing incidence for 1968-1977 with 1988-1997 there were significant increases in the incidence of bone tumors (rate ratio 1.72, 95% CI 1.10-2.68), testicular tumors (rate ratio 1.64, 95% CI 1.16-2.32), thyroid cancer (rate ratio 2.63, 95% CI 1.37-5.02), and malignant melanoma (rate ratio 2.04, 95% CI 1.36-3.08). Survival rates improved significantly (P < 0.001) over the study period; 5-year survival rates over the three time periods 1968-1977, 1978-1987, 1988-1997 for all cancers were 45% (95% CI 41%-49%), 62% (95% CI 58%-65%), and 74% (95% CI 71%-77%) respectively. Survival rates improved and there were significant increases in incidence for specific cancers in young adults in the North of England. Further research is required to identify the reasons for changing incidence and to investigate the late effects of treatment among survivors.

Second primary tumors in children and young adults in the North of England (1968-99)

This study describes the risk of second malignancy in patients diagnosed with cancer under the age of 25 years, registered on the Northern Region Young Person's Malignant Disease Registry. Incidence rates were calculated to describe the occurrence of second malignancies, rate ratios were estimated to compare rates between subgroups. Standardized incidence ratios (SIRs) were calculated for comparison with a reference population. There were 4,072 children and young adults diagnosed with a first malignancy from 1968 to 1999, of whom 68 had a second malignancy (including basal cell carcinomas and meningiomas). The incidence rate of second malignancy is 1.7 per 1,000 survivor person-years (95% CI: 1.4, 2.2), reflecting a four-fold increased risk of malignancy compared with the general population. The rate of second malignancy was non-significantly higher for those diagnosed during young adulthood rather than childhood (RR = 1.2, 95% CI: 0.7, 2.0), significantly higher in females than males (RR = 1.8, 95% CI: 1.1, 3.0) and significantly lower for those diagnosed in more recent years (RR = 0.4, 95% CI: 0.2, 0.8). In contrast, the SIRs indicated that children were at substantial increased risk; whilst males and females, and those diagnosed in earlier and later time periods, were at equivalent risks. There is evidence of a sustained increased risk of second malignancy in those treated for primary cancer, especially those diagnosed in childhood; with no evidence that this risk is reducing.

Fathers' occupational contacts and risk of childhood leukemia and non-hodgkin lymphoma.

There is evidence to suggest that childhood leukemia and non-Hodgkin lymphoma have an infective etiology. We investigated the risk of childhood leukemia and non-Hodgkin lymphoma in relation to paternal occupational contact with other individuals, a surrogate for potential exposure to infection. We carried out a case-control study using 792 cases from the Northern Region Young Persons' Malignant Disease Registry, United Kingdom, 1968-1997, and 2 large population-based control groups. Paternal occupations at birth were classified as having standard, high, or very high levels of contact. Conditional logistic regression was used in the analysis. There was an increased risk of childhood leukemia and non-Hodgkin lymphoma in those children whose fathers' occupational contacts were high or very high compared with standard (odds ratio = 1.3; 95% confidence interval = 1.0-1.5). The excess risk in the very high group was most pronounced for 245 cases of acute lymphoblastic leukemia aged 2 to 5 years at diagnosis (1.5; 1.1-2.1). The risk with paternal occupational contacts was greater in rural areas, although it was also present in urban areas. The risks of leukemia and non-Hodgkin lymphoma were also higher among the offspring of men employed as policemen, sales representatives, or teachers. Our findings are consistent with the hypothesis of an infective etiology for childhood leukemia and non-Hodgkin lymphoma, and they add to the evidence that infections could be transmitted to children by adults.

Skin cancer in children and young adults

Little population-based data has been published about skin cancers in children and young adults. In this study, 200 cases of melanoma and non-melanoma skin cancers diagnosed under 25 years of age in the North of England from 1968–1995 were obtained from the Northern Region Young Persons’ Malignant Disease Registry. The incidence was 1.2 cases per million per year for children (aged 0–14 years) and was 13 cases per million per year for young adults (aged 15–24 years). Melanoma accounted for 138 cases, of which 16 were in subjects aged < 15 years at diagnosis. The incidence of melanoma increased in females at a rate of 5.6 per million per decade (95% confidence interval [CI] 2.2–8.9, P = 0.002), largely due to an increased incidence of primary lower limb tumours. The incidence for males was unchanged. Survival improved significantly over time for both males and females (P ≤ 0.02). Of the 62 patients with non-melanoma skin cancers, 66% were diagnosed with primary non-basal cell carcinoma, 13% with dermatofibrosarcoma protuberans, 10% with squamous cell carcinoma and 11% with other tumours. Two cases were iatrogenic second malignancies following treatment for an earlier primary brain tumour. The incidence of non-melanoma skin cancers was significantly higher during 1982–1995 than during 1968–1981 (rate ratio 1.7, 95% CI 1.0–2.8). There were three deaths from non-melanoma skin cancer, and the overall 5 year survival rate was 98% (95% CI 89–100%). The reason for the increasing incidence of both melanoma and non-melanoma skin cancer in young people is unknown, but it is likely that ultraviolet exposure plays an aetiological role. It is important that families continue to be advised of the need for vigilance with regard to childhood sun exposure.

Thyroid cancer in children and young adults in the North of England. Is increasing incidence related to the Chernobyl accident?

Population-based data on thyroid carcinomas was obtained from the Northern Region Young Person's Malignant Disease Registry to analyse the incidence of thyroid cancers in young people (<25 years) in the North of England for the period 1968 and 1997 and to assess if changes in incidence were consistent with the spatial and temporal distribution of the fallout from the Chernobyl nuclear accident. We compared incidence rates for differentiated (papillary or follicular) thyroid carcinomas 1968–1986 with those for 1987–1997. There were 75 cases of thyroid carcinoma diagnosed over the study period, of which 63 were differentiated carcinoma and 12 were medullary carcinoma. There were 26 young adults (15–24 years) diagnosed with differentiated thyroid carcinoma in the 19-year period 1968–1986 and 30 in the subsequent 11 years 1987–1997, Age standardised rate (ASR) 3.0 versus 6.5, respectively (rate ratio 2.2, 95% confidence interval (CI): 1.3–3.6). There were three children (aged <15 years) diagnosed with differentiated carcinoma in the period 1968–1986 and four in the period 1987–1997, ASR 0.2 versus 0.6 (rate ratio 2.7, 95% CI: 0.6–12.1). Regression models showed a significant increase in the incidence of thyroid cancer after the Chernobyl accident (P=0.002). In Cumbria, the area receiving the heaviest fallout in the UK, the increase in incidence was much greater (rate ratio 12.19, 95% CI 1.5–101.2). These temporal and spatial changes in incidence are consistent with a causal association with the Chernobyl accident although a greater effect in the younger rather than the older age group would have been anticipated. However, factors including improvements in ascertainment and earlier detection of tumours may also have contributed to the increasing incidence. Further collaborative international studies are needed to investigate changes in the incidence of thyroid cancer in children and young adults.

Neuroblastoma: changing incidence and survival in young people aged 0-24 years. A report from the North of England Young Persons' Malignant Disease Registry.

Population based data for neuroblastoma in children and young adults under 25 years at diagnosis were ascertained from the Northern Region Young Persons' Malignant Disease Registry for the period 1968-1995. Age-standardised incidence rates were calculated (ASR) and changes in incidence and survival were investigated. Over the study period 144 patients were registered, of these 136 were children under 15 years at diagnosis (median age: 2.2 years, ASR: 8.6 cases per million children per year), and 8 were 15-24 years (ASR 0.6). Incidence of childhood neuroblastoma in the North of England increased significantly over time; ASRs were 5.8 for 1968-1981 and 9.5 for 1982-1995 (rate ratio: 1.6, 95%; CI 1.2-2.3). The increase in incidence was seen in both infants and older children, and in both low stage and advanced disease. Overall 5 year survival was 15% for 1968-1981 and 40% for 1982-1995 (P < 0.0001). Significant improvements in survival were documented across different stage and age-groups, including those over 1 with stage 4 disease (0% versus 18%, P < 0.0001). Further research is needed to investigate the reasons for the increasing incidence of neuroblastoma.

Incidence and survival for cancer in children and young adults in the North of England, 1968-1995: a report from the Northern Region Young Persons' Malignant Disease Registry.

The Northern Region Young Persons’ Malignant Disease Registry records information on young people under 25 years old diagnosed with cancer in the Northern Region of England. Incidence and survival rates were calculated for children and young adults diagnosed with cancer between 1968 and 1995. There were 2099 (M:F 1.28:1) children (age 0–14 years) and 2217 (M:F 1.23:1) young adults (15–24 years) diagnosed with a first cancer between 1968 and 1995. The age-standardized rate (ASR) for childhood cancer was 121 per million 0 to 14 year-olds per year. For young adults the ASR was 175 per million 15 to 24 year-olds, per year. Incidence of childhood cancer increased over time at a rate of 12 extra cases per million children, per decade (P< 0.001). In young adults incidence rates increased by 16 extra cases per million 15 to 24 year-olds, per decade (P< 0.001). For childhood cancer 5-year survival was 42% for those diagnosed 1968–1977, 57% for 1978–1987 and 71% (95% CI 67–75) for 1988–1995. Survival for young adults over the three periods was 45%, 62% and 73% (95% CI 70–78) respectively. The cumulative risk of developing cancer before the age of 25 is 1 in 285. Over the 28-year period there were significant improvements in survival and modest increases in incidence in both children and young adults. © 2000 Cancer Research Campaign

Pineal tumours in the north of England 1968-93.

The records of 38 patients under 25 years of age presenting with pineal tumours between 1968 and 1993, identified from the Northern Region Children's and Young Adults' Malignant Disease Registry, were analysed retrospectively with regards to clinical presentation, diagnostic approach, treatment strategy, and outcome. The overall five year survival was 45%. Fifteen patients had a histological diagnosis: six with germinomas, three with teratomas, three with astrocytomas, and three with pinealoblastomas. One patient had a definitive diagnosis of teratoma made on the basis of raised tumour markers (alpha fetoprotein). Treatment consisted of surgery (87%) (ventriculoperitoneal or atrial shunt and/or biopsy), and/or radiotherapy (82%), and/or chemotherapy (26%). Those patients with a tissue diagnosis appeared to have a more favourable outcome, especially after 1976 when treatment was determined by tumour type (five year survival for those with a tissue diagnosis was 91% v 51% for those without, 95% confidence intervals 74 to 100% and 26 to 75%). This study suggests that tissue diagnosis allows more appropriate treatment to be delivered for children with pineal tumours resulting in improved survival. Referral to a centre with neurosurgery, radiotherapy, neuropathology, and paediatric oncology collaboration is essential.