Sex differences in associations between birth characteristics and childhood cancers: a five-state registry-linkage study.

BACKGROUND: There is a well-recognized male excess in childhood cancer incidence; however, it is unclear whether there is etiologic heterogeneity by sex when defined by epidemiologic risk factors. METHODS: Using a 5-state registry-linkage study (cases n = 16,411; controls n = 69,816), we estimated sex-stratified odds ratios (OR) and 95% confidence intervals (95% CI) between birth and demographic characteristics for 16 pediatric cancers. Evidence of statistical interaction (p-interaction < 0.01) by sex was evaluated for each characteristic in each cancer. RESULTS: Males comprised > 50% of cases for all cancers, except Wilms tumor (49.6%). Sex interacted with a number of risk factors (all p-interaction < 0.01) including gestational age for ALL (female, 40 vs. 37-39 weeks OR: 0.84, 95% CI 0.73-0.97) and ependymoma (female, 40 vs. 37-39 OR: 1.78, 95% CI 1.14-2.79; female, ≥ 41 OR: 2.01. 95% CI 1.29-3.14), birth order for AML (female, ≥ 3rd vs. 1st OR: 1.39, 95% CI 1.01-1.92), maternal education for Hodgkin lymphoma (male, any college vs. < high school[HS] OR: 1.47, 95% CI 1.03-2.09) and Wilms tumor (female, any college vs. HS OR: 0.74, 95% CI 0.59-0.93), maternal race/ethnicity for neuroblastoma (male, black vs. white OR: 2.21, 95% CI 1.21-4.03; male, Hispanic vs. white OR: 1.86, 95% CI 1.26-2.75; female, Asian/Pacific Islander vs. white OR: 0.28, 95% CI 0.12-0.69), and paternal age (years) for hepatoblastoma in males (< 24 vs. 25-29 OR: 2.17, 95% CI 1.13-4.19; ≥ 35 vs. 25-29 OR: 2.44, 95% CI 1.28-4.64). CONCLUSIONS: These findings suggest etiologic heterogeneity by sex for childhood cancers for gestational age, maternal education, and race/ethnicity and paternal age.

The association between sex and most childhood cancers is not mediated by birthweight.

BACKGROUND: Male sex is associated with an increased risk of childhood cancer as is high birthweight. Given that sex determination precedes birthweight we conducted a mediation analysis to estimate the direct effect of sex in association with childhood cancer tumor type with birthweight as the mediator. METHODS: Cases (n = 12,632) and controls (n = 64,439) (ages 0-14 years) were identified from population-based cancer and birth registries in Minnesota, New York, and Washington states (1970-2014). An inverse odds weighting (IOW) mediation analysis was used to estimate odds ratios (OR) and 95% confidence intervals (95% CI) as the measure of association between sex and cancer. RESULTS: A significant indirect effect was observed for sex and lymphoid leukemia, mediated by birthweight (indirectOR: 1.03; 95% CI: 1.02-1.04). We observed significant direct effects for male sex and lymphoid leukemia (directOR: 1.16; 95% CI: 1.08-1.25), Hodgkin lymphoma (directOR: 1.48; 95% CI: 1.22-1.81), Burkitt lymphoma (directOR: 5.02; 95% CI: 3.40-7.42), other non-Hodgkin lymphoma (directOR: 1.42; 95% CI: 1.18-1.70), intracranial embryonal tumors (directOR: 1.49; 95% CI: 1.26-1.76), hepatoblastoma (directOR: 1.90; 95% CI: 1.40-2.59), and rhabdomyosarcoma (directOR: 1.47; 95% CI: 1.19-1.81). There were also inverse associations for extracranial GCTs (directOR: 0.41; 95% CI: 0.26-0.63) and thyroid carcinoma (directOR: 0.35; 95% CI: 0.25-0.50). CONCLUSION: Significant direct effects for sex and numerous childhood cancer types suggests sex-specific factors such as differences in gene expression from the autosomes or the X chromosome, rather than birthweight, may underlie sex differences in tumor risk.

Infant birthweight and risk of childhood cancer: international population-based case control studies of 40 000 cases.

BACKGROUND: High birthweight is an established risk factor for childhood leukaemia. Its association with other childhood cancers is less clear, with studies hampered by low case numbers. METHODS: We used two large independent datasets to explore risk associations between birthweight and all subtypes of childhood cancer. Data for 16 554 cases and 53 716 controls were obtained by linkage of birth to cancer registration records across five US states, and 23 772 cases and 33 206 controls were obtained from the UK National Registry of Childhood Tumours. US, but not UK, data were adjusted for gestational age, birth order, plurality, and maternal age and race/ethnicity. RESULTS: Risk associations were found between birthweight and several childhood cancers, with strikingly similar results between datasets. Total cancer risk increased linearly with each 0.5 kg increase in birthweight in both the US [odds ratio 1.06 (95% confidence interval 1.04, 1.08)] and UK [1.06 (1.05, 1.08)] datasets. Risk was strongest for leukaemia [USA: 1.10 (1.06, 1.13), UK: 1.07 (1.04, 1.10)], tumours of the central nervous system [USA: 1.05 (1.01, 1.08), UK: 1.07 (1.04, 1.10)], renal tumours [USA: 1.17 (1.10, 1.24), UK: 1.12 (1.06, 1.19)] and soft tissue sarcomas [USA: 1.12 (1.05, 1.20), UK: 1.07 (1.00, 1.13)]. In contrast, increasing birthweight decreased the risk of hepatic tumours [USA: 0.77 (0.69, 0.85), UK: 0.79 (0.71, 0.89) per 0.5 kg increase]. Associations were also observed between high birthweight and risk of neuroblastoma, lymphomas, germ cell tumours and malignant melanomas. For some cancer subtypes, risk associations with birthweight were non-linear. We observed no association between birthweight and risk of retinoblastoma or bone tumours. CONCLUSIONS: Approximately half of all childhood cancers exhibit associations with birthweight. The apparent independence from other factors indicates the importance of intrauterine growth regulation in the aetiology of these diseases.

Thyroid cancer incidence in highly observant Jewish neighborhoods in metropolitan New York City.

BACKGROUND: Thyroid cancer incidence in New York State has increased rapidly in recent years, particularly in New York City and its surrounding metropolitan area. In 2007 among white non-Hispanics, incidence rates were about 40% higher in the New York City metropolitan area than in the rest of the state. Here we explore the extent to which living in neighborhoods with a high percentage of highly observant Jews may be associated with this pattern. METHODS: We identify neighborhoods with concentrations of highly observant Jewish persons based on the use of Yiddish among children and the location of Orthodox synagogues. Thyroid cancer risk is modeled as a function of living in such a neighborhood, adjusting for age, sex, and other factors. The model was repeated for small (<2 cm) and large (≥2 cm) tumors to assess the role of diagnostic improvements in driving the spatial-temporal patterns. RESULTS: A moderate association with thyroid cancer was found among those living in Jewish neighborhoods and downstate New York. A lesser association was found among those who live in neighborhoods of high levels of people born in Russia, Belarus, or Ukraine. Similar elevated rate ratios were seen for small and large tumors in Jewish neighborhoods, providing evidence against differences in diagnostic practices in this group. Smaller tumors were more pronounced among women and persons diagnosed more recently. CONCLUSIONS: The associations found do not seem to be diagnostically driven, but rather due to environmental, genetic, or cultural factors in the highly observant population of New York State.

Childhood cancer in relation to parental race and ethnicity: a 5-state pooled analysis.

BACKGROUND: Children of different racial/ethnic backgrounds have varying risks of cancer. However, to the authors' knowledge, few studies to date have examined cancer occurrence in children of mixed ancestry. METHODS: This population-based case-control study examined cancer among children aged <15 years using linked cancer and birth registry data from 5 US states from 1978 through 2004. Data were available for 13,249 cancer cases and 36,996 controls selected from birth records. Parental race/ethnicity was determined from birth records. Logistic regression analysis was used to examine the association of cancer with different racial/ethnic groups. RESULTS: Compared with whites, blacks had a 28% decreased risk of cancer (odds ratio [OR], 0.72; 95% confidence interval [95% CI], 0.65-0.80), whereas both Asians and Hispanics had an approximate 15% decrease. Children of mixed white/black ancestry also were found to be at decreased risk (OR, 0.71; 95% CI, 0.56-0.90), but estimates for mixed white/Asian and white/Hispanic children did not differ from those of whites. Compared with whites: 1) black and mixed white/black children had decreased ORs for acute lymphoblastic leukemia (OR, 0.39 [95% CI, 0.31-0.49] and OR, 0.58 [95% CI, 0.37-0.91], respectively); 2) Asian and mixed white/Asian children had decreased ORs for brain tumors (OR, 0.51 [95% CI, 0.39-0.68] and OR, 0.79 [95% CI, 0.54-1.16], respectively); and 3) Hispanic and mixed white/Hispanic children had decreased ORs for neuroblastoma (OR, 0.51 [95% CI, 0.42-0.61] and OR, 0.67 [95% CI, 0.50-0.90], respectively). CONCLUSIONS: Children of mixed ancestry tend to have disease risks that are more similar to those of racial/ethnic minority children than the white majority group. This tendency may help formulate etiologic studies designed to study possible genetic and environmental differences more directly.

Surveillance of prophylactic mastectomy: trends in use from 1995 through 2005.

BACKGROUND: Prophylactic mastectomy, prophylactic oophorectomy, and antiestrogen chemoprevention are currently the only available methods for breast cancer risk reduction. To the authors' knowledge there is little published information regarding the prevalence of prophylactic mastectomy for the primary prevention of breast cancer among high-risk women or for the prevention of subsequent tumors among women with breast cancer. METHODS: The objective of the current study was to examine the frequency of prophylactic mastectomy in New York State between 1995 and 2005 using mandated statewide discharge data combined with data from the state cancer registry. RESULTS: Identified were 6275 female residents of New York State receiving prophylactic mastectomy; 19% had no identifiable personal history of breast cancer (including women with lobular carcinoma in situ) and 81% had a personal history of breast cancer (84% with invasive disease and 16% with ductal carcinoma in situ). The increased use of prophylactic mastectomy over time was found to be more pronounced among women with breast cancer compared with those without. Women who underwent prophylactic mastectomies were more likely to be younger and white and to have private insurance compared with women who underwent therapeutic mastectomies and compared with all women with breast cancer. The International Classification of Diseases, Ninth Edition, Clinical Modification diagnostic code for prophylactic mastectomy introduced in 1995 was found to have low sensitivity for identifying prophylactic mastectomies in coded discharge data. CONCLUSIONS: The results of the current analysis demonstrate that, although the discharge data alone are inadequate for surveillance purposes, combining these data with the cancer registry data allowed for the detailed examination of the prevalence of prophylactic mastectomies. Mastectomy among high-risk women for cancer prevention appears to be relatively uncommon, but the use of contralateral mastectomy in women with breast cancer is increasing.

Cancer risk among children with very low birth weights.

OBJECTIVE: The risk of hepatoblastoma is strongly increased among children with very low birth weight (<1500 g). Because data on very low birth weight and other childhood cancers are sparse, we examined the risk of malignancy with very low birth weight in a large data set. METHODS: We combined case-control data sets created by linking the cancer and birth registries of California, Minnesota, New York, Texas, and Washington states, which included 17672 children diagnosed as having cancer at 0 to 14 years of age and 57966 randomly selected control subjects. Unconditional logistic regression analysis was used to examine the association of cancer with very low birth weight and moderately low birth weight (1500-1999 g and 2000-2499 g, respectively), compared with moderate/high birth weight (>or=2500 g), with adjustment for gender, gestational age, birth order, plurality, maternal age, maternal race, state, and year of birth. RESULTS: Most childhood cancers were not associated with low birth weights. However, retinoblastomas and gliomas other than astrocytomas and ependymomas were possibly associated with very low birth weight. The risk of other gliomas was also increased among children weighing 1500 to 1999 g at birth. CONCLUSIONS: These data suggested no association between most cancers and very low birth weight, with the exception of the known association of hepatoblastoma and possibly moderately increased risks of other gliomas and retinoblastoma, which may warrant confirmation.

Parental age and risk of childhood cancer: a pooled analysis.

BACKGROUND: Few risk factors for childhood cancer are well-established. We investigated whether advancing parental age increases childhood cancer risk. METHODS: We assessed the relationship between parental age and childhood cancer in a case-control study using pooled population-based data. Our pooling was based on linked cancer and birth registry records from New York, Washington, Minnesota, Texas, and California. Subjects included 17,672 cancer cases diagnosed at ages 0-14 years during 1980-2004 and 57,966 controls born during 1970-2004. Individuals with Down syndrome were excluded. Odds ratios and 95% confidence intervals were calculated by logistic regression for the association between parental age and childhood cancer after adjustment for sex, birth weight, gestational age, birth order, plurality, maternal race, birth year, and state. RESULTS: Positive linear trends per 5-year maternal age increase were observed for childhood cancers overall (odds ratio = 1.08 [95% confidence interval = 1.06-1.10]) and 7 of the 10 most frequent diagnostic groups: leukemia (1.08 [1.05-1.11]), lymphoma (1.06 [1.01-1.12]), central nervous system tumors (1.07 [1.03-1.10]), neuroblastoma (1.09 [1.04-1.15]), Wilms' tumor (1.16 [1.09-1.22]), bone tumors (1.10 [1.00-1.20]), and soft tissue sarcomas (1.10 [1.04-1.17]). No maternal age effect was noted for retinoblastoma, germ cell tumors, or hepatoblastoma. Paternal age was not independently associated with most childhood cancers after adjustment for maternal age. CONCLUSIONS: Our results suggest that older maternal age increases risk for most common childhood cancers. Investigation into possible mechanisms for this association is warranted.

Perinatal risk factors for neuroblastoma.

Neuroblastoma is the most common cancer among infants, suggesting an etiologic role for prenatal factors. In this case-cohort study, neuroblastoma cases (n = 529) diagnosed between 1985 and 2001 were identified from the New York State Cancer Registry and were matched to the electronic birth records for 1983-2001 from New York State and New York City. Controls (n = 12,010) were selected from the same birth cohorts. Analysis was stratified by age at diagnosis, with one to six months (younger infants), seven to 18 months (older infants), and older than 18 months (older children) analyzed separately. Perinatal exposure data was obtained from the birth certificates. No risk factors were identified to be consistently associated with risk across all three age groups. Generally, more risk factors were identified as associated with neuroblastoma among younger infants relative to older ages, including high birth weight, heavier maternal gestational weight gain, maternal hypertension, older maternal age, ultrasound, and respiratory distress. Among older infants, low birth weight was associated with increased risk while heavier maternal gestational weight gain was protective. In the oldest age group, first born status, primary cesarean delivery, prolonged labor and premature rupture of the membranes were associated with increased risk.

Effects of randomization methods on statistical inference in disease cluster detection.

Monte Carlo methods are commonly used to assess the statistical significance of disease clusters. This usually involves permuting the observed outcome measure, such as the rate of disease, across the geographic units within the study area. When the variance of the disease rates is heterogeneous, however, randomizing the disease rate across the geographic units results in over-estimating the p-values in areas of low variance and under-estimating the p-values in areas of high variance. This bias results in under-ascertainment of clusters in urban areas and over-ascertainment of clusters in rural areas. As an alternative, randomizing the number of cases of disease or deaths proportional to the population at risk preserves the variance structure of the study area, therefore resulting in unbiased statistical inference. We compare results from randomizing rates with those from randomizing case counts, using county-level prostate cancer mortality data for the United States and ZIP-Code level prostate cancer incidence data for New York State, using the local Moran's I statistic.

Maternal and infant birth characteristics and hepatoblastoma.

Hepatoblastoma is a rare embryonal tumor with unknown etiology. The authors conducted a case-cohort study using public health surveillance data sets to examine perinatal risk factors for hepatoblastoma. Hepatoblastoma cases (n = 58) diagnosed between 1985 and 2001 were identified from the New York State Cancer Registry and were matched to electronic birth records for 1985-2001 from New York State, excluding New York City. Controls (n = 6,056) were selected from the birth cohorts for the same years. Having a birth weight less than 1,000 g was associated with a strongly increased risk of hepatoblastoma (relative risk (RR) = 56.9, 95% confidence interval (CI): 24.0, 130.7). After adjustment for birth weight, a moderately increased risk of hepatoblastoma was found for younger maternal age (<20 years vs. 20-29 years: RR = 2.5, 95% CI: 1.0, 5.5), presumptive use of infertility treatment (RR = 9.2, 95% CI: 2.1, 31.5), maternal smoking (RR = 2.1, 95% CI: 1.0, 4.2), and higher maternal prepregnancy body mass index (body mass index of 25-29 vs. 20-24: RR = 2.9, 95% CI: 1.2, 7.6).

Cancer incidence patterns among adolescents and young adults in the United States.

OBJECTIVE: The purpose of this study was to examine age-specific cancer incidence patterns among adolescents and young adults (ages 15--49). METHOD: Cancer incidence data for 1995--1999 from 22 population-based central cancer registries, covering about 47% of the US population, were used. Relative frequencies and average annual age-specific incidence rates per 100,000 person-year were computed for the five-year age groups from age 15--19 years through 45--49 years. Tests of significance for comparison were at a level of p<0.05. RESULTS: The age at crossover from a predominance of non-epithelial cancers to a predominance of epithelial cancers during adolescence and young adulthood varied by gender and race. Epithelial cancer became the predominant type of tumor after age 40 years among males while it was the predominant type after age 25 years among females. There was also a shift in the top five cancer types with increasing age, which varied by race and gender. Epithelial cancers of the thyroid, breast, ovary, and cervix uteri started to increase sharply among young women in their 20s while among males epithelial cancers rarely occurred until the early 30s (ages 30--34). Cancers of the female breast, colon and rectum, and lung began to occur at an earlier age and increased more sharply among blacks than among whites. However, the incidence rates of epithelial thyroid and ovarian cancers rose more quickly among whites than blacks. Non-Hodgkin lymphoma and soft tissue sarcoma (excluded Kaposi's sarcoma) increased with age among both whites and blacks but the rates were significantly higher among blacks than among whites. Both Kaposi's sarcoma and testicular cancer incidence increased with age and peaked in the early 30s (ages 30--34). The former was significantly higher among blacks than whites while the latter was significantly higher among whites than blacks. Cervical cancer incidence leveled off when white women reached their 30s, but for black women the rate continued to rise with advancing age. Cutaneous melanoma rates were significantly higher among females than among males between the ages of 15 and 39. Conclusion Cancer incidence patterns among adolescents and young adults are distinctive. Specific cancer prevention and control strategies should be targeted accordingly and tailored to their specific needs.

Geographic patterns of prostate cancer mortality and variations in access to medical care in the United States.

BACKGROUND: Striking geographic variation in prostate cancer death rates have been observed in the United States since at least the 1950s; reasons for these variations are unknown. Here we examine the association between geographic variations in prostate cancer mortality and regional variations in access to medical care, as reflected by the incidence of late-stage disease, prostate-specific antigen (PSA) utilization, and residence in rural counties. METHODS: We analyzed mortality data from the National Center for Health Statistics, 1996 to 2000, and incidence data from 30 population-based central cancer registries from the North American Association of Central Cancer Registries, 1995 to 2000. Ecological data on the rate of PSA screening by registry area were obtained from the 2001 Behavioral Risk Factor Surveillance System. Counties were grouped into metro and nonmetro areas according to Beale codes from the Department of Agriculture. Pearson correlation analyses were used to examine associations. RESULTS: Significant correlations were observed between the incidence of late-stage prostate cancer and death rates for Whites (r = 0.38, P = 0.04) and Blacks (r = 0.53, P = 0.03). The variation in late-stage disease corresponded to about 14% of the variation in prostate cancer death rates in White men and 28% in Black men. PSA screening rate was positively associated with total prostate cancer incidence (r = 0.42, P = 0.02) but inversely associated with the incidence of late-stage disease (r = -0.58, P = 0.009) among White men. Nonmetro counties generally had higher death rates and incidence of late-stage disease and lower prevalence of PSA screening (53%) than metro areas (58%), despite lower overall incidence rates. CONCLUSION: These ecological data suggest that 10% to 30% of the geographic variation in mortality rates may relate to variations in access to medical care.

Incidence of noncutaneous melanomas in the U.S.

BACKGROUND: Description of the epidemiology of noncutaneous melanoma has been hampered by its rarity. The current report was the largest in-depth descriptive analysis of incidence of noncutaneous melanoma in the United States, using data from the North American Association of Central Cancer Registries. METHODS: Pooled data from 27 states and one metropolitan area were used to examine the incidence of noncutaneous melanoma by anatomic subsite, gender, age, race, and geography (northern/southern and coastal/noncoastal) for cases diagnosed between 1996 and 2000. Percent distribution by stage of disease at diagnosis and histology were also examined. RESULTS: Between 1996 and 2000, 6691 cases of noncutaneous melanoma (4885 ocular and 1806 mucosal) were diagnosed among 851 million person-years at risk. Ocular melanoma was more common among men compared with women (6.8 cases per million men compared with 5.3 cases per million women, age-adjusted to the 2000 U.S. population standard), whereas mucosal melanoma was more common among women (2.8 cases per million women compared with 1.5 cases per million men). Rates of ocular melanoma among whites were greater than eight times higher than among blacks. Rates of mucosal melanoma were approximately two times higher among whites compared with blacks. CONCLUSIONS: In contrast to cutaneous melanoma, there was no apparent pattern of increased noncutaneous melanoma among residents of southern or coastal states, with the exception of melanoma of the ciliary body and iris. Despite their shared cellular origins, both ocular and mucosal melanomas differ from cutaneous melanoma in terms of incidence by gender, race, and geographic area.