Recall of diet during a past pregnancy.

The authors conducted a study of women's ability to recall diet during a past pregnancy. For a prospective study, women completed self-administered food frequency questionnaires (FFQs) before and during pregnancy (1989-1992). These women, mostly White and well-educated, were contacted 3-7 years later (1996-1997) for a retrospective dietary assessment performed by either telephone interview (n = 154) or self-administered FFQ (n = 115). Energy-adjusted Pearson correlations ranged from 0.10 to 0.49 for the telephone interview group and from 0.02 to 0.67 for the self-administered questionnaire group. When participants' intakes were ranked, quintile agreement (within one quintile) between original diet and recalled diet ranged from 60% to 69% in the telephone interview group and from 69% to 79% in the self-administered questionnaire group. Correlations and percentages of agreement were higher among women who used the same questionnaire for both dietary assessments than among those who used different questionnaires. These results suggest that diet during pregnancy is recalled with similar accuracy as or perhaps slightly lower accuracy than adult diet generally. This may reflect, in part, the influence of current (nonpregnancy) diet on recall of past (pregnancy) diet. While the results of this study may not be generalizable to those obtained from other populations, to the authors' knowledge it is the first study of recall of diet during pregnancy.

Cyclophosphamide-based, seven-drug hybrid and low-dose involved field radiation for the treatment of childhood and adolescent Hodgkin disease.

PURPOSE: The outlook for children and adolescents with Hodgkin disease (HD) is excellent with combined modality therapy. However, the long-term toxicities of multiagent therapy and radiation therapy remain of concern for these patients with curable disease. In an attempt to reduce long-term toxicities while preserving excellent cure rates, we developed a combined-modality protocol using a modified seven-drug hybrid and low-dose (2,000 cGy) involved field radiation therapy (RT). The hybrid used cumulative doses of alkylating agents and anthracyclines that were lower than those used in previous four-drug regimens and substituted a less leukemogenic agent, cyclophosphamide, for nitrogen mustard. PATIENTS AND METHODS: From 1991 through 1994 a cyclophosphamide, vincristine, procarbazine, and prednisone/adriamycin, bleomycin, and vinblastine hybrid was used to treat 29 patients with HD. Median age was 12 years (range 6-16 yrs). Patients who were postpubertal with early stage disease as determined by surgical staging were excluded. Treatment consisted of four cycles of therapy for stages I and IIA, six cycles for stages IIB and III, and eight cycles for stage IV. Twenty-two patients also received low-dose RT to areas of bulky disease. RESULTS: Twenty-eight patients (97%) had a complete response to chemotherapy. Five patients experienced relapse; two died from disease 27 and 29 months after initial diagnosis; three received additional therapy and are alive with no evidence of disease. Follow-up for all other patients is a median of 56 months (range 24-78 mos) from cessation of therapy and all have remained disease-free. At 5 years follow-up, actuarial disease-free survival is 82%, and the overall survival is 93%. There have been no clinically significant cardiac or pulmonary toxicities and no secondary malignancies. CONCLUSIONS: This therapy has resulted in 5-year overall survival and disease-free survival rates similar to regimens using higher doses of alkylating agents, anthracyclines, and radiation. Longer follow-up will be necessary to fully evaluate disease-free survival, organ damage, and quality of life.

Chemoreduction and local ophthalmic therapy for intraocular retinoblastoma.

PURPOSE: To study the effectiveness of combined systemic chemotherapy and local ophthalmic therapy for retinoblastoma with the goal of avoiding enucleation and external-beam radiation therapy (EBRT). PATIENTS AND METHODS: This was a prospective, nonrandomized, single-arm clinical trial. Seventy-five eyes were followed in 47 children. Patients were treated with a six-cycle protocol of vincristine, etoposide, and carboplatin. Most (83%) also received ophthalmic treatment (cryotherapy, laser photocoagulation, thermotherapy, or plaque radiation therapy) during and/or after the chemotherapy. RESULTS: With a median follow-up of 13 months, event-free survival was 74%, with an event defined as enucleation and/or EBRT. Six children required EBRT in seven eyes (9%); five required enucleation of one eye (7%); five required a combination of EBRT and enucleation in six eyes (8%). Reese-Ellsworth groups 1, 2, and 3 eyes had excellent results, with avoidance of EBRT or enucleation in all 39. Treatment of groups 4 and 5 was less successful, with 33% of six eyes and 53% of 30 eyes, respectively, requiring EBRT and/or enucleation. Toxicities from chemotherapy were mild and included cytopenias (89%), fever and neutropenia (28%), infection (9%), and gastrointestinal symptoms, dehydration, and vincristine neurotoxicity (40%). No patients developed a second malignancy, metastatic disease, renal disease, or ototoxicity. CONCLUSION: In retinoblastoma patients with Reese-Ellsworth eye groups 1, 2, or 3, systemic chemotherapy used with local ophthalmic therapies can eliminate the need for enucleation or EBRT without significant systemic toxicity. More effective therapy is required for Reese-Ellsworth eye groups 4 and 5.

The descriptive epidemiology of craniopharyngioma.

OBJECT: In this report the authors describe the epidemiology of craniopharyngioma. METHODS: The incidence of craniopharyngioma in the United States was estimated from two population-based cancer registries that include brain tumors of benign and borderline malignancy: the Central Brain Tumor Registry of the United States (CBTRUS) and the Los Angeles county Cancer Surveillance Program. Information on additional pediatric tumors was available from the Greater Delaware Valley Pediatric Tumor Registry (GDVPTR). The overall incidence of craniopharyngioma was 0.13 per 100,000 person years and did not vary by gender or race. A bimodal distribution by age was noted with peak incidence rates in children (aged 5-14 years) and among older adults (aged 65-74 years in CBTRUS and 50-74 years in Los Angeles county). Survival information was available from GDVPTR and the National Cancer Data Base (NCDB), a hospital-based reporting system. In the NCDB, the 5-year survival rate was 80% and decreased with older age at diagnosis. Survival is higher among children and has improved in recent years. CONCLUSIONS: Craniopharyngioma is a rare brain tumor of uncertain behavior that occurs at a rate of 1.3 per million person years. Approximately 338 cases of this disease are expected to occur annually in the United States, with 96 occurring in children from 0 to 14 years of age.

Maternal diet during pregnancy and risk of brain tumors in children.

Nine studies of childhood brain tumors and maternal diet during pregnancy have focused on foods related to the N-nitroso-compound(NOC) hypothesis. An association between frequent consumption of cured meat by pregnant women and increased risk is a consistent finding in most of the studies. The data on fruit and vegetable consumption are less consistent, but suggest decreased risk. Studies that assess all aspects of maternal diet during pregnancy are needed to determine whether the observed associations remain after adjustment for other aspects of diet. Such comprehensive studies also may elucidate other dietary factors that affect the risk of brain tumors in children.

Paternal age and sporadic neurofibromatosis 1: a case-control study and consideration of the methodologic issues.

Sporadic neurofibromatosis 1 (NF1) occurs in the absence of a family history of the disease and usually results from a new mutation in the germ cell of one of the parents, most commonly the father. Older paternal age may increase the risk for a new germinal NF1 mutation, but the results of studies to address this question conflict. We investigated paternal age in sporadic NF1 by using a case-control study design. Patients who were seen at two specialty NF clinics in Houston, Texas, born between 1970 and 1992 and living in the Houston area and surrounding counties, were studied. Birth certificates with information on the father were found for 89 cases. For each case, two birth certificates were chosen at random from the same year and county of birth. In this way, the control group of 178 individuals was formed. Fathers of patients with NF1 were 1.5 years older than fathers of control subjects at the birth of the child, but the difference was only of borderline statistical significance (P = 0.07). This paternal age difference was not changed by adjustment for socioeconomic status or maternal age. These and previous data are consistent with either a small paternal age effect in sporadic NF1 or a bias such as that resulting from the selection of cases and/or controls.

The descriptive epidemiology of craniopharyngioma.

The incidence of craniopharyngioma in the United States was estimated from two population-based cancer registries that include brain tumors of benign and borderline malignancy: the Central Brain Tumor Registry of the United States (CBTRUS) and Los Angeles county. Information on additional pediatric tumors was available from the Greater Delaware Valley Pediatric Tumor Registry (GDVPTR). The overall incidence of craniopharyngioma was 0.13 per 100,000 person years and did not vary by gender or race. A bimodal distribution by age was noted with peak incidence rates in children (aged 5-14 years) and among older adults (aged 65-74 years in CBTRUS and 50-74 years in Los Angeles county). Survival information was available from GDVPTR and the National Cancer Data Base (NCDB), a hospital-based reporting system. In the NCDB, the 5-year survival rate was 80% and decreased with older age at diagnosis. Survival is higher among children and has improved in recent years. Approximately 338 cases of craniopharyngiomas are expected to occur annually in the United States, with 96 occurring in children from 0 to 14 years of age.

Increasing incidence of childhood cancer: report of 20 years experience from the greater Delaware Valley Pediatric Tumor Registry.

As little is known about the aetiology of cancer in children, analysis of time trends may be useful. Recent data on time trends for paediatric cancers are very limited. We report here on trends in the incidence of 15 categories of cancer in children under 15 years of age from 1970 to 1989, using data from the Greater Delaware Valley Pediatric Tumor Registry in the US. Total cancer incidence increased 1% per year (P < 0.001). Neither acute lymphocytic leukaemia, acute myelocytic leukaemia, nor total leukaemia incidence changed significantly. In contrast, the incidence of central nervous system (CNS) tumours rose 2.7% per year (P < 0.001). All three subgroups of this category, glioma, primitive neuroectodermal tumor (PNET)/medulloblastoma, and other CNS tumours, showed increases. For glioma and PNET/medulloblastoma, trends differed by age, race, and/or gender. Among the other childhood cancers, significant increases were observed for non-Hodgkin lymphoma and neuroblastoma. For osteosarcoma and retinoblastoma, no overall change in incidence was observed, although decreases were observed in some age and race subgroups. The rise in CNS tumour incidence confirms previous reports from the US and Great Britain. The lack of change for acute lymphocytic leukaemia conflicts with other data from the US, but diagnostic changes appear to explain at least part of the discrepancy. The increase in neuroblastoma has also been observed in Great Britain. In contrast to our finding, investigators in the US and Great Britain have reported no rise in non-Hodgkin lymphoma. Analyses for more of the childhood cancers from other registries would aid in detecting and interpreting incidence trends in recent years.

Relationship between Fc receptor IIA polymorphism and infection in children with sickle cell disease.

OBJECTIVE: Despite penicillin prophylaxis and vaccination, infection with encapsulated organisms remains a leading cause of morbidity and death in children with sickle cell disease. The role of Fc receptors in the clearance of encapsulated organisms is well documented. The His(H)-Arg(R) polymorphism at amino acid 131 of the Fc gamma RIIA receptor alters binding affinity for human IgG2 and influences infection with encapsulated organisms in children without sickle cell disease. We hypothesized that the genotype for high-affinity human IgG2 binding (H/H131) is underrepresented in children with sickle cell disease who had encapsulated organism infection. DESIGN: We studied 60 black children with sickle cell disease from four participating centers who had a history of encapsulated organism infection. Genomic DNA from peripheral blood was subjected to amplification by polymerase chain reaction and to sequence analysis for identification of the Fc gamma RIIA genotype, and the genotype distribution was then compared with our data from ethnically matched control subjects. RESULTS: Contrary to our hypothesis, the H/H131 genotype was overrepresented in all individuals (p = 0.046) and in particular in the 11 individuals with a history of Haemophilus influenzae type b infection (64% H/H131, 27% H/R131, 9% R/R131; p = 0.002), in comparison with ethnically matched control subjects (14% H/H131, 60% H/R131, 26% R/R131). In the 51 individuals with a history of Streptococcus pneumoniae infection, the genotype distribution was not statistically significantly different from that of the control population. CONCLUSIONS: The H/H131 Fc gamma RIIA genotype is overrepresented in black children with sickle cell disease and a history of H. influenzae type b infection but not in those with S. pneumoniae infection.

Informing subjects of epidemiologic study results. Children's Cancer Group.

OBJECTIVE: To assess the feasibility and process of providing feedback to parents regarding the results of epidemiologic research, in particular to look at the importance and clarity of the information provided, parental reactions to the results, and utilization of the data provided. METHODOLOGY: Parents who participated in an epidemiologic study of pediatric brain tumors (patient and control mothers) were sent a letter summarizing the results of the study and the Parent Study Results Survey to complete and return. The final sample used for analyses was 109 (patient) and 90 (control) mothers. Analyses were conducted to determine differences between patient and control mothers and differences among subsets defined by educational level and vital status of the patient. RESULTS: Mothers rated the importance and clarity of the information very highly, although patient mothers were more likely than control mothers to want more information and a telephone contact. Patient and control mothers were similar in reported sadness, anxiety, and being overwhelmed, but patient mothers felt less satisfied and relieved. Patient mothers expressed feeling more guilt nad anger than control mothers, although even the levels among the patient mothers were only moderate. Close to half of all mothers commented on the inconclusiveness of the study results. Nearly all mothers indicated they would suggest that other parents participate in epidemiologic research. CONCLUSIONS: It is valuable to many parents that they receive information about results of research in which they have participated. We found little evidence of strong negative effects to a detailed feedback letter. We recommend that evaluative data be used to guide the process of informing research participants about study results and that investigators consider making feedback letters a standard part of research protocols.

p53 gene mutations in pediatric brain tumors.

We investigated the frequency of p53 mutations in 47 pediatric brain tumors of various histologic subtypes that were collected over a period of 5 years. The specimens included 15 primitive neuroectodermal tumors (PNETs), 17 low grade astrocytomas, one anaplastic astrocytoma, three glioblastomas (GBMs), one mixed glial tumor, eight ependymomas, one choroid plexus carcinoma, and one gangliocytoma/ganglioneuroma. Mutations were identified by single strand conformation polymorphism analysis of exons 4-8 and verified by sequencing. Mutations were present in 2 of 3 cases of GBM, but not in 17 low grade astrocytomas (P = 0.02, Fisher's exact test). One GBM demonstrated a germline GGC to AGC transition (gly to ser) at codon 245 with loss of the wild-type allele. A second GBM contained a CGG to TGG transition (arg to trp) at codon 248, also with loss of the wild-type allele, but normal tissue was not available for comparison. In addition, one of 15 PNETs retained heterozygosity but demonstrated a somatic CGT to TGT transition (arg to cys) at codon 273. p53 mutations were absent in other histologic subtypes and in two cases with multiple primary cancers. These data are consistent with earlier findings that p53 mutations are rare in PNETs, which are primarily pediatric tumors. In contrast to adult gliomas, p53 mutations in pediatric gliomas appear restricted to the GBMs. The lack of p53 mutations in pediatric low grade astrocytomas suggests not only histological differences, but also a different molecular pathogenesis in adult and pediatric patients.

Neurologic and other disorders in relatives of pediatric patients with CNS tumors.

Abnormalities of embryogenesis and nervous system development may cause or contribute to the development of childhood brain tumors. To identify genetic or environmental factors that may be associated with etiologies of childhood central nervous system tumors, we examined family histories of 165 children with such tumors for the presence of neurologic disorders, including neural tube defects, mental retardation, seizures, and central nervous system tumors, as well as other cancers and birth defects. Only 1 patient, with the neurofibromatosis-Noonan syndrome, was confirmed to have an underlying syndromic diagnosis associated with central nervous system tumorigenesis. Families of 2 probands with posterior fossa primitive neuroectodermal tumors reported relatives with olivopontocerebellar atrophy. Although increased incidences of study disorders were not identified in this population, it is possible that within individual families one or more of these disorders is related to childhood central nervous system tumorigenesis.

Cognitive deficits in long-term survivors of childhood medulloblastoma and other noncortical tumors: age-dependent effects of whole brain radiation.

Twenty-four children, aged 1.5-20 yr at diagnosis, with noncortical brain tumors, primarily medulloblastoma, have been followed for 3-4 yr for intellectual status. All the children received craniospinal irradiation, and 19 of 24 received chemotherapy as well. For the group as a whole. Full Scale IQ fell from 104 at baseline to 91 at final follow-up. Children younger than 7 yr at diagnosis showed a significant decrease in IQ as early as year 1, and all changes from baseline to years 3 and 4 were significant. In contrast, children older than 7 yr at diagnosis did not show a significant IQ change from baseline to year 3 or 4. The Spearman correlation coefficient between IQ change and age at diagnosis from baseline to year 4 was 0.57 (P = 0.003). This study supports the hypothesis that children treated with whole brain radiation at a younger age have more severe cognitive impairment than those treated at a later age. Limitations in sample size and duration of observations do not permit us to identify whether a true plateau occurs 2-4 yr after irradiation versus a continued progressive decline in intellectual performance. Moreover, we cannot at this time distinguish between a true dementing process versus failure to acquire new cognitive skills at a rate comparable to age-matched peers.

Risk factors for astrocytic glioma and primitive neuroectodermal tumor of the brain in young children: a report from the Children's Cancer Group.

We conducted a matched case-control study to investigate risk factors for the two most common types of brain tumors in children, astrocytic glioma and primitive neuroectodermal tumor (PNET). Since the study focused on gestational exposures, we restricted it to young children because these exposures would be expected to act early in life. Parents of 155 astrocytic glioma cases, 166 PNET cases, and controls identified by random digit dialing completed telephone interviews. Few associations occurred with the hypothesized risk factors, which were gestational exposure to alcohol, hair coloring products, farms, and substances containing N-nitroso compounds (passive smoking, makeup, incense, new cars, pacifiers, baby bottles, beer). Of the products studied that contain N-nitroso compounds, only beer was associated with a significantly increased risk of either tumor type [odds ratio (OR) for PNET = 4.0; 95% confidence interval (CI), 1.1-22.1; P = 0.04]. Elevated ORs for PNET were observed for farm residence of the mother during the pregnancy (OR = 3.7; 95% CI, 0.8-23.9; P = 0.06) and of the child for at least a year (OR = 5.0; 95% CI, 1.1-46.8; P = 0.04). Significant associations with astrocytoma were observed for mother's use of kerosene (OR = 8.9; 95% CI, 1.1-71.1; P = 0.04) and birth by Caesarean section (OR = 1.8; 95% CI, 1.1-3.2; P = 0.03). History of miscarriage was associated with a lower risk of PNET (OR = 0.5; 95% CI, 0.3-0.9; P = 0.02).(ABSTRACT TRUNCATED AT 250 WORDS)

Maternal diet and risk of astrocytic glioma in children: a report from the Childrens Cancer Group (United States and Canada)

N-nitroso compounds and their precursors, nitrites and nitrates, have been hypothesized as risk factors, and vitamins C and E, which inhibit N-nitroso formation, as protective factors for brain tumors. A case-control study of maternal diet during pregnancy and risk of astrocytoma, the most common childhood brain tumor, was conducted by the Childrens Cancer Group. The study included 155 cases under age six at diagnosis and the same number of matched controls selected by random-digit dialing. A trend was observed for consumption of cured meats, which contain preformed nitrosamines (a class of N-nitroso compounds) and their precursors (adjusted odds ratio [OR] for highest quartile of intake relative to lowest = 1.7, P trend = 0.10). However, no strong trends were observed for nitrosamine (OR = 0.8, P = 0.60); nitrite (OR = 1.3, P = 0.54); nitrate (OR = 0.7, P = 0.43); vitamin C (OR = 0.7, P = 0.37); or vitamin E (OR = 0.7, P = 0.48). Iron supplements were associated with a significant decrease in risk (OR = 0.5, 95 percent confidence interval = 0.3-0.8). The effect of several dietary factors differed by income level, making interpretation of the results difficult. Future research should investigate the effect of dietary components not assessed in this study, as these may explain the disparate effects by income level. The results of this study provide limited support for the nitrosamine hypothesis.

Family history of cancer and seizures in young children with brain tumors: a report from the Childrens Cancer Group (United States and Canada).

The occurrence of cancer and neurological disorders in first- and second-degree relatives of children in the United States and Canada diagnosed with brain tumor before age six was investigated. A pair-matched case-control study with 155 astrocytoma and 166 primitive neuroectodermal tumor (PNET) cases was performed. Cases were identified through the Childrens Cancer Group. Controls were selected by random-digit dialing and matched to cases on age, race, and telephone area code and exchange. Childhood cancers were more common in PNET relatives compared with the general population (standardized incidence ratio [SIR] = 2.5, 95 percent confidence interval [CI] 1.1-4.8, P = 0.02) and with control relatives (odds ratio [OR] = 3.0, CI = 0.5-30, P = 0.29). For astrocytoma, nonsignificant excesses of brain tumor, leukemia/lymphoma, and childhood cancer occurred among case relatives compared with control relatives, but not compared with the general population. Astrocytoma cases were significantly more likely than controls to have a relative with seizures (OR = 2.5, CI = 1.2-4.9, P = 0.009), especially childhood seizures (OR = 3.4, CI = 1.2-12, P = 0.02), epilepsy (OR = 3.0, CI = 0.9-13, P = 0.08), and febrile convulsions (OR = 4.5, CI = 0.9-43, P = 0.07). A family history of stroke was not a risk factor for either type of brain tumor. These results suggest that some childhood brain tumors may result from a genetic susceptibility and that some risk factors may affect childhood astrocytoma and PNET differently.

Relation between maternal diet and subsequent primitive neuroectodermal brain tumors in young children.

BACKGROUND: It has been hypothesized that a high dietary intake of nitrosamines and their precursors, nitrites and nitrates, is a risk factor for brain tumors. Vitamins C and E inhibit the formation of nitrosamines and thus may be protective. METHODS: We conducted a case-control study of maternal diet and the risk of primitive neuroectodermal tumors of the brain in children. The case patients were under the age of six years at diagnosis in 1986 to 1989. The controls were selected by random-digit telephone dialing and were matched for age and race to 166 case patients. Telephone interviews with the mothers included questions on the frequency of consumption of alcohol, vitamin and mineral supplements, and 53 foods during pregnancy. RESULTS: Significant protective trends were observed for vegetables (odds ratio for the highest quartile group for intake relative to the lowest, 0.37; P for trend = 0.005), fruits and fruit juices (odds ratio, 0.28; P = 0.003), vitamin A (odds ratio, 0.59; P = 0.03), vitamin C (odds ratio, 0.42; P = 0.009), nitrate (odds ratio, 0.44; P = 0.002), and folate (odds ratio, 0.38; P = 0.005). A nonsignificant trend of increasing risk was observed for nitrosamine (odds ratio, 1.65; P = 0.15). The use of iron (odds ratio, 0.43; P = 0.004), calcium (odds ratio, 0.42; P = 0.05), and vitamin C (odds ratio, 0.35; P = 0.04) supplements at any time during the pregnancy and the use of multivitamins during the first six weeks (odds ratio, 0.56; P = 0.02) were associated with decreased risk. In multivariate analyses, folate, early multivitamin use, and iron supplements generally remained protective. CONCLUSIONS: These results do not support the hypothesis that nitrosamines have a role in the development of primitive neuroectodermal tumors in young children, but they do suggest that certain other aspects of maternal diet can influence the risk.

Three- and four-year cognitive outcome in children with noncortical brain tumors treated with whole-brain radiotherapy.

Cognitive function and school achievement were studied prospectively over 3 to 4 years in 19 children treated for brain tumors with whole-brain radiotherapy; 14 of 19 also received adjuvant chemotherapy. For the group as a whole, mean IQ fell from a baseline of 104 to 92 at follow-up (p < 0.01). Age was inversely correlated with change in IQ over time (r = 0.71; p < 0.001). Children younger than 7 years at diagnosis had a mean IQ loss of 27 points, while children over 7 years at diagnosis showed no significant decrease in IQ. Decline in IQ occurred between baseline and year 2 of follow-up; none could be documented between years 2 and 4. All children younger than 7 years at diagnosis were receiving special education at follow-up; 50% of the children over 7 years at diagnosis were receiving supplemental educational services.