As noted in C3 Vol 12, No 3 , the COG Epidemiology Steering Committee is piloting the   establishment of a North American Pediatric Cancer Registry, called the Childhood Cancer Research Network (CCRN) (COG protocol AADM01P1). This research registry involves the active consent of parents (and children, if appropriate) to release names and contact information to the CCRN. This will include all newly diagnosed patients at COG institutions. In addition, consent would be obtained to allow for potential future contact to consider taking part in non-therapeutic research studies. Each study would be separately consented by the individuals conducting the research. We have randomly selected 23 (approx. 10%) COG institutions in the US and Canada for this pilot study. All institutions obtained IRB approval, and thus far, 887 parents/children have been approached for this protocol. Of these, 858 (96.7%) have agreed to release of personal identifiers and future contact, 17 (1.9%) have agreed to release of personal identifiers only, and 12 (1.4%) refused both consent levels. We have recently received funds from the National Cancer Institute to continue with feasibility studies for the CCRN. One aspect of this pilot testing will involve the random selection of 100 mothers of children with cancer who agreed to recontact.   Each mother   will be asked to take part in a brief telephone interview and obtain a buccal cell sample from herself and her child (if feasible) for submission to a central repository for future studies of disease susceptibility. It is anticipated that given the success of this protocol, a comprehensive proposal will be submitted to the National Institutes of Health to open the protocol throughout COG to form the foundation of what will eventually be a North American Pediatric Cancer Research Registry.   This will be a major resource for the conduct of future studies of the etiology of childhood cancers.

As we have noted in the past, the use of random digit dialing for the selection of controls in pediatric cancer case-control studies in North America is getting increasingly more difficult. With the use of cell phones, caller identification, and answering machines, along with the persistence of telemarketing, the participation and characterization of a control base is becoming nearly impossible. In recognition of the development of the CCRN, it will be important to consider new ways in which controls can be obtained for future studies. The Epidemiology Committee has submitted a grant proposal to the NCI to test the feasibility of obtaining alternative control groups including birth certificate controls and cousin controls.   Stay tuned.

FAS and furious!

 

FAS is a cell surface receptor involved in apoptotic signal transmission in many cell types, including cells of the immune system.   FAS belongs to the family of tumor growth factor receptors and binding to the receptor by FAS ligand triggers receptor trimerization and subsequent assembly of the death inducing signaling complex.   This mechanism of FAS mediated apoptosis is believed to be involved in the removal of auto reactive lymphocytes during the normal development of the immune system.   The FAS mediated apoptotic pathway is also dysregulated in several immune system malignancies resulting in down regulation of apoptosis and subsequent assistance of the malignant clone.   BCL2 transgenic mice with targeted expression in myeloid cells develop a myeloproliferative condition analogous to human chronic myelomonocytic leukemia; however, they rarely go on to develop acute leukemia. When mice constitutively expressing BCL2 are crossed onto a FAS -/- background, however, approximately 15% develop AML, implicating FAS mediated apoptosis in the pathogenesis of AML [Traver D et al   Immunity 1998; 9:47 -57] . In a new report Sibley et al [Cancer Research 2003; 63:4327-4330] have genotyped two polymorphisms (G1377A and A670G) in the human FAS promoter in cases (n=471) and controls (n=931) from a case-control study of adult AML.   The data show a significantly increased risk of AML associated with heterozygous and homozygous variants at position -1377 (Odds Ratio (OR)=1.69; 95% CI=41.32 to 2.16). Extended haplotype analysis revealed that the -1377A -670A haplotype was significantly associated with disease (OR= 6.72; 95% CI=3.13 to 14.51). Analysis of genotypes across cytogenetic categories showed no association with particular AML subtype, suggesting that the risk associated with polymorphisms in the FAS promoter occurs in all kinds of AML. In this study the authors demonstrated biological plausibility confirming that the polymorphism study occurred within transcription binding sites and bind transcription factors. They show that the presence of an adenine residue at position 1377, associated with increased risk of leukemia, reduces SP1 binding compared with the guanine residue.  

 

COMMENT: This study adds one more gene to the increasing numbers shown to contribute to risk of AML.   This study has a number of strengths: firstly, the number of cases is large and well-characterized; secondly, the method of genotyping was rigorously verified; lastly, biological plausibility has been demonstrated. It will be of interest to investigate the FAS polymorphism in childhood AML.

Stella M. Davies

Something brewing with Ewing 's

 

Ewing 's sarcoma (ES) and peripheral primitive neuroectodermal tumors (PPNETs) are cancers that are of bone or soft tissue origin.   Cancer cells in ES and PPNETs typically have an   EWS genetic rearrangement and are sometimes studied together, as the Ewing sarcoma family of tumors (ESFT).   ES in the United States is almost exclusively a disease of white children.   Also, rather than occurring mainly in the long bones, as does osteosarcoma, ES occurs throughout the skeletal system.   Together these facts suggest a genetic predisposition to ES, though no familial syndrome has yet been described.   Apart from these descriptive data, case-control studies have, with some consistency, suggested that farm exposures and a history of hernia are associated with ES [Cope J et al, Med Ped Oncol 2000; 34:195-199] .   Recently, two reports from an Australian case-control study on farming [Valery et al. Cancer Causes and Control 2002; 13:263-270] and other exposures [Valery et al. Int J Cancer 2003; 105:825-830] have offered some corroboration of earlier findings.

 

The overall study included 106 ESFT cases diagnosed between 1991-96 in patients   up to forty years of age. The analysis included only those cases whose telephone number was present in the national calling list that served as the population base. Most cases were identified through registries, though those in one state came from clinical trial records and word of mouth. Three hundred forty-four controls, frequency matched by age and state of residence, were recruited through random digit dialing of numbers on the calling list.   Exposure data were obtained through interview, mostly of the parents, though in some instances only from participants' themselves.  

 

According to the first paper on farming, neither parental occupation nor single categories of chemical exposure significantly affected ESFT risk. However, having had at least one parent in farming around the time of pregnancy was suggested to increase risk (OR=1.8; 95% CI: 0.7-4.3), as was having had at least one parent who handled any chemical (OR=2.1; 95% CI:1.2-3.4). In a sub-analysis by age, the risk of ESFT for farm residency was significantly doubled for subjects 0-20 years old (OR = 2.0; 95% CI: 1.0-3.9) but not among subjects 21-40 years old (OR=0.7; 95% CI: 0.2-2.3).

 

The second paper reported on a variety of medical and other factors not related to parental occupation.   The OR for having had any hernia was 3.1 (95%: 1.2-7.6); OR for inguinal hernia (3.6; 95% CI: 1.0-12.8) and inguinal hernias operated on (5.6; 95% CI: 1.3-24.5) were even more pronounced.   Results for other medical conditions or procedures, as well as for pregnancy-related factors, were mainly null.   There was a significant inverse trend with age at first shave among boys (p = 0.01), though not with age at first menstrual period among girls (p = 0.62).   For boys, the OR comparing a first shave at fourteen years or less to first shave at seventeen years or greater was 4.1 (95% CI: 1.4-12.1).   Results for other signs of puberty, such as the appearance of pubic or underarm hair, were not significant. Lastly, the investigators reported a significant increasing trend with increasing exercise (p = 0.04).  

 

COMMENT: This study had the advantages of a clear sampling frame and high participation rates. Dis-advantages were the collection of data by interview, which raises the possibility of recall bias for certain exposures, and small sample size, which decreased precision. The suggestions of increased risk of ESFT with farm   residency or work were not significant, but warrant interest due to similar findings in previous studies.   We would not expect recall bias to be a major influence in those results. On the other hand, the association of ESFT with combined parental chemical exposure could be due to recall bias, since they are obvious candidates for parental concern. The associa-tion of hernias with ESFT was most intriguing, given that the association   was strong, recall is likely to be accurate (especially those operated on), and consistent with three previous studies that reported on the topic. The inverse association of ES with secondary sexual characteristics has also been observed before. The   literature on ESFT is small but the consistency of the reports, at least as regards farming and hernias, suggests that future research on these tumors may be especially promising.   

Logan G. Spector

C3 Quarterly Newsletter
Children's Cancer Research Fund
Epidemiology Research Unit
Division of Pediatric Epidemiology
Clinical Research
University of Minnesota
420 Delaware St. SE, Box 422
Minneapolis, MN 55455
pedsepi@umn.edu

Editors: 
Stella M. Davies, MD, PhD, and Julie A. Ross, PhD