Norway

Environmental and Genetic Causes of Birth Defects
(Project: # ES049027 )
Wilcox, Allen J (wilcox1@niehs.nih.gov) - NIEHS
Abstract: Summary: Facial clefts are one of the most common birth defects, affecting 2 out of every thousand babies. While the main causes of facial clefts are unknown, it is obvious that genetics plays a strong role. We've shown that families with one member affected by clefts are 40 times as likely to have a baby with a cleft. Environmental factors are also presumed to play a role in clefts. For example, clefts are easily produced in experimental animals exposed to teratogens. It is likely that humans vary in their genetic susceptibility to teratogens that cause clefts. More than a decade ago, we anticipated that genetic susceptibility would become an important area of epidemiologic research at NIEHS. Accordingly, we selected facial clefts as a condition with both genetic and environmental causes, and we began in 1992 to develop a study to address the causes of clefts. In 1996 we launched a population-based case-control study of facial clefts in Norway. (Norway has one of the highest rates of cleft lip and palate in the world.) The field phase was completed in 2002. We enrolled 88% of all babies with facial clefts born in Norway between 1996 and 2002 (574 cases), and 76% of eligible control infants (763) selected randomly from the population. Mothers of these infants provided detailed information on occupational and other exposures during pregnancy, as well as on nutrition, personal habits and medical history. Biological samples for DNA analysis were collected from cases and controls as well as their mothers, fathers and siblings. These total nearly 4000 people. DNA has been extracted from these samples and is now ready for genetic analysis. In the course of carrying out this study, we developed and published a new statistical strategy for the analysis of genetic data in case-parent triads that has been widely adapted. We have demonstrated the application of this new method in a preliminary analysis of 262 case-parent triads. We are working intensively on analyses of folic acid, cigarette smoking and hazardous occupations, and their effects on the risk of facial clefts. Parents? occupational exposures We have information on mother?s and father?s employment during the first trimester of pregnancy. Previous studies have identified occupational risks for clefts among service-related industries, agriculture and manufacturing. We coded employment using standard Nordic occupation and industry classifications. In an analysis led by Ruby Nguyen, my post-doctoral fellow, employed women who worked in manufacturing were found to have a slightly higher risk than other employed women (OR=1.3, 0.73-2.3). Adjustment for smoking, folate intake, income, year, and maternal education reduced this risk slightly (aOR=1.2, 0.65-2.3). This excess risk was present only among babies with cleft palate only (CPO) (aOR=1.9, 0.82-4.2). When we restricted the sample to women working full time, the aOR for CPO increased somewhat (aOR=2.7, 0.95-7.8). After excluding cleft palate cases that also had non-facial birth defects, the adjusted risk was 4.7 (1.6-7.3). With only 51 exposed women, we were unable to consider specific chemical exposures. If this association is not due to chance, it may reflect chemical exposures or perhaps unmeasured effects of shift-work or noise. A previous European study had reported increased risk with mother?s employment in manufacturing, but only for CLP and not CPO. Previous studies had identified beauty care workers as having increased risk for CL+P. There was an increased risk in our data as well, although the number of exposed mothers was small and the estimate imprecise (aOR=1.9; 0.57-6.2). After excluding cases with other defects, the aOR was 6.2 (1.0-38). We also asked women about their husbands? occupation and industry. While we don?t expect substantial associations with father?s exposures, this information has been reported in previous studies and will be incorporated into our analysis. Folic acid and clefts Low maternal folate causes neural tube defects, and folate supplementation reduces the risk. Researchers have proposed that folate might also help to prevent facial clefts. This hypothesis is not without some biological plausibility (the mid-facial structures of the lip and palate are formed from the same embryonic neural crest cells that create the neural tube), but epidemiologic data have been inconclusive. With this hypothesis in mind, we designed our case-control study to collect detailed folate exposure data during pregnancy. There are two main sources of folate: vitamin supplements and diet. We asked women about their use of folate supplements before and during pregnancy. Women who took supplements were asked to send an empty bottle so that we could confirm the dose. We also asked for bottles from other vitamin supplements, some of which contained unreported folate. To be consistent with earlier studies, we chose for our analysis to define the exposure window as the month before pregnancy and the first two months of pregnancy. We used the food frequency questionnaire to estimate dietary sources of folate in the first trimester. Dose categories for folate supplementation were chosen a priori on the basis of the frequency distribution of dose, with modes at 100 ug and 400 ug per day. Half of control women took no supplement, about one-quarter were in the low dose group, and one-quarter were in the high dose group. We see strong effects of folate on the risk of cleft lip with or without cleft palate (CL+P). We adjusted these risks by smoking, dietary folate, income, mother?s education, and year of case birth. Adjustment weakened the association with CPO, but had virtually no effect on the CL+P associations. Women who took 250 ug or more had an aOR of 0.60 for CL+P (95% CI = 0.42 ? 0.85) compared with all others. Excluding clefts with other defects did not change the results. Other vitamin supplements were unrelated to either type of cleft. Dietary folate had similar effects. The figure shows crude (unadjusted) risks. There was a consistent dose-response pattern across quartiles for cleft lip and palate, and a weaker pattern for cleft palate only. Once again, adjustment had little influence. Adjusting for the same variables as above and adding folate supplement to the model, women in the highest quartile had a risk of 0.67 for CL+P (0.47 ? 0.97) compared with the lowest quartile. The risk for CPO among women in the highest dietary quartile was 0.74 (0.48 ? 1.14). Once again, excluding clefts cases with other defects did not change the results. The two sources of folate were nearly independent in their effects. When we combine the two sources, the risk of CL+P was 0.43 (0.27 ? 0.69) among folate users (250+ mg) who were above the median in dietary folate, compared with women who took less than 250 mg folate and were below the dietary median. For cleft palate only, this risk was 0.56 (0.33 ? 0.96). These results are robust to minor changes in the exposure window. These data provide the strongest evidence to date that folate deficiency is a cause of facial clefts. We plan to pursue this finding in more detail through the assay of folate metabolism genes, which will allow us to look within genetically susceptible groups. Clinical classification of facial clefts We compiled clinical data from the two hospitals in Norway that treated facial clefts between 1967 and 1998, and combined these with data from the Medical Birth Registry to provide a complete population-based record of facial clefts for Norway. Using clinical data from more than 3500 cases, we then provided a detailed description of these defects by clinical severity, at a level of unprecedented detail for population-based data. These data provide a reference not just for clinical centers but for epidemiologists exploring clusters.

Epidemiologic Study of Reproductive Outcomes and Environment
(Project: # ES044003 )
Wilcox, Allen J (wilcox1@niehs.nih.gov) - NIEHS
Abstract: Summary: In the mid-1980s we carried out a prospective study of early pregnancy in which we enrolled 221 health women who were planning to become pregnant. These women collected daily urine specimens for up to six months. We've assayed these specimens to describe the hormonal events of the menstrual cycle and early pregnancy. 155 women became clinically pregnant during the study, while 44 had pregnancies that ended so early that the pregnancies were detectable only by assay of urinary human chorionic gonadotropin. This unique study has been called a landmark, and continues to provide a rich resource for the description of the earliest stages of pregnancy. (More than 30,000 urine samples are still being stored.) We've published more than 40 papers from this study over the past two decades, some of which have led to new understanding of the fundamental processes of conception and early pregnancy. In addition, we have continued to make use of large population registries in order to pursue basic questions on pregnancy and maternal and infant health. We have worked especially closely with Norwegian colleagues in the analysis of the Norwegian Medical Birth Registry. Last year's progress. It is well known that clinical miscarriages can be accompanied by heavy bleeding. However, there has been no description of the vaginal bleeding that accompanies loss of pregnancy earlier in gestation. Such bleeding is typically mistaken for ordinary menstrual bleeding, since women usually do not realize they are pregnant. We used menstrual diary data from 42 women with early losses to assess the amount of bleeding with early losses, compared with the women's normal (non-conception) menstrual cycles. On average there was no increased bleeding with early loss. Unexpectedly, the earliest losses were accompanied by less bleeding than average, while later (but still subclinical) losses had slightly heavier bleeding. Reasons for bleeding less than the usual menstrual cycle with the earliest losses are unknown. We speculate that this may reflect suboptimal priming of the endometrial lining that contributes to the loss itself. We have pursued our interest in preeclampsia using data from the Norwegian Medical Birth Registry. Preeclampsia remains one of the most enigmatic of all diseases, manifest only in pregnancy and "cured" by delivery of the fetus. While a familial risk has been recognized, it has not been clear how much of the risk is from genes carried by the mother, or genes carried by the fetus. We demonstrated that both play a distinct role. This may provide direction for further studies that focus on candidate genes for preeclampsia. In another study of preeclampsia, we showed that induction of delivery (done to protect mothers with preeclampsia from developing more severe disease) has apparently benefited the fetus as well as the mother, at least in terms of the fetus' survival. We used data from Norway to show that fetal mortality associated with preeclampsia has dropped dramatically over recent decades, as doctors increasingly intervened to deliver the pregnancy prematurely. Such preterm delivery might be expected to put the newborn at some increased jeopardy, but we saw no evidence of higher mortality during the neonatal period with preeclampsia. This does not rule out the possibility of later problems from preterm delivery, but it does provide some preliminary reassurance that current medical practices to protect mothers with preeclampsia are not producing a compensatory cost in the newborn. Public Health or Environmental Health Significance: Successful reproduction is highly valued by most couples (as evidenced by the thousands who resort to advanced technologies in order to conceive). For society as a whole, there are many costs of reproductive failure, including the long-term care required by individuals born with disabilities. The underlying goal of our research program to understand the causes of reproductive problems, in order to avert problems that are preventable. Research Accomplishments: Mothers lose about 25% of their pregnancies before the women even know they are pregnant. We have showed that the bleeding with such pregnancy losses is, on average, indistinguishable from the bleeding of ordinary menses. We also showed that the medical practice of preventing the advance of maternal preeclampsia by delivering women before the baby has reached term has actually benefited the fetus as well. Fetal mortality associated with preeclampsia has declined over recent decades as intervention has become more common. We found no evidence that the babies being delivered early suffered increased mortality during the neonatal period.

Genetic and Environmental Factors in Childhood Respiratory Asthma
(Project: # ES049019 )
London, Stephanie (london2@niehs.nih.gov ) - NIEHS
Abstract: Asthma has long been known to run in families. More recently, linkage studies have been conducted to identify specific genes involved. However, this work has gone slowly with multiple regions of linkage being identified. It is clear that gene-environment interactions are important in determining who develops asthma. Linkage studies conducted to date have not been designed to properly explore interactions. Given the complex nature of asthma, it is clear that candidate gene association studies will be necessary to identify disease genes and interactions with environmental factors. The role of air pollution in asthma exacerbation has long been appreciated. Ozone, an oxidant air pollutant, is one of the agents that can worsen existing asthma. Recent data suggest that ozone and other air pollutants may play a role in asthma etiology, but there are few data. Recent linkage studies in animals are beginning to identify genes that are involved in specific physiologic responses to ozone and other inhaled pollutants. This project includes several studies the role of genetic polymorphisms, environmental factors and their interaction in the etiology of childhood respiratory disease. The primary project is a family study of genetic susceptibility to asthma in a highly ozone exposed population, Mexico City. This study uses the case-parent triad design. We are actively conducting genetic analyses with these samples. I have established collaborations with two birth cohorts to study early life factors in the etiology of asthma. In the ALSPAC study birth cohort, we are examining the relation between breastfeeding and objective measures of atopy and asthma. I am also working Norwegian and other European asthma researchers to establish an asthma research program within the Norway Mother and Child Cohort (called MoBa), a population based cohort of 100,000 pregnant women in Norway who are being followed until their children reach adulthood. The asthma program will be focused around gene-environment interaction. NIEHS/DIR partially supports the MoBa study with the goal of enabling such add-on studies. As background for developing this project, I am analyzing data from the cohort to follow-up well established findings from the literature that children of later birth order are at decreased risk for asthma. This has been ascribed to greater childhood infections in later born children under the so-called "hygiene hypothesis". However, the in utero environment may also be different for later born children than for their older siblings. Because MoBa is one of the few studies including women who have enrolled for more than one pregnancy, I am examining whether exposures relevant to asthma risk in the offspring change from one pregnancy to the next and whether the atopic status of the mother changes across pregnancies. The other cohort that I have established collaboration with is ALSPAC, a birth cohort of 12,000 women in Avon, UK. We are examining the relation between breast feeding and the later development of asthma and atopy using objective measures of these phenotypes.

Pesticides and Parkinson's Disease in the Agriculture H
(Project: # ES044008 )
Hoppin, Jane (hoppin1@niehs.nih.gov) - NIEHS
Abstract: The Norwegian Mother and Child study is a cohort study of pregnant women and their children in Norway. NIEHS is contributing to this effort by creating of a biological specimen repository to evaluate environmental exposures and health effects to pregnant women and their unborn children. As of July 31, 2006, 51,472 women had contributed biological samples (blood and urine) for the NIEHS portion of the study; 16,000 women were enrolled this fiscal year. These samples have been collected at 41 hospitals across Norway including urban centers and the Arctic regions. The enrollment target is 100,000 women. At the current recruitment rate we anticipate 80,000 women with blood and urine samples collected at 17 weeks gestation to evaluate for environmental exposures by the end of 2007. During the past fiscal year, we published a manuscript on issues related to urine storage and environmental exposures (phenols, phthalates, non-persistent pesticides) and have a manuscript in press regarding the biological specimen bank. In collaboration with Norwegian investigators, we are implementing a quality assurance experiment and protocol to assess storage conditions for a variety of parameter. Results of the initial quality assurance measures were presented at a professional meeting in October 2004 and a manuscript is currently in preparation.

Serum Organochlorine Levels and Primary Liver Cancer: a Nested Case-Control Study
(Grant: # R01ES014662 )  Publications
Engel, Lawrence S (engell@mskcc.org ) - Sloan-Kettering
Abstract: Although polychlorinated biphenyls (PCBs) and certain organochlorine pesticides induce liver tumors in animals, there is limited supporting epidemiologic data because of methodological challenges. There has also been little epidemiologic research into the risk associated with exposure to the low levels of aflatoxin, a known liver carcinogen, typically found in developed countries. We propose to investigate these relationships using already-collected serum and data from two large, population-based, prospective cohorts of men and women (total subjects =451,000) who provided blood samples and risk factor information between the mid- 1960s and mid-1970s. Follow-up and case ascertainment in these cohorts, one in Norway and one in Northern California, is very high and data on potential confounders are available. The aims of this study are to determine whether higher serum concentrations of organochlorines are related to increased risk of liver cancer; whether serum aflatoxin biomarkers are related to this risk; and if a synergy exists for these factors with each other or with HBV or HCV infection. We will include all cases of primary liver cancer diagnosed one or more years after blood draw (n=278) and a random sample of controls matched 3:1 to cases by cohort, age at blood draw, date of blood draw, sex, race, and county of residence (n =1,112 subjects total). We will measure serum concentrations of 39 PCB congeners, 11 organochlorine pesticides, and aflatoxin- albumin adducts, and assess HBV and HCV infection status using banked serum. We will also examine a second serum sample collected from a subset of subjects an average of 8 years after the primary sample. We will obtain information on potential confounders via linkage with previously collected health examination data (collected at blood draw), as well as clinical, census, and birth registry data available for each cohort. To our knowledge, this study would be the first to examine liver cancer risk in relation to body burden of PCBs. To be valid, such a study requires prospectively collected biospecimens because liver disease may affect the metabolism and subsequent elimination of organochlorines. The serum samples in this study were collected around the time that these organochlorines were banned in the U.S. and Norway and, thus, should reflect peak or near-peak levels of these chemicals in the general population of these countries. Although the production and use of these chemicals are banned in many countries, human exposure continues because of bioaccumulation in the food chain, ongoing use of organochlorine pesticides in less developed countries, leakage from equipment and landfills, and environmental persistence and redistribution. This study offers not only an efficient opportunity to examine the association of organochlorines and low-level aflatoxin exposure with risk of liver cancer, but also one of the few opportunities to study it in a methodologically rigorous way.

The Generation R Cohort Study as an NIEHS Resource
(Project: # ES101575 )
Longnecker, Matthew (longnec1@niehs.nih.gov) - NIEHS
Abstract: More epidemiologic data are needed on the relation of background-level exposure to man-made chemicals with short half-lives to child development. The embryonic and fetal stages of development are periods of heightened susceptibility to effects of xenobiotics. For xenobiotics with short half-lives, measurement has been a challenge in the past, but new technologies now support better exposure assessment. Multiple urine specimens are frequently the medium of choice for assessing exposure to such agents. Recent data suggest: a) that exposure of pregnant women to background-levels of bisphenol A results in increased risk of pregnancy loss, increased risk of obesity among offspring, and earlier age at menarche among female offspring, b) that background-level of exposure of pregnant women to nonpersistent pesticides can result in reduced birthweight and impaired neurodevelopment in offspring, and c) that background-level exposure to phthalates may be having adverse reproductive effects and increased risk of allergic diseases such as asthma. We seek to increase our capacity to study the relation of background-level exposure to chemicals with short half-lives to pregnancy outcomes and child development. To achieve this goal we supported collection of multiple urine specimens during pregnancy in the Generation R study (described below). The plan was to support collection of urine 3 times during pregnancy for the mothers of 2,500 children in the cohort. Generation R is an ongoing prospective study of 10,000 children who will be followed from early fetal life to young adulthood, and aims to study how factors and events during pregnancy and early childhood can affect growth, development, and health in later life. All pregnant women in Rotterdam who expect to give birth between June 2002 and June 2006 are invited to participate, together with their partner. This study has been set up by the Erasmus Medical Center. The multidisciplinary characterization of the cohort, starting in early pregnancy, will produce a database containing biological, medical, genetic, psychological and community-related data which can be used to address a wide spectrum of research questions. The research questions have been subdivided as growth and physical development, cognition and behavior, illnesses and accidents, and utilization of health care resources. The advantage to studying this cohort, e.g., over the Norway Mother and Child Cohort Study, is that we will have multiple urine specimens, including one during the first trimester, enhancing our ability assess exposures, especially during organogenesis. Furthermore, the outcome assessment in the Generation R cohort is more intensive and standardized than in Norway. In February of 2004, NIEHS support enabled an increase in the number of urine specimens collected from each pregnant woman from 1 to 3 (at 12, 20, and 30 weeks of gestation). As each pregnant woman presents for an ultrasound examination of her fetus, she provides a spot urine specimen that is divided into three 20ml aliquots and frozen at -20 degrees C in polypropylene containers. Two of these aliquots are reserved for collaborative studies with NIEHS. Although our primary interest was in women with a complete set of three urines (1/trimester, n=2,500 planned), for logistic reasons all women presenting for ultrasound provided a urine specimen that was stored. Last year's progress: Enrollment is now complete. We have a complete set of 3 urines for 2,025 women, 2 urines for 970 women, and 1 urine for 356 women. We had 100 third trimester urine specimens sent to a laboratory to check the levels of exposure in this population. The laboratory results have now been obtained. In general, the levels of phthalates, nonpersistent pesticides, and bisphenol A are similiar to those reported in other populations in developed countries. The levels of two antiandrogenic phthalates, MEHP and MnBP, were higher than in Swan's study where reduced anogenital distance was associated with phthalate exposure. The levels of total dimethyl alkyl metabolites (of organophosphate pesticides) were higher than in the Chamacos study (Salinas, California); total diethyl metabolites were similar. Levels of bisphenol A were about the same as in Calafat's NHANES report from 2005. Thus, examinations of health outcomes in relation to these contaminants may be worthwhile. Future studies will depend on the success of follow-up of subjects in the Generation R study, which outcomes are ascertained in the cohort, and future reports of health effects by other investigators. A report on the levels of contaminants is being prepared.

The Norwegian Mother and Child Study: Environmental Spec
(Project: # ES044008 )
Hoppin, Jane (hoppin1@niehs.nih.gov) - NIEHS
Abstract: The Norwegian Mother and Child study is a cohort study of pregnant women and their children in Norway. NIEHS is contributing to this effort by creating a biological specimen repository to evaluate environmental exposures and health effects to pregnant women and their unborn children. As of July 31, 2005, 35,082 women had contributed biological samples (blood and urine) for the NIEHS portion of the study; 13,000 women were enrolled this fiscal year. These samples have been collected at 41 hospitals across Norway including urban centers and the Arctic regions. The enrollment target is 100,000 women. At the current recruitment rate we anticipate 40,000 women with blood and urine samples collected at 17 weeks gestation to evaluate for environmental exposures by the end of 2005. During the past fiscal year, we completed a manuscript on issues related to urine storage and environmental exposures (phenols, phthalates, non-persistent pesticides). In collaboration with Norwegian investigators, we are implementing a quality assurance experiment and protocol to assess storage conditions for a variety of parameters. This will take approximately 2 years to complete. Results of the initial quality assurance measures were presented at a professional meeting in October 2004 and a manuscript is currently in preparation. We are currently designing a study to use the NIEHS samples and link to the questionnaire data from the mother and the Medical Birth Registry data to assess health effects of a common environmental contaminant, probably phthalates.