Polymorphisms in genes within the IGF-axis influence antenatal and postnatal growth.

Two pregnancy cohorts were used to investigate the association between single-nucleotide polymorphisms (SNPs) in genes within the insulin-like growth factor (IGF)-axis and antenatal and postnatal growth from birth to adolescence. Longitudinal analyses were conducted in the Raine pregnancy cohort (n = 1162) using repeated measures of fetal head circumference (HC), abdominal circumference (AC) and femur length (FL) from 18 to 38 weeks gestation and eight measures of postnatal height and weight (1-17 years). Replications of significant associations up to birth were undertaken in the Generation R Study (n = 2642). Of the SNPs within the IGF-axis genes, 40% (n = 58) were associated with measures of antenatal growth (P ⩽ 0.05). The majority of these SNPs were in receptors; IGF-1R (23%; n = 34) and IGF-2R (13%; n = 9). Fifteen SNPs were associated with antenatal growth (either AC or HC or FL) in Raine (P ⩽ 0.005): five of which remained significant after adjusting for multiple testing. Four of these replicated in Generation R. Associations were identified between 38% (n = 55) of the IGF-axis SNPs and postnatal height and weight; 21% in IGF-1R (n = 31) and 9% in IGF-2R (n = 13). Twenty-six SNPs were significantly associated with both antenatal and postnatal growth; 17 with discordant effects and nine with concordant effects. Genetic variants in the IGF-axis appear to play a significant role in antenatal and postnatal growth. Further replication and new analytic methods are required in order to better understand this key metabolic pathway integrating biologic knowledge about the interaction between IGF-axis components.

Associations between aggressive behaviour scores and cardiovascular risk factors in childhood.

OBJECTIVE: To examine the influence of aggressive behaviour scores on cardiovascular disease (CVD) risk factors throughout childhood. METHODS: This study utilized cross-sectional and longitudinal data from the Western Australian Pregnancy Cohort (Raine) Study (n = 2900). Aggressive behaviour scores were derived from the Child Behavior Checklist/4-18(CBCL), Youth Self-Report/11-18 (YSR) and Teacher Report Form/6-18 (TRF). CVD risk factors included body mass index (BMI), blood pressure, fasting lipids and homeostasis model of insulin resistance (HOMA-IR). RESULTS: Girls with higher aggressive behaviour scores had higher BMI from 10 years of age (P ≤ 0.001), higher BMI trajectories throughout childhood (P = 0.0003) and at 14 years higher HOMA-IR (P = 0.008). At the 14-year survey, this equated to a difference of 1.7 kg/m2 in the predicted BMI between the extreme CBCL scores in girls (top 5% (CBCL ≥ 17) vs. CBCL score = 0). Boys with higher aggressive behaviour scores had higher BMI at 5 years (P = 0.002), lower diastolic pressure at 14 years (P = 0.002) and lower systolic blood pressure trajectories throughout childhood (P = 0.016). CONCLUSION: Aggressive behaviour influences BMI from early childhood in girls but not boys. If this association is causal, childhood offers the opportunity for early behavioural intervention for obesity prevention.

Fat mass and obesity-associated obesity-risk genotype is associated with lower foetal growth: an effect that is reversed in the offspring of smoking mothers.

Fat mass and obesity-associated (FTO) gene variants are associated with childhood and adult obesity; however, the influence of FTO polymorphisms on foetal growth is unknown. Associations between the FTO variant rs9939609 and the foetal growth trajectories, maternal pregnancy weight gain, anthropometric measures at birth and body mass index (BMI) at age 14 years were assessed in 1079 singleton-birth Australian Caucasians. Analyses were repeated in 3512 singleton-birth Dutch Caucasians. The rs9939609 obesity-risk AA genotype was associated with symmetrical intrauterine growth restriction; an effect reversed in mothers who smoked during pregnancy. The effect increased over time and was modified by maternal smoking for head circumference (P = 0.007), abdominal circumference (P = 0.007), femur length (P = 0.02) and estimated foetal weight (P = 0.001). The modification of the association between the AA genotype and birth anthropometrics by maternal smoking was consistent across birth weight (P = 0.01) and birth length (P = 0.04) and neonatal day 2 anthropometry. Consistent associations were replicated in the Generation R cohort. Maternal pregnancy weight gain matched the pattern of birth weight and was independent of placental weight. In adolescents, the AA genotype was associated with increased BMI-adjusted-for-age in males (P = 0.00009), but no effect was detected in females. A variant in the FTO gene influences foetal growth trajectories in the third trimester, early postnatal growth and adiposity in adolescence. Maternal smoking during pregnancy reversed the direction of association of rs9939609 on foetal growth, which was probably mediated by maternal energy intake. The detection of genetic variants associated with foetal growth has the potential to identify novel molecular mechanisms underlying growth and targeted early life intervention.

AN EM COMPOSITE LIKELIHOOD APPROACH FOR MULTISTAGE SAMPLING OF FAMILY DATA.

Multistage sampling of family data is a common design in the field of genetic epidemiology, but appropriate methodologies for analyzing data collected under this design are still lacking. We propose here a statistical approach based on the composite likelihood framework. The composite likelihood is a weighted product of individual likelihoods corresponding to the sampling strata, where the weights are the inverse sampling probabilities of the families in each stratum. Our approach is developed for time-to-event data and can handle missing genetic covariates by using an Expectation-Maximization algorithm. A robust variance estimator is employed to account for the dependence of individuals within families. Our simulation studies have demonstrated the good properties of our approach in terms of consistency and efficiency of the genetic relative risk estimate in the presence of missing genotypes and under different multistage sampling designs. Finally, an application to a familial study of early-onset breast cancer shows the interest of our approach. While it confirms the important effect of the genes BRCA1 and BRCA2 in these families, it also shows that incorrect inference can be made about this effect if the sampling design is not properly taken into account.

Segregation analysis of prostate cancer in France: evidence for autosomal dominant inheritance and residual brother-brother dependence.

Four segregation analyses concerning prostate cancer (CaP), three conducted in the United States and one in Northern Europe, have shown evidence for a dominant major gene but with different parameter estimates. A recent segregation analysis of Australian pedigrees has found a better fit of a two-locus model than single-locus models. This model included a dominantly inherited increased risk that was greater at younger ages and a recessively inherited or X-linked increased risk that was greater at older ages. Recent linkage analyses have led to the detection of at least 8 CaP predisposing genes, suggesting a complex inheritance and genetic heterogeneity. To assess the nature of familial aggregation of prostate cancer in France, segregation analysis was conducted in 691 families ascertained through 691 CaP patients, recruited from three French hospitals and unselected with respect to age at diagnosis, clinical stage or family history. This mode of family inclusion, without any particular selection of the probands, is unique, as probands from all previous analyses were selected according to various criteria. Segregation analysis was carried out using the logistic hazard regressive model, as incorporated in the REGRESS program, which can accommodate a major gene effect, residual familial dependences of any origin (genetic and/or environmental), and covariates, while including survival analysis concepts. Segregation analysis showed evidence for the segregation of an autosomal dominant gene (allele frequency of 0.03%) with an additional brother-brother dependence. The estimated cumulative risks of prostate cancer by age 85 years, among subjects with the at-risk genotype, were 86% in the fathers' generation and 99% in the probands' generation. This study supports the model of Mendelian transmission of a rare autosomal dominant gene with high penetrance, and demonstrates that additional genetic and/or common sibling environmental factors are involved to account for the familial clustering of CaP.

Regressive threshold model for familial analysis of complex diseases with variable age of onset.

Few models for segregation (or combined segregation-linkage) analysis have been developed to account for variable age of onset. The unified model (UM) can only take into account age at examination. In the logistic hazard model (LHM), Abel and Bonney ([1990] Genet. Epidemiol. 7:391-407) incorporated survival analysis concepts into the regressive logistic model of Bonney ([1986] Am. J. Med. Genet. 18:731-749), but interpretation of familial dependence parameters is difficult. In this article, we extended the regressive threshold model (RTM) proposed by Demenais ([1991] Am. J. Hum. Genet. 49:773-785) to account for a variable age of onset of complex diseases. This model assumes an underlying liability to disease and is more general than the original logistic formulation, since the phenotypes of each individual's antecedents can be adjusted for their own genotypes and covariate effects. The variation of risk with age can be expressed as a general step function, and variants of the model have been proposed by imposing different types of constraints among the time-dependent thresholds. The performances of the three models (UM, LHM, and RTM) were compared in the context of segregation analysis of family data generated with variable age of onset. All analysis models were robust with respect to false conclusion of a major gene, and the best results were obtained under RTM. The power to detect the major gene was higher under LHM than RTM, but the best fit of the estimated cumulative age-dependent penetrance with respect to the true value was obtained under RTM. This new model may thus prove helpful in contributing to identification of genes underlying complex diseases, since it can easily include linked marker loci and linkage disequilibrium.

[Evaluation of a cervical cancer screening campaign: reflections on the experience in Martinique]. / Dispositifs d'évaluation d'une campagne de dépistage du cancer du col de l'utérus: réflexions à partir de l'expérience martiniquaise. Comité de Pilotage de la Campagne de dépistage.

A pilot cervix cancer screening is organized in Martinique since 1991, as part of the programmes of the Fonds National de Prévention d'Education et d'Information Sanitaire of CNAMTS. Evaluation of the programme is conducted by the cancer registry, which includes the measurement of the impact, quality of the Pap smear test, quality of action and efficiency of the programme. Evaluation is a key part that allows to manage the screening programme. According to the results of the organized screening in Martinique, modalities of the evaluation will be discussed with respect to the pilot programme organization, involvement of participants, indicators to collect and the organism in charge of the evaluation.

Patterns of familial aggregation of three melanoma risk factors: great number of naevi, light phototype and high degree of sun exposure.

BACKGROUND: Besides melanoma susceptibility genes and shared environmental exposures, part of the familial clustering of cutaneous malignant melanoma (CMM) might be due to familial aggregation of melanoma-associated phenotypes. Our goal was to assess the patterns of familial aggregation of three melanoma risk factors: great number of naevi (GNN), light phototype (LP) and high degree of sun exposure (HDSE). METHODS: Familial aggregation of GNN, LP and HDSE was investigated in 66 French families with at least two CMM cases and was measured by the asssociation of the relatives' traits with the probands' traits, using the generalized estimating equations approach. The probands were the melanoma cases leading to ascertainment of the families, subdivided into cases (with the trait studied) and controls (without the trait). RESULTS: We found significant evidence for familial aggregation of GNN only among sibs (OR = 3.7, 95% CI : 1.4-10.5, P = 0.01), of LP among blood relatives (OR = 3.8, 95% CI : 1.8-8.0, P = 0.004) and of HDSE among blood relatives (OR = 4.5, 95% CI : 2.1-9.9, P < 0.001) and spouses (OR = 44.3, 95% CI : 5.1-382.2, P < 10(-3)). These results suggest that genetic factors might account for the clustering of GNN and LP and shared environment for the aggregation of HDSE. The GNN clustering was lower in families with increasing numbers of CMM (>/=3 cases) or presence of p16 mutations, the opposite being observed for LP and HDSE. Moreover, the familial aggregation of LP was significantly lower in families with highly sun-exposed members. CONCLUSION: Melanoma might not only result from specific genetic and environmental factors but also from those underlying melanoma-associated traits involving complex gene-gene and gene-environment interactions.

[The PROGENE study, the French project of genetic analysis of familial prostatic cancer: recruitment and analysis]. / Etude PROGENE, project français d'analyse génétique du cancer de la prostate familial: recrutement et analyse.

OBJECTIVES: To initiate a genetic linkage study in order to localize one or several predisposition gene(s) for hereditary prostatic cancer (PC), as various epidemiological studies have demonstrated a possible family aggregation in about 25% of cases. A family segregation study [14] has also shown that a genetic predisposition, with autosomal dominant transmission and high penetrance (88% at 85 years) could be responsible for 9% of all PC. METHODS: A national collection of families with at least 2 cases of PC allowed: 1) identification of families with hereditary forms of PC, 2) creation of a constitutional DNA bank after collecting blood samples from subjects belonging to these families, and 3) a simulation study of genetic linkage analysis prior to microsatellite genotyping. RESULTS: From July 1994 to September 1995, we included 67 families (180 cases of PC). Another 45 families are currently being included. 24 of these 67 families (89 PC, 54 survivors) satisfied at least one of the criteria defined in the study by CARTER et al. for hereditary forms of familial PC. Two families were also included as the 3 patients with PC were second degree relatives. A total of 26 families therefore presented a hereditary form, 18 of which (73 PC, 46 survivors) were considered to be informative for a genetic linkage study (lod score = 4 for theta = 0.001 with an 8 allele marker). The constitutional DNA of 271 individuals of these informative families was extracted from circulating cells obtained from blood samples, immortalized lymphocytes, and the genotyping was initiated for 216 microsatellite markers distributed throughout the genome, an average of every 20 cM. CONCLUSION: Although the recruitment allowed us to identify many informative families for an inherited risk of PC, the predictive study suggested a high probability for localization of a predisposition gene by genetic linkage analysis. It would therefore be possible to identify, within the families concerned, the subjects carrying the genetic anomaly and consequently at high-risk of PC. Finally, the demonstration of the locus would allow cloning and identification of the gene (s) involved.

Genetic and epidemiological risk factors for a malignant melanoma-predisposing phenotype: the great number of nevi.

Multiple genetic and epidemiological factors are involved in the etiology of cutaneous malignant melanoma. The phenotype defined by a great number of nevi (GNN) is consistently reported as being a major risk factor for melanoma and is likely to be under genetic control. As part of a large genetic and epidemiologic survey of melanoma in France, our goal was to understand the pattern of inheritance of GNN in 295 nuclear families ascertained through 295 melanoma probands. The GNN phenotype was defined as having 50 or more nevi vs. less than 50 nevi. Four percent of the 295 families included at least 2 GNN cases. We found that sun exposure during holidays and propensity to sunburns were significantly associated with GNN, independent of the occurrence of melanoma. Four segregation analyses of GNN were conducted, using the class D regressive logistic model, which differed according to the sampling procedure chosen (with and without the melanoma probands) and the inclusion or not of the significant risk factors. Three of these analyses led to the detection of a recessive-like major factor, which did not fit a Mendelian pattern of inheritance in two of them. Our results are discussed with respect to the low familial clustering of GNN, the type of analysis conducted, the characteristics of the model used, and the complexity of the mechanisms underlying the GNN phenotype.