White matter hyperintensities and gray matter lesions in physically healthy depressed subjects.

OBJECTIVE: Previous studies reported that depressed subjects had more white matter hyperintensities on magnetic resonance imaging scans than control subjects, but the subjects had cerebrovascular disease risk factors. This study used subjects with a history of recurrent major depression and matched comparison subjects, screened to exclude cerebrovascular disease risk factors, to determine whether depressed subjects had more white matter hyperintensities and other lesions. METHOD: A semiautomated volumetric computer program was used to compare numbers and volumes of white matter hyperintensities, basal ganglia lesions, and total lesions in 24 women with a history of recurrent major depression and 24 comparison subjects case-matched on age and education and group-matched on height. In addition, images were measured with the use of a validated categorical scale. All subjects were screened to exclude cerebrovascular disease risk factors. RESULTS: There were no significant differences in the total volumes or total numbers of lesions. However, multiple linear regression showed a significant correlation of age and depression with number of lesions; this was accounted for by a greater number of small lesions (diameter < or = 0.4 cm). CONCLUSIONS: These findings suggest that cerebrovascular disease risk factors most likely mediated the relationship between depression and white matter hyperintensities seen in previous studies. However, the independent effect of depression, as well as an age-by-depression interaction, for small lesions suggests a causal role of depression in certain types of white matter pathology irrespective of other cerebrovascular disease risk factors. The volumetric method used in this study may be more sensitive than other methods in determining lesion characteristics and correlations with clinical variables.

Familiality and heritability of subtypes of attention deficit hyperactivity disorder in a population sample of adolescent female twins.

OBJECTIVE: Attention deficit hyperactivity disorder (ADHD) is a highly heritable but clinically heterogeneous syndrome. The study examined the familiality and heritability of ADHD subtypes as defined by DSM-IV and by latent-class analysis in a population sample of adolescent female twins. METHOD: To determine which elements of ADHD cluster together, latent-class analysis was applied to data obtained from parents on the 18 DSM-IV ADHD symptoms in 4,036 female twins age 13-23 years in a population sample identified from the registry of all births in Missouri for the years 1968-1996. Relative risk and odds ratios were used to assess within-subtype and between-subtype familiality and heritability of both DSM-IV and latent-class ADHD subtypes. RESULTS: Latent-class analysis was most compatible with the existence of three mild and three severe classes of ADHD symptoms in the general population. The three severe classes showed moderate overlap with DSM-IV ADHD subtypes. The primarily inattentive and combined subtypes of DSM-IV ADHD co-clustered within families. The primarily hyperactive/impulsive DSM-IV subtype and the individual latent-class analysis subtypes did not co-cluster. Subtypes defined by both approaches were highly heritable. CONCLUSIONS: Unlike DSM-IV subtypes of ADHD, latent-class ADHD subtypes appear to be independently transmitted in families. These classes may be more appropriate targets for molecular genetic studies of ADHD.

Genetic heterogeneity in familial hypobetalipoproteinemia: linkage and non-linkage to the apoB gene in Caucasian families.

Familial hypobetalipoproteinemia (FHBL) is an autosomal dominant disorder of lipid metabolism characterized by extremely low plasma levels of apolipoprotein B (apoB), and total-, and low-density lipoprotein (LDL) cholesterol. Various truncated forms of apoB have been found to cosegregate with the FHBL phenotype in more than 30 kindreds. By contrast, no truncated forms of apoB protein were detected with sensitive immunoblotting in the plasmas of any of the 6 kindreds reported here. Individuals with apoB levels in the 5th centile for their age and sex were considered as affected with FHBL. Linkage analysis was performed using 3 microsatellite markers flanking the apoB gene (D2S131, D2S149, and D2S144), a 3' variable number of tandem repeats (VNTR) marker and one intragenic marker. Two-point linkage of FHBL was established to the 3' VNTR marker with a combined maximum LOD score of 8.5 at theta = 0 for 5 of the 6 families. Maximum LOD scores for flanking microsatellite markers were 5.0, 2.4, 1.3, 1.2 and 2.1 for these kindreds (D, T, De, C and Z, respectively). A test of homogeneity differentiated the 6th family (F kindred) from the other five. LOD scores of -25.2 at the 3' VNTR and -7.8 at the intragenic apoB/Xbal marker at theta = 0 excluded linkage to the apoB gene in the F kindred. These kindreds demonstrate the heterogeneity of FHBL and also offer the possibility to investigate as yet undescribed mutations of apoB, resulting in alterations of apoB metabolism. The F kindred may shed light on novel gene(s) contributing to the low apoB-phenotype.

Lack of association of dopamine D4 receptor gene polymorphisms with ADHD subtypes in a population sample of twins.

Attention-deficit hyperactivity disorder (ADHD) is a common, highly heritable syndrome of childhood characterized by problems with inattention, hyperactivity, and impulsivity. A variety of case control and family-based transmission distortion genetic studies of ADHD have focused on the possible involvement of polymorphisms of the DRD4 receptor gene. The majority of studies have examined the association of variously defined ADHD with an exon 3 polymorphism containing a variable number of imperfect 48 base pair repeats. Recently, McCracken et al. [2000: Mol Psych 5:531-536] reported an association of the DSM-IV primarily inattentive ADHD subtype with a 5' 120 base pair repeat polymorphism in the DRD4 gene. In this report, we test for the possible association of these two polymorphisms with population-derived samples of DSM-IV ADHD subtypes. Furthermore, we extend previous studies by testing for associations with ADHD subtypes derived from latent-class analysis of interview responses. In contrast to most, but not all, previous studies, we failed to demonstrate any significant association of the exon 3 7-repeat allele with ADHD. Nor did we replicate the association of the 5'120 base pair repeat polymorphism. We do find a significant association of the exon 3 3-repeat allele with a novel talkative/impulsive latent-class-defined subtype of ADHD.

No association of the dopamine transporter gene 3' VNTR polymorphism with ADHD subtypes in a population sample of twins.

Dopamine pathway genes have been the subject of a variety of studies testing the association of candidate genes and liability for attention-deficit hyperactivity disorder (ADHD). Due to the known effects of stimulant medications such as methylphenidate on the dopamine transporter, a variety of case control and family-based transmission distortion genetic studies of ADHD have focused on DAT1 polymorphisms. The most widely reported positive finding has been with a variable number of tandem repeats (VNTR) polymorphism of unknown function in the 3' untranslated region of the DAT1 gene. In this report, we test for association of alleles of this polymorphism with ADHD using population-derived samples of twins. We use the transmission disequilibrium test and ADHD subtypes defined by both DSM-IV and latent class criteria. We fail to demonstrate any significant association or trend for association of any of the VNTR alleles with any of the variously defined ADHD subtypes.

Association between DQB1 and cervical cancer in patients with human papillomavirus and family controls.

OBJECTIVE: The role of human leukocyte antigen (HLA) DQB1 alleles and human papillomavirus (HPV) as contributing factors to invasive cervical cancer was investigated. To overcome problems of misleading causal inferences common in traditional case-control studies, a family-based test, the transmission/disequilibrium test, was used. METHODS: Ninety-six patients with pathologically confirmed invasive cervical cancer were ascertained. Human papillomavirus types were determined in 80 patients, of whom 81.25% were HPV-positive, and 18.75% were HPV-negative. Deoxyribonucleic acid was extracted from samples, taken from patients and their parents, and sequenced to determine DQB1 genotypes. Nuclear family data were used to test whether the DQB1 locus is associated with invasive cervical cancer while controlling for high-risk HPV-positive patients. The transmission/disequilibrium test evaluates whether the frequency of transmission of parental marker alleles to their affected offspring deviates from the expected Mendelian frequency of 50%. RESULTS: The HLA DQB1 locus showed evidence for allelic association with invasive cervical cancer in high-risk HPV-positive patients (P = .006). The transmission/disequilibrium test showed that the DQB1*0303 allele was transmitted to high-risk HPV patients more often than expected by chance, chi2(1) = 8.0, P = .005 (P = .035 when correcting for multiple tests). Tests of association were negative when applied to all 96 patients, irrespective of HPV status. No significant differences were found in the distribution of the DQB1 alleles among HPV-positive patients compared with those who were HPV-negative, indicating that HLA alleles are not associated with susceptibility to HPV infection. CONCLUSION: These results suggest that the DQB1*0303 allele increases the risk for invasive cervical cancer in women who are HPV-positive.

Cancer among first-degree relatives of probands with invasive and borderline ovarian cancer.

OBJECTIVE: The familial clustering of ovarian, breast, endometrial, colon; and prostate cancer was compared in first-degree relatives of probands with invasive and borderline ovarian cancer to determine coaggregation. METHODS: Probands (n=392), who had been patients in the Division of Gynecologic Oncology at Washington University, were ascertained consecutively. Family history on 2192 first-degree relatives was collected by personal interviews of the probands and other family members. Estimates of prevalence of cancers in first-degree relatives of the two proband groups were compared. Survival analysis was used to examine the age-at-onset distribution of each cancer in relatives of invasive probands versus relatives of borderline probands. RESULTS: Among the relatives were 24 cases of ovarian cancer, 46 cases of breast cancer, 13 cases of endometrial cancer, and 25 and 28 cases of colon and prostate cancer, respectively. There were no significant differences in the prevalence of any of these cancers in relatives of the invasive and borderline probands. Cumulative lifetime risk estimates did not differ between the relatives of the two groups for any cancers. Age-at-onset of ovarian cancer did not differ between probands with positive family histories of the five cancers and those with negative histories. The inability to reject the null hypothesis of no differences in the first-degree relatives of our two study groups might be from insufficient power to detect small differences, given our sample size. CONCLUSION: These results suggest that relatives of patients with invasive and borderline ovarian cancer might share similar cancer risks and age-at-onset distributions.

Increased prevalence and earlier onset of mood disorders among relatives of prepubertal versus adult probands.

OBJECTIVE: To compare the familial clustering of affective disorders among first-degree relatives of prepubertal versus adult probands with mood disorders. METHOD: The Family History-Research Diagnostic Criteria (FH-RDC) assessment instrument was used to obtain data on first-degree relatives of all probands. Logistic regression was used to assess the strength of the age of the proband to predict FH-RDC diagnoses in relatives. Survival analysis was used to examine the age-at-onset distribution of first FH-RDC diagnosis while controlling for year of birth. RESULTS: The prevalence of major affective disorders was more than two times higher among first-degree relatives of child probands versus those of adult probands even when controlling for birth cohort. Cumulative risk for lifetime major mood disorders in first-degree relatives of child probands was significantly higher than for those of adult probands. CONCLUSION: These analyses further support that ascertainment of families through affected children identifies pedigrees with a higher proportion of affected relatives than ascertainment through affected adults.

Factor and latent class analysis of DSM-IVADHD symptoms in a school sample of Brazilian adolescents.

OBJECTIVE: To evaluate the validity of the multidimensional construct proposed by DSM-IV for the diagnosis of attention-deficit/hyperactivity disorder (ADHD) in a school sample of young Brazilian adolescents. METHOD: An instrument including all 18 DSM-IVADHD symptoms was administered to 1,013 students aged 12 to 14 years at 64 state schools by trained research assistants. Each symptom was rated on a Likert scale with five levels of severity (never, almost never, sometimes, frequently, and always). RESULTS: Using an exploratory factor analytic approach (principal components analysis), two factors were extracted. Factor I (hyperactivity-impulsivity) comprised eight DSM-IV hyperactive-impulsive symptoms with loadings > or =0.40. Factor II (inattention) included also eight DSM-IV symptoms of inattention. The two factors explained 34% of the total variance and had an interfactor correlation of 0.45. Latent class analysis demonstrated similar classes in males and females, but class structures were markedly different from previous analyses of parent report data. CONCLUSION: The findings support the appropriateness of the multidimensional construct introduced by DSM-IV in the diagnosis of ADHD in a different culture but emphasize the possible impact of different reporters on the results of structural model-testing.

Latent class and factor analysis of DSM-IV ADHD: a twin study of female adolescents.

OBJECTIVE: In an attempt to validate the current DSM-IV criteria for attention-deficit/hyperactivity disorder (ADHD) in females and to determine whether symptoms are continuously distributed or categorically discrete, the authors performed factor and latent class analysis on ADHD symptom data from a large general population of adolescent female twins (1,629 pairs). METHOD: A structured diagnostic assessment of DSM-IV ADHD was completed with at least one parent of 1,629 pairs by telephone. ADHD symptoms from 1,549 pairs were subjected to latent class and factor analysis. RESULTS: Latent class and factor analyses were consistent with the presence of separate continuous domains of inattention (ATT), hyperactivity-impulsivity (H-I), and combined ATT with H-I problems. Severe latent classes corresponding to the predominantly inattentive, predominantly hyperactive-impulsive, and combined types were identified with lifetime prevalence estimates of 4.0%, 2.2%, and 3.7%, respectively. Membership in the severe ATT class predicted academic problems, family problems, and referral to health care providers. Membership in the H-I and combined classes also predicted impaired social relationships. CONCLUSIONS: These results suggest that DSM-IV ADHD subtypes can be thought of as existing on separate continua of inattention, hyperactivity-impulsivity, and combined type problems. Membership in any of there severe ADHD latent classes did not preclude academic excellence, but it was associated with different types of impairment and health care-seeking behavior. These data have implications in the areas of diagnosis, classification, treatment, and research.

Evaluation of ADHD typology in three contrasting samples: a latent class approach.

OBJECTIVE: To identify subtypes of attention-deficit/hyperactivity disorder (ADHD) and characterize them as either categorical or continuous; to investigate familial resemblance for ADHD among sibling pairs; and to test the robustness of all results by using contrasting data sets. METHOD: Latent class analysis was applied to the ADHD symptom profiles obtained from parents or best informant about their offspring in 3 samples: a population-based set of female adolescent twins (724 monozygotic pairs, 594 dizygotic pairs) and male (N = 425) and female (N = 430) child and adolescent offspring ascertained from high-risk alcoholic families. RESULTS: Latent class analysis revealed 2 categories of clinically significant ADHD which were replicated in all 3 study groups: a subtype with high endorsements of ADHD inattention symptoms and a second combined type with high endorsements of both inattention and hyperactivity-impulsivity items. Both appeared to be continuous across all 3 data groups. The high-risk families contained a class in which members heavily endorsed the ADHD "fidget" item but not other ADHD items. A large proportion of the monozygotic sibs (80%) versus a smaller proportion of dizygotic sibs (52%) were assigned to the same latent class. Among the high-risk children and adolescents, 51% of the female and 41% of the male siblings were concordant for class membership. CONCLUSIONS: The pattern of latent classes suggested that ADHD consists of an inattentive and a combined subtype, within each of which lies a dimensional domain. These analyses further support that genetic factors are significant determinants of latent class membership.

Note on linkage analysis when the mode of transmission is unknown.

A major difficulty in a linkage analysis arises from the necessity of specifying the mode of inheritance prior to analysis. For a complex disease, such as those encountered in psychiatric illnesses, the mode of inheritance is generally not known in advance. Consequently, some estimation procedure is often combined with linkage analysis to circumvent this. We discuss several precautions that should be taken when using traditional statistical testing methods: correction of the likelihood for the method of sampling families and the computation of the lod score. We analyze simulated data with pedigrees selected under a sampling scheme approximating single ascertainment. In this situation, the severity of the above problems is attenuated.

A bit about haplotypes: using binary digits to code tightly linked loci.

The advent of recombinant DNA techniques has resulted in the detection of a large number of polymorphic marker loci many of which are not useful for linkage studies because of their low degree of polymorphism. However, when no apparent recombination exists between several closely linked markers, the amount of 'information' available can be significantly increased by establishing haplotypes from those loci; that is, haplotyping increases the number of heterozygotes at marker loci. Haplotyping can be problematic when more than one of the loci involved in the haplotyping are heterozygous and the phase of the haplotypes cannot be inferred from the data. We present a method for recoding the phenotypic marker data for pedigree members that will circumvent this difficulty.

Two-locus models of disease.

Most complex diseases have not been amenable to genetic analysis under the assumption of single locus or multifactorial models. Consequently, interest has turned to the consideration of the properties of oligogenic models. i.e., genetic models involving a small number of genes. Nine two-locus models of disease, representing both epistatic and heterogeneous genetic models, are investigated: three models of heterogeneity and six models of epistatis. For each model we derive formulas for the recurrence risk to various classes of relatives in terms of penetrances and gene frequencies. We also develop formulas for the components of variance for the epistatic models in terms of the same genetic parameters. The range of penetrances and the associated gene frequencies that predict a predetermined value for the population prevalence and recurrence risk to the sibling of proband are calculated for various rates of the prevalence and risk to sibs. It is found that for many of these genetic models, there is a very limited range of penetrances that fit a particular set of assumed risks. Estimated population prevalence and risks to sibs and monozygotic twins for bipolar and schizophrenia illness are used to test for compatibility with expected values for recurrence risks under these models.

Testing the utility of mod scores and sib-pair analysis to detect presence of disease susceptibility loci.

Linkage analyses and association studies were employed to detect disease susceptibility loci leading to elevated Q1 levels in Problem 2B. Phenotypes were defined to be the dichotomous affection status, the quantitative value for Q1, and Q1 adjusted for covariates. The method of mod-scores (for the dichotomous phenotype) and the Haseman-Elston sib-pair test on the dichotomous and quantitative phenotypes were used to screen for linkage of disease susceptibility genes to 367 markers. These analyses were performed on a sample ascertained from the first 60 replicates. The mod-score method detected linkage to MG1, MG2, and MG3 with scores of 1.5, 5.0, and 1.6 respectively. Sib-pair analysis using quantitative phenotypes signaled linkage only to the area surrounding MG1; the dichotomous phenotype detected linkage only to MG2. Association studies used ANOVA on all founders in the first 60 replicates and ASSOC on the ascertained families and on a subset of families from the 60 replicates but only confirmed an association to MG1. In conclusion, the mod-score method may be a useful tool for genomic screens.

Covariates in linkage analysis using sibling and cousin pairs.

Using simulated data from GAW 12, problem 2, we further develop a novel technique to detect and use significant covariates in linkage analysis. The method, first introduced by Rice et al. [Genet Epidemiol 17(Suppl. 1):S691-5, 1999], uses logistic regression to model perturbation in sharing as a function of covariate levels. The original method allows use of all sib pairs (concordant affected, concordant unaffected, and discordant). Here we extend this method to include cousin pairs in analysis.

Clustering methods applied to allele sharing data.

Here we focus on using clustering methods to disentangle the interacting factors that lead to the presentation of complex diseases. Relative pairs are placed in discrete subgroups, or classes, based upon their pattern of allele sharing at a sequence of markers and on concomitant risk factors. The relationship between the locus information and the affectation status of the relative pairs within each subgroup then can be assessed. Cluster analysis (CLA) and latent class analysis (LCA) were applied to sibling allele sharing data from GAW11 simulated data, and to an existing Alzheimer's disease (AD) dataset. Both methods were able to identify markers linked to all 3 disease loci in the GAW11 data. LCA and CLA also replicated regions of chromosomes identified in an analysis of the AD data using affected-sib-pair methods. These analyses indicate that classification tools may be useful for detecting susceptibility genes for complex traits.

TDT with covariates and genomic screens with mod scores: their behavior on simulated data.

We describe an extension to the TDT (transmission/disequilibrium test) which allows for more than two marker alleles and for covariates measured on the parent or offspring. We also describe a systematic genomic search where the mod score (maximized lod score) is computed for each marker under constraints on the population prevalence or penetrances of a single locus.

Mutational analysis of the nicotinic acetylcholine receptor alpha 4 subunit gene in attention deficit/hyperactivity disorder: evidence for association of an intronic polymorphism with attention problems.

Recent studies suggest the presence of genetically distinct subtypes of attention deficit/hyperactivity disorder (ADHD) and that attention problems can be treated with receptor subtype selective nicotine agonists. In this study, individuals with two independent familial subtypes of ADHD defined by latent class analysis were systematically screened for sequence variations in the coding regions and intron/exon junctions of the nicotinic acetylcholine receptor alpha 4 subunit gene (CHRNA4). Common polymorphisms were used for transmission disequilibrium test (TDT) analyses. A significant association was found for a 5' intron 2 single nucleotide polymorphism and severe inattention problems (P = 0.007, effect size = 4, 95% CI 1.3-14.1). The location of the polymorphism is compatible with it affecting pre-mRNA stability or splicing.

Linkage analysis of a complex disease: application to familial Alzheimer's disease.

Evidence for linkage of the Alzheimer's gene to markers on chromosomes 19 and 21 was assessed using single-locus and two-locus models of inheritance. Families were divided into groups determined by their average age at onset. The youngest group produced higher lod scores for markers on chromosome 21 while an older group showed evidence for linkage to markers on chromosome 19. Two-locus models of disease were used to analyze the youngest group for linkage to pairs of markers on chromosome 21 and an older group with markers on chromosome 19.