Twin Study (CATSS - DOGSS - HORSES)

A. INTRODUCTION
Childhood neurodevelopmental problems (NDPs, i.e. attention deficits and hyperactivity (AD/HD), autism spectrum disorders (ASD), i.e. social interaction deficits, language impairments, and repetitive/stereotyped behaviors, tic disorder, motor dyscoordination (DCD), and learning disabilities (LD)) . The study cohort was initially assessed at their 9th and 12th birthdays (the CATSS study), and now NDP screen positive twins pairs and controlls are having clinical examinations when 15 years old (the DOGSS study). Large investments have been made in equipment and training of 30 clinicians and 8 site coordinators. Currently we have clinically examined 120 twin pairs. this stage of the longitudinal project, we aim to identify genetic variants and their interaction with environemntal susceptibility behind NPDs and their development into mental disorders, substance abuse and criminality. For this purpose, collaboration has been formally initiated with one of the leading molecular genetic laboratories in the world (Professors Thomas Bourgeron and Marion Leboyer, Pasteur Institute and University of Paris XII), creating a translational project encompassing in-depth clinical assessments, epidemiology, and molecular neuroscience.

B. BACKGROUND
NDP may be associated with an increased risk for later mental health problems, substance abuse, and criminality (referred to as “negative outcome” in this text). Problem constellations associated with NDPs affect at least 5 % of all children in severe forms (Kadesjo 2000), they persist in most cases into adult age and give rise to considerable functional and psychosocial impairments (Biederman 2006), including increased mortality (Jokela et al 2009). About a third of children with hyperactivity and inattention will develop conduct disorder (CD), a problem complex that in its turn leads to a very considerable risk for all types of adult mental disorders, substance abuse and aggressive criminality (Kim-Cohen et al 2003). It is less well established how social interaction and communication problems (defined on the autism spectrum), learning problems, and behavioral stereotypies are associated with mental disorders and the risk for substance abuse and criminality in adult age (Gilmour et al 2004).
Five main questions call for consideration in order to understand how NDPs influence later negative outcomes. (1) Childhood hyperactivity and inattention (AD/HD) overlap to a substantial degree with other problem types (e.g., autism spectrum disorders and learning disorders). Specific risk constellations for the different forms of adolescent and adult negative outcomes remain to be identified. (2) It is known that both NDPs and negative outcomes are over-represented among boys, but not whether there are gender-specific developmental trajectories. (3) Both genetic and environmental factors are known to influence NDPs and negative social outcomes. Using identical twins discordant for NDP (and thus controlling for genetic effects) provides unique opportunities to identify environmental risk factors empirically. (4) Genetic effects are of high importance for psychiatric disorders (Lichtenstein et al 2009) and recent advances have identified gene variants (both single nucleotide polymophisms, SNPs and structural changes on the chromosomes (copy number variants; CNVs), (Mefford et al 2008; ISC 2008). However, today we have only a very limited understanding of the genes contributing liability to psychiatric diseases in general and more specifically to NDPs. (5) Psychosocial adversities are known to affect adjustment in both childhood and adulthood, and may act in concert with NDPs to create an increased risk for negative outcomes. Gene by environment interactions are most certainly involved in this process (Rutter & Silberg 2002).
The study will also be a founding-stone for future follow-up studies of the cohort aiming at clarifying the role of NDPs for mental health problems surfacing in the late adolescence and early adulthood (e.g., mood disorders, schizophrenia, other psychoses) (Stahlberg et al. 2004), substance abuse disorders (Cloninger et al. 1981), severe aggressive antisocial personality disorders (Soderstrom et al. 2005), and their impact on working life, general health, and specific forms of crimes, such as violent or sexual crimes. The 18-years follow-up is currently being planned, and further waves of studies will be performed every three years well into adulthood.

C. OVERALL AIM
To identify genetic and environmental risk factors for the development of NPDs into negative outcomes.

D. RESEARCH QUESTIONS
1. Which specific aspects of NDPs are risk factors for negative outcomes?
2. Are there gender-specific patterns of associations between NDPs and negative outcomes?
3. Are there specific environmental risk factors for the development of NDPs into negative outcomes?
4. Are there specific genetic risk factors for the development of NDPs into negative outcomes?
5. Is the relationship between molecular genetic variants, NDPs and negative outcomes dependent on interactions with specific environmental factors?

E. RESEARCH DESIGN, METHODS, PROGRESS REPORT
Participants to this study are recruited from the cohort of all Swedish twins turning 15 years of age (n≈1400 pairs per year).
The base line CATSS-study (data already collected)
a. CATSS-telephone interview At age 9 and 12, all twins in Sweden are recruited to the ongoing Child and Adolescent Twin Study in Sweden (CATSS). Parents of these twins participate in a telephone interview developed for studying NDPs and other child psychiatric problems (the Autism – Tics, AD/HD and other Comorbidities inventory, A-TAC (Hansson et al. 2005); current response rate=80%; current n=8,000 interviews). The A-TAC assesses both categorical diagnoses according to the DSM-IV and patterns of dimensional measures of symptoms.
b. CATSS-questionnaire Pairs in which at least one of the twins screened positive for autism, ADHD, oppositional defiant disorder, conduct disorder, tic disorder, obsessive compulsive disorder or eating disorder (8% of the children in 14% of the twin pairs) and 5% control twin pairs have been followed-up with a questionnaire, which measures different aspects of the environment.
The ongoing DOGSS-study
a. Questionnaires to all families in Sweden with 15 year old twins All families with twins turning 15 years during the years 2008-2011 are given questionnaires to measure beginning negative outcomes in the adolescent twins, including parent and self-report questionnaires on emotional and behavioural problems among children (The Strengths and Difficulties Questionnaire (SDQ); a 25-item behavioural screening (Goodman,2000), as well as instruments measuring prodromal states of schizophrenia, bipolar disorder, substance abuse, criminality, and psychopathic personality traits (e.g., Loewy et al 2005; Andershed et al.2002; Luby et al 1999, Junger-Tas et al. 1994). Questionnaires also include measures of micro-social environments, focusing on parenting and parent-child relationships (e.g., inconsistency, harshness, and coercion; Hetherington & Clingenpeel 1992), scales measuring child monitoring and marital processes. Psychosocial and contextual environments, such as stressful events, SES, and school environment are measured in both the parental and child questionnaires.
b. Clinical examinations Families who were screen-positive or controls at age 12 (or 9 for those twins born after June 1995) in the CATSS (thus also included in the CATSS questionnaire follow-up) as well as twins in whom the new wave of questionnaires indicated that problems have developed since the first screening (about 1 % or 10 pairs yearly) are contacted. Consenting families have a clinical examination at one of three sites, one in Stockholm (60 pairs yearly), one in Malmö (20 pairs yearly), and one in Göteborg (30 pairs yearly). Two specially trained psychologists, blind to all previous information and to the results of the examination of the co-twin, perform the clinical interviews with teenagers and parents. The clinical interviews are specifically targeted to measure NDPs and multiaxial mental disorders, including K-SADS (Kaufman 1997), ASDI (Gillberg et al, 2001), WISC-IV (Wechsler, 2006), computerized test and a structured status by the PARIS proforma (Gillberg & Coleman 2003). Questionnaires tap into personality (J-TCI, Luby et al 1999), sexuality, bullying, criminality, gambling, traumatic experiences, friendship, substance and alcohol abuse and school performances. The results are finally validated by a clinical expert (senior child psychiatrist) and compared to life-time medical records. Blood and saliva are collected from the twins and saliva from the parents.
c. Molecular genetic analyses We will use two complementary approaches: a whole genome genotyping and a sequenced based mutation screening of candidate genes. Using an Illumina platform, we plan to genotype 660 000 single nucleotide polymorphisms (SNPs) covering the human genome. This specifically targets genes, tag SNPs, CNV, and other high-value genomic regions, offering one of the most comprehensive genotypings available on a single microarray. The genotyping step will be followed by a whole genome association analysis, aiming to identify susceptibility genes to NDPs by comparing the frequency of SNP/CNV in concordant and discordant MZ and DZ twins as well as in a control population. The association analyses will be conducted using the PLINK, gPLINK, and Haploview softwares, with modules for ascertaining quality control, detecting population stratification, and performing association tests. In addition, we will use the results of the Illumina genotyping to detect CNVs associated with NDPs. QuantiSNP, an objective Bayes Hidden-Markov Model, is able to detect and accurately map CNV using SNP genotyping data. The candidate genes screened for mutations include genes for synapse formation and development (e.g. Jamain et al 2003; Durand et al 2007), circadian rhythms (Melke et al 2008) and for the metabolism of peptide transmitters and sex steroids (Westberg & Eriksson 2008).
d. Progress report of the ongoing study The 30 clinicians trained in the instruments applied now assess about 2-3 twin pairs per week. The assessments started in May 2008 but has covered the entire cohort of 2008. Today, we have complete clinical evaluations of 120 twin pairs and their parents.We have registered the data for the first 38 individuals. Of these, 6 has been diagnosed with ASD, 8 with ADHD, 4 with language disorders, 4 with stress disorders, 2 with ODD and 1 with a depressive disorder; suggesting that the inclusion criteria identifies the types of cases we have defined for the study. In addition, we have evidence that virtually all individuals diagnosed as cases in the CATSS express some psychiatric subthreshold pathology, indicating the existence of familial (genetic or environmental) comorbidity. 3 of the twin pairs so far entered into the databse were control pairs. The response rate has been 70% and the gender distribution is 4:3 in favor of boys. We have chosen to establish a comprehensive databse using the Filemaker software and a special interface to enter data according to the written protocols. The databse contains more than 8000 variables per individual, each entered by two independent data assistants and cross-checked, giving access to very detailed assessments of mental health and antisocial behaviour categories. .
The current project Up to now we have achieved our originally planned goals. Due to the increasing scale of molecular work, the cohort size initially planned for will be insufficient for the new methods that have become available since the first planning of this project. We now intend to collect clinical data from another 220 twin pairs during a 2-year period to include a minimum sample of 500 clinically assessed pairs in a total study population of 10 528 twin pairs. Needless to say, it is a huge investment of time and money to start up clinical research sites, and the system we have established gives us a unique possibility to pursue this study, which has both a closer clinical methodology and a broader psychosocial perspective than any other international child and adolescent psychiatric study we know of.
Our study has attracted considerable international interest, and we have been approached by several research groups. The molecular genetic work described above has been planned together with one of the world-leading molecular genetic laboratories, which was first to identify specific molecular genetic causes for autism (mutations in genes encoding neuroligins and other synaptogenetic proteins, Jamian et al, 2003; Durand et al, 2007).

F. Relevance, Gender Aspects
NDPs with negative outcomes constitute one of the most important obstacles for adaptation in the societal and professional life and are associated with mental disorders and treatment needs, unemployment, poor family adjustment and criminality, especially inter-personal aggressive and violent crimes. Girls have much lower prevalences of NDPs but poorer outcome than boys, sometimes developing extremely difficult problem constellations that indicate different developmental trajectories. Understanding the genetic and environmental causes of these problems is essential in order to understand the development of antisocial carreers and to develop effective preventive measures.