IMAGINE-2: Stratifying Genomic Causes of Intellectual Disability by Mental Health Outcomes in Childhood and Adolescence

In England, there are over a million people with learning disabilities, a quarter of whom are children of school age. Most moderate to severe intellectual disability (ID) has a genetic cause. In order to identify those genetic risks, the NHS is now routinely screening the DNA of children who have significant developmental delays. Being informed that their child's ID is caused by a genetic change is of value to parents. But, at present, we can rarely use that information to advise on best management of behavioural and educational issues, or to reduce the risk of poor mental health outcomes. Our study aims to fill that gap in knowledge.

Our IMAGINE-ID programme of research began in 2014. By 2019 we had recruited nearly 3500 UK families whose child has ID due to a genetic cause. Using a combination of online interviews, questionnaires, and face-to-face meetings with families, we built up a comprehensive picture of those children's strengths and weaknesses. We discovered there was a far greater risk of severe behavioural and emotional problems than was previously recognised. Whilst children with ID from the general population are about six times as likely to have problems of this nature, the risk is over thirty times greater if the disability has a genetic cause.

We also discovered that children whose genetic risk was inherited had more severe emotional and behavioural problems than those in which the equivalent change occurred by chance. Perhaps parents who carry the genetic anomaly could be mildly affected by it, although they do not share the same degree of disability as their child? They are more likely than other families participating in our research programme to live in disadvantaged circumstances with overcrowding, poor quality housing, and unemployment. Adverse social circumstances would contribute to parenting difficulties and exacerbate their child's problems. We need to learn more about these important points of vulnerability. Families at risk could be identified sooner, and supported more effectively in future, if we understood more about the processes that led to their difficulties. These questions will be addressed by our new research.

We will follow up and interview all participants 5 years after our initial assessment, to ask: first, have the mental health issues we uncovered in the previous study persisted? Second, if they have persisted, or improved, what are the medical, educational and environmental factors that have changed since we first met those families?

Most children we saw in the first study were between 6 and 13 years of age. During our follow-up, many will be entering adolescence or early adulthood. That is a time when the risks of some mental health problems become substantially greater. We will be endeavouring to discover whether the young person's behavioural and emotional adjustment, or their risk of emerging mental health disorders, is influenced by the educational, medical or other support their families have received over the past 5 years. We will be looking for clues that pinpoint those children with the best and worst outcomes.

More than one in three children in IMAGINE-ID had an Autism Spectrum Disorder. A quarter had ADHD, and a similar proportion had either severe anxiety or serious challenging behaviour. What was the impact on those children's educational progress? To what extent were those conditions recognised and treated by their local medical and mental health services? To gather that information, we will supplement what we learn from parents in the course of our follow-up interviews with nationally collected records on the children's education (from the National Pupil Database) and on their medical history (from the NHS Hospital Episode Statistics Database). We will also use information from medical records to learn more about the strong association between ID with a genetic cause and seizures, which affect up to 70% of children in the IMAGINE-ID cohort

Our genotype-first study IMAGINE-ID recruited 3402 children from UK NHS Regional Genetics Clinics, whose intellectual disability (ID) is associated with either a pathogenic Copy Number Variant (CNV-75.9%) or Single Nucleotide Variant (SNV-23.6%) or both (0.5%). 84% of consented families provided standardized assessments of their child's mental health, functional adaptation, medical and educational history, and family/social environment. We have data on 4,054 genetic variants, logged for 3,398 individuals, representing 3,117 CNVs (1,534 in recognised syndromes, 1,583 unique) and 944 SNVs in 278 different genes. 57% of IMAGINE-ID participants had one or more psychiatric diagnoses: Autism Spectrum Disorders (38%); ADHD (24%); anxiety disorders (12%); challenging behaviour (13%). 31% had a history of seizures, 57% take anticonvulsant medication. Little specificity was found between genotype and diagnostic phenotype, implying multiple developmental processes lead to broadly similar outcomes. Defining potential subtypes of complex neurodevelopmental disease, based on genetic risk, will be facilitated by the observation of longitudinal trajectories of neurodevelopment into adolescence and early adulthood. We aim to conduct a 5-year longitudinal study of the cohort with annual on-line phenotyping of all participants, supplemented by intensive face-to-face assessments of over 500 young people with highly pathogenic CNV that were intensively studied in our first phase. We have already found that environmental risk factors are key to child adjustment, potentially accounting for the majority of variance in outcomes. In our intensive family study, we will test hypotheses about potential mediating variables: for instance, we know children with an inherited CNV are at greater risk, indicating the importance of family environment. We will measure the role of educational and medical support by linkage to the National Pupil Database and NHS Digital's Hospital Episode Statistics.

The short-term impact is upon the domains of clinical practice and quality of life for affected families. Potential economic benefits include more efficient targeting of resources and reduced familial disruption. Longer-term, knowledge arising from genetic discoveries could support partnership with industry through targeted drug development. We liaise with stakeholders, including parent-support organizations, and hold focus groups/conferences at which potential benefits are discussed.

In the UK, specialist genetic investigations for children with ID rarely translate into specific recommendations for behavioural management or prognostic statements about mental health outcomes. Our PPI has shown families would strongly welcome such knowledge. By linking genotypic data on specified SNV/CNV with standardized phenotypic data, our study has generated clinically valuable information. Unusual behaviour patterns or emotional disorders associated with ID are often ascribed to poor parenting practices. Recognising common disorder-specific patterns is the first step to reassuring parents, and educating clinicians/social support staff. Impact will reduce self-blaming and stress, with resultant improved quality of life for affected families. Impact timescale < 5 years.

Child behaviour can be reliably and easily measured across time, and may independently predict future symptoms and psychiatric disorders, including the process by which disturbed emotions and behaviour can undermine family/individual quality of life. Our findings already indicate opportunities for intervention. We aim to enhance parent and child wellbeing (e.g. by promoting their mental health, reducing school exclusions, limiting risk of parental separation), which may have a beneficial economic impact. Impact timescale <5 years.

Education and Care Professionals:
Intellectual disability implies global impairments in cognitive skills and functional adaptation. Nevertheless, some developmental trajectories may be preserved. We aim to parse the wide range of ID syndromes by CNV/SNV subtyping. Stratification may allow the identification of rare variant disorders in which islets of ability are preserved. Gaining such knowledge about specific CNV/SNV has implications for education planning, and fostering the maximisation of individual potential. Such discoveries, supplemented by data retrieved from the National Pupil Database and Hospital Episode Statistics could inform policy on management. Information on environmental factors influencing, for instance, the emergence of challenging behaviour (when linked to specific genotypic risks) could point to the value of early interventions, thus reducing risk of transfer to residential care and the associated costs. Impact timescale <5 years.

Clinical Pharmacology Developmental pathways:
Neurodevelopmental risk encompasses (inter alia) autism, schizophrenia, and epilepsy. We lack a lifespan perspective on risk related to genotype. In our accelerated longitudinal cohort study, linking genotype to phenotype could lead to pathway and network analysis of complex 'omics data on the basis of identified genotypes. Potential collaborations could lead, in longer term, to academic-pharmaceutical industry partnerships and the discovery of causal mechanisms and potential novel drug targets for intervention. Impact timescale <10 years.

Training of skilled researchers:
Within the framework of the program we have fostered the academic skills of junior staff, several of whom have already obtained PhD Fellowships (Wellcome/NIHR). Impact timescale <5 years.

International collaborations:
We have established the foundations of international partnerships with The Centre for Applied Genomics (Univ. Toronto), The Autism and Developmental Medicine Institute (Geisinger Health System, Lewisburg), The Simons Simplex Collection Global Consortium, among others (see appended letters of support). Impact timescale:<5 years.