Genetic Heritability and Shared Environmental Factors Among Twin Pairs With Autism
To be published July 2011 in Archives of General Psychiatry
Authors: Joachim Hallmayer1, Sue Cleveland1, Andrea Torres1, Jennifer Phillips1, Brianne Cohen2, Tiffany Torigoe2, Janet Miller2, Angie Fedele2, Jack Collins3, Karen Smith3, Linda Lotspeich1, Lisa Croen4, Sally Ozonoff5, Clara Lajonchere2, Judith Grether3, Neil Risch6
1 Stanford University, 2 Autism Genetic Resource Exchange (AGRE), 3 EHIB/DEODC/CDPH,4 Division of Research, Kaiser Permanente, 5 University of California Davis/MIND Institute, 6 University of California San Francisco
Summary of article:
This article is the first of several anticipated reports from the California Autism Twins Study (CATS), a collaboration among scientific and research staff from six academic/research institutions. (Listed above) The research was funded through a grant from the National Institute of Mental Health (NIMH) and a supplemental grant from Autism Speaks, with Dr. Joachim Hallmayer (Stanford University) as the Principal Investigator (PI/Director) and lead scientist on the project.
The CATS study was designed to estimate genetic heritability of autism spectrum disorders (ASD) in a large, representative population using current diagnostic criteria. 'Heritability' can be defined as the degree to which a particular trait or set of traits is inherited (genetic). Genetic heritability of autism has previously been estimated to be as high as 90% based on prior twin studies which were conducted about 20 years ago. Since then, there have major changes in the way autism is diagnosed. In our study we were able to assess more than twice the number of twin pairs then in the previous studies combined. The CATS study also took advance of the California wide databases, so it is representative of the population at large. These early estimates of the genetic heritability of ASD have contributed substantially to funding priorities of the National Institutes of Health and other agencies.
For the CATS study, twin children with ASD who were born in California between the years of 1987-2004 were invited into the study. Participation included behavioral and cognitive assessments, as well as interviews with the parents. Assessments were conducted by Stanford University staff or by staff from Autism Speaks' Autism Genetic Resource Exchange (AGRE). Sophisticated statistical procedures (i.e. statistics & math) were used to estimate genetic heritability and heritability based on shared co-twin environment. It is important to note that in this study, it was not possible to define the relevant environmental time period (i.e., early prenatal, late prenatal, neonatal, etc.) or to identify specific components of the shared co-twin environment.
We found that among identical twins, the concordance rate for ASD (i.e., both twins in a pair have ASD) is similar to the concordance rates reported in the earlier ASD studies. However, an important new finding from our study is that the ASD concordance rate for fraternal pairs is considerably higher than reported in the earlier studies (these prior studies included very few fraternal twins). When we analyzed CATS data using statistical models, we concluded that, in addition to heritability based on shared genetic factors, there is also substantial heritability based on environmental factors that are shared by the co-twins in a pair. Our estimates of genetic heritability and heritability attributed to the shared co-twin environment are similar for males and females.
The CATS study is the largest autism twin study yet reported (192 twin pairs in this initial analysis) and these results are likely to guide future funding priorities for research and stimulate studies of environmental factors that may contribute to ASD.
Q&A
1. What is the study about?
This study was designed to investigate the extent to which autism and autism spectrum disorders (ASD) are genetically inherited. The best way to evaluate the genetic heritability of any disorder is to study identical and fraternal twin pairs to see if only one or both twins in the pair have the disorder. If both twins have the disorder, they are considered to be "concordant" for the disorder; if only one of the twins has the disorder, they are considered to be "discordant". To determine the genetic heritability, we compared how many identical twin pairs both have ASD vs. how many fraternal twin pairs both have ASD. We also took into account the sex of the twins and some other demographic characteristics. Using statistical models, the heritability through shared genes can be estimated, as well as the heritability through shared co-twin environment.
2. How was the study done?
The majority of the families were initially contacted by the California Department of Developmental Services (DDS), in cooperation with their Regional Centers. If they gave permission, the families were then contacted by the CATS project staff at Stanford University. The CATS staff at Stanford contacted the families to provide detailed information on the study, to determine eligibility, and to obtain written consent from the parents. For participating families, both twins in the pair were assessed by study staff using an extensive battery of standardized ASD and other behavioral and cognitive assessments. Blood or saliva was collected to determine if the twins were identical or fraternal. For this initial report, we analyzed data on 192 twin pairs, including 54 identical twin pairs and 138 fraternal twin pairs.
Data analysis took into account whether the twins were identical or fraternal, the sex of the twins, and a few other demographic characteristics. We were able to determine which pairs are concordant and which are discordant for ASD and also which pairs are genetically identical and which are not (fraternal twins share about 50% of their genes, which is the same as non-twin siblings). Using statistical models, we estimated the degree to which autism is impacted by genetic factors (i.e. hereditability) and the degree to which autism is impacted by shared environmental factors.(i.e., non-genetic) It is important to note that in this study, we were not able to study specific environmental factors that may contribute to the risk of ASD.
3. Why was the study done?
The genetic heritability of ASD has been estimated to be as high as 90%, based on prior twin studies which were conducted about 20 years ago. Since that time there have been major changes in the way autism is diagnosed. In our study we were able to assess more than twice the number of twin pairs then in the previous studies combined and we were also able to assess a larger number of fraternal twin pairs. We conducted the CATS study to determine current estimates using the largest yet reported group of twins who are representative of the larger population.
4. What did the study find?
We found that the concordance rate among identical twins is comparable to that reported in the earlier published studies (50-70%), but that the concordance rate among fraternal twins is considerably higher than previously reported (about 35%). All statistical models demonstrated that both genetic and shared environmental components are significant with regard to ASD, an estimated 38% for genetic heritability and 58% for the shared environment. We also found that the relative contributions of shared genes and shared environment are similar for males and females.
Our study provides rigorous evidence that the influence of genetic factors contributing to ASD may have been seriously overestimated in the previous studies. Most importantly, our statistical modeling suggests that the environment shared between co-twins has a strong impact on autism. A limitation of this study is that we were not able to further define the relevant environmental time period (i.e., early prenatal, late prenatal, neonatal, etc.) or to identify specific components of the shared co-twin environment (i.e., non-genetic factors) that may contribute to ASD in some children.
As with all research, further studies are needed to replicate our findings and to deepen scientific understanding about genetic and environmental factors that contribute to ASD in some children.
Additional reports will be prepared using detailed clinical data collected in the CATS study to look further at specific behavioral and cognitive characteristics in relation to genetic and environmental heritability.