You are here:
-A A +A

IAN INTERVIEW: Dr. Walter Kaufmann on the Importance of Twins in Autism Research

Interactive Autism Network at Kennedy Krieger Institute
Date Published: 
October 6, 2009

Dr. Walter Kaufman of the Kennedy Krieger InstituteWalter E. Kaufmann, M.D., is the Director of the Center for Genetic Disorders of Cognition and Behavior at Kennedy Krieger Institute. He is also a Professor of Pathology, Neurology, Pediatrics, Psychiatry, and Radiology at the Johns Hopkins University School of Medicine.

IAN: Do autism spectrum disorders (ASD) run in families?

DR. KAUFMANN:  It depends. Scientists are finding that some children with ASD come from families with no other relative showing any sign of ASD, much less a diagnosis. On the other hand, there are absolutely families where the first diagnosed child (called a “proband,” in genetics terms) is the tip of the iceberg. Some or all siblings, parents, even grandparents, cousins, etc., may have one or more autistic features, like repetitive behaviors or communication problems, known as the “broad autism phenotype.”

IAN:    Scientists don't refer to "identical" twins, meaning twins from the same egg, and "fraternal" or nonidentical twins," meaning twins from different eggs. What language do scientists use? What could different types of twins tell us about ASD?

DR. KAUFMANN: You’re right. “Identical” twins share all the same DNA and because they come from the same fertilized egg, they’re called “monozygotic” twins, or MZ. Fraternal twins should technically share as much DNA as any nontwin sibling pair, developing from two different fertilized eggs and known as “dizygotic,” or DZ twins, for short.

By comparing MZ (“identical”) and DZ (“fraternal”) twins, we are able to assume that while both types of twins have extremely similar environments, only one type (MZ) have the same genes. This is helpful in figuring out what roles genes and the environment may play in determining who is at risk for ASD during pregnancy and beyond.

IAN: What’s so special about twins?

DR. KAUFMANN: As any family with multiples knows, there are many special things about twins, including their relationship with each other.

On a research level, twins are unique because they serve as what we call “controls” for each other. For example, if we were to compare a 9-year-old brother with his 4 year old brother, we have to wonder what role different environments could have played in determining why one brother has ASD and the other doesn’t. (One is first-born compared with second-born, or what if the family had moved before mom’s second pregnancy?) Twins help us to decrease these sort of unknown “what-ifs” and, through various types of statistical modeling, can help guide us to look at potential causes of ASD.

IAN: Has research with twins been helpful in any other conditions?

DR. KAUFMANN: Absolutely. In psychiatry, there has been extensive research on schizophrenia. Because we know that MZ twins have extremely high concordance rates – that is, the chance that both twins or neither twin have the disorder is high – we know that schizophrenia is highly determined by genetic factors.

IAN: What have we learned from studies of twins and ASD? What are the limits and benefits of these past studies and do we need any more?

DR. KAUFMANN: Past studies of twins with ASD have been very interesting. As in schizophrenia, researchers have found that ASD is in fact highly concordant among twins. More sophisticated analysis looking at the complex gene and environment interactions within families suggests that ASD are highly heritable in many families. This means, related to your first question, that some families have ASD “in the genes” and that, even accounting for the fact that children learn behaviors from their parents and are heavily influenced by their environments, the likelihood or susceptibility for developing an ASD is “hard-wired.”

The reason we need to do more studies with twins is to try and figure out what happens in twins, especially MZ, who are not concordant. Why did one twin develop ASD and the other didn’t? This could be due to a change that happened to the genes early on in pregnancy or maybe to developing brain tissue later during pregnancy or later.

Another reason to pursue twin studies in ASD is to examine the role of gender. We know that boys are much more likely to have an ASD than girls by about 4:1; this chance is even higher for Asperger’s (10:1). In the past, we have not had enough information about girls to use as a comparison.

IAN: Does being a twin make you more likely to be autistic?

DR. KAUFMANN: Great question. Some research in Europe early this decade did suggest that perhaps being a twin put you at higher risk for developing an ASD, just as being male puts you at higher risk. However, later studies in the Scandinavian countries, which have very good complete records for all of their citizens, suggest that being a twin (including twins who are the result of assisted reproductive technology, like in vitro fertilization) is not a risk factor for ASD. But studies are ongoing.

IAN: If one twin has autism, what are the chances that the other one will?

DR. KAUFMANN: This depends on the type of twin and the sex of twins. For identical (MZ) twins, the concordance rate in all studies has been at least 85%. That means that almost 9 out of 10 times, if one MZ twin has an ASD, the other will, too, regardless of gender. For fraternal (DZ) twins, the concordance rate has varied over the years, especially with the increasing rates of diagnosis and the way that diagnosis has expanded to label people as ASD who may not have been recognized as such in the past. These rates should technically be the same among DZ twins as for any nontwin sibling, about 5-10%. Twins are a little different, because there is not the issue of “stoppage rules,” where a family may decide not to have another biological child because of the risk of ASD. In the case of twins, this tendency wouldn’t be relevant.

IAN: I know someone who has identical twins; one has autism and one does not. Why is that?

DR. KAUFMANN: That is certainly not typical, based on what we know. Perhaps they both had genes making them susceptible to developing an ASD, but for some reason, the affected twin had an epigenetic change due to chance, or maybe due to an environmental exposure in utero, or perhaps after birth. Epigenetic changes are very small changes to genes that can turn them “on” or, more typically, “off.”

In this way, although both twins’ genetic maps are nearly identical, a random detour sign appeared in the affected twin’s genes, skipping a street entirely and resulting in an ASD.

I hope this family is considering being part of IAN or other autism research, like the Autism Genetic Resource Exchange (AGRE;

IAN: I know someone else who has fraternal (nonidentical) twins; both have autism. Is that unusual?

DR. KAUFMANN: This is a more likely scenario, given the risk of ASD among siblings that I mentioned earlier. Currently, in a family with one child with ASD, the risk that each subsequent child would develop an ASD is around 5-10%. This number is a very rough estimate, and it changes depending on sex, of course, as well as the type of ASD – severe versus mild, with mental retardation or another syndrome, like Fragile X, or not.

IAN: Previously, there have been only a small number of twin sets to study, 50 or so. Now, thanks to the IAN Project, there are nearly 300! How will having more twins make a difference to researchers? What might we learn?

DR. KAUFMANN: IAN is a wonderful resource for the research community. Indeed, IAN has one of the largest collections of families with twins, with at least one member with ASD, in the world. We are able to make generalizations about patterns of environment and genes that were just not possible before. In addition, because IAN families have been so generous with their time, we have information about parents and unaffected nontwin siblings, which is a fantastic opportunity for those of us working to uncover why some people have ASD and others do not.

IAN: Thank you, Dr Kaufmann.

DR. KAUFMANN: You are most welcome. And thank you and IAN families for getting us closer to unraveling the mysteries of the autism spectrum disorders, so we can prevent, diagnose, and treat ASDs better and sooner.

Please rate the helpfulness of this article: 
Average: 4.7 (3 votes)