By Jill Escher and Denise Haas
One of the most frustrating things about autism is, well, the meaninglessness of “autism.”
It doesn’t tell you much about your kid’s symptoms. He could be a college graduate or severely cognitively impaired. She could have radiant health or suffer seizures and gastrointestinal distress. Autism can involve any number of brain and health abnormalities. And of course the word conveys nothing about cause. Sadly, for many, autism is something of a fake diagnosis, just a placeholder referencing some level of social-communication-behavioral impairment threading though a slew of wildly varying conditions.
Slowly, however, genetic findings are giving some families long-awaited answers, sometimes bearing implications for treatment, trajectory, and co-morbidities. Families affected by similar mutations are connecting, supporting each other and working together toward improving their children's lives.
Despite the fact that chances of a definitive finding are still small—probably a bit less than 10%—some of us, like author Denise, hit the jackpot, unleashing a flood of insights about our children’s conditions. Also, even if nothing in the DNA looks fishy right away, you might get a genetic diagnosis in the coming years as the research continues to progress. And finally, and perhaps most important, by giving your child’s DNA (and/or brain, which we'll also discuss) to science, you will further the noble research quest to plumb the mysteries of the autisms.
Causes of autism
We'll start with an overview of what research has discovered so far about causes of autism. Please note this is a constantly evolving field and though our discussion reflects the current literature, nothing we say here is a fixed reality.

First, the “somatic insult” group. Under this umbrella would fall fetal or early life injury, such as prematurity, hypoxia at birth, prenatal infection or immune response, or prenatal exposure to certain drugs such as valproic acid (often used in the treatment of epilepsy). It’s not known what percentage of autism cases can be attributed to somatic insults but it’s probably safe to say somewhere between 10 and 20%. Though it's difficult to be definitive, clinicians can often make an educated guess about whether any of these non-genetic stressors played a role.

Second, there’s the “identifiable genetic glitch” category, with that nearly 10% chance a clinician might hand you a genetic cause. There are several types of genomic errors in this category.
Rare parent-inherited mutations sometimes knock out an important gene function, as can happen, for example in “consanguineous” families where parents are first cousins. Rare inherited mutations seem to account for up to 5% of autism cases.
“Point mutations” typically occur de novo in the child’s genome, meaning the glitch happened anew in one of the parent’s gametes. These small mutations usually impair a bit of a gene, misguiding the creation of a protein necessary for normal development and/or function. It’s estimated these may account for 6% of cases. Denise’s daughter was found to have one of these de novo mutations.
Genetic errors called “copy number variations” refer to stretches of DNA that have been deleted or copied, usually de novo in the child. These “CNVs” account for perhaps up to 5-10% of cases.
It’s important to emphasize the genetic errors in this identifiable glitch category are extremely rare, in the sense that no single mutation accounts for more than 1% of all cases of autism, and usually the frequency is much lower. So while this category of causation has seen solid progress in the past decade it’s also the land of diminishing returns.
Keep in mind that while these rare mutations are typically disease causing, seldom is risk restricted to autism. They often confer risk for other pathologies as well, including epilepsy, ADHD, schizophrenia, anxiety, and intellectual disability, not to mention abnormal physical traits. And these errors tend to cause more severe, not mild, impairments.

The third category of autism causation is what we think of as “genomic mischief that may be in there but we can’t really detect it (yet).” Your kid might have a mystery susceptibility or blooper lurking in his genome but given the limits of current technologies no clinician can say for certain. Hopefully in the coming years we will learn more about these glitches, which include the following:
“Somatic mosaicism.” In the case of autism, somatic mosaicism refers to mutations present in all or part of the brain but not present in other parts of the body. These mutations occur very early in embryonic development rather than in the egg or sperm of a parent like the other de novo mutations. Clinicians, however, typically can’t detect somatic mosaicism in your kid because, well, you can’t get into the tissues of the brain. Instead, these mosaic patterns have been found in post-mortem (meaning the individual has died and donated his brain to science) tissue. It’s thought that 3-6% of cases result from this phenomenon.


Her autism became apparent at around two years old in 1997; she had pretty normal development to that point, but started slipping away. With her regression, she lost all language and much of the fine motor skills she had acquired. She received a diagnosis of autism at 27 months. Some notable childhood traits included a large forehead, abnormal bone development, joint hypermobility, late development of gross motor skills, an abnormal EEG, and sensory challenges (she hated the sound of a large outdoor fountain that was near our home). As a baby, she did not seem bothered by her motor challenges; she was happy and content to walk at a 13 months holding onto our hands, but did not walk alone until 17 months. Unfortunately, by age two it was apparent that her skills had stealthily declined and she experienced the disturbing loss of eye contact and initiated stereotypic behaviors like spinning and odd hand movements often seen in autism.
Like most people diagnosed with autism she was tested for chromosomal abnormalities as a child — this test was normal. A test for Rett Syndrome also came back normal.
Fourteen years later, in 2011, our family was able to get an appointment with a clinical geneticist interested in autism. He mentioned genetic testing as a future plan. Then it took until 2014 to arrive at being considered for exome sequencing. This is a genetic test that looks for mutations in that 1% protein-coding portion of the genome. Insurance gave clearance after a few months. We did the testing as a "trio," meaning that my husband, myself, and daughter all submitted blood samples. We did the testing through Stanford, where I work.
After a few months the results came back... with nothing important to tell us. There was a missense mutation found, but it was not known to be pathogenic, even I had that same mutation.
Then more than a year later, in Dec 2015, our family got an interesting phone call from the geneticist. A de novo germline mutation was found in our daughter's genome. The mutation was on chromosome 1, and concerns a gene called GNB1.
In 2015, a team at Columbia University identified the first few cases of individuals with a mutation in GNB1. Columbia then queried genetic repositories worldwide and found a total of 13 individual with this mutation. The Columbia researchers published a paper sharing their amazing findings in April 2016.
During our fateful call in 2015, our geneticist also ventured to add that he thought the GNB1 mutation was the cause of our daughter’s autism. Wow, autism explained – in our case, the explanation is an identified rare disease that has autism as one of its features.
I’ve always been driven, but now I have a mission with a solid direction. I now have knowledge of what is causing my daughter's many challenges and symptoms. Rather than addressing that amorphous thing we call "autism," I can instead direct my attention to her particular disorder and its biological pathways. Rather than treating autism, we can support research on pharmaceutical treatments that target blocking or unblocking receptors and signals that aren't working in our daughter's specific syndrome.
This mutation is noted to be disease causing (though it does not cause autism symptoms in all cases) and we know the mutation affects many bodily systems. Through the years, she has had many symptoms of some sort of unknown underlying illness – a general malaise that could not be pinpointed. We now know that her chronic hematological abnormalities through the years likely stem from the mutation and unfortunately, is an ominous symptom to follow as she ages. We can explore what her mutation does to the body’s chemistry that may influence her seizures. Her treating physicians can help us to use medicine to prevent potential health problems that are unique to people who have this mutation. She fits a particular mold, which we now recognize and better understand.
My short-term hope is to find the cause of seizures that many of the individual with the mutation experience. These seizures are often life-threatening and must be avoided. In the long term, I hope research unveils the cell signaling that is dictated by the GNB1 gene which will advance our understanding of the pathogenesis of the mutation and human biology in general.
Like everything in autism, nothing about this mutation is straightforward. Individuals with glitches in GNB1 present with a spectrum of features, from verbal to non-verbal, ambulatory to non-ambulatory, and having sensory challenges of differing degrees depending on where the mutation is located on the gene. Knowing that defects within one single gene can seed such a dizzying array of phenotypes should give us pause when we even think of “autism” as a unitary disorder.
But we also have a great deal in common, and connecting with other GNB1 families has been nothing short of amazing. I get emotional just writing about it. I see photos of the young children with this mutation, and I worry. Many have not yet developed seizures but unfortunately may do so as they age. We must assist these parents so that they can take preventive measures to avoid their child’s exposure to seizures. I feel empowered with this newfound information and motivated to nth degree to understand more.
Now, I am not a Facebook fan (in fact, I am a bit of a hater) but I must say that for our families, it’s been the best thing EVER. A parent set up a GNB1 disorder private group that now includes families from more than 10 countries. We share all sort of information about our kids that help us evaluate trajectory and interventions. We are also working on a GNB1 website that is about to go live (gnb1.org)
I want to encourage all autism families to undergo genetic testing. We must remember that as parents, we help guide our loved one’s care – so please ask for testing and follow-up on the process. Ask who you should call to follow-up on the authorization process and call monthly (I would like to encourage follow-up calls – the insurance process can be difficult and clinicians have large caseloads). Do not feel shy about asking for testing, and insisting — you are your loved one's absolute best advocate.
To get your child’s genes tested you have a few options: you can seek evaluation by your primary doctor such as a psychiatrist who is up to date on autism genetics, get referred to a clinical geneticist, and/or enroll in a genetics research project.
When you receive a diagnosis of autism your clinician may order a test for Fragile X syndrome and also a test known as a chromosomal microarray (CMA). The CMA can detect copy number variations, those CNVs discussed above. Also, depending on clinical evaluation, a test called whole exome sequencing, which looks at that 1% of genes that code for proteins, or tests for specific genes might be ordered as well. A test called whole genome sequencing, which looks at both protein-coding genes as well as the other 99% of the genome, is not in standard clinical use at this time.
Sometimes a physician will refer you to a genetics clinic, such as those at Stanford or UCSF. Sometimes he or she will order the tests from his own practice.
Stanford: The Medical Genetics Clinic at Stanford Children's Health sees children, adolescents & adults with autism and other developmental disorders for a full genetic work-up.
http://www.stanfordchildrens.org/en/service/genetics/medical-genetics
UCSF: The UCSF genetics clinic is here: https://www.ucsfbenioffchildrens.org/clinics/medical_genetics_and_genomics/. TheUCSF speciality clinic for neurogenetic conditions is here: http://anp.ucsf.edu/patientcare/neurogenetic
Are these tests covered by insurance? There’s no easy answer to that, but with the diagnosis and CPT codes given by the provider, you can call your insurance company to check if the tests are covered. Note to the many of us who have older kids who received a diagnosis long before the clinically relevant findings became more robust—it’s definitely not too late to go back to your clinician and seek a referral for testing. In fact we think it’s very important that you try—you may be missing out on very important information about your older or adult child.
Another important avenue to consider is participating in an autism research project. SPARK, sponsored by the Simons Foundation, is one such study.* It’s entirely free and it’s so easy to participate you can do everything from home. No blood draw is needed, instead the affected individual and other family members spit in a tube which is then sent to the study in a pre-paid package. Now if your kid can't spit (as is the case with our kids), you might be able to extract enough saliva with a swab included in the kit. If not, you might need to wait until the study allows for blood draws instead.
According to SPARK, through whole exome sequencing, their analyses reveal an autism-relevant mutation about 8-10% of the time. If one is found, the program will report the information back to you. SPARK has already enrolled over 13,000 families and is seeking to enroll 50,000. The more families that enroll, the more discoveries that can be made.
Enrolling is easy, just check out SPARK at sparkforautism.org.
Conclusion
Sometimes genetic results hit a home run on a personal level, as with Denise. Other times your contribution may reveal something later, as the research evolves. But in all cases your contribution will help sow seeds for future scientific discoveries. Please consider how your family can add to our growing understanding of the autisms by seeking genetic testing, enrolling in Autism BrainNet and participating in genetic research such as SPARK.
Jill Escher and Denise Haas are members of the board of Autism Society San Francisco Bay Area. They are both parents of children with autism.
* Disclosure: Both Jill and Denise are active proponents of autism research. Jill serves as a volunteer on the SPARK community advisory council and independently funds research into genetic and non genetic inheritance in autism and related disorders. Denise is the co-founder of the GNB1 Foundation.