A widely shared post says a large study found Phelan-McDermid syndrome, a genetic condition in which most patients also meet the criteria for autism, to be "much more common than previously thought." The study is real, the number is striking, and the practical lesson behind it matters for a lot of autistic people and families.
What Phelan-McDermid Syndrome Is
Phelan-McDermid syndrome (PMS) is caused by a deletion or mutation affecting the SHANK3 gene on chromosome 22. SHANK3 builds part of the scaffolding of synapses, the connections between neurons. Losing one working copy leads to a recognizable but variable picture: low muscle tone in infancy, delayed or absent speech, intellectual disability, and often seizures, sleep problems and kidney or gastrointestinal issues. The majority of people with PMS also meet criteria for autism spectrum disorder, and SHANK3 changes are thought to account for up to about 1% of all autism cases. That makes it one of the most common single-gene causes of autism known.
The Real Study
The source is Levy and colleagues (2026), published in Autism Research by the Seaver Autism Center at Mount Sinai (first author Tess Levy, a genetic counselor; senior author Joseph Buxbaum). Rather than counting known patients, they pooled genetic-testing results from nearly 180,000 autistic individuals across ten sources: commercial labs (GeneDx, Labcorp, Ambry Genetics), the SPARK research cohort, the Autism Sequencing Consortium and several children's hospitals.
They then modeled how many cases the raw counts must be missing, adjusting for people never tested, tests that do not cover SHANK3 well, and people with PMS who do not meet autism criteria. The result: an estimated 13.7 cases per 100,000 people, about 1 in 7,300, which would mean more than 45,000 people in the United States alone. Previous estimates were far lower.
One honest caveat: this is a modeled estimate built on assumptions about who goes untested, not a direct census. The direction of the finding, that PMS is substantially underdiagnosed, is robust. The exact figure carries uncertainty, as the authors themselves acknowledge.
Why So Many Cases Are Missed
The gap between known and estimated cases comes down to one thing: most people with autism or developmental disability are never offered genetic testing. When testing is offered, insurance barriers get in the way, or the test used does not adequately examine SHANK3. Tess Levy's recommendation is blunt: every child with autism should undergo genetic testing, because a genetic diagnosis changes what comes next.
What a Consumer DNA Test Cannot Tell You
This is the part our readers most need to hear. If you have raw data from 23andMe, AncestryDNA or a similar service, it cannot rule out Phelan-McDermid syndrome. Those services use SNP arrays that read a fixed set of common variants; they do not reliably detect deletions of a chromosome region or rare point mutations in SHANK3, which is exactly what causes PMS. The same limitation applies to any analysis built on that raw data, including ours. Detecting PMS requires clinical genetic testing: a chromosomal microarray to find deletions, and exome or gene-panel sequencing to find SHANK3 mutations, ordered through a doctor or genetic counselor.
Why a Diagnosis Changes Things
A PMS diagnosis is not just a label. It guides medical monitoring (kidneys, seizures, the risk of regression in adolescence), informs family planning, and opens the door to condition-specific research. A randomized, placebo-controlled trial is currently testing lithium in PMS, which we covered in our article on lithium and autism, and other targeted therapies are in development.
For transparency: the study was supported by CureSHANK, a patient foundation, and Neuren Pharmaceuticals, a company developing a PMS treatment. That does not undermine a prevalence model, but it is worth knowing when you read that "45,000 people" headline.
If autistic traits are part of your life, or your child's, our free Autism AQ-50 screening test is a sound first step. And if autism comes with intellectual disability, very delayed speech or seizures, ask a doctor specifically about clinical genetic testing.
Sources: Levy T, et al. (2026). "Prevalence of Phelan McDermid Syndrome Estimated To Be ~1:7300 Using a Multisource Model." Autism Research; GeneReviews: Phelan-McDermid Syndrome-SHANK3 Related (NCBI); NCT04623398, lithium vs placebo in Phelan-McDermid syndrome.