Diagnosis usually arrives after the damage is already done
Genetics plays a central role in psychiatric and neurodevelopmental disorders — but a risk variant alone tells a clinician almost nothing about whether, or how severely, a person will actually develop disease. Here's the full scale of that gap, and why it persists.
Most psychiatric and neurodevelopmental disorders — including autism spectrum disorder, schizophrenia, and major depressive disorder — are only diagnosed after symptoms become severe. By then, relationships, careers, and neurodevelopment have already been affected for years without intervention.
A European perspective
The burden is just as stark closer to home. Mental ill health affects more than one in five people across OECD and EU countries — one of the most significant public health and economic challenges in Europe today. Anxiety, depressive, and alcohol use disorders alone are projected to reduce healthy life expectancy by 2.5 years between 2025 and 2050, account for roughly 6% of total EU health expenditure, and cost an estimated 1.7% of GDP annually.
Independent European sources add further texture: years lived with disability due to mental disorders rose by roughly 31% between 2019 and 2021 across European countries, mental disorders are now the leading cause of death among people aged 15 to 29 in Europe (driven largely by suicide), and an estimated 235,000 deaths per year across Europe are linked to mental and behavioral disorders.
The diagnostic blind spot: incomplete penetrance
The field already knows genetics plays a central role. But knowing someone carries a risk-associated variant tells us almost nothing about whether — or how severely — they will actually develop disease. Two people with the same genetic mutation can have radically different outcomes: one may develop profound disability, while the other shows no symptoms at all. This phenomenon — incomplete penetrance — is one of the most clinically costly blind spots in genomic medicine, and it currently has no solution.
Neurodevelopmental disorders, including ASD, affect approximately 1 in 100 individuals worldwide and carry an estimated annual economic burden exceeding €250 billion in the EU. A defining and poorly understood feature of these conditions is this same incomplete genetic penetrance — individuals carrying identical pathogenic variants, such as mutations in the Sonic Hedgehog (SHH) signaling pathway, a central regulator of forebrain development, can have dramatically different clinical outcomes.
Three compounding failures
- No human-relevant penetrance prediction tool exists. Animal models fail to capture human-specific neurodevelopmental variability, leaving genomic diagnoses clinically uninterpretable for families and physicians.
- No molecular patient stratification for clinical trials. Pharmaceutical companies cannot identify which variant carriers will respond to interventions — a major contributor to clinical trial failure rates above 90% in CNS indications.
- No scalable biomarker platform. Current academic multi-omics workflows are one-off and non-standardized, and cannot be commercialized at the throughput diagnostic labs or pharma partners actually need.
A related but distinct problem: treatment selection
A second, related need — and the one most existing commercial competitors have chosen to address first — is drug-response prediction rather than disease-risk prediction. Roughly 63% of major depressive disorder patients try multiple medications before finding one that works, and around a third don't respond even after two rounds of treatment. That's a real, monetizable problem — but it's conceptually different from penetrance and risk stratification, which is the problem NeuroPenetrance is built to address. This distinction matters commercially: it's the main reason the closest-looking competitor in the market isn't actually a direct competitor (more on this on the Platform page).