Brain GAP: The MISDIAGNOSIS

Brain GAP: The MISDIAGNOSIS 

AUTHOR: KJ Lavan

We are misdiagnosing the problem in brain health.

Everyone is focused on innovation.

New drugs.

New biomarkers.

Earlier detection.

And yes, the science is moving fast.

Lecanemab and donanemab have shifted Alzheimer’s care from symptom management toward disease modification in early-stage disease. 

But 2025 appropriate-use recommendations for donanemab make clear that treatment requires biomarker confirmation, careful patient selection, MRI monitoring, and structured clinical pathways. (ScienceDirect)

Blood-based biomarkers are also accelerating the field. A 2025 Nature Medicine study showed that automated plasma p-tau217 testing can identify Alzheimer’s pathology with high accuracy, while also emphasizing the need for two-cutoff approaches across real-world settings. (Nature)

A 2025 JAMA Neurology cohort study of nearly 3,000 cognitively unimpaired individuals found that plasma p-tau217 can help identify preclinical Alzheimer’s. 

But that confirmatory PET or CSF testing may still be needed when very high certainty is required. (JAMA Network)

And in 2026, longitudinal p-tau217 research pushed the frontier even further: blood-based models may help estimate when Alzheimer’s symptoms are likely to emerge years before clinical onset. (Nature)

So yes, for the first time, we can increasingly see the disease before it becomes irreversible.

But here’s the UNCOMFORTABLE TRUTH:

We are solving the WRONG constraint.

Because the real bottleneck is no longer discovery.

It is integration.

• Diagnostics don’t connect to longitudinal care pathways

• Biomarkers don’t automatically translate into treatment access

• Infusion models don’t align with workforce capacity

• Reimbursement still rewards late-stage intervention over prevention

• Data systems remain fragmented across providers, payers, caregivers, and platforms

• Health systems are not built for decades-long cognitive risk management

So what happens?

We get precision science trapped inside fragmented systems.

That is not a science failure.

That is an architecture failure.

The 2024 Lancet Commission estimated that up to 45% of dementia cases may be preventable or delayable by addressing 14 modifiable risk factors across the life course. (The Lancet)

That finding should have changed the global operating model for brain health.

Because if nearly half of dementia risk is shaped before diagnosis, then brain health cannot remain organized around late-stage clinical rescue.

It must become infrastructure.

↳ Detection infrastructure.

↳ Prevention infrastructure.

↳ Reimbursement infrastructure.

↳ Care-navigation infrastructure.

↳ Data infrastructure.

↳ Caregiver infrastructure.

↳ Policy infrastructure.

We have built components.

We have not built the system.

And until we do, progress will continue to look like:

• breakthroughs that don’t scale

• access determined by geography and income

• innovation that outpaces implementation

• biomarkers without pathways

• therapies without system readiness

• precision medicine without precision infrastructure

The next chapter in brain health will not be defined only by who discovers the next molecule.

It will be defined by who builds the integration layer across the ecosystem.

Because biology does not scale itself.

Infrastructure does.

And right now…

No one owns that layer.

#systems #infrastructure #innovation #data

AUTHOR: KJ Lavan