Neurofilament light (NfL) is a blood test that measures a protein released when nerve cells become damaged, potentially helping detect Alzheimer’s disease and other forms of dementia earlier than traditional methods. The test works by analyzing blood samples—not cerebrospinal fluid or imaging—making it far simpler and less expensive than PET scans or MRI evaluations. While NfL itself has not yet received independent FDA approval as a standalone diagnostic test, it is increasingly incorporated into multi-biomarker blood panels that the FDA has begun clearing, and it shows promise as a screening tool that could identify dementia-related brain changes before significant cognitive symptoms appear.
The significance lies in speed and accessibility. A patient can get their blood drawn in a primary care office, wait days instead of weeks for results, and spend a few hundred dollars instead of thousands. For families concerned about memory loss or cognitive decline, NfL testing offers a non-invasive way to assess whether Alzheimer’s-related pathology might be present in the brain—information that previously required either a risky lumbar puncture or expensive neuroimaging.
Table of Contents
- What Is Neurofilament Light and Why Does It Matter for Dementia Detection?
- Current FDA Approvals and the Multi-Biomarker Landscape
- Diagnostic Accuracy—What the Numbers Actually Show
- Cost, Pricing, and What Insurance Currently Covers
- Who Can Get Tested and When It’s Clinically Recommended
- Recent Clinical Trials and Emerging Evidence
- Platform Variability and the Need for Assay-Specific Interpretation
What Is Neurofilament Light and Why Does It Matter for Dementia Detection?
Neurofilament light is one of the structural proteins that make up the cytoskeleton of nerve cells. When neurons are damaged—whether from Alzheimer’s disease, frontotemporal dementia, Lewy body dementia, or other neurological conditions—these proteins leak into the bloodstream. Measuring the concentration of NfL in blood plasma can therefore reflect the degree of ongoing axonal damage occurring in the brain. A higher NfL level generally indicates more neurodegeneration, though the severity of symptoms does not always correlate directly with the biomarker level. What makes NfL particularly valuable is its ability to differentiate between different types of dementia.
For example, patients with Alzheimer’s disease show distinct NfL patterns compared to those with frontotemporal dementia or even psychiatric conditions that might mimic cognitive decline. A research cutoff of 38.6 pg/mL can distinguish Alzheimer’s from frontotemporal dementia with 76.7% sensitivity and 88.4% specificity—meaning if this threshold is used, nearly 77 out of 100 people with Alzheimer’s would test positive, while roughly 88 out of 100 people with frontotemporal dementia would test negative. This specificity is critical because misdiagnosis can lead to wrong treatment approaches. Researchers have also observed that NfL correlates well with cerebrospinal fluid (CSF) biomarkers and neuroimaging findings. The value of this correlation is practical: a patient who cannot tolerate a lumbar puncture or cannot afford a PET scan now has a blood-based alternative that reflects similar pathological processes. Over the past several years, NfL has moved from research curiosity to clinical tool, though implementation remains uneven and largely dependent on which multi-biomarker panel a testing company or healthcare system offers.
Current FDA Approvals and the Multi-Biomarker Landscape
This is where clarity is essential: Neurofilament light does not currently have its own FDA clearance as a standalone diagnostic test. However, regulatory momentum is building for tests that include NfL as one component of a larger biomarker panel. In May 2025, the FDA cleared the Lumipulse G pTau217/β-Amyloid 1-42 test made by Fujirebio Diagnostics, designed for use by neurologists and dementia specialists. In October 2025, Roche received FDA clearance for the Elecsys pTau181, which is positioned differently—as a test that primary care physicians can order to rule out Alzheimer’s-related amyloid pathology in patients with cognitive complaints. The most significant development for NfL specifically is Quanterix’s multi-biomarker test, for which the company submitted an FDA 510(k) application in February 2026. This test combines five biomarkers including neurofilament light, phosphorylated tau variants, and amyloid-beta.
The rationale behind multi-biomarker panels is sound: no single protein tells the complete story. A patient might have some Alzheimer’s-related pathology but also have other contributing factors, and a panel approach can help quantify the relative contribution of different disease processes. A key limitation is that the diagnostic thresholds (cutoff values) for NfL vary by assay and laboratory platform. One lab’s “normal” range might be 20.3 pg/mL, while another uses 18 pg/mL as the upper limit. This platform variability means that a patient tested at one facility and then re-tested at another might receive conflicting results if the labs use different assays. Healthcare providers ordering these tests must be aware of which platform the lab uses and interpret results accordingly.
Diagnostic Accuracy—What the Numbers Actually Show
Clinical research has generated specific accuracy metrics for NfL in dementia detection. Against healthy controls, NfL achieves an Area Under the Curve (AUC) of greater than 0.95 on most studies, which indicates excellent discrimination. Using a cutoff of 20.3 pg/mL, NfL identified Alzheimer’s disease with 76.7% sensitivity and 95.5% specificity in one major study. In practical terms, this means if 100 people with Alzheimer’s were tested, about 77 would correctly test positive (true positives), but about 23 would test negative (false negatives)—potentially providing false reassurance. Among 100 healthy controls, about 95 would correctly test negative, but 5 would test positive despite having no disease. For frontotemporal dementia, the numbers shift.
When distinguishing genetic forms of FTD from primary psychiatric disorders (like depression or anxiety that can resemble cognitive decline), a cutoff of 22.1 pg/mL achieved 78.5% sensitivity and 86.2% specificity, with an AUC of 0.908. For behavioral frontotemporal dementia specifically, the performance improved further: 88% sensitivity and 86% specificity using a cutoff of 13.3 pg/mL and an AUC of 0.95. These differences highlight an important reality: NfL is more reliable at detecting some conditions than others. The warning embedded in these statistics is that no single cutoff value applies universally. An NfL result of 25 pg/mL might be elevated for detecting Alzheimer’s disease but could still be within a “normal” range for detecting behavioral FTD. This is why NfL results must be interpreted by someone familiar with the specific clinical context—the patient’s age, symptoms, imaging findings if available, and family history. A patient with a borderline NfL result and minimal symptoms might not warrant the same intervention as a patient with a similar NfL level and progressive memory loss.
Cost, Pricing, and What Insurance Currently Covers
One of the major advantages of blood-based biomarker testing is affordability relative to alternatives. The anticipated out-of-pocket cost for the Lumipulse test is $500 to $1,000, compared to a PET scan which costs $3,000 to $6,000 and requires a radiology facility and appointment. Commercial NfL tests available directly are even less expensive: Lucent Diagnostics’ phosphorylated tau 217 test costs around $300, while Quest Diagnostics’ AD-Detect panel is available for $399. For those willing to use direct-to-consumer services, BetterBrain’s “blueprint plan” includes over 50 biomarkers plus cognitive assessment and a consultation for $399. However, insurance coverage remains the bottleneck.
An advisory panel recommended Medicare reimbursement at $130 per test, but Medicare coverage is currently limited to symptomatic patients—those already experiencing cognitive decline or memory problems. Pre-symptomatic screening, which would catch dementia-related brain changes before symptoms emerge, is not covered by insurance at this time. This creates a two-tier system: patients with symptoms may eventually get coverage, while those in the worried-well category must pay out of pocket. A cost-effectiveness analysis presented at the 2026 AD/PD conference found that blood biomarker triage testing identified 98.2% of PET-positive patients (those with confirmed Alzheimer’s pathology on imaging) at a lower average cost per diagnosis: $8,868 with blood testing versus $10,345 for PET screening alone. This economic argument is driving insurance companies and health systems toward broader adoption, though expansion of coverage remains in progress. For now, patients should clarify with their healthcare provider whether their insurance will cover the test and what out-of-pocket costs to expect.
Who Can Get Tested and When It’s Clinically Recommended
Both the FDA-approved tests are currently authorized for patients aged 55 and older who are experiencing memory problems or cognitive decline. This age cutoff reflects the research populations on which the tests were validated and the fact that Alzheimer’s disease most commonly manifests in people over 55, though early-onset Alzheimer’s can occur in younger individuals. The indication—”memory problems or cognitive decline”—means the test is designed for symptomatic patients, not for cognitive screening in asymptomatic people. The current clinical use cases include helping a primary care physician decide whether to refer a patient to neurology, guiding a neurologist toward a dementia diagnosis, or helping clarify whether cognitive symptoms are attributable to Alzheimer’s pathology versus other causes. In one real-world scenario, a 62-year-old woman notices difficulty finding words and occasionally forgets appointments.
Her primary care physician can order an NfL test; if the result is elevated, the next step might be specialist referral or MRI. If the result is normal, it reduces the likelihood of Alzheimer’s-related pathology and might prompt investigation into other causes like depression, medication side effects, or thyroid dysfunction. A significant caveat is that insurance will not cover NfL testing for cognitive screening in asymptomatic people. If a 58-year-old with no cognitive symptoms but significant family history of Alzheimer’s wants to know their status, they would need to pay out of pocket. This limitation reflects both the current state of research (long-term outcome data for asymptomatic biomarker-positive people is still being collected) and insurance companies’ caution about expanding coverage before evidence of benefit in presymptomatic populations is stronger.
Recent Clinical Trials and Emerging Evidence
A significant ongoing study is the NAVAIDD trial at Monash University, which is assessing the diagnostic utility of plasma NfL across all-cause dementia—not just Alzheimer’s disease. The study is estimated to complete in 2027 and should provide broader real-world data on how well NfL performs when clinicians don’t already have a suspected diagnosis. This is important because in routine clinical practice, patients don’t always arrive with a clear diagnostic hypothesis. Some present with mixed cognitive and behavioral symptoms, and NfL’s ability to guide differential diagnosis in that messy, real-world context is what clinicians most need to know.
The blood-based biomarker approach offers practical advantages beyond simple cost savings. A 78-year-old with early memory concerns can get answers through an office visit and a simple blood draw, without needing to visit a PET imaging center, worry about radiation exposure, or spend hours in an MRI machine. For elderly patients with mobility limitations, anxiety about enclosed spaces, or other barriers to imaging, a blood test removes friction from the diagnostic pathway. For healthcare systems in rural areas where PET or specialized MRI is not available, blood biomarker testing could become the initial screening tool, with imaging reserved for cases where diagnostic uncertainty remains after biomarker results.
Platform Variability and the Need for Assay-Specific Interpretation
One of the most important practical limitations that clinicians and patients should understand is that NfL reference ranges and decision thresholds are platform-specific. The blood test for NfL can be run on several different laboratory platforms—Lumipulse, Elecsys, Simoa, and others—and each platform produces slightly different absolute values. A plasma NfL concentration of 25 pg/mL measured on one assay might correspond to 30 pg/mL on another.
This variability arises from differences in antibodies, calibration standards, and instrument sensitivity. In practical terms, this means that if a patient is tested for NfL at one laboratory and receives a result of 28 pg/mL, and then is tested months later at a different laboratory using a different platform, comparing the raw numbers directly is unreliable. The laboratory reports should reference platform-specific cutoff values and interpretation ranges, and if longitudinal tracking is planned—checking NfL levels over time to see if they’re rising—it is best done using the same platform and laboratory. Healthcare providers must be aware of which specific test and platform was used, read the detailed lab report, and interpret results in that context rather than applying generic cutoff values from research literature without knowing which assay was used in the original studies.
