The New Alzheimer's Blood Tests

In over twenty years of practicing geriatric psychiatry, few developments have given me as much hope as what's happening right now in Alzheimer's diagnostics.

For most of my career, the standard dementia workup meant an initial MOCA screen, further neuropsychological testing and a non-contrast brain MRI—valuable for staging and ruling out other causes, but unable to confirm the underlying pathology. Spinal taps and PET scans could provide further confirmation, but were typically reserved for diagnostically challenging cases, particularly when differentiating Alzheimer's from other dementias like frontotemporal dementia or Lewy body disease. Most patients never had access to that level of certainty.

That's changing.

In 2025, the FDA cleared the first blood tests to aid in diagnosing Alzheimer's disease. A simple blood draw can now provide information that previously required a $4,000 brain scan or a needle in the spine.

But like any new tool, these tests come with nuance. They're not perfect. They don't replace clinical judgment. And understanding what they can and can't do matters enormously.

This post is my attempt to make sense of the new landscape—for fellow clinicians who consult with dementia specialists, for therapists and functional medicine practitioners working with aging patients, and for patients and families who want to understand their options.

The numbers tell the story: about half of patients with cognitive impairment are either undiagnosed or misdiagnosed. Fewer than 8% are identified in the early disease stage—mild cognitive impairment (MCI) or early dementia—when treatments are most effective.

We now have disease-modifying therapies—lecanemab and donanemab—that can slow progression in early-stage Alzheimer's. But these drugs only work if we identify patients early enough. And they come with significant risk factors that limit patient selection.

As I wrote in my 2025 year-in-review, the convergence of precision diagnostics and new therapeutics is reshaping how we approach brain health. Blood tests are a key part of that shift.

For most of medical history, Alzheimer's was confirmed only at autopsy! Then came structural imaging—CT and MRI—to rule out other causes of memory loss and cognitive decline and identify atrophy patterns, though these detect downstream damage, not the underlying pathology.

CSF biomarkers via lumbar puncture gave us direct measurement of amyloid and tau with ~90-95% accuracy, but required specialist access and an invasive procedure. Amyloid and tau PET imaging visualized pathology in living patients—the gold standard—but at $3,000-5,000 per scan with limited availability.

Now, blood-based biomarkers offer the same molecular targets from a simple blood draw. 2025 marked the first FDA clearances for clinical use.

Not all tau biomarkers are equal.

pTau217 has emerged as the most specific blood marker for Alzheimer's pathology. Unlike pTau181—which reflects general neurodegeneration—pTau217 is tightly linked to the amyloid and tau changes specific to Alzheimer's disease. It rises early in the disease cascade, often coinciding with the first detectable amyloid on PET imaging.

In meta-analyses, plasma pTau217 achieves approximately 90% sensitivity and 87% specificity, with diagnostic accuracy (AUROC >0.92) approaching that of CSF testing and PET imaging.

Importantly, pTau217 helps distinguish Alzheimer's from non-Alzheimer's dementias—frontotemporal dementia, Lewy body dementia, vascular dementia, and primary age-related tauopathy (PART)—where levels remain low. This differential diagnostic value is critical, since these conditions require different treatment approaches.

This is the biomarker driving the current revolution.

Two FDA-cleared blood tests arrived in 2025:

Lumipulse G pTau217/Aβ1-42 Ratio (Fujirebio) — Cleared May 2025 for specialty settings. In clinical trials, 92% of positive results and 97% of negative results matched PET/CSF findings. Designed as a diagnostic aid ("rule-in"). Note: Recent real-world data has shown some variability; performance may be lower in patients over 80 years of age.

Elecsys pTau181 (Roche) — Cleared October 2025 for primary care. A negative result (97.9% NPV) provides high confidence that cognitive symptoms aren't due to Alzheimer's pathology. Designed as a "rule-out" tool, but is population dependent.

Neurocode ALZpath Dx — A laboratory-developed test using ultrasensitive Simoa technology. Their three-panel approach adds NfL (neuronal injury) and GFAP (neuroinflammation) to pTau217. This combination helps differentiate Alzheimer's disease from non-Alzheimer's dementias—including frontotemporal dementia, vascular dementia, and mixed pathologies—and enables longitudinal tracking of disease progression or treatment response. Apollo Health's BrainScan integrates this panel with Dr. Dale Bredesen's ReCODE protocol.

Blood tests don't replace the full dementia workup—they streamline it.

Imaging still matters. CT and non-contrast MRI remain essential for ruling out other causes of memory loss and cognitive decline—tumors, strokes, normal pressure hydrocephalus, vascular dementia—and for assessing atrophy patterns. Volumetric MRI quantifies hippocampal and cortical volume loss. ASL (Arterial Spin Labeling) is an advanced MRI that can look cerebral perfusion by targeting water in arterial blood. FDG-PET reveals metabolic patterns that help differentiate Alzheimer's from frontotemporal dementia or Lewy body disease. Amyloid PET remains the gold standard for visualizing plaque burden, and tau PET correlates with disease stage and symptom severity.

CSF testing via lumbar puncture varies in accuracy: CSF p-tau217 approaches 95% sensitivity and specificity, but traditional markers (Aβ42, total tau, p-tau181) are less reliable—specificity drops to 66-77% for distinguishing Alzheimer's from other dementias in clinical practice. Specialist access required.

The practical shift: blood tests can reduce the need for CSF and PET by 80-90% in appropriate patients, reserving invasive and expensive testing for indeterminate cases. Amyloid PET runs $3,000-5,000 at specialized centers; CSF analysis costs $1,000-1,500. Blood biomarkers range from $300-1,500—a fraction of the cost, available through a simple draw at most any lab.

This is where things get interesting for those of us practicing functional and integrative psychiatry.

Amyloid accumulation may begin up to 20 years before symptoms appear. The conventional model waits for symptoms, then diagnoses. The prevention paradigm flips this.

For patients focused on brain health optimization, cognitive longevity, and healthspan, these tests open new possibilities:

Cognitive risk assessment for the asymptomatic: Not everyone seeking these tests has memory complaints. Some are asymptomatic individuals who want to understand their cognitive wellness—the same way they might pursue a coronary calcium score or full-body scan for preventive insight. This is particularly relevant for those with a family history of Alzheimer's or dementia, or known genetic risk factors like APOE ε4 carrier status. While population-level screening isn't yet guideline-supported, an individualized brain health screening as part of a comprehensive functional assessment can provide meaningful risk stratification and a baseline for longitudinal monitoring.

Baseline and serial testing: Consider establishing a baseline pTau217 in your 50s, then monitoring every 5-10 years. Rising levels—even within "normal" range—may signal early changes worth addressing. We don't wait for heart attacks to check cholesterol; why wait for dementia to assess brain biomarkers?

Root-cause intervention: A positive or trending biomarker is a call to action. The Bredesen Protocol (ReCODE) addresses the metabolic, inflammatory, and lifestyle factors contributing to neurodegeneration—insulin resistance, chronic inflammation, sleep dysfunction, cardiovascular risk, nutrient status, and toxic burden.

Monitoring response: The three-panel approach (pTau217 + NfL + GFAP) isn't just diagnostic—it tracks whether interventions are working. If you implement metabolic and lifestyle changes, you want to see NfL and GFAP trend down over time.

Therapeutic eligibility: Blood tests identify candidates for lecanemab and donanemab, which require biomarker confirmation.

For the longevity-focused patient, these tests represent a shift from reactive medicine to proactive brain health—the same philosophy we apply to cardiovascular and metabolic optimization. The goal isn't just early diagnosis; it's preventive cognitive care.

These tests are powerful, but not perfect.

Technical limitations:

A word on interpretation:

A biomarker result is only as useful as the clinical context surrounding it. I've seen the critique of some functional medicine approaches: order a broad panel, hand the patient a report, call it a day. That's not medicine—that's data without meaning.

A pTau217 level doesn't tell you whether memory loss stems from early Alzheimer's disease, mild cognitive impairment progressing to dementia, depression, sleep apnea, thyroid dysfunction, vascular contributions, or medication effects. It doesn't tell you whether the patient in front of you is a candidate for lecanemab or would be better served by aggressive cardiovascular management and sleep optimization. It doesn't account for their goals, values, or what a diagnosis would mean to them.

The test is one data point. A skilled clinician integrates it with history, examination, cognitive testing, imaging, labs, and—most importantly—the patient's story. If you're pursuing testing, work with a physician trained to interpret these biomarkers in the context of comprehensive, individualized assessment.

That's the foundation of any evidence-based approach, conventional or integrative.

For clinicians: Blood tests improve diagnostic efficiency—rule-out in primary care, rule-in for specialty confirmation. Multi-marker panels add value for differential diagnosis and monitoring. Verify your lab's performance before implementation.

For therapists and mental health professionals: Cognitive symptoms in older adults warrant biomarker workup. Don't assume depression is the whole story. Early detection changes the care planning conversation.

For patients and families: Blood tests for dementia diagnosis are now available through your physician—no spinal tap required. A positive result isn't the end of the story; it's information that opens doors to action, planning, and intervention while options remain.

For the prevention-focused: If you're optimizing metabolic health, tracking biomarkers, and thinking about healthspan and cognitive longevity, these tests belong in your toolkit. Whether you're asymptomatic but concerned about family history, an APOE ε4 carrier wanting to understand your risk, or simply proactive about brain health screening, consider baseline cognitive biomarker testing in your 50s as part of a comprehensive assessment. It's the same philosophy you'd apply to cardiovascular or metabolic health—and ideally guided by a clinician who can integrate the results into a personalized prevention strategy.

Five essential resources for those who want to go deeper:

After two decades of practicing geriatric psychiatry, I've sat with countless families in that difficult moment when cognitive decline becomes undeniable. Too often, we've been limited to describing what we observe rather than explaining what's happening biologically.

These blood tests don't solve everything. But they bring us closer to the kind of medicine I've always wanted to practice—detecting problems earlier, explaining them more clearly, and acting while there's still time.

For those of us working at the intersection of functional psychiatry, brain health, and longevity medicine, this is the diagnostic foundation that makes proactive care possible.

But let's not lose sight of what matters most. Amid precision diagnostics and biomarker panels, the most important assessment tool remains the time we spend listening to our patients and helping them make sense of what these numbers mean.

The tests are new. The art of understanding isn't.

The mission of Psychiatry 3.0 continues: bridging traditional psychiatry with emerging, evidence-based approaches—without losing the art of understanding at the center. The science is advancing. The training is catching up. And the opportunity to do better for our patients has never been greater.

What topics would you like me to explore? Reply or comment, and let me know. Your input shapes where this goes next.

And if you find the Psychiatry 3.0 newsletter valuable, consider forwarding it to a colleague or friend. Sharing helps grow this evidence-informed community and keeps the conversation growing.