Imaging vs Blood Biomarkers: Two Thresholds Shape Clinical Triage

Imaging and blood biomarkers are complementary: use blood tests for low-barrier screening and triage, and imaging for confirmation, localization and staging. Plasma p-tau217 tests have reached AUCs up to 0.97 in research cohorts, and the FDA has cleared a plasma pTau217/ß-Aβ1-42 ratio test. Some results still land in a gray zone that needs confirmatory imaging.

TL;DR:

• Use two thresholds for plasma p tau 217; intermediate results leave about 13% to 40% of patients needing PET or CSF confirmation.
• FDA cleared Lumipulse ratio testing is indicated for adults 55 and older with cognitive impairment; its pivotal study found 91.7% positive and 97.3% negative agreement.
• Blood first screening can reduce confirmatory PET scans by roughly 10% to 40%, depending on threshold selection and patient risk.
• Before relying on imaging biomarkers across sites, standardize acquisition and analysis because thresholds remain unsettled and scanner and vendor differences can affect results.

What do blood and imaging biomarkers each measure?

Blood biomarkers are soluble molecules that reflect disease activity: how much of a protein is circulating, and how it changes over time. Imaging biomarkers are measurements extracted from scans, which report where disease sits, how far it extends and how tissue behaves. The two answer different questions, which is why they work best together.

Common blood markers include:

• p-tau species and Aβ ratios in neurodegeneration
• Neurofilament light chain (NfL) for neuronal injury
• Troponin and BNP for cardiac stress
• D-dimer for thrombosis

Common imaging outputs include:

• PET standardized uptake values (SUVs)
• Volumetrics from MRI or CT
• Apparent diffusion coefficient (ADC)
• Late gadolinium enhancement (LGE) on cardiac MRI
• Global longitudinal strain (GLS) on echocardiography

Imaging provides localization and regional distribution that a tube of plasma cannot, as reviews of Alzheimer biomarkers describe. Imaging modalities provide different kinds of information, so selection depends on the clinical question. A clinician-facing guide to choosing a scan can help with modality selection.

How well do blood biomarkers perform, and where do they fall short?

Plasma p-tau217 tests have shown AUCs as high as 0.97 in research cohorts and about 0.92 in real-world cohorts. With stratified thresholds, predictive values reach the 82% to 94% range.

In the FDA’s pivotal study, 91.7% of positive Lumipulse results had amyloid plaques on PET or CSF, and 97.3% of negative results were negative. Strong numbers, yet not a verdict for every patient.

A single cutpoint produces unacceptable misclassification. Even high-performing assays leave about 13% to 40% of patients in a gray zone, which typically needs confirmatory PET or CSF testing.

A blood result should be interpreted alongside the clinical picture and any needed confirmatory testing. Escalate when:

• The result is borderline
• The picture is discordant with symptoms
• The finding will drive a therapy decision

Assay variability matters here. Cutpoints and reporting practices should be considered when interpreting results.

What does imaging add, and why is validation hard?

Imaging contributes spatial mapping, staging and tissue characterization. Its derived biomarkers include:

• SUVs for tracer uptake
• Volumetrics for structure size
• ADC for cellularity
• LGE for myocardial scar or inflammation
• Strain for myocardial function

Quantitative imaging (QI) development moves through analytical validation, qualification and utilization. Reproducible clinical use depends on standardization across acquisition and analysis. Many imaging biomarkers still lack universally accepted thresholds and face inter-scanner and inter-vendor variability. For a deeper look at standards, see our piece on IBSI standards for quantitative imaging biomarkers.

Imaging still anchors decisions where anatomy matters. Imaging can contribute to therapy monitoring and anatomic detail.

How do you combine them in a staged workflow?

Here is a practical flow, built on the complementary model in which blood tests filter and imaging confirms:

1. Assess pretest probability from symptoms, history and risk profile.
2. Run a blood biomarker with two thresholds.
3. If the result rules in, move to imaging for confirmation, localization and staging.
4. If the result falls in the gray zone, order CSF or PET.
5. After treatment starts, monitor with the combination that fits the disease: serial blood markers for frequent checks, imaging for structural or regional change.

Integrating blood screening can reduce confirmatory PET scans by roughly 10% to 40%, depending on thresholds and patient risk. Radiology experts likewise suggest staggered selection, with blood tests prioritizing patients for resource-intensive scans.

Weigh these factors at each step: clinical context, assay performance, test availability, therapeutic consequences and the need for spatial detail.

Pro Tip: Standardize assays and report formats across your sites so longitudinal results stay comparable.

Where does this workflow apply in practice?

• Alzheimer’s disease: Use plasma p-tau217 for triage, then amyloid PET or CSF for confirmation and therapy qualification. The FDA-cleared Lumipulse ratio is indicated for adults 55 and older with cognitive impairment.
• Traumatic brain injury: The choice of follow-up imaging after a negative CT should be based on clinical judgment.
• Cardio-oncology: Troponin and BNP plus echocardiography with GLS screen for cardiotoxicity, while cardiac MRI adds tissue characterization and myocarditis diagnosis, as multimodality imaging reviews describe.
• Oncology: Imaging biomarkers such as volumetrics and SUVs can contribute to oncology assessment. See our guide to oncology diagnostics for CROs and pharma. See our guide to oncology diagnostics for CROs and pharma.

What should a lab and clinic have in place?

A dependable blood-to-imaging pathway needs more than a good assay. Our checklist:

• Validated assays with documented cutpoints and performance
• Rapid turnaround, so gray-zone patients reach confirmatory testing quickly
• Reliable specimen pickup and handling
• Standardized reports that show thresholds and interpretation
• LIS and EHR integration
• Trial-support services for sponsors and CROs

Our services are intended for clinics and research teams seeking actionable results. If you are selecting assays, planning specimen logistics or integrating results into your records, our team can help. Our therapeutic-area diagnostic support examples show how this works across disease areas.

Where is this field heading, and what stands in the way?

One area of interest is multimodal integration, including the use of radiomics to analyze imaging features. Emerging trends in diagnostic radiology discuss quantitative imaging and radiomics. The barriers are practical: assay variability, imaging standardization, reimbursement and access. My view is that the field will move fastest where labs and radiology groups agree on shared protocols before the algorithms arrive. Adopt standardized protocols, join prospective validation studies, and use two-threshold approaches wherever they are validated.

— Kohealth Labs

How do we support integrated biomarker programs?

Ask our team about specialty testing, specimen logistics, turnaround and trial support for your blood-to-imaging pathway. Whether you run a practice or a study, we work with you on assay selection, logistics and results delivery. See what we test, review our clinical trials services, and contact us to plan your next protocol.

This article is general information, not a substitute for advice from a qualified doctor. Consult a qualified healthcare professional about your own circumstances before acting on anything here.

FAQ

Is imaging or a blood biomarker better for diagnosis?

Neither is better in every case. Blood tests screen and triage with a low barrier to access, while imaging confirms, localizes and stages disease. Imaging and fluid biomarkers complement rather than compete.

What is the two-threshold approach?

It uses a lower and an upper cutpoint instead of one. Results below or above them are definite, and results in between go to confirmatory PET or CSF testing. This lowers false positives and negatives compared with a single cutpoint.

Can a blood test replace amyloid PET?

Not entirely. Blood tests can reduce how many patients need PET, but a meaningful share of patients stays in an intermediate zone that still benefits from confirmatory PET or CSF testing.

Why does imaging still matter if blood tests are accurate?

Blood tests cannot show where disease sits or how far it has spread. Imaging supplies that spatial and tissue detail, which staging, treatment planning and monitoring depend on.

Sources

• Pmc
• FDA Clears First Blood Test Used in Diagnosing Alzheimer’s Disease | FDA
• Complementary utility of plasma biomarkers and Aβ‐PET for diagnosis, risk‐stratification, and treatment monitoring in Alzheimer’s disease

For more information about our diagnostic testing, clinical research laboratory services, or partnership opportunities, visit https://kohealthlabs.com or contact KoHealth Labs at (469) 792-8439 and ask for Amy Patterson.