Why IVD product decisions need to start before the science finishes
Step-change clinical value in diagnostics, such as earlier detection, more precise patient stratification or continuous monitoring, originates in the biomarker. But the product architecture determines how much of that value reaches the patient.
The life sciences sector generates a steady stream of promising biomarkers, yet few become products a clinical laboratory can run routinely. The gap is rarely the biomarker itself. It lies in the sample preparation, assay development and manufacturing needed to turn a validated signal into an in vitro diagnostic (IVD) product.
Why the safer path wins by default
Most R&D teams qualify a new biomarker against the platform and workflow they already operate. It’s a rational choice: validated workflows, established regulatory pathways and existing laboratory infrastructure all lower the commercial risk. But those platforms were designed for laboratory standardization and efficiency, not around any one biomarker. When a biomarker’s own requirements (sample type and stability, abundance in the sample, molecular form and biological variability) are never specified independently, the cost of compromise goes unmeasured. Where the fit fails outright, it tends to surface only after the science is complete, when changing course is slowest and most expensive.
A framework for when to diverge
To understand when it’s right to diverge you can use a two-dimensional framework to assess each biomarker: the clinical value it creates, and the workflow change needed to deliver it. Fitting onto an existing platform remains the right call when that platform can meet the biomarker’s requirements. Divergence is justified where it can’t, and where the clinical value is high enough to warrant the change.
Three biomarkers where the science works, but the product doesn’t yet
- Alpha-synuclein seed amplification assays can flag Parkinson’s disease potentially years before motor symptoms, but depend on cerebrospinal fluid or skin biopsy, with no accessible, standardized sample route.
- Metagenomic sequencing could identify almost any pathogen and its resistance profile in one test, but resistance-gene databases remain incomplete and turnaround is not yet fast enough for acute care.
- Tumor-informed ctDNA testing for minimal residual disease received its first FDA companion diagnostic approval in May 2026, yet each assay is still built per patient in the manufacturer’s own laboratories.
In each case, the barrier is product development, not biomarker science.
Start the product conversation early
In practice, this means specifying what a biomarker will require of a product while qualification is still under way, and letting that specification shape the qualification work itself. Sample route, assay format, consumable architecture, setting of use and time to result then stay open as choices rather than becoming constraints.
Continue exploring emerging biomarkers
Go deeper into this area and discover how to:
✓ Map your translational risk. Weigh clinical value against the workflow change needed to deliver it.
✓ Pinpoint where translation stalls. Understand what is holding back alpha-synuclein SAA, metagenomic AMR prediction and tumor-informed MRD testing, and what would close each gap.
✓ Know when to diverge. Judge when a biomarker justifies departing from an established platform, and when it doesn’t.
✓ De-risk the route from biomarker to IVD. Define what the sample, assay and manufacturing process must deliver.
Read our in-depth paper, Emerging Biomarkers and the Case for System Divergence, to find out more.
Connect with CDP
Our Life Sciences and Diagnostics team works with clients to develop innovative, high-impact devices. Reach out today, and a member of our team will get in touch.