What to Demand From a CNS Biomarker Service Provider

Selecting a CNS biomarker service provider? Learn the four criteria that matter most: regulatory validation, reproducibility, tissue provenance, and clinical trial readiness.

A biopharma team picks a biomarker vendor for a Phase II ALS trial based on a sensitivity spec sheet — femtomolar detection, a clean standard curve, a compelling webinar. Eighteen months and several hundred thousand dollars of sample processing later, the data can’t distinguish drug effect from assay noise, and nobody can say with confidence whether the signal in a given tube ever came from a neuron. The instrument was never the problem. The vendor’s provenance story was.

Key Takeaways

• Sensitivity and LOD claims are the easiest thing for a biomarker vendor to demonstrate and the least useful thing for a sponsor to evaluate on.
• Regulatory-grade validation (GLP, CLIA) and inter-assay/inter-site reproducibility data are the real predictors of whether a biomarker will hold up across a multi-site trial.
• Tissue-of-origin, or provenance, is the criterion vendors talk about least and sponsors should ask about first — a highly sensitive assay for a non-specific analyte still produces noise, not signal.
• The field’s L1CAM debate is a case study in why provenance claims need independent validation, not vendor assertion.
• A four-part evaluation framework — regulatory status, reproducibility data, provenance evidence, and context-of-use fit — gives sponsors a vendor-agnostic scorecard.

The Vendor Evaluation Problem No One Has Solved

Neurology trial sponsors now have more biomarker vendor options than at any point in the field’s history: large-panel CROs offering CSF and plasma testing, specialty labs built around a single analytical platform, and a growing set of extracellular vesicle (EV)-based providers promising blood-based access to CNS biology once thought to require a lumbar puncture. The vendor landscape has scaled faster than the evaluation frameworks sponsors use to navigate it.

Most vendor selection still defaults to a familiar comparison: assay sensitivity, limit of detection (LOD), turnaround time, price per sample. These numbers are easy to request and easy to compare across a spreadsheet. They are also the wrong starting point. A biomarker assay can be extraordinarily sensitive and still be measuring the wrong thing — detecting a protein at subfemtomolar concentrations does nothing for a trial if that protein’s concentration in blood doesn’t reliably track the biology the trial is designed to detect.

This matters more in neurology than almost any other therapeutic area, because the central diagnostic challenge in CNS drug development isn’t detecting a signal in blood — it’s knowing whether that signal actually originated in the brain or spinal cord, or whether it’s confounded by peripheral tissue, age, renal clearance, or comorbidity. A sponsor evaluating vendors on sensitivity alone is optimizing for the wrong variable.

Start With Regulatory Status, Not Marketing Claims

The first filter should be documentation, not narrative. “Peer-reviewed” and “published” are marketing claims; GLP and CLIA are compliance statuses with audit trails behind them; and FDA’s Biomarker Qualification Program (BQP) evaluates something different from either — not a lab’s competence, but whether a specific biomarker is fit for a specific, stated purpose. Sponsors should ask vendors to be precise about which of these three, if any, applies to the assay in question, and should be able to point to actual documentation rather than a claim on a slide.

FDA’s guidance is explicit that these are separate questions. Qualification through the BQP means a biomarker has undergone a formal regulatory process to ensure that it can be relied on to have a specific interpretation and application in medical product development and regulatory review, within a stated context of use [2]. Critically, a biomarker is qualified independently of the specific test used to measure it — a biomarker cannot be qualified without a reliable means to measure it, but FDA clearance of a testing device does not itself indicate qualification of the biomarker for drug development use [2]. In other words, a vendor’s assay can be immaculately validated and the underlying biomarker still not qualified for the context of use the sponsor needs — and vice versa.

Sponsors chasing an overly broad claim of biomarker utility should also know this is one of FDA’s most commonly cited submission deficiencies: qualification requests that try to cover multiple patient populations or disease states at once, rather than a single, well-defined patient population and disease state [3], are routinely flagged. A vendor who pitches a biomarker as broadly applicable “across neurodegeneration” rather than tied to a specific context of use is describing a marketing position, not a regulatory one.

Separately, any vendor running biomarker assays to support IND-enabling or later-phase work should be validating those assays against FDA’s bioanalytical method validation framework, which was originally issued in 2001, updated in 2018, and establishes the standards that pharmaceutical companies, CROs, and analytical laboratories must follow to ensure their methods produce reliable, scientifically defensible data across the full arc of drug development [4]. Ask for the validation report, not the summary paragraph.