In laboratory medicine, reference materials are the foundation of trust. They are well-characterized samples used to validate that a diagnostic test performs as expected - measuring the right thing, at the right sensitivity, with acceptable accuracy. Every clinical chemistry analyzer, every PCR-based infectious disease test, and every tumor marker assay relies on reference materials for calibration and quality control.

Prenatal screening, despite its rapid technological advances, has a significant gap in this area. And it's quietly holding back the entire field.

What Reference Materials Do

A reference material is essentially a "known answer" sample. When a lab runs a reference material through their testing pipeline, they should get a predictable result. If they don't, something is wrong with the assay, the reagents, or the process.

Reference materials serve several critical functions:

• Assay validation - Proving a new test works before it reaches patients
• Quality control - Ongoing monitoring that an established test continues to perform correctly
• Proficiency testing - Comparing performance across different laboratories running the same or similar tests
• Regulatory compliance - Meeting the evidentiary requirements of bodies like the FDA, MHRA, and EU IVDR

Organizations like NIST (National Institute of Standards and Technology) and WHO produce reference materials for many clinical assays. For prenatal cfDNA screening, no equivalent exists at scale.

Why Prenatal Screening is Different

Several factors make creating reference materials for NIPT uniquely difficult:

Mixture complexity. Unlike most diagnostic samples, prenatal cfDNA is inherently a mixture of two genomes - maternal and fetal. The fetal fraction varies between patients (typically 5-20%) and changes throughout pregnancy. A reference material needs to capture this two-genome mixture at controlled ratios, which is far more complex than a single-genome reference.

Condition rarity. For common trisomies (21, 18, 13), clinical samples are available in reasonable numbers. But as NIPT expands to cover microdeletions, sex chromosome aneuploidies, and single-gene disorders, the conditions being screened for become vanishingly rare. Collecting enough positive samples for conditions affecting 1 in 10,000 to 1 in 100,000 births is impractical through normal clinical workflows.

Sample instability. Cell-free DNA is fragile. Plasma samples degrade over time, even when stored properly. Creating a stable, distributable reference material from real cfDNA samples presents significant logistical challenges - the material needs to survive shipping, storage, and handling across multiple labs while maintaining its original properties.

Ethical constraints. Collecting large volumes of blood from pregnant individuals specifically for reference material creation raises ethical questions. Unlike a blood bank draw, these collections serve commercial or research purposes and involve a vulnerable population.

Platform diversity. Different NIPT providers use different sequencing platforms and bioinformatics approaches. A reference material ideally needs to work across all of them - Illumina whole-genome sequencing, targeted SNP-based methods (like Natera's), and single-molecule counting approaches (like BillionToOne's). This cross-platform compatibility requirement adds another layer of complexity.

The Downstream Consequences

The absence of standardized reference materials creates real problems:

Validation gaps. When a lab develops a new NIPT panel or adds a new condition, they often validate against whatever clinical samples they can access. This ad hoc approach means validation rigor varies enormously between providers. Some may validate a rare condition screen on as few as 5-10 positive samples - hardly the statistical power needed for clinical confidence.

Regulatory uncertainty. The EU IVDR framework, which came into full effect recently, requires more rigorous performance evaluation for IVD devices, including NIPT. Without standardized reference materials, labs struggle to meet these requirements consistently. This is particularly acute for smaller labs that lack the clinical networks to accumulate rare samples.

Limited comparability. Without a common reference, it is difficult to compare the performance of different NIPT providers. Published sensitivity and specificity figures are often derived from different study populations, different sample sets, and different definitions of "positive." Patients and clinicians have no easy way to evaluate which test actually performs better.

Innovation bottlenecks. Novel approaches - whether new bioinformatics algorithms, new sequencing chemistries, or new screening targets - all require validation data. The reference materials gap means promising innovations sit in development longer than they should.

Emerging Solutions

The field is beginning to address this gap through several approaches:

Synthetic and computational reference materials. Computationally generated cfDNA data that mimics real clinical samples offers a scalable alternative. Synthetic datasets can be produced for any condition, at any fetal fraction, in unlimited quantities. While they don't replace the need for clinical validation entirely, they can provide a standardized baseline for benchmarking and quality control. Companies like Eabha are building exactly this - using genomic AI to generate biologically realistic synthetic cfDNA that can serve as reference material for labs developing and validating prenatal screens.

Cell-line-based mixtures. Some groups are creating physical reference materials by mixing DNA from well-characterized cell lines at controlled ratios. This approach produces real DNA that can be processed through standard laboratory workflows, though it doesn't perfectly replicate the fragment characteristics of genuine cfDNA.

Consortium efforts. Industry and academic groups are beginning to collaborate on shared reference datasets. These initiatives recognize that the reference materials problem is pre-competitive - all providers benefit from having better validation tools, even if they compete on everything else.

What Needs to Happen

Solving the reference materials problem requires coordinated action. Standards bodies need to define what "good" reference materials look like for prenatal cfDNA testing. The industry needs to agree on minimum validation requirements. And technology providers need to build the tools - both physical and computational - to produce reference materials that are accessible, affordable, and comprehensive enough to cover the expanding range of conditions being screened.

The prenatal screening field has made remarkable progress in the last decade. Ensuring that progress is built on a solid validation foundation is the next critical step.