The answer to this critical normalization challenge lies in the Multiple of the Median (MoM) method. Maternal serum biomarker concentrations are normalized by dividing the patient’s raw immunoassay result by the median value observed for that exact gestational age in a healthy reference population. This raw MoM is then statistically adjusted for maternal weight, ethnicity, diabetic status, smoking, and other physiological variables before being fed into a multivariate risk algorithm. The final output is a standardized, platform-independent risk score for conditions like Down syndrome, Trisomy 18, and open neural tube defects.
MoM transformation removes the overwhelming influence of gestational age on biomarker levels, turning a fluctuating concentration into a stable, comparable ratio. This allows a single action limit—such as 2.5 MoM—to be valid across different analysers, reagent lots, and clinical sites. However, the entire risk calculation chain depends on highly reproducible IVD raw materials and meticulously validated local median curves.
Why Raw Biomarker Concentrations Cannot Be Used Directly
Biomarkers like AFP, hCG, PAPP‑A, uE3, and Inhibin A change dramatically from week to week during pregnancy. Using a single reference range would falsely flag many normal pregnancies as high risk or miss true positives.
The Gestational Age Problem
A concentration of 40 IU/mL for free ß‑hCG might be perfectly normal at 10 weeks but dangerously high at 16 weeks.
Without correction, clinicians would need a different reference interval for every single gestational day, creating impossible recall demands and requiring massive sample collections to build each weekly normal.
Platform and Reagent Lot Variability
Different immunoassay analysers and even different reagent lots from the same manufacturer can produce systematically different absolute concentration readings.
These analytical biases would make risk cut‑offs meaningless if left uncorrected. MoM normalisation absorbs most of this platform‑to‑platform variation.
The Core MoM Calculation: Dividing by Gestational‑Age‑Specific Medians
The heart of the normalisation process is a simple ratio, but its accuracy depends on a robust reference median curve.
Establishing the Reference Median Curve
Each laboratory must calculate its own gestational‑age‑specific median for every biomarker by testing a large representative population of normal, unaffected pregnancies.
For example, the lab measures AFP in thousands of samples from women with singleton, chromosomally normal fetuses and plots the median concentration at each completed week (or day) of gestation.
Calculating the Raw MoM
Once the median curve is in place, a new patient’s raw concentration is divided by the median value for her exact gestational age.
Raw MoM = Patient’s analyte concentration / Median concentration for that gestational age
A value of 1.0 MoM means the result is exactly at the expected centre for normal pregnancies.
This step eliminates gestational age as a confounder and makes the number directly comparable across all stages of pregnancy.
Adjusting MoM for Maternal and Pregnancy Covariates
A raw MoM does not yet account for physiological factors that systematically shift biomarker levels. Failing to correct for these would introduce systematic false‑positive or false‑negative risks for large sub‑groups of women.
Weight, Race, and Diabetes Corrections
Higher maternal weight dilutes serum biomarkers, artificially lowering the raw MoM. The adjustment applies a weight‑based correction factor to bring the MoM back toward what it would have been in a woman of average weight.
Similarly, race has a pronounced effect—PAPP‑A levels are approximately 50% higher in women of Afro‑Caribbean ethnicity compared to Caucasian women. Insulin‑dependent diabetes also shifts baseline values for several markers.
Smoking, ART, and Twin Pregnancies
Smoking can increase Inhibin A by roughly 60% and elevate PAPP‑A, while assisted reproductive technologies (ART) alter the endocrine environment.
Twin pregnancies double the circulating concentrations of most fetoplacental products. The MoM must be adjusted using validated factors so that the risk calculation treats the pregnancy as a single‑twin appropriate scenario rather than flagging an outlier.
Translating Adjusted MoM into Clinical Risk
The adjusted MoM is not the final risk. It becomes one input into a multivariate Gaussian algorithm that combines information from several biomarkers.
The algorithm calculates a likelihood ratio for each marker profile and multiplies it by the woman’s a priori risk (based on maternal age or prior history).
The resulting personalised probability is then compared to a universal decision threshold—commonly a 1:250 cut‑off for Down syndrome screening—regardless of which analyser or lot performed the assays.
The Critical Role of IVD Raw Materials and Calibration
For diagnostic manufacturers and assay developers, MoM‑based screening reliability starts with the raw materials.
Lot‑to‑lot consistency in antibodies, calibrators, and chemiluminescent substrates is non‑negotiable.
If a new reagent lot shifts median concentrations by even a few percent, the entire laboratory’s median curve becomes inaccurate, and every patient MoM drifts. That drift directly changes the final risk probability, potentially altering clinical management. Validated reference standards and rigorous method comparison studies are essential to preserve MoM stability over time.
Understanding the Trade‑offs and Limitations
As robust as the MoM system is, it is not immune to real‑world challenges. Acknowledging these pitfalls is essential for clinical laboratories and kit manufacturers alike.
Dependence on Representative Reference Populations
The median curve is only as good as the population it was built from. If a lab serves a demographically unique community but uses a median derived from a different population, systematic bias can creep into every MoM.
Local validation and periodic re‑assessment of medians are mandatory.
Platform‑Specific Biases and Lot Drift
While MoM normalisation hides many analytical differences, it cannot completely erase large inter‑platform biases when no internationally recognised primary reference material exists for an analyte.
Labs that change analysers or manufacturers must re‑establish their own medians; otherwise, MoMs can shift subtly and corrupt clinical cut‑offs.
Incomplete Covariate Adjustments
Current adjustments cover the major known variables, but research continuously uncovers new influences (e.g., extreme maternal age, certain medications).
Any unrecognised confounder that operates in only a subset of patients will produce MoMs that do not reflect true biological risk, potentially degrading screening performance at the edges.
How to Ensure Robust MoM‑Based Screening in Your Setting
The universal principles are the same, but the specific actions differ depending on your role.
- If your primary focus is clinical laboratory management: Audit your median curves annually using local population outcomes and re‑validate any new reagent lot before releasing results. Document every covariate adjustment source.
- If your primary focus is IVD assay development: Invest in high‑purity recombinant or native antigens, multi‑year lot‑bridging studies, and provide clear guidelines to your laboratory customers on how to establish and maintain their local gestational medians.
- If your primary focus is manufacturing raw materials for prenatal kits: Guarantee lot‑to‑lot immunological reactivity within narrow specifications and offer reference standard characterisation data. Your consistency is the first line of defence against MoM drift and misclassified patient risk.
MoM normalisation is a finely engineered system that transforms a volatile biological signal into a trustworthy clinical decision tool—but only when every link in the chain, from raw material to population median, is meticulously maintained.
Summary Table:
| MoM Normalization Step | Key Variable / Factor | Function & Clinical Impact |
|---|---|---|
| Raw MoM Calculation | Gestational Age | Divides raw concentration by exact gestational age median to eliminate timeline bias. |
| Covariate Adjustment | Weight, Ethnicity, Diabetes, Twins | Corrects for physiological dilution, baseline shifts, and volumetric changes across patient sub-groups. |
| Multivariate Risk Output | Integrated Algorithm | Combines adjusted MoMs into a standardized risk ratio for Down syndrome and neural tube defects. |
| Raw Material Consistency | Antigens, Antibodies & Calibrators | Prevents median curve drift and lot-to-lot bias, safeguarding overall clinical assay reliability. |
Ensuring lot-to-lot consistency and preventing MoM drift in prenatal screening starts with uncompromising raw material quality. CamelBio provides diagnostic manufacturers, labs, and research institutes with one-stop access to premium IVD raw materials, technical services, and consulting—covering every stage from concept to clinic. Contact us today to optimize your assay reproducibility and elevate clinical accuracy.