Setting Up an sST2 Immunoassay: Antigen, Antibody Pairs and Calibrators

Updated 2026-09-28

A practical sequence for groups building an sST2 assay for the first time, from antigen format through to curve stability.

Antigen choice comes first

Recombinant antigen gives lot traceability and reproducibility; native material may be closer to the circulating form but is harder to source consistently. For calibrators, the decisive question is whether the format your antibody pair recognises is exposed — tagged versus carrier-free changes antibody accessibility.

Antibody pairing and matrix

The pair sets your linear range and sensitivity. On the calibrator side, carrier proteins such as BSA can introduce non-specific signal, which is the practical argument for carrier-free antigen in calibrator preparation.

Why lot consistency decides your curve

A drifting standard curve is more often traceable to antigen lot variation than to the antibodies. Re-validate a small side-by-side on each new lot and record the lot number with every curve.

Sanity checks before you trust the curve

A curve that fits well can still be measuring the wrong thing, and the checks that catch this are cheap. Spike a known amount of antigen into the matrix you intend to use and compare the recovered concentration against the amount added: recovery outside roughly 80–120% points at matrix interference or at a calibrator whose composition differs from the sample. Dilute a high sample serially and check that the measured concentration falls in proportion — non-linear dilution is the classic signature of an interfering substance in the sample rather than a problem with the standard. Run a zero-antigen blank on every plate so you can see background drift as it happens rather than after a batch has been reported. And test the low end deliberately: many pairs look excellent at mid-range and fail where clinical decisions are actually made, because non-specific binding dominates the signal once the true concentration approaches zero.

Documenting the assay so it survives a change of hands

Assay development data tends to be held by one person, and the most expensive event in a small laboratory is that person leaving. The defence is unglamorous: record the antigen lot number, the antibody lot numbers, the coating and detection concentrations, the buffer composition, the matrix and the incubation conditions for every curve you treat as a reference. Note which lot was used to qualify the assay and keep an aliquot of that lot aside, so a future problem can be traced by comparison rather than by memory. Guidelines for immunoassay validation in a regulated context, such as the industry perspective published for bioanalytical work, set out a structure worth borrowing even in a research setting, because the questions they ask — accuracy, precision, selectivity, dilution linearity, stability — are the questions that will be asked of your data eventually.

Ask a technical question