Abstract
We describe a novel generic method to derive the unknown endogenous concentrations of analyte within complex biological matrices (e.g. serum or plasma) based upon the relationship between the immunoassay signal response of a biological test sample spiked with known analyte concentrations and the log transformed estimated total concentration. If the estimated total analyte concentration is correct, a portion of the sigmoid on a log-log plot is very close to linear, allowing the unknown endogenous concentration to be estimated using a numerical method. This approach obviates conventional relative quantification using an internal standard curve and need for calibrant diluent, and takes into account the individual matrix interference on the immunoassay by spiking the test sample itself. This technique is based on standard additions for chemical analytes. Unknown endogenous analyte concentrations within even 2-fold diluted human plasma may be determined reliably using as few as four reaction wells.
Highlights
Standard curves are used in immunoassays to interpolate unknown concentrations of the analyte of interest within biological test samples by relative quantification using the observed signal response
Standard curves were constructed from signal response data derived from human serum samples spiked with known concentrations of cortisol standards, taking into account the endogenous and exogenous cortisol concentrations (both known via the use of certified reference materials (CRMs) from the National Institute of Standards and Technology (NIST)), using 4-parameter logistic (4-PL) curve fitting (MSD Discovery Workbench v3.0 software)
Despite overestimation in analyte recoveries (i.e. 138–151% and 137– 148%, respectively, for male and female NIST sera; Table 1) when the endogenous concentration of cortisol is interpolated from the standard curve, recovery is not concentration-dependent and the recovery range is narrow (≤13%) given that the acceptance range for the recovery is usually 80–120% which extends over 40%
Summary
Standard curves are used in immunoassays to interpolate unknown concentrations of the analyte of interest within biological test samples by relative quantification using the observed signal response. If a test sample is spiked with standards, and the estimate for the unknown endogenous analyte concentration is correct, the correlation between the logarithm of the total analyte concentration and the logarithm of the response should produce a sigmoidal curve that includes an approximately linear region. Working within this linear range means that the relationship between total concentration and signal response can be treated as a simple linear regression. Can deliver good results with as few as four spike levels if the repeatability precision is sufficiently high
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call