Jim Walters (Presenter)
MilliporeSigma
Authorship: James J Walters(1), Jacqueline Day(1), Judy Boland(1), Ryan Schofield(2), Dean Carlow(2)
(1) MilliporeSigma, Applied Research and Development, St. Louis MO, (2) Memorial Sloan Kettering Cancer Center, Department of Laboratory Medicine, New York NY
| Short Abstract The objective is to determine the utility of a well-defined, stable, animal-free matrix for use as a serum (or stripped serum) substitute in clinical assays. The synthetic surrogate matrix was prepared with 2% rHSA expressed from rice. It was tested in commercially available IVD ELISA kits with two different analytes (β-2 microglobulin and thyroglobulin), and in direct comparison to human serum in LC-MS based clinical assays for methotrexate, testosterone and estradiol in blood. For all comparative analyses the synthetic surrogate serum matched the performance of the matrix provided by the kit manufacturers, pooled human serum and stripped human serum with no interferences observed and with correlations of R2 > 0.998. We have formulated a simple, animal-free, analyte-free matrix which has been shown to be suitable as a calibrator/blank matrix as well as a patient sample diluent. |
Long Abstract
Introduction:
The objective is to determine the utility of a well-defined, stable, animal-free matrix for use as a serum (or stripped serum) substitute in clinical assays. The potential clinical benefits would include the complete absence of both clinically relevant endogenous analytes as well as any potentially harmful blood-borne pathogens. This, along with the ability to manufacture with precise lot-to-lot consistency could prove clinically useful as a blank, calibrator and control matrix as well as a sample diluent. The matrix under evaluation was prepared with 2% recombinant human serum albumin (rHSA) expressed from an animal-free rice expression system. Such a synthetic surrogate serum matrix could provide more consistently reliable clinical performance compared to animal-derived matrices for both ELISA and LC-MS based clinical assays.
Methods:
The ‘synthetic surrogate serum’ was tested in commercially available IVD ELISA kits with different analytes including beta-2 microglobulin and thyroglobulin. Blank and mock samples were prepared by spiking respective analyte reference materials at varying concentrations into the synthetic surrogate serum and comparing results to spiked pooled human serum, spiked stripped serum and spiked matrix provided by the kit manufacturers. The synthetic surrogate serum was also tested in direct comparison to pooled human serum in LC-MS based clinical assays for measuring blood levels of methotrexate (MTX), testosterone and estradiol. Several comparisons were performed including measuring spiked human serum samples and patient samples using either pooled human serum or synthetic surrogate serum calibrators. Synthetic surrogate serum was also evaluated as a sample diluent as compared to stripped human serum.
Results:
For the ELISA kits, several concentrations of analyte were tested, spanning the range of each of the respective kits. For all analytes, measurements showed interferences when pooled human serum was utilized indicating endogenous levels of target analyte. This was also observed when stripped serum was utilized with the thyroglobulin ELISA kit. As expected, no interferences were observed in any of the kits when the synthetic surrogate serum was used. Further, synthetic surrogate serum matched the performance of the matrix provided by the kit manufacturers indicating a suitable matrix for these ELISA kits. In the LC-MS based MTX assay, comparison of pooled human serum calibrators used to calculate synthetic serum substitute calibrator concentrations was described by the Deming equation y=1.048x - 6.14 with a correlation coefficient of R2 = 0.9995. Comparison of the values assigned to patient samples when using calibrators from pooled human serum or synthetic surrogate serum was described by the Deming equation y = 0.963x + 3.52 with a correlation coefficient of R2 >0.999. A clinical testosterone and estradiol assay were also used to compare synthetic surrogate serum vs stripped human serum as a patient sample diluent. For testosterone 10 specimens were compared over a range of 16.9 to 6,219.32 pg/mL and was defined by a Deming regression of y= 1.00x + 0.55 and a correlation coefficient of 0.9988. For estradiol, 10 specimens were compared over a range of 7.52 to 2555.97 pg/mL and was defined by Deming regression of y = 0.99x + 11.76 and a correlation coefficient of 0.9987.
Conclusion:
We have formulated a simple, animal-free matrix which has been shown to be suitable as a calibrator/blank matrix as well as a patient sample diluent. This has been shown in commercially available IVD ELISA kits as well as in LC-MS based clinical assays for both endogenous targets and therapeutic drug monitoring. Unlike typical animal based matrices, this synthetic surrogate serum offers a reproducibly defined formulation and is completely void of any analytes of interest or potentially harmful blood-borne risk factors.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |