Development and Validation of a Highly Informative Immuno-Enzymatic Analysis for the Determination of Free Prostat-Specific Antigen

• Published in: Innovative biosystems and bioengineering Vol. 3; no. 4; pp. 220 - 231
• Main Authors: Komar, Anatolii, Kozerecka, Oksana, Besarab, Olexandr, Galkin, Alexander
• Format: Journal Article
• Language: English
• Published: Igor Sikorsky Kyiv Polytechnic Institute 10-12-2019
• Subjects: enzyme immunoassay; free prostate-specific antigen; validation
• ISSN: 2616-177X; 2616-177X
• DOI: 10.20535/ibb.2019.3.4.185877

Background. Prostate-specific antigen (PSA) is a key marker used in the monitoring of patients with prostate hyperplasia, risk of deve­lopment and cancer of prostate. Clinical laboratory diagnostics for such patients are implemented using appropriate serological tests. Enzyme immunoassay kits are bioanalytical products that are subject to the requirements for medical devices for in vitro diagnostics. Thus, the urgent task of modern analytical biotechnology is improving bioanalytical, technical, and economic indicators of diagnostic kits, as well as their validation, as a prerequisite for the admission of such products to the market. Objective. Scientific substantiation of the composition and technology of highly informative enzyme-linked immunosorbent assay (ELISA) test-kit for the quantitative determination of free PSA in human serum (blood), as well as its bioanalytical validation at the time of manufacture and expiry date. Methods. A previously prepared and characterized set of monoclonal antibodies (mAb) to different PSA epitopes was used to construct ELISA. Peroxidase conjugates of mAbs were synthesized by periodate oxidation. We used the WHO Prostate Specific Antigen International Standard and the human serum with different PSA amount. The validation of the developed ELISA was performed according to the conventional scheme used for quantitative bioanalytical methods. Results. At the first stage of the work, the determination of the optimal mAbs configuration was carried out for the development of a non-competitive sandwich ELISA. The mAbs studied were characterized by the following indicators: activity in ELISA (relative to PSA, its complex with α1-antihymotrypsin (α1-ACT) and related protein kallikrein-2), isotype, titer, affinity constant, comparative epitope specificity, degree of PSA enzymatic activity inhibition, source of origin (Balb/c or NZB mouse). Based on these characteristics, the principle of using mAbs to construct ELISA for the determination of various forms of PSA (free molecule and one associated with α1-ACT) was formed. For different ranges of PSA concentrations, it was assessed the feasibility of sharing different mAbs to determine free PSA. At the second stage of the work, a protocol for the quantitative ELISA was formulated (determination of the number of reagent components, optimal time, and temperature indicators for all stages of the analysis). At the final stage of the study, we carried out validation of the developed ELISA, de­termining the following indicators: diagnostic specificity, linearity, limit of detection and quantification (analytical sensitivity), precision, correctness (accuracy). Conclusions. The antibodies of the epitopes P2 and P4 (and different specificity groups) are the most optimal combination for the sandwich ELISA to determine free PSA. The sorption-detection ability of different mAb pairs correlates with their affinity. Antibodies 26B9, 21B7, and 11G5 of the P2 epitope (group II specificity) were the best antibodies for use in solid phase sorption, and antibodies 14C8 and 21D7 of the epitope P4 (group I specificity) were the best ones for detections. The pair of high-affinity mAbs 26B9 and 14C8 possessed the most pronounced sorption-detection properties. The combined use of the mAbs 14C8 and 21D7 enzyme conjugates leads to increased sensitivity in the study of PSA concentrations in the range of 1–10 ng/ml. Validation characteristics were determined at the time of kit release and at the end of the shelf life (1 year). The average value of diagnostic specificity is 100.6%. The technique is linear in the range of 0.1–30 ng/ml, the uncertainty of the calibration graph is insignificant. The estimated detection limit is 0.001782 ng/ml, and the limit of quantification (analytical sensitivity) coincides with the reproducibility limit and is 0.0054 ng/ml. The correctness (systematic error) is 0.03 ng/ml and is statistically insignificant.