B-293 Enzyme linked immunosorbent assay development and full validation for the quantification of the antibody conjugated fraction of an antiFAP ADC in human serum

Abstract

Abstract Background Cancer remains a global public health issue. Despite recent progress in early diagnosis and treatment strategies, development of anti-tumor therapies with reduced systemic toxicity remains a challenge. Antibody-drug conjugates (ADCs), a new, highly-potent drug class are synthesized by linking a small molecule cytotoxic drug to a tumor specific antibody, are an emerging treatment option for many cancers. Despite advances in ADCs, study of the pharmacokinetics (PK) is challenging. A complete PK profile requires at least three analytical methods to determine: conjugated antibody, total antibody, and free drug by Ligand Binding Assays (LBA) and liquid chromatography/tandem mass spectrometry (LC-MS/MS) respectively. Objective To develop and validate an ELISA method to quantify the conjugated antibody fraction of an anti-fibroblast activation protein (FAP) ADC against solid tumors in human serum. Methods Conjugated antibody assay: This indirect sandwich ELISA utilizes human FAP as capture antigen (Abcam) with a proprietary rabbit polyclonal anti-drug IgG as detection antibody. The signal is amplified by the use of a second antibody conjugated to Horse Radish Peroxidase (Abcam). Absorbance is measured at 450nm. Immunoassay method validation was done in compliance with the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) M10 which establishes the limit for precision in CV% and accuracy in Relative Error (RE%) at ± 20% each (± 25% for LLOQ and ULOQ). Additionally, the total error (TE%) for accuracy and precision batches, should be ≤ 30% (≤ 40% at LLOQ and ULOQ). Results The weighted calibration curves were linear over a concentration range 0.5 to 100 µg/mL. A four-parameter regression model with a 1/Y2 weighting factor was adopted to best fit the OD450 values and concentrations. The intra- and inter-assay accuracy and precision for RE% and CV% was <12%, with a TE% < 23% for all concentrations tested. The quantification limit of 0.5 µg/mL complies with ICH criteria with a RE%, CV% and TE% of -10%, 8,5% and 15% respectively. Selectivity and specificity test fall within the criteria indicating that no individual matrix or co-medication affects the samples. The results from the dilution linearity obtained were an RE% of 11, -16 and -6 for 1:50, 1:100 and 1:200 respectively. Furthermore, not hook effect was observed as the dilution QC results gave above the quantification limit. Finally, the analyte was found to be stable at -20° and -80°for 148 days, at room temperature for 24hr and after 5 freeze/thaw cycles with a maximum RE and CV of 12 and 14% respectively in all the stability samples. Conclusions We have developed and optimized an ELISA method for measuring the antibody conjugated fraction of the ADC in serum. This method complies with the criteria of the ICH50 for selectivity, specificity, accuracy and precision, dilution linearity and hook effect in a concentration range between 0.5 - 100 µg/mL. Further, serum specimens are stable for 24h at room temperature, 5 freeze/thaw cycles and 148 days at -20 and -80 °C.