A sensitive and robust bioanalytical assay for pharmacokinetic analysis of ROR1xCD3 bispecific T cell engager (NVG-111) in a first-in-human study.

Abstract

e19505 Background: Receptor tyrosine kinase-like Orphan Receptor-1 (ROR1) is widely expressed on hematological and solid tumors. NVG-111, a first in class humanized tandem scFv ROR1xCD3 bispecific antibody elicits potent killing of ROR1+ tumor cells in vitro and in vivo. This bispecific T-cell engager (TCE) is being evaluated in a first in human, Phase I trial in patients with relapsed/refractory chronic lymphocytic leukemia (CLL) and mantle cell lymphoma (MCL). The predicted therapeutic dose and steady state serum concentrations of NVG-111 were estimated by allometric scaling using relevant doses from a murine PK study. To assess free drug levels in patients following 21 days of continuous infusion of NVG-111, a bespoke, sensitive pharmacokinetics (PK) assay with high levels of specificity and sensitivity was developed. Methods: Anti-idiotype (anti-ID) antibodies directed to anti-ROR1 (αROR1-ID) and anti-CD3 (αCD3-ID) were generated by mouse immunization or by phage display from customized libraries. A proof-of-concept sandwich ELISA assay was developed using αCD3-ID to capture NVG-111 and detection by biotinylated hROR1-streptavidin-HRP. Gyrolab and Quanterix Simoa high sensitivity ELISA platforms were used to detect NVG-111 by αCD3-ID capture and αROR1-ID detection. The mesoscale discovery electrochemiluminescence assay (MSD-ECLA) was developed using a reversed format; NVG-111 capture with αROR1-ID and detection with αCD3-ID. Results: Allometric scaling predicted a theoretically relevant therapeutic dose and steady state serum concentration of 1ng/mL NVG-111 in humans, which was just at the level of sensitivity of a conventional ELISA under non-matrix conditions. Transferring the format to Quanterix Simoa had limited success due to high background levels in all configurations evaluated. The Gyrolab platform increased sensitivity to 75pg/mL, but suboptimal individual human sera matrix selectivity limited assay validity. Assessment of MSD-ECLA provided the best signal/noise, enhanced human disease and healthy sera selectivity, and a dynamic sensitivity range of 250pg/mL to 32ng/mL, which enabled the development of a GCLP qualified PK assay. The MSD-ECLA assay was employed to measure NVG-111 concentrations in CLL or MCL subjects dosed with 0.3-30µg/day NVG-111. MSD-ECLA detected drug in patients receiving NVG-111, with a range of steady-state serum concentrations (Cavg.ss) of 168-610pg/mL. This was in-line with the predicted drug levels from the single species allometric scaling, albeit with observed levels being marginally lower than expected. Conclusions: Development, custom optimization and validation of a highly sensitive MSD-ECLA PK assay has enabled GCLP-compliant measurement of circulating NVG-111 in CLL or MCL patients treated with at least 10µg/day cIV NVG-111.