High-Throughput Automated Data Analysis workflow for ADC Biotransformation Characterization

Antibody-drug conjugates (ADCs) have garnered renewed attention after decades of research to fine-tune their components, maximizing efficacy while minimizing toxicity. Understanding structural integrity of ADCs in vivo is critical for lead optimization. Despite of its importance, currently ADC biotransformation has not been broadly studied. High resolution accurate mass data analysis of biotransformed products has been a major bottleneck because of manual and time-consuming analyte identification process which can take days to weeks of analysis time by expert users. We developed a streamlined data analysis workflow that enables automated peak identification using various commercial software tools significantly improving data processing efficiency. A linker-payload biotransformation library was created for each new molecule and then combined with antibody sequence information for peak matching. As a proof of concept, we tested this workflow across different payload and linker types: an example using a topoisomerase I inhibitor-conjugated ADC and a comparison to a historical in vivo ADC biotransformation dataset for a pyrrolobenzodiazepine-conjugated ADC. Using this automated workflow, we were able to rapidly identify major biotransformation species that were previously found manually including loss of linker-payload, maleimide ring hydrolysis, cysteinylation at deconjugated site and partial linker-payload cleavage. Overall, this improved data-analysis workflow has demonstrated its superb effectiveness in streamlining ADC biotransformation identification and enabled quantification that was highly comparable to previously obtained results, thus demonstrating its comparability. This advancement can now positively impact drug development by substantially reducing data analysis time thus enabling faster design-test-analyze cycle times which are critical in early drug discovery setting.