(1249-B) Comparison of Increasingly Automated Versions of an Electrochemiluminescence Assay Method for Regulated Assay Support

Abstract: As technology continues to advance, it is becoming increasingly necessary for assays assessing vaccine candidates lasting multiple phases of clinical trials to adapt. A common ligand-binding serological assay for measuring the immune response to these candidates uses the electrochemiluminescence (ECL) MesoScale Discovery (MSD) platform. Over four years a single ECL assay of an investigational multivalent vaccine was followed through three qualifications, two studies, and several preliminary experiments, advancing innovative automation technologies at every step.

Standard manual bench executions, integrated liquid handling centralizing around Hamilton StarPlus and Star systems, and an integrated robotic platform have been developed. Qualified methods showed many automation benefits including eliminating >5,000 pipette transfers for 128 samples per run and increased longest walkaway time from 25% to 75% from the Hamilton StarPlus to Star methods. Studies were performed manually and using the Hamilton Star method utilizing a validated, in-house LIMS for real-time clinical sample tracking and data analysis of all serotypes in < 15 seconds. The automated method reduced sample turn-around-time from 6 weeks to 10 days, expediting the vaccine candidate lifecycle. The incorporated sample dilution flexibility for retest, reflex, and repeat testing enabled 100% of clinical sample processing to be performed on the Hamilton Star system compared to the possibility of only 10% being performed on the Hamilton StarPlus system. Preliminary experiments assessed pre-diluted sample stability and more advanced instruments for an integrated, multi-equipment robotic platform controlled by an in-house built driver. This enabled process enhancements to potentially improve throughput to 192 samples without sacrificing the 75% longest walkaway time. Throughout this assay’s lifespan, many challenges have been overcome. The stigma of automation not equating to manual processing was disproved as clinical samples fell within 1.4-fold when compared using Deming regression analysis between manual and Hamilton StarPlus methods, equipment flexibility allowing for the option of integrated versus separate incubator to be used for the Hamilton Star method, and automating based on limited reagents and sample volumes. This poster compares the evolving workflows from manual to increasingly automated ECL assay used in a regulated environment.