On Accelerating Bioanalytical Workflow Automation

Bioanalysis is critical to pharmaceutical R&D in terms of providing pharmacokinetic, pharmacodynamic, toxicokinetic, and immunogenicity data. The bioanalysis lab work itself is a meticulous, tedious, lengthy, and error-prone process if all done manually. A huge effort has been put forward to automate bioanalysis workflow, mostly in bioanalytical sample analysis production, but relatively recently method (assay) development as well. However, my observation is that in the past decade, the adoption of bioanalytical workflow automation in the bioanalytical industry has been slow. By bioanalytical workflow automation, I am referring to the continuous automation of the majority of the bioanalytical wet experiments from calibration curve preparation, QC and dilution QC preparation, study sample pre-dilution, extraction, etc., as opposed to step specific automation such as microfluidic disc-based ligand binding assay coating, binding, washing and detection steps.

Three reasons could be causing the slow adoption. First, for many virtual or start-up companies, the bioanalytical function is simply outsourced. Bioanalytical lab automation is out of their sight and hence out of their mind. Second, given the current state of commercially available automation instrumentation, extensive customization/ scripting is still needed to make them suitable for bioanalytical workflow (as opposed to step) automation. The ROI equation for the skills needed makes this level of customization unfavorable for small to medium bioanalytical organizations. Third, at established bioanalytical CROs where this level of customization/scripting could be potentially cost-effective and provides a positive ROI, the lack of internal know-how (bioanalysis + computer programming knowledge)is slowing the progress. 

Despite the slow adoption, it is beyond doubt that the need for bioanalytical workflow automation is still very clear, present, and relevant. First and foremost,for mid to large biopharmaceutical R&D organizations, some internal bioanalytical work is still necessary because of the quick data turnaround for decision making which outsourcing cannot provide. For example, non-GLP dose range-finding studies’ bioanalysis portion is often conducted internally in mid to large bioanalytical organizations. Another example of internal bioanalytical work is method/assay development. 

Design of Experiment type of systematic approaches for large and/ or small molecule bioanalytical assay development are better achieved through workflow automation. 

Secondly, at bioanalytical institutions where investment in workflow automation can be justified, either at mid to large pharma or CRO, the capacity crunch due to the pandemic-induced labor shortage could really use some help from automation. With the high staff turnover rate during the pandemic, bioanalytical institutions are facing more challenges hiring, pouring more resources in training new hires, and retaining bioanalytical institutional knowledge, all of which could be alleviated by bioanalytical workflow automation.

With the high staff turnover rate during the pandemic, bioanalytical institutions are facing more challenges hiring, pouring more resources in training new hires, and retaining bioanalytical institutional knowledge, all of which could be alleviated by bioanalytical workflow automation

“Despite the slow adoption, it is beyond doubt that the need for bioanalytical workflow automation is still very clear, present, and relevant”

The previously published successful approaches to bioanalytical workflow automation through extensive customization/scripting of commercially available general-purpose automation instrument has been proven not widely applicable during the past decade. Thus, the only seemingly realistic approach is for lab automation instrument companies to develop special-purpose bioanalytical workflow automation instruments. These need to be differentiated from the flock of niche, proprietary fluidics/ microfluidics disc-based instruments that focus on partial tasks of bioanalysis such as antibody capturing, washing, and detection. The latter has the drawbacks of high barrier of entry, high pass through costs, a high lot to lot variability of consumables, emphasis on sensitivity but not necessarily the general applicability of small and/or large molecule bioanalytical workflow, e.g. calibration curve preparation, QC and dilution QC preparation, sample pre-dilution, extractions, liquid handling, etc. The successful bioanalytical workflow automation instruments should provide the above workflow automation and be special-purpose in the sense that they address specific bioanalysis needs, but also general-purpose in the sense that they should automate the general bioanalytical workflow with varying parameters. After the pandemic, the pharma market outlook for such AI-enabled lab automation technology is brighter than ever.