Ahead of his session at Pharma IQ Live: SmartLab Digital, we asked Henry Charlton, Commercial Director in Biologics at IDBS, about moving away from paper-processes and making data capture and analysis in the lab more efficient.
Pharma IQ: What are the main challenges when it comes to collecting and analyzing data?
Additionally, lots of organizations still use quite old-fashioned methods for archiving, so there is a reliance on paper and Excel in approximately 50 percent of biopharma development companies. That means you need an archivist, a physical location in which your laboratory notebooks can reside, and a system for determining which experiment is stored and in what location, so that years later people can go back and find those experiments.
This can be a problem. Often the experiment and the person who performed it are kept in the corporate memory of the individuals who work there. If they leave, the retrieval of that information can be very hard, almost impossible. Further down the line this can lead to the need for re-work – repetition of previously performed experiments. At that point, when regulatory information is being brought together with the preparation of the biologics licence application, it can delay the preparation and submission of the documentation which in turn results in lost revenue of high-value products, costing hundreds of thousands of dollars for each day of delay.
Pharma IQ: Why do so many rely on paper processes given that this is not very efficient?
HC: One reason is familiarity. Moving away from the old ways of doing things – whatever they are – represents a challenge for organisations. Systemic change within organizations is normally driven by a shock or sudden wake-up that reflects the poor fit of an existing process. There must be hundreds of publications that look at organizational change and change management must be part of a transition to a more intelligent and efficient data management method.
Increasingly we are seeing that people have moved away from paper but haven't yet addressed two of its core problems, which is how to have compliant data capture and retention and leveraging it to gain insights from it.
Pharma IQ: What can be done to make data capture and analysis more efficient?
HC: One approach is applying FAIR data principles (Findable, Accessible, Interoperable and Reusable). This basically democratizes the use of data within organizations. It means that your data management system will allow you to access and find data, and you can reuse it so that it's not lost forever.
There are certain elements that need to be embedded into the quality of that data to ensure this happens. One of the terms we quite often use is “contextualized data”. This means that, if something is in a flat file, even if you search for it, a machine can’t associate any meaning with the data you have found. When the data is contextualized, you have an opportunity to retrieve it more effectively and understand what it is describing.
This means collecting information in a consistent format. An example is Excel, which has numerous different date formats using combinations of letters, spaces, numbers, hyphens and full stops. That can make searching an Excel file very hard unless everyone has agreed to use the same protocol and you know exactly which date format everyone is using – the likelihood is that without an overarching system, they won’t do that - people express their preferences and use different date formats across an organization.
There is also the opportunity to move away from siloed datasets, by which we mean that the machine where the experiment is performed becomes the data storage system. If somebody uses a chromatography system, then every experiment they perform on a certain molecule might well be recorded on that operating system, but it wouldn't necessarily communicate with other platforms across the laboratory, so it effectively stands in a silo and that can make it very hard to find for months or even years later. If data is taken out of that piece of equipment, that's very often done manually, which is prone to error when it's transcribed.
One way to solve this is to have a data management system that can communicate with all the instruments and equipment used in the development laboratory, bringing the data into a central point so that users can then go back and find it based on which experiments they performed and when.
Pharma IQ: What are your predictions for how smart labs will operate in the future?
HC: I think we’ll be looking at things like predictive maintenance so you can eliminate inefficiencies like instrument logbooks and having call outs for when machines need servicing or if they're falling out of calibration. You can also think about the supply chain of the instruments themselves, so that reagents are ordered automatically rather than going to the fridge and finding that you haven't got the right buffers.
There will be increased use of voice-powered technologies such as LabVoice, for example using dictation to instruct lab robots or record results. Another area that we focused on with our Polar HTPD solution is high-throughput process development, whereby you can screen very large numbers of cell lines within a bioprocess with a very small volume of material, but do far more experiments than you could a decade ago. In that way, you get much closer to understanding the process and hopefully identifying those cell lines and process steps with the highest yield, and therefore the most robust process with lowest cost of goods for the manufacturing process.
SmartLab Digital 2022 will feature speakers from Novo Nordisk, GSK and IDBS explaining the importance of integration and how to leverage AI, automation and the cloud to ensure your lab is agile, scalable, and fully equipped for the future. The event takes place on March 8–9, 2022. For more information and to register visit SmartLab Digital 2022.