How to improve CMC productivity through digital lab transformation – your questions answered

In today’s pharma industry, the growing complexity and diversity of therapeutic product lines as well as the acceleration of clinical trial processes, has made the organization of chemistry, manufacturing, and controls (CMC) increasingly critical on the path to commercial product launches. Speeding innovative and successful new drugs to patients will therefore require the end-to-end, digital transformation of CMC processes.

In this free SelectScience^® webinar, now available on demand, BIOVIA product manager at Dassault Systèmes, Dennis Curran, discusses a digital laboratory solution that enables cross-domain integration and real-time, consistent, synchronized, and standardized information flow from R&D to manufacturing.

Read on for highlights from the live Q&A discussion or register now to watch the webinar at a time that suits you.

Can you handle all the recipes from lab to manufacturing?

DC: Following the concepts I mentioned previously, a recipe has a construct of materials, process steps, and equipment dependencies. The ability to handle all those recipes is enabled by our ability to have master data, have our customers define, ‘here's my specific site, my specific equipment’, and map that to it.

The process definition is: ‘Here are my vocabularies to do these, this set of operations.’ We have standard ones, but when a customer has a specific recipe that may be out of the box, such as a new domain that we haven't hit yet, we can build those process definitions, those process parameters and setpoints.

So yes, we can handle all the recipes from lab to manufacturing — we have a large variety out of the box. If there's a standard ontology, we want to use that because that's faster dissemination of knowledge, but if there's a specific new area, new domain, what we do is we build the foundation first.

Here's the process definition, here are the dependencies, then we'd layer on the process steps on top of that. Our target would be to create repeatable stages so that we can lower the burden of development, validation, and test. As I mentioned, we're looking for areas of commonality, and I called out the USP as a specific example, where we want to target having all of the 5,000 USP methods available out of the box.

Next, we're looking at the ASTM methods for customers who might have different material testing needs. We've evaluated those, and we know we can build those ASTM methods as standard, out of the box. We are looking at USP content, we are looking at EP and JP, Japan Pharmacopoeia, and building those into the system.

So, yes, the system is designed with the extensibility to handle all of the recipes from making substances, end products, tablets, combination products as well as testing those given the S88 structure.

Can we use JMP, GraphPad, etc., through BIOVIA?

DC: Yes, we can absolutely support that exchange. From a design of experience standpoint, JMP, GraphPad, we have a lot of integrations with that type of thing. SAD allows me to identify these parameters. When I do a design of experiments and I want to vary perimeters, I can send those parameters or that definition off to that third-party application, and then bring back the number of experiments or the new setpoints.

On which data format are you standardizing?

DC: We're committed to supporting several different formats. It's not a singular answer. There's so much variety globally, especially in the data formats. We're completely supportive of Allotrope’s data standardization initiative around analytical devices, the Allotrope Data Format (ADF). We have the standard ability to write and read in that ADF format. But in many cases, the data exchange could be simpler than that. So, we also support basic AnIML or XML exchanges. If a transaction needs to be fast, an exchange with an instrument, then CSV is also something we do a lot of. We're committed to supporting multiple standards. We really like the ADF format, but there are other competing standards, and we will be supporting those for the foreseeable future.

LC-MS involves data collecting, data analysis, and report. How is this type of experiment handled?

DC: This is that third type of data exchange that I mentioned. It's not just a rich file export, but there's a lot of rich information in that. For BIOVIA, we don't want to replace the capabilities that the LC-MS data analysis software has, we want it to continue to do its job. We want to exchange, and provide it with any method information, any sample information. But then, when the experiment is completed, we want the results back — allow the LC-MS software to do what it does well, and we might want the aggregation and the results. We could take this in the format of sample IDs, aggregated results.

For LC-MS specifically, we have a lot of customers who want to see traces from the spectra. We can store those as pictures. If that data stays in the LC-MS system, we have a hyperlink that allows this. We'll take the sample ID according to the material, and then the results, the identification, any sort of specific information that LC-MS gives us, and then have a hyperlink back to that source system.
If that source system doesn't persist, we're going to put the raw data, the instrument data, in a secure location that we can hyperlink into for any follow-up investigation.

To learn more, watch the webinar on demand >>

SelectScience runs 10+ webinars a month across various scientific topics, discover more of our upcoming webinars>>