In Pitfalls to Avoid For Determining a Product’s Shelf Life Q&A, The Windshire Group’s Chief Stability Scientist discusses Excel trending and the amount of time needed to properly analyze stability data.
Q: What are some of the pitfalls that you think companies out there need to avoid related to determining shelf-life?
A: The use of excel spreadsheets is one issue that I see frequently. Many professionals try to use Excel trending in their regulatory submissions. If you have a very stable product and you don’t have ANY degradation byproducts or trends, FDA will probably sign off on that, but if you have a product that’s degrading, you can’t just use Excel spreadsheets. Most products, including biologics, have some degradation and will need a more rigorous evaluation of the data using something other than Excel.
You also have to consider the confidence intervals associated with any trends and the statistics behind it. I’ve also seen companies that only do their Excel trending on assay and related substances, and that’s because they think, well, as long as I have potency and purity, that’s good enough. That’s NOT good enough. There’s been more than one product I’ve worked on where dissolution or moisture was the rate limiting factor, stability indicating factor. You need to look at all the product attributes evaluated on stability.
So, those are the two big things that I see people make mistakes. “Oh, yeah, we’ve done trend analysis,” but it’s done in Excel. There are no statistics. “Oh well, we looked at assay and purity. It looks great.” That is not good enough.
Q: What is the proper amount of time it takes to analyze the amount of stability data that you have?
A: It depends on how much data is available. I’ve done it in as little as two weeks for the statistical analysis, but I think an appropriate safety margin for a full blown program is to plan on a minimum of six weeks from the time of contract signing to the time that you have all your reports. Even this may not be enough time if a new issue is discovered during analysis that needs to be assessed and dealt with. You can’t be two weeks from an NDA filing and believe you’re going to have enough time to do a proper evaluation of your trend analysis.
So, when I’m doing an NDA, I start considering doing all my analysis at least two months prior to NDA filing dates. Preferably six months, because even if you’re just putting your samples up?, you can at least start setting up the database. Then as the data comes in, you can start plugging the data? and evaluate early trends. In that way, the turnaround time at the back end can be as little as a week if you’ve done all your homework up front. For more information on stability data analysis and management, email firstname.lastname@example.org.
Process, Manufacturing and Stability Product Presentation Announcement – Dr. James Blackwell, President of The Windshire Group, gave a presentation on Stabilityshire– compliant stability data analysis/ reporting and other leading quality products and solutions on October 5th, 2016 in Gillette Stadium, Foxboro, MA. For more information on getting the most value from your process, manufacturing and stability data, please click here.