Back at the end of 2025, I was asked to join a Pistoia Alliance meeting to talk to them about SEND and efficacy and data re-use. To misquote Jerry Maguire, “You had me at SEND.” I’ll talk to pretty much anyone about it, so I happily jumped on a call despite knowing very little about what I was getting myself into.
What I found was a group of people who had looked at what had been achieved through standardizing toxicology data and thought, we’d like some of that. Not because they needed SEND for a regulatory submission, but because standardized data have value in their own right.
While I’ve long believed this, it was genuinely inspiring to hear others look at what we have accomplished with toxicology data and want the same benefits for themselves. For years, I’d faced resistance to SEND, with many organizations adopting it largely because the FDA forced their hand. Yet here was a group who actually wanted standardization, because they could see the value in being able to bring studies together, search and compare data, and ultimately uncover scientific insights that are much harder to find when those data are scattered across different formats and systems.
Oh, and there was also an increasingly familiar two-letter acronym involved: AI.
The team was interested in how AI could be used with efficacy data and saw standardization as an important part of that. After all, if we want AI to help us search, analyze and identify patterns in scientific data, giving it consistently structured data seems like a pretty good starting point.
There was, however, one slight problem. While I know a thing or two about SEND, I knew next to nothing about efficacy data.
Fast forward around 12 months and I’m now deep into the project, working with people who do understand efficacy data as well as some fellow SEND nerds, and trying to work out how oncology efficacy results could be represented using SEND.
We’re now working towards publishing what we’ve developed. Importantly, this isn’t a CDISC initiative. The idea is to do the heavy lifting first, work through the science and demonstrate that the approach can work. Personally, I’d then love to see CDISC pick it up, because that brings the industry recognition and acceptance that comes with a formal standard.
So, hopefully in the new year, look out for something from the Pistoia Alliance regarding how to represent efficacy studies in SEND.
次回まで
マルク
