5 Takeaways From our Fireside Chat with Sanofi

There’s a lot of uncertainty in clinical trials. Which sites have eligible patients? How many can they enroll? This site performed well last time, does that mean they’ll perform well again? What if they’re running five other trials like mine?

AI can now unlock real-time patient access data that addresses many of these questions. At SCOPEX, our cofounder and Chief Strategy Officer Liz Beatty sat down with Sanofi’s Global Head of Clinical Project Operations & Dossier Delivery Kim Hawkins to discuss how we’re working together to leverage new real-time data to reshape feasibility and cross-asset planning and drive faster, higher enrollment across the portfolio. 

Here are five of the biggest takeaways from their conversation. 

‍

1. Feasibility has a trust problem. Real-time patient data can help fix it.  

One of the biggest challenges with traditional feasibility is that it’s always been nearly impossible for sites to know how many patients they can enroll. It would take hundreds of hours of chart review to find out—so sites estimate, and sponsors treat those estimates with skepticism. 

Now, sites can use AI to review patient charts, find out exactly how many eligible patients they have for a trial, and confidently share that data with sponsors up front. “We're moving away from the days where it's a feasibility questionnaire, you're writing down how many patients you think you can enroll, we take it at the sponsor [side], we cut it by a third, we cut it by half, whatever it is. We're seeing a more robust predictability, which is a win-win for both sides,” said Kim. 

What this means: sponsors choose sites based on real-time patient need, sites enroll quickly and predictably, and patients get access to the right trials for them. 

‍

2. Real-time patient access gives sites a boost in a highly competitive field.

Non-performance is costly, and sponsors are reluctant to take a gamble on sites that may not be able to enroll. The ability to prove patient access up front can mitigate these concerns and make a site more attractive in feasibility. 

“From a sponsor's perspective, you can imagine having the ability to verify what sites are saying with real data could certainly give a site a very competitive edge in terms of being selected for a study. There's so much competition out there these days, that can really make a big difference,” said Kim. 

‍

3. New data won’t just redefine site selection on individual trials; it opens up new possibilities for how sponsors plan across assets.

Sponsors also face increased competition finding sites for their trials—not only from other sponsors, but from other trials internally. Study teams don’t always have visibility into other studies internally, which makes it challenging to strategically partner with sites across assets.

Liz and Kim spoke about how Sanofi and Inato teamed up to take a different approach, moving away from a trial-by-trial view and toward a portfolio view. It started with a respiratory program: Sanofi shared a handful of upcoming respiratory studies and invited sites to share patient access data to be selected for one or more opportunities. The result: Sanofi saw 3x the site interest by sharing multiple studies and unlocked new data early in their process to help them partner with sites across assets based on real patient need.  

‍

4. Results > rhetoric

“Can we shorten that time in between activation and first subject screened?...Can we increase the productivity of sites? Can sites effectively enroll more patients than they would have before they had those tools available to them?” 

Kim opened the conversation by saying Sanofi doesn’t adopt AI just to try something new. So when it came to her work with Inato, she wasn’t interested in theoreticals: she wanted to know if together we could actually bend the enrollment curve, accelerate trials, and contribute to Sanofi’s goal of halving development time by 2030.

‍

And the (initial) results are in. Liz and Kim shared a recent COPD case study where sites used AI patient pre-screening to streamline screening and enrollment. Some key results:

• Sites using the tool screened their first patient 33% faster (17 days sooner) compared to non-AI users
• 100% of sites who adopted the tool are actively enrolling
• 80% of sites adopting the tech are high-performers
• And one site–after 108 days of no screening activity–turned to Inato and screened their first patient just 8 days later

‍

5. Efficiency is a major benefit of AI implementation, but it’s not the only measure of success.

A lot of the conversation about AI is focused on efficiency. And efficiency matters—we all want to bring breakthrough therapies to more patients, sooner—but it isn’t the only value AI unlocks.

Liz shared that sites using AI patient pre-screening have a queue of thoroughly pre-qualified patients ready to go as soon as they’re selected. So they’re not only ready to hit the ground running, but they’re also seeing much lower screen fail rates.

“This is a huge efficiency gain for sites, but actually what we hear from them is that it's making the patient experience better. There's a lot of disappointment when you bring in someone to your center, they think they might be a good fit for the trial, you're excited to talk to them about it, and then you realize going deeper into their chart and their history that actually they don't qualify.” 

‍

Sponsors: get in touch to learn more about how you can use real-time patient data to strategically partner with sites and accelerate enrollment across your portfolio! 

‍