Sites To Sponsors: Have A Little More Trust

A conversation with Atlas Clinical Research CEO Mark Scullion

We talk about trust within healthcare as it relates to providers’ ability to establish trust with patients. Specific to clinical research, we often talk about trust in the context of pharmaceutical companies engaging with underserved and often minority communities, to encourage their participation in trial studies. However, given cases of historical malpractice (think Tuskegee, Henrietta Lacks, and others), hesitancy and skepticism are warranted.

Yet still, we don’t often discuss trust as it’s shared among drug development partners. While conference programs and paper topics abound in strengthening partnerships among sites, sponsors, and CROS, specifically by improving communication, trust as a mechanism for improvement doesn’t pop up nearly as often.

This is precisely what Atlas Clinical Research CEO Mark Scullion thinks will be the difference in executing future clinical trials more efficiently. In this interview, Scullion suggests that trust should be the foundation of a proper site-sponsor relationship and posits that site networks are at an advantage to do just that.

Clinical Leader: What roles do the sponsor and the site each have when capacity planning for a trial?

Mark Scullion: Fundamentally, sponsors decide the number of patients they need to address the scientific question the study is set up to answer. Then, the sponsor, in some cases a CRO, selects the number of sites they need to achieve the patient count. They'll then ask sites to tell them how many patients they could enroll.

Now here's where it starts to get interesting. The sponsor or CRO might not think those numbers credible and might discount them. Historically, a sponsor or CRO might take a certain percentage off that number, and this is where the divergence can begin.

The site then develops its own recruitment and engagement plan. Management of capacity at a site is a balance of having the right number of trained people at the right times to crucially give really good service to the patients and collect quality data. Sites will plan their resourcing against the work they have to do, and the work they have to do is determined by the number of patients. Therefore, it can be a disappointment for the sponsor if the site doesn't do what it has said, or a disappointment for the site if the sponsor reduces the number of patients because it does not trust the number the site gave it.

You’ve hit on an age-old sponsor belief that sites oversell their ability to recruit patients. Is that still happening and, if so, how can we get it back to a baseline where the number given by the sites is a truthful estimate?

You're right to use the word truthful because the perception is that sites are not being truthful. However, in my experience, that's not the case.

On the CRO and the sponsor side, you expect a reasonable number of patients per site, per month. Sites are expecting the same thing; they know their patient populations and believe they can recruit a certain number of patients. Things can change, though. From the time the site expresses interest until the study launches, there might turn out to be a competing study for example. And so there are reasons why that study becomes more difficult to enroll and it ends up doing less. Obviously, protocols are often amended along the way, too and those changes may impact the ability of a site to enroll.

On the other side, to be fair about it, the site very much wants to participate because it's interested scientifically and thinks patients will benefit. So, there may be some cases where the site says, “I think I can figure that out. My numbers say five; I'm being told seven, and I think I can do seven.” It might end up falling short, but I don't think it's about candor. I think it's about being inexact. If it was an exact science, we wouldn't have any trouble.

What would the ideal situation look like between the site and sponsor/CRO to execute capacity planning in a way that satisfies both parties?

Raising the level of mutual trust as the first thing. If the sponsor says “We need this many patients and the site truly can commit to it,” can the sponsor give the site what it needs? If we could then get more certainty as a site, what are the expectations? How much time do we have to do it, and do we have the financial support necessary to do it? Once that is established, the trust becomes a little bit stronger. We speak a great deal of candor, transparency, and open communications – if we have that, it helps, but it's not enough.

“The sponsor” is often referred to like it's a single entity, but there is no such thing as “insert sponsor name”. There are different teams in different therapeutic areas working similarly, but not always exactly alike. So, if you're at a site, you may have participated in studies for that specific sponsor prior and still get asked the same questions about facilities, fridges, storage, or other things. In short, you're burning a lot of time and effort inefficiently. It would make sense, particularly where a sponsor has worked with a site or network frequently, if we can move away from repeated qualification visits. For example, if we can get to a point where sponsors can quickly say to a group of sites upfront, “We want you to recruit X percent of the portfolio or Y number of patients,” and the network figures it out and does it, that would take away some of the inefficiencies and redundant resources.

You’re talking about portfolio-based allocations. How would that work?

With a standard study, a sponsor may select 50 sites and then do 50 qualification visits, 50 budget negotiations, and all the rest. If they had instead decided to choose 25 sites within a site network, there's a more strategic approach because you're distributing the clinical studies across a group of research sites that have proven capabilities, performance histories, and resource availabilities. That method allows the site group and the sponsor to collectively reduce costs and increase speed because, with fewer active sites, that equates to a lower cost for selection, qualification, and activation.

That's direct cost avoidance savings for the sponsor on a portfolio allocation basis, too. It gives sites leeway to pre-identify patients with certainty that they're going to be participants in that study and to expedite the fulfillment of enrollment targets.

This matters, especially given the recent changes in guidance on the diversity profile for trials. It isn't possible to just go to an underrepresented population after you've received a protocol. You have to put in the work over a long period of time to build trust in those communities. You can do that if you know that certain studies might be a match. In return for that more defined portfolio allocation, the site offers more predictability and efficiency.

You might think that seems highly aspirational and improbable, and that sponsors are not going to do that. I’m trying to set us in that direction, for sponsors to put more faith in what the sites believe they can do — or they wouldn't have said it. If you can do that as a sponsor community, a little bit like CRO-sponsor relationships of 15 years ago — with aggregation comes more innovation, more mutual trust, and, hopefully, a better experience for everyone.

For each study and each site, a sponsor company collects information on site feasibility, making their efforts duplicative and time-consuming. How does a site network infrastructure alleviate some of the burden?

A typical large pharma company is operating between 1,000 to 2,000 sites in a region. You can’t really have a relationship with that many sites, but if there's a grouping of sites, you can have that relationship. The sponsor has assurances of SOP oversight, quality management, and infrastructure. It is a little like working with a CRO across a portfolio: a sponsor wasn't always necessarily comfortable but eventually got assurance that they knew how the relationship was going to work because it was tailored to them.

With a site network, it’s the same thing. If we can have greater strengths of process and technology, resource management gets easier. It also enables us to bring in new physicians. If we say on the front end that these are standards for data flow, quality, oversight, and staff retention, then we can say, “Trust us to do this.”

It's a managed risk. It’s like when a sponsor opens a new country that the study team has never worked in before. You're still under the same oversight team with the same quality and oversight metrics; they're willing to try it, and once they do try — with strong performance — they are willing to do it again.

A couple of weeks ago, the FDA revised its draft guidance on the diversity action plan. Have any sponsors talked to you yet about a diversity action plan and how you might help fulfill it?

As a community, sponsors recognize the need to have a more diverse representation. Even going back to the last FDA guidance, they're saying to sites, “Are you sufficiently familiar with your community? What are your modalities for bringing in patients from more diverse backgrounds?” And maybe two or three years ago they would've said, “We need X percent diverse patients on this study.” Now, to a greater extent, they're selecting based on past performance — have you recruited a broad, diverse population?

However, if a group like ours spends time, effort, and money to build relationships in a community, and then in a year no studies arrive or they are canceled or capped, it's almost worse than having never asked. If you're going to build those relationships and tell people they should participate, and then you don't deliver on it, it becomes difficult.

We've got underrepresented populations for several reasons, in some cases, because the industry just hasn't put too much effort into it. But there's also a significant trust hurdle for historically good reasons. So, we have to first establish trust within our underrepresented communities. That's going to take some time and if we aren't looking after that trust and nurturing it sufficiently, it's going to backfire.

Understanding the diversity action plan and the data that the sponsors have to collect and submit to the FDA, does that require anything different from the site level?

Historically, the site might not necessarily have looked at the number of patients being prescreened and screened to see whether there was a diverse portfolio. Going back two or three years, have increasingly done it but still filled their allocation as quickly as possible to meet the sponsor’s timeline. Now, we're doing both of those things. We're trying to discover what that prescreen pipeline of patients looks like and prioritizing balance as it goes, rather than quickly getting the first X number of patients in and then doing the balancing after the fact.

About The Expert:

Mark Scullion is the CEO of Atlas Clinical Research, a therapeutically driven clinical research site network built on centralized process excellence and integrated technology. Mark has spent almost 20 years in the clinical research industry and brings a unique understanding of the site selection process and value needs of pharmaceutical sponsors. Prior to taking the helm at Atlas Clinical, he served as the global head of trial monitoring for Novartis AG and worked for 15 years in CRO.