September 30, 2019
Diane Carozza, WCG vice president of clinical strategic services, and Danya Kaye, director of business development for R&D and innovation at Inspire, share their thoughts on a variety of issues related to successful site feasibility practices.
Question: What are the pros and cons of conducting blinded versus unblinded initial site feasibility studies?
Carozza: When we begin speaking with either the steering committee members or key opinion leaders or thought leaders that sponsors tend to work with going through the protocol feasibility process, my answer is 100 percent that the unblinded outreach yields the better. But when we actually do site feasibility within WCG, we see a difference in months when we had those blinded outreach efforts versus the ones where the sponsor is identified. We just recently had an instance where we were able to begin and end a whole site feasibility outreach effort in a matter of two weeks, and it was because we knew that the sites were very interested in working with this sponsor that they know is doing very novel research in their area of expertise.
Kaye: At Inspire, as an online social health network that partners with pharma and CRO organizations and research services, we’re not doing site feasibility, but we do a lot of protocol feasibility, and from a protocol feasibility perspective, it’s kind of the opposite. We’re typically doing blinded outreach to collect unbiased patient perceptions of a protocol design. So, we don’t want the biases or perceptions associated with a specific pharmaceutical company to be masking the types of data that we collect on the actual protocol. The exception to that is if you’re looking to collect perceptions on competitive trials and you know, for example, that a patient community is very well aware of maybe just a few pharmaceutical companies working in that space, then maybe it makes sense to unblind it to understand where their perceptions are of your pharmaceutical organization versus some of the competitor trials.
Question: Beyond feedback on patient perspectives that you receive from sites, what other insights can be garnered from doing a more general patient insights initiative?
Kaye: When you’re collecting patient perspectives, in terms of hearing directly from sites what patients are saying, you’re really only getting one part of the story because you’re only getting a small percentage of patients that actually even make it to the site or are under the care of the specific investigators that you might be working with. There’s a whole community, globally, that can really be leveraged to understand patients in a more holistic way.
There are a couple of different common initiatives or themes of data that we collect. One is collecting patient feedback on protocol design. How do you minimize the patient burden? How do you ensure that patients are going to be willing to enroll in this study?
We can also look at a patient insights initiative as it relates to patient relevant trial endpoints. So, identifying information that is most important to patients related to treatment, benefits, risk, burden and how to best communicate information to them to support their decisionmaking. This, along with all of the patient voice insights projects or patient voice incorporation, is something that the FDA is now looking for in developing guidances.
Question: What are the most important questions to include in the site feasibility questionnaire?
Carozza: The complexity of protocols these days really necessitates that we look at experience in a different way. It’s no longer just about how many trials you’ve done, but really which ones you have done relative to the study that you are being evaluated for.
Of course, we do want to be asking about depth and breadth of experience within the therapeutic area and indication, but we also need to be thinking about expanding our pool of potential investigators and sites with personnel who are really passionate about the research and the patients, especially in some of these novel approaches.
But if we’re talking about the questionnaire itself, use the site profile information that you have with whatever partner you are working with. Here at WCG, we’re working very hard to expand the data that we’re collecting on our investigators and our institutions and facilities so that the questionnaires can really be very focused on just the study-specific information that we would need to ask in the moment.
So, of course things like estimated enrollment on the patient funnel for the target population, any concerns about the study design or how much work that a site does in the indication annual basis — those would definitely be the objective questions that I would always want to see asked in an operational feasibility process. Use and reuse as much as you can when it is facility and investigator profile-related. Consider it more of data collection than question asking.
Question: What is the best way to get a site to give actual data on their feasibility questionnaire and not just guess?
Carozza: An important aspect of this is for sites to know that, more and more, their qualification is being targeted based on the historical performance metrics that may exist in a public forum. We need to move in the direction of having sponsors come prepared to discuss the expectation based on that enrollment projection that they have worked out. I think the site has to be prepared then to be able to show evidence through their access to their EHR or to their locally kept databases. I just think there’s a risk for the site to guess, because we’re starting to all get more informed. And so now, maybe, we have to start to shift the way we ask the question, where sponsors have to share what their knowledge is, that sites may not have that perspective.
So, I challenge sponsors to be more transparent about the methodologies that they’re using to come up with their enrollment targets, and sites to be extremely honest and transparent. Because I’ve been doing this a very long time, and I will tell you, every time the sponsor will say they would rather have a site say that they can get five patients, and they turn out to be five valuable patients that will make it to the end of a trial, than for them to claim that they can get 10.