“Failing to plan is planning to fail.”
This describes in one sentence why clinical trial feasibility is so important for having a successful clinical trial on budget, on time. The challenge is it’s not so simple.
Clinical Trial Feasibility is a process that usually takes a lot of time and data. Sponsor companies spend on average a few months on this step. Where CROs are usually pressured to conduct the indication feasibility (competitive landscape, patient prevalence, recruitment rates etc.) in a matter of approx.10 days and then site feasibility (sites’ previous performance, ongoing status, and willingness to participate in the study) in no more than a month or two. And these timelines are the short answer to:
What’s wrong with Clinical Trial Feasibility?
Let’s have a closer look at the main feasibility components and the data behind them. This is another reason why we should rethink our feasibility strategy.
Indication Feasibility:
1.1 Competitive landscape.
The way to find this out is to look at the landscape of other clinical trials in the same indication and eligibility criteria (patients’ profile). Analyzing clinical trial registries gives you a good high-level idea of where trials are being conducted.
If you are looking to see such a competitive clinical trial map you can reach us out and we’ll be happy to provide it for any indication/TA.
Competition though comes not only from other clinical trials. If you need patients enrolled, then all available treatment options and what local doctors and institutions prefer are also competitors. Most of the time even bigger than other clinical trials.
Failing to understand the local standard of care is a huge gap in any competitive landscape analysis.
1.2 Patient prevalence.
I wish we had unified EHR standards and access across all countries. This would save us so much time and effort in the drug development space. I know this is a utopia at the moment. Yet, I also know someday we’ll realize that by sharing some of our personal data (in a secure way) we can get to real discoveries about how to be healthier and live longer.
Back to reality. For most indications, we have publications and observations on the prevalence of a disease in different countries. They provide approximate numbers of how many patients there are with a certain disease.
Not to mention that most trials don’t need just any patient with the disease but very particular ones. The more we go into gene therapies and other personalized treatments, knowing how many patients and where they are will be more difficult.
We can either stop relying on this metric and find a different way to conduct clinical trials (a more decentralized way) or try to get into consideration when consuming prevalence data.
TrialHub has prevalence data for more than 5000 indications in 70 countries. Let us know if this can be helpful.
1.3 Recruitment Rates.
This may be at the center of what’s wrong with clinical trial feasibility. Recruitment Rates or also known as enrollment benchmarking should provide an overview of the expected speed of the clinical trial.
Similar to competition, clinical trial registries can be used for gathering historical data and calculating the recruitment rates. Yet, this is not always accurate as you first need to make sure all clinical trials that you are looking for benchmarking are similar to your protocol.
The other thing is that there are other criteria such as timing and country selection. One clinical trial might have performed quite well but maybe back then there was less competition or fewer treatment options. There are also other “if’s” when considering recruitment rates.
One recommendation: try to look at all trial performance indicators – recruitment rate, number of sites involved, countries involved, number of days in delay, etc. And all this in retrospect (when the trial started and when it was completed).
These indicators will give you a good high-level picture of the possible pitfalls related to the performance of the future clinical trial. This is also something you can find on TrialHub.
Site feasibility:
The same old (but gold) site surveying approach. Why I say gold, because so far this is the only way you can get information about the motivation of the site, availability of patients, and other important aspects for the sites to be able to participate in the clinical trial. That is unless you are working with a site consultant who knows the sites and their capacity and performance.
The challenge is that sites either reply too late or never at all to your surveys. It helps for CROs that have built a close relationship with the sites and can push for a quicker response but what about the ones that have not yet?
The other challenge is how accurately the site survey is answered. Most surveys are filled by investigators or study coordinators who at the end of the day are incentivized to conduct clinical trials at their site, especially if this is a specialized research center and this is their primary business.
Again, I recommend a combined approach:
- Get information about the availability of patients. At least about the patient journey – do patients go to this site to get treatment/diagnosis for the indication?
- Get information about their research potential based on historical data + other trials performed at the site at the moment. Have they performed similar clinical trials?
- Conduct a short interview/call with the investigator to see what they think about the protocol. Ask their perspective on what could be the challenges for recruitment/retention.
Clinical trials are “black swans” and no projections can be fully accurate. But trying to get clinical trial feasibility right by understanding the limits and overcoming the data challenges is a must. Otherwise, companies risk losing money and time on clinical trials that not only get delayed but fail to provide the necessary proof that they are onto a new promising treatment/diagnosing solution.
