In this section, I'm going to be talking about how the design of a clinical trial influences recruitment and site selection. One of the most important things that affects recruitment is the reason for doing the study in the first place. Trials of experimental treatments for regulatory approval, sometimes have an advantage because the only way you can have access to that treatment is to join a clinical trial, that is especially significant for serious condition that don't have treatments. Again, COVID-19 is a great example of a disease that had no treatments and no preventative treatments. It was easier to enroll people in those trials because there was no approved treatment. Because the pandemic was really gripping the world, there was a strong societal need for a vaccine and push for vaccine. Where as when you're doing comparative effectiveness trials of approved drugs, those drugs are available to patients outside of the trial, even if they haven't been approved for that particular indication. Position in the US is allowed to use most licensed drugs for any indication regardless of the regulatory approval. For example, corticosteroids are commonly used for all problems from poison ivy to auto-immune diseases. Yet if you actually look at the label for a corticosteroid drug, most of those conditions will not be on the label. However, insurance coverage can be an issue if a drug is not approved for a particular indication. Non-inferiority trials are particularly challenging because there is an established control that we know is affected and we're trying to show that the experimental treatment is at least as good as a control. What makes that experimental treatment attractive to patients or clinicians? Is there something about the experimental treatment that may have advantages over the control treatment? For example, is there an easier route of administration, or a better dosing schedule, or fewer side effects or lower cost? Another important factor is, it was invested in the success of the trial. You would expect investigators to be invested, but is this trial important to patients? Is it important to other providers? Will this provide evidence for their practice or their advocacy groups that are promoting the trial? Is it important to the founders or potentially for insurance coverage? Those are all related to whether there are unmet needs for a particular disease that they are not good treatments, whether providers may have a number of treatments to choose from, but no evidence as to what is the best treatment. Trials can also be critical in deciding what treatments will be covered by insurance. The objectives of the trial flow down into the actual design of the trial. How are we going to design this trial so that it's acceptable to patients and providers? Based on the ethical principles that should be equipoise, that the benefit risk profile of the control treatment and the experimental treatment are not known to be different and there's a genuine uncertainty about which treatment is better. What is the control group going to be? Is going to be an active control group so that all participants are assured there getting active treatment, or we're going to use a placebo control or delayed treatment, which may make some people anxious and feel like they would be getting good medical care if they joined a trial. Sometimes we can get around that problem by having the new experimental treatment combined as an adjuvant with the existing standard of care versus not having any treatment at all. In that case, we would compare the combination of the standard of care with the new treatment versus the standard of care alone. A critical issue in the design that has significant influence on recruitment is whether we decide to mask, or blind the interventions. Masking means that we disguise the treatments so you can't tell the difference between the control group and the experimental group. That's a hard sell for people, especially when we're talking about placebo controls. It makes people uncomfortable to think that not only do they not know their treatment, but their clinician may not know what treatment they're on. Obviously, sample size is a huge factor in recruitment efforts. How many people are we going to need to recruit for a trial? Sample size is dependent on a number of factors including what do we choose as the primary outcome? How big are the anticipated differences between the control and the experimental treatment? Are there important sub-groups that we need to analyze and have sufficient numbers to get a credible answer for a specific group of patients? How much do we think we're going to have to inflate the sample size that we calculated that we need to do the trial for missing data or incomplete data? Another design factor that has a critical influence on recruitment is the specific eligibility criteria used for the trial. Eligibility criteria, again, relate back to ethical principles that we need to limit the risks and make sure there's a potential for benefit for patients who enroll in the trial. We also need to monitor eligibility criteria to see which ones turn out to be barriers for the trial. Finally, another design element is the protocol or how often are the patient's going to need to come in to be seen for the trial? What procedures are going to be done? Are those invasive procedures that involve risk? Are they procedures that are above and beyond clinical care so that they are done solely for research and don't influence the patient's clinic care? How long are we going to ask a participant to be in a trial? What's the length of follow-up and how is that going to impact their lives? Other factors that are related to the design include recruitment sites. Is this a single center trial versus a multicenter trial? What type of recruitment site is it? Is it a site where people receive clinical care or perhaps a specialized site where experiments are done that's commonly used in phase 1 clinical trials or sometimes in clinical trials that involve comparing diets? People may go to specific research centers. Resources and personnel available for the trial are an important factor in the ability to recruit patients, in terms of how much time do personnel have to devote to this, and what are the incentives for sites to enroll patients in a particular trial? What's the incentives for individuals? Very often those are financial incentives, but there can be other incentives. Finally, what's the consent process that we designed for the trial. This is going to be talking to the patient. Is that a trusted individual? What kind of document have we produced? Is it a 20-page technical document that no one's going to read or are we really making the effort to describe the trial in language that most people can understand and what's supporting materials and are there to explain what the trial is about? What the patient's commitment to the trial would be? Those are some of the factors related to the design of the trial. We also have to consider what sites we select to be in the trial. To start, we have to think about what does a successful site need. Create a profile. What kind of investigators do you need. What kind of patient population. That will help you know what type of site you want to recruit. When you're recruiting sites, you need to look at the expertise at that site as well as their track record. The best predictor of future performance is past performance usually. But you shouldn't be too dogmatic about that rule because sometimes the patient population can be a bit different. When a site couldn't recruit well for one trial, it may be able to recruit well for another trial that has different eligibility criteria. Again, staffing, the expertise and availability. How many patients are seen at a site? Do they have the volume to support a clinical trial? Do they have the infrastructure available for research in terms of the specific procedures and equipment that's needed, and in terms of the regulatory oversight that's needed? The selection procedures should be competitive. You should be taking an objective look at different sites to see what sites are best suited for this particular trial. Sometimes, you can rely on existing networks. Existing networks are advantageous because there is an existing infrastructure for research and personnel have likely been trained and are more ready to launch a study. You can consider whether you want to have international sites or not, and there's good reasons to include international sites, but you have to recognize that they come with additional regulatory and logistical hurdles. Finally, site selection can be a continuous process throughout a trial. Often, you will need to add sites because you aren't meeting your recruitment goals and you may want to drop sites that are not able to recruit and are just using resources without contributing to the trial. You don't want to make applying for being a part of a clinical trial too onerous for site. You can streamline things and get information about the site by fairly short directed questions, like the example I'm showing you here. The first part is asking about site characteristics. This particular form is looking for sites that are able to enroll people with chronic obstructive pulmonary disease, COPD. We look at the patient population and also whether they have the facilities that are going to be needed for this particular trial. Here's an example of a research network from my own experience for a rare disease, non-infectious uveitis, which is considered rare because it affects less than 200,000 individuals per year in the US, but it's still an important disease because it frequently emerges when people are young adults in their working years and it's often a chronic condition that can be a significant threat to vision. In this particular network, we have several sites in the US, but we've also recruited sites in Canada, were working with the National Health Service in the UK. We have sites in Australia, and because of the recruitment challenges we actually brought on sites in India. Sometimes these international sites have the advantage that the structure of their healthcare system tends to funnel these type of patients to fewer clinics. Any one clinic in India may see many more patients than a typical clinic in the US. Finally, in selecting sites, you should be cognizant and take into account that there's going to be a time lag at each site to get prepared for a trial. The diagram on this slide is an illustration of how variable that time can be. There are sites listed on the y-axis and the number of days to different milestones for approval, like days to IRB approval, days to get your staff certified. This can be quite variable across sites and needs to be taken into account in the planning phase. In summary, recruitment potential is an essential element of the design of a trial. You can design a beautiful trial in concept. But if you can't get people enrolled in the trial, it's not going to be a successful trial. You need to have a screening and evaluation process for sites, and you have to be prepared to add and drop sites as you go through the trial. Finally, you always have to expect the unexpected and be prepared to deal with those type of obstacles.
