And it gets worse. The main reason trials are terminated in the first place is due to a low accrual rate, according to a report from GlobalData. The enrollment efficiency rate — the number of patients planned for the study divided by the number of patients that enrolled — in these trials was below 40 percent at all phases. The lowest enrollment efficiency was in Phase III at 32.1 percent and the highest in Phase IV at 39.8 percent.
Damning figures indeed. Let’s explore the extent of the under-enrollment problem, and look at tactics to address it.
The past 18 years has seen the number of active clinical trials rise dramatically from 2,119 in 2000 to around 280,000 in 2018, but the enrollment figures have not kept pace. Participants in clinical trials have decreased from around 105,000 in 2015 to fewer than 26,000 in 2016, writes the team at patient recruitment services provider Clariness.
There are a few common areas where sponsors are struggling. For a start, they don’t have the staff and resources to provide sufficient empathy and attention to patients. Additionally, elderly patients are not being recruited nor are those patients with mobility issues. These gaps need to change for enrollment levels to pick up.
Some trials may have a relatively successful recruitment drive, boosting sponsors’ confidence that the study will progress according to plan. However, the reality is often quite different, according to health communications agency Couch.
Despite successful recruitment runs, approximately 25 percent of patients do not stay enrolled until the end of the trial. This results in an unplanned-for financial burden as trials are extended to add patients to the study. But more troubling is the lives that might have been saved or improved had the study been successfully completed.
There's a correlation between the increased number of active clinical research studies and decreased patient participation. Sponsors and investigators say the sheer number of clinical trial options available is negatively impacting enrollment, writes Jack Kaufman at MobiHealthNews.
Other reasons for low enrollment include patient burden, difficulty finding patients with specific or unusual conditions which leads to a lack of sufficient numbers of patients for a single trial to be statistically valid, and a lack of recommendations from primary healthcare providers.
Tech startups are attempting to address these issues, says Kaufman. Examples include Deep 6 AI, which uses artificial intelligence to analyze clinical data and output new clinical data points to match patients to clinical trial criteria, and PatientWing, an online recruitment platform with mobile-friendly landing pages and embeddable forms that improve the patient application process.
Meeting enrollment targets require both sponsors and sites to undertake certain responsibilities. Trial sites, for instance, should ascertain during the feasibility questionnaire whether they are equipped to deal with the disease indication, have a suitably sized patient population to draw from and what their prior enrollment performance in similar studies indicates, writes Kristina Lopienski at Forte Research.
Sponsors, on the other hand, need to ensure the goal of the questionnaire is clear and that the site is given enough time to respond to it. One way of streamlining response time is use previously collected data, eliminating time spent completing information already available to sponsors in their databases.
Sites have added responsibilities, says Lopienski. They should undertake checks to ensure they don’t open trials that won’t enroll. This is accomplished by a break-even analysis — the lowest number of patients needed to warrant opening the study. They should also avoid overestimating enrollment potential and not be afraid to refuse to open the trial should they doubt their ability to enroll patients.
While not possible for all trials, adapting the design can help improve enrollment as it ensures patients are more eligible to participate. Adaptive trials use pre-specified changes based on accumulating data to include or exclude patients, write researchers led by Dr. Eric H. Rubin at the American Society of Clinical Oncology. This is often more efficient than multiple sequential trials, they add.
Enrichment designs also offer potential for improved enrollment. Patients are eligible if the test for a biomarker performed on a tumor or blood specimen is positive. However, these different designs bring challenges such as greater complexity of protocol and a lack of flexibility, the researchers conclude.
Enrolling the right number of patients for the duration of the study is key, but recruiting patients without medical insurance could be problematic both practically and ethically.
On a practical level, uninsured patients tend to be from lower socio-economic backgrounds and might struggle to afford travel to sites as well as take time off from work to attend study sessions, says cardiologist George Marzouka.
Additionally, there are ethical implications to consider. Marzouka cites the example of a low-income patient that enrolls in a trial. At the end of the study, their data will have helped bring a treatment to market but, without insurance, that patient will unlikely have continued access to the new drug. Marzouka says this situation could be considered exploitative of these uninsured patients.
Fixing the diversity issues in clinical research could also lead to improved enrollment rates. Consider the discrepancy between the number of African American patients and white patients in studies say Caroline Chen and Riley Wong at ProPublica.
One out of five people with multiple myeloma in the U.S. is African American and black Americans are twice as likely to be diagnosed with blood cancer than white Americans. And 24 of the 31 cancer drugs approved by the FDA since 2015 had fewer than 5 percent of black patients participate in the trials.
Targeting groups that are most affected by the disease and thus best suited to the treatment may not guarantee better enrollment but it will certainly boost the chances.
Rare diseases are difficult to recruit for, which means enrollment will also be a challenge. Sponsors should start by looking at foundations and patient advocacy groups that deal with a specific disease for help, writes biochemist Ramani A. Aiyer.
Another useful tool in the recruitment and enrollment process is rare disease patient registries, he says. These provide useful information about the disease, patients’ histories, tests and procedures, and lists of support groups. The databases also grant access to people that are more interested and eligible to participate in that particular area of study. Of note too is that patients with rare diseases are often few and scattered around a country or the world.
Aiyer offers a caution to sponsors and trial recruiters making use of these registries: Treat this information as sacred and treat patients and their data with respect.
Only two percent of patients in the U.S. know about trials and that they can participate in them. So lack of awareness is a big problem affecting enrollment and retention of patients, says CRO consultant Willie Muehlhausen. But another issue is that the process of joining a trials is unnecessarily challenging.
Technology such as electronic consent forms can simplify the process, he adds. Not only can tech make it easier for patients to agree to participate, the forms can be written so patients understand exactly what it is they’re agreeing to.
Part of the problem with patient enrollment is that study investigators are falling victim to the “one and done” syndrome. This refers to the trend in which trial investigators work on just one trial before quitting. The result is a negative impact on research, writes Suzanne Caruso, vice president of clinical solutions at clinical research review services provider WCG.
She cites a survey shows that 54.2 percent of investigators fell into the “one and done” category. The result is greater demand from sponsors on a smaller pool of investigators.
Caruso says these figures are important because successful patient enrollment is directly linked to the experience levels of study investigators. Indeed, investigators that have completed at least four studies had non-enrollment rates of 13 percent while those with less experience recorded rates of 25 percent.
The take-away is that sponsors should try to have experienced investigators working on their trials. But they must also focus on designing flexible trials with broad eligibility criteria, greater diversity and inclusivity to reduce patient burden. Doing so will likely see enrollment rates rise.
Images by: rawpixel/©123RF Stock Photo, eunika/©123RF Stock Photo, kuprevich/©123RF Stock Photo