How Robust Data Technology is Transforming the Design and Execution of Viable Clinical Trials

Mike Gallup

Ashley Schwalje

While incorporating real-world data (RWD) and real-world evidence (RWE) across the drug development lifecycle is getting a lot of attention, the application of such insights gleaned from rigorous analysis to the design and execution of clinical trials has yet to be widely adopted, despite its potential to address some of these persistent challenges.

A paradigm shift that leverages advanced tech-enabled data analytics can aid pharmaceutical companies in overcoming common clinical trial delays related to randomization and a host of other trial design and execution issues, such as trial eligibility criteria, predicted responses, estimated sample size, endpoint selection, and increased diversity.

Tapping into the power of data analytics to design better, more viable clinical trials enable pharmaceutical companies to answer research questions faster and more accurately, avoid common study delays and failures, depends on the trifecta of RWD, RWE, and leading-edge technology.

Here are seven steps to “getting clinical trials right”:
1. Develop a solid clinical research strategy via viable, early clinical development plan— using a global data-driven approach from the start rather than institutional knowledge only.
2. Think broadly about your research questions and avoid being too narrow in your target patient population and eligibility. Prioritizing any one question or criteria in the design of your clinical trial increases the risk associated with the study.
3. Use RWD to understand the true unmet patient needs and how best to address them.
4. Design patient-focused representative clinical trials that remove or reduce potential operational burdens that can cause participants to drop out of a study prematurely.
5. Identify investigators and clinical sites, including wider HCP referral networks, with access to representative patient populations who will be most impacted.
6. Find, educate, and engage eligible participants using varied networks, channels, and outreach methods to increase the likelihood of enrollment success.
7. Use predictive technology to enable you to be agile to pivot and change the course of action while simultaneously managing the risk, budget, and timeline.
Data-first approach to clinical trial randomization
Randomization can deter selection bias while also safeguarding against potential accidental bias. However, one of the top contributors to clinical trial delays is failing to recruit enough patients at speed with a suitably inclusive eligibility criteria to randomize patients fast enough. With ready access to more robust data and evidence, pharmaceutical companies can avoid potential clinical trial delays before they happen.

Here are four common issues that result in clinical trial delays or failures and how to address them head-on:

● Issue 1: Protocol design is too exclusionary or not representative of the patient population affected by the disease. If the clinical trial is focused on an intervention that overly impacts a certain patient population, ensuring that the segment is included—or avoiding criteria that may inadvertently exclude them—is critical.

Solution: Use RWE/D findings that incorporate direct patient feedback to identify and enroll more diverse patient populations and expand trial eligibility.

● Issue 2: There is an over-reliance on existing site and investigator relationships. This is problematic for numerous reasons, including returning to the same patient population pools time and again instead of expanding reach to pursue the patients who would most benefit from the drug.

Solution: Use available benchmarking metrics, past performance, RWD, and predictive analytics to select the most appropriate clinical sites and investigators. The ideal patient population for the study should drive site selection rather than opting for patients simply because they may be easier to enroll in a study.

● Issue 3: There is limited time and resources to vet, onboard, and oversee multiple patient recruitment vendors.

Solution: Leverage technology solutions and partners to cast a wide-enough net to secure pre-vetted and diverse populations. The combined power of data and technology can make vetting and managing vendors more efficient and effective.

● Issue 4: Sponsors use up their recruitment budget without understanding the channels that are performing best.

Solution: Access real-time performance analytics to reallocate budgets to high-performing channels. This data not only alleviates frustration but enables the team to operate more efficiently.

One recent, high-profile example of this situation in the real world happened with Moderna’s Phase III COVID-19 vaccine trial. FDA emergency use approval of the vaccine was at risk because of low trial enrollment and lack of diversity among participants. Moderna needed to rapidly expand access to recruitment vendors to reach and enroll new and diverse patient populations. They leveraged data-driven technology and one-to-many outreach to recruit the right participants quickly. Within two months, enrollment in the study doubled and diversity enrollment increased from 24 to 37 percent.3
Clinical trial design and complexity
One example of how a clinical trial design resulted in significant negative consequences happened during the development of PD-1 inhibitors for first-line non-small cell lung cancer (NSCLC).4 The first-to-market drug failed its pivotal trial because the population enrolled in the clinical trial was too broad. As a result, the closest competitor leap-frogged—despite being at a timeline disadvantage—with more precise clinical trial inclusion criteria. All the work that went into being first-to-market was for naught and the situation could have been avoided if the appropriate data and intelligence had been used to inform decision-making during the clinical trial design.

In this case, enrolling a less-than-ideal patient population in the study translated into billions of dollars lost. Had the pharmaceutical company accessed better data and insights, it would have identified a more representative patient population for the study that would have helped more patients with NSCLC.

Clinical trial design is complex, with numerous yet consequential decisions to be made at every step. Most novel drug development studies pursue the path of randomized controlled trials (RCT) because they produce the gold standard data that regulators require to verify the efficacy and safety of treatments. However, there is extensive variation—or forks in the path—in how to conduct RCTs. These variations can lead an organization to quick success or down a dead-end path, resulting in opportunities for other drugs to get to market first. As a result, it is vital to consider the impact of different design elements, including placebo control, active comparators, synthetic control arms, complex adaptive and cross-over designs, among others.

To bring an innovative drug to patients, studies need to deliver relevant, quality data and insights so regulators can better understand the treatment’s safety and efficacy in an already highly competitive environment for the determination of value and the gold standard.

Understanding the global landscape of clinically, demographically, and geographically relevant patients, as well as utilizing RWD and RWE, play an important role in designing a viable clinical trial by driving more informed answers to these critical design questions.

While still underutilized, data analytics and evidence provide pharmaceutical companies with robust insights into the disease, target patient populations, standard of care (SOC) as comparators and treatment pathways to drive more accurate and faster decision-making related to study design. Although the use of RWD and RWE is unlikely to usurp the role of RCTs, it can be a secret weapon companies can rely on to augment their research.
Lessons learned COVID
First, the pandemic highlighted how important it is for individuals to participate in clinical trials. As a result, nearly 68 percent of adults with a condition are more likely to participate in a clinical trial.5 However, evidence suggests interest is already declining to pre-pandemic levels, increasing the challenge further of identifying patients for continually increasing trial numbers.

Second, with a significant decline in face-to-face encounters, the pandemic accelerated the demand for and adoption of technology. Pharmaceutical companies had no choice but to turn to emerging methods such as RWD, RWE, and remote technology to continue running existing studies and launch new ones during the pandemic.

Remote digital technology proved that it could be effective for clinical trials, and much like telemedicine for managing patient care, is likely here to stay—transforming the future of the industry.

The growth in the drug development pipeline turbocharged by the pandemic means future demand for trials is set to explode, making data-based decision-making and the increased use of RWE more important than ever.

Leveraging data and insights to design and plan better, more efficient clinical trials, and tapping into innovative recruitment solutions, is a potential game-changer in accelerating time to market ahead of the competition.

References

1. Dagenais S, Russo L, Madsen A, Webster J, & Becnel L (2021). Use of real-world evidence to drive drug development strategy and inform clinical design. Clinical pharmacology & therapeutics. 111(1), 77-89. doi: https://doi.org/10.1002/cpt.2480
2. Suresh K P. An overview of randomization techniques: An unbiased assessment of outcome in clinical research. J Hum Reprod Sci [serial online] 2011 [cited 2023 Jan 27];4:8-11. Available from: https://www.jhrsonline.org/text.asp?2011/4/1/8/82352
3. Ivarrson M (2021, April 15). Diseases don’t discriminate—neither should clinical trials. In ModernaTX.com. Retrieved from: https://www.modernatx.com/en-US/media-center/all-media/blogs/diseases-dont-discriminate-neither-should-clinical-research
4. Liu A (2022, September, 21). Esmo: Amgen’s Lumakras confirmatory lung cancer data leave door open for KRAS competitors. In Fiercepharma.com. Retrieved from: https://www.fiercepharma.com/pharma/esmo-amgens-lumakras-lung-cancer-confirmatory-trial-data-are-here-fanning-kras-battle
5. Saarony G (2021, September 21). A race against time: How COVID-19 changed drug development. In ACRP.net. Retrieved from: https://acrpnet.org/2021/09/21/a-race-against-time-how-covid-19-changed-drug-development/