Contributed Commentary by Barbara Argibay Gonzalez 

August 2, 2024 | Clinical trials are a crucial step to advancing healthcare and bringing new treatments to patients. However, drug development is a lengthy, expensive, and high-risk proposition, and, according to PhRMA, a process that can take up to 15 years with an average cost exceeding $2 billion per new medicine. Those factors raise the stakes for clinical trials, which even under the best of circumstances are challenging and require complex planning to ensure positive outcomes. While those challenges may be daunting, sponsors and innovators can now take advantage of advanced technologies that can help keep a trial on track and avoid common missteps and roadblocks. 

Jump-starting Early Planning  

Trial planning can be extremely complex and time-consuming, requiring high precision at every stage of the process, especially at the beginning. That makes it critical to efficiently identify robust, reliable data sources and to aggregate data into insights that can yield: 

• Enhanced understanding of the competitive landscape 
  
• Identification of sites and investigators with the requisite capacity and expertise 
  
• Realistic patient enrollment benchmarks 
  
• Participant populations that meet diversity criteria 
  
• Reduced number of protocol amendments 

In recent years numerous advanced technologies have arisen to support trial planners by facilitating early planning. Clinical business intelligence software platforms that can analyze, measure, and rank trial and site data are available.  These digital tools help manage key factors such as site capacity, investigators’ therapeutic area experience, and engagement with competitors, thereby enabling timely decision-making and supporting the entire study workflow, including protocol design. A recent study (currently a preprint) by the Tufts Center for the Study of Drug Development (Tufts CSDD) assessed that protocol amendments are becoming increasingly common in early-phase studies (phases I and II), more than three-quarters of amendments—77%—are considered unavoidable due to regulatory agency requests and changes to study strategy. In addition, the total average time to implement an amendment has nearly tripled during the past decade. The researchers found that including patients’ and investigative sites’ input in the protocol design improved clinical trial performance and reduced the number of protocol amendments.  Thus, clinical trial sponsors, armed with such clinical business intelligence platforms could also identify KOLs and sites early enough to make them participants of the protocol design. 

Facilitating Patient Recruitment and Retention

Failure to engage patients early in the trial design process is one of the most common roadblocks to successful clinical research. An analysis of all terminated trials within the Clinical Trials Database implicated low accrual rate as the determining factor in 55% of such terminations. And according to a 2020 article in Perspectives in Clinical Research, Mira Desai reports that globally, more than 80% of trials fail to meet enrollment deadlines, often leading to study extensions and/or addition of new sites. Similarly, insufficient retention of participants through study completion can compromise data collection and analysis.   

Clinical IT platforms can flag and keep the history of changes to help trial managers closely monitor site recruitment numbers throughout a study. They can also aid in the design of protocols with realistic eligibility criteria, thereby helping to avoid the roadblock of overly restrictive criteria. Sophisticated platforms can also mitigate the risks of other common contributors to low recruitment, including: 

• Competitive trials recruiting patients at the same site  
  
• Limited site experience with complex trials requiring multiple procedures on a very tight schedule 
  
• Inconvenient geographic location, making it difficult for participants to reach the facility; this is a key concern if multiple visits are required 
  
• Lack of diversity within the medical team. Patients need to feel comfortable with the doctors who are treating them. Notably, data published in the 2020 Journal of Primary Care Community Health indicate that both female and male patients tend to prefer a same-gender primary care physician; this preference is especially pronounced among male patients. 

Insights into the factors affecting patient engagement can help improve communications with patients about trial details and can aid in incorporating patient input into protocol design. Those benefits can make participants feel valued, reinforcing efforts to facilitate patient engagement and retention.  

Streamlining Feasibility Analysis

The site-selection process should be informed by a thorough and diligent feasibility analysis, one that evaluates sites’ ability to mitigate patient enrollment and retention issues while also minimizing delays and controlling costs. Given the increasingly competitive clinical trial landscape, with numerous sponsors seeking to secure the participation of a finite number of high-performing sites, it is vital to verify a site’s capacity and availability to recruit for an upcoming trial. However, the feasibility analysis is something that should be reassessed during the course of the study, since new clinical trials will start in those sites and their capacity will be altered.  

Increasingly, trial planners are employing advanced IT to facilitate scenario planning, leveraging the power of predictive analytics. Some of these technologies incorporate simulation tools that use aggregated, continually updated historical trial data to analyze and optimize assumptions about patient profiles and enrolment targets such as country selection, estimation of patients-per-site-per-month, comparative of treatment lengths, study historical overruns and endpoints’ selection. The more sophisticated tools incorporate feasibility algorithms that, among other capabilities, pinpoint optimal times for trial initiation, identify competing studies taking place at preferred sites within enrolment periods, and recommend alternative sites and/or scheduling dates. These algorithms also aid in evaluating a site’s previous experience with a given sponsor, as well as its track record in successfully enrolling a diverse cadre of patients in trials within a selected disease category.  

Today’s competitive landscape IT analytics tools can also facilitate thorough identification of investigators, evaluating factors such as current involvement in clinical trials, experience with trials of similar complexity, and experience with specific interventions, whether drugs, medical devices, or cell therapy, for example. These applications can also examine investigators’ scientific publications and conference attendance to pinpoint their clinical interests, to evaluate their global and local networks and how those have evolved over time. Such thorough analysis can strengthen the connection between investigator, sponsor, and patient, increasing the likelihood of a successful trial. As Feehan and Garcia-Diaz have observed in their spring 2020 publication in the Ochsner Journal, “Investigators are not only responsible for producing high-quality, meaningful, scientific research, but they are also responsible for maintaining public trust.” 

Leveraging Big Data and AI

In clinical research, unfettered access to data—and the software to make sense of it—translates into well-designed clinical trial protocols and better alignment of study requirements with research objectives. Through improved site selection and recruitment, and better insight into patient populations and their shared traits, artificial intelligence (AI) and machine learning (ML) can turn data into valuable tools to refine research questions. 

AI is making enormous inroads into the pharmaceutical and medical research markets with the promise of uncovering more robust data insights, promoting health equity, identifying and mitigating risks, improving decision-making, and increasing efficiency. Unfortunately, the adoption of AI has been slow because of concerns around compliance with Health Insurance Portability and Accountability Act (HIPAA) strictures, as well as with similar requirements in other world markets. 

As AI and ML become more operational in clinical trial design, recruitment, and management, governing bodies are working to protect the interests of those involved in clinical research. Over the past few years, the European Commission, the US Food and Drug Administration (FDA), the National Institute of Standards and Technology (NIST), and the World Health Organization (WHO) have released policies to help guide compliance professionals in their use of AI. 

The WHO, for example, recently published its guidance on Ethics & Governance of Artificial Intelligence for Health. The result of 18 months of deliberation among leading experts in ethics, digital technology, law, and human rights, the WHO report identifies the ethical challenges and risks of using AI in healthcare practices and outlines six consensus principles to ensure AI works to the public benefit: 

• Protecting human autonomy 
  
• Promoting human well-being and safety and the public interest 
  
• Ensuring transparency, explainability, and intelligibility 
  
• Fostering responsibility and accountability 
  
• Ensuring inclusiveness and equity 
  
• Promoting AI that is responsive and sustainable 

Healthcare IT is advancing fast, and AI will be unequivocally part of its growth, however it is our responsibility to make sure it remains patient centered and not exclusively technology focused.  

Ensuring a Diverse Patient Population

Although the FDA has previously published guidance encouraging diversity in clinical trials, sponsors have made little progress in improving the representation of racial/ethnic subgroups, particularly in oncology clinical trials. The FDA’s Drug Trials Snapshots Summary Report of trials supporting the approvals of 15 novel drugs indicated for various types of cancer in 2021, the majority of patients (72-95%) were White.