How Leading Cancer Centers Overcome Trial Accrual Challenges

Poor clinical trial accrual is not an “outlier” issue, with 1 in 4 trials failing to enroll enough participants to draw valid conclusions. However, given the urgent need for more novel treatments and cutting-edge research, leading cancer centers are exploring new ways to tackle the problem head-on. 

This article explores how they are doing this – and why it is so important. It will help leaders within cancer centers rethink their clinical trial design and recruitment strategies, as well as distill the latest research on the impact of poor trial accrual. 

Expect to learn: 

• How many trials are forced to extend their recruitment period 
• Why trial accrual is linked to patient retention issues 
• What novel technology can do to support more effective trial design and recruitment 

Cancer Clinical Trial Accrual Problems: An Overview 

How Serious Are Trial Accrual Problems? 

The best way to assess trial accrual is simply to look at the statistics: research shows that 90% of trials end up extending their trial recruitment period, while 18% of trials are still failing to meet 50% of their accrual targets 3 years after initiation. Additionally, 55% of trials are terminated due to low accrual rates, with average enrollment at less than 40% for phase III and IV trials.  

Efforts to more accurately predict accrual or adapt recruitment strategies have generally been ineffective. Cancer centers still struggle to fill trials or design inclusion criteria that will be more accessible – and that has a disastrous impact on patients, researchers, and the organizations themselves. 

Why is Poor Accrual an Issue? 

Poor trial accrual leads to three primary problems: 

1. Wasted Time and Resources 

Running a clinical trial costs millions of dollars and thousands of man-hours – and poor patient accrual can render it all worthless. The best-case scenario is the trial results are less generalizable, but low enough accrual can lead to the trial being canceled or paused part-way through. 

Recent figures suggest 20% of trials fail to meet their accrual goals, while nearly a third (29%) of trials that don’t meet accrual goals close specifically because they didn’t accrue enough patients. The net result? Time and resources are wasted without any research findings being produced or patients helped. 

This can have particularly serious consequences for organizations’ relationships with pharmaceutical firms. Failure to complete research a company is relying on for product development may lead them to end the relationship – and cut off a vital stream of funding. 

2. Reduced Care Quality 

While many studies conclude that participation in clinical trials leads to better medical outcomes, there is no definitive proof that this is the case. However, it is true that trials do lead to what one expert calls “intangible” benefits, such as a perception of being in safe hands and accessing the best possible treatment. 

Poor trial accrual, therefore, not only potentially reduces patients’ healthcare outcomes and survivorship odds – it also means less access to quality care. This is particularly true when viewed over the long term: low trial participation slows the development of novel interventions and, therefore, stalls the progress of cancer treatment.  

We can see this most clearly by comparing the trajectory of cancer treatments for children (who are far more likely to participate in trials) than adults: between 1975-95, mortality rates for cancer patients aged 19 or younger dropped 2.6% every year – a far faster improvement than adults have benefited from. 

3. Increased Patient Churn 

The negative impact on patient care is best illustrated by the retention issues caused by low trial accrual. Studies consistently show that up to 70% of US cancer patients want to participate in clinical trials, yet just 7.1% are offered the opportunity to do so. This leads to a lot of frustration, with a growing number seeking better trial options at other organizations. 

In summary: Poor trial accrual is harmful across multiple dimensions, and cancer centers have a powerful incentive to solve the problem. So, how have they approached that problem? 

__ 

How Have Accrual Problems Been Addressed Historically? 

Low clinical trial accrual rates have been reported within the oncology community since at least 1990. Efforts to redress the issues have included: 

• Awareness Campaigns: Promoting the importance and availability of clinical trials more widely within healthcare organizations - based on the theory that low accrual was the result of providers not offering patients the opportunity to participate. 
• Improved Access: Reevaluating “friction points” such as funding or logistical barriers that make it harder for patients to commit to participating.  
• Predictive Models: Developing statistical models that predict trial accrual based on data rather than projections from the disease team (DT).  

Each has delivered some benefit, but these methods have not generally been scalable. While predictive models consistently outperform DT in accurately predicting accrual, most cancer centers lack the kind of real-time center-specific required to effectively implement them.  

Similarly, research published in the Journal of Clinical Oncology revealed that simply making providers aware of clinical trials at the point-of-care could increase accrual. However, the researchers pointed out that the solution inevitably lost utility over time because the data was not updated frequently enough. 

That is why a growing number of cancer centers believe technology is the ultimate solution to trial accrual issues – and have increasingly embraced novel solutions to improve trial design and recruitment. 

The Future of Trial Accrual 

Recent years have seen a revolution in the usage of cancer registry data. With the emergence of large language models (LLMs) and natural language processing (NLP) technologies capable of analyzing huge datasets, the possible use cases for registry data have expanded – and there has been growing investment in solutions purpose-built to unlock those use cases. 

For hospital-based registries (HBRs), this has clear implications for clinical trials. Registry data can provide a comprehensive overview of the available patient population; it would only require the right system and interface to leverage that overview and leverage it across multiple areas of trial design and recruitment. 

Research has already begun to validate this possibility: a recent scoping review shows that LLMs can be used to parse trial eligibility criteria far more efficiently – and suggests that adoption of more advanced NLP models would be highly beneficial to accelerate research and reduce wastage. Studies from Oxford University have similarly found that LLMs can identify suitable patients based on eligibility criteria with high levels of accuracy. 

However, it is important to remember this is a highly sensitive area. Errors in trial matching can be even more harmful than accrual failures, and many evaluators will be resistant to outsourcing decisions to machines – even if it saves them a lot of time and effort. As a result, research in the International Journal of Medical Informatics argues NLP models should be used in-tandem with skilled human evaluators. 

Inspirata’s Trial Navigator perfectly illustrates what this approach would look like – and presents an optimistic future for trial accrual. The tool uses NLP to ingest and analyze huge quantities of patient data and provide a range of user-friendly insights. This helps cancer centers: 

• Improve Trial Design: Enable principal investigators to evaluate how changes to the inclusion criteria will impact trial feasibility. This lets them construct criteria that meet research requirements while maximizing the likelihood of meeting accrual targets. 
• Improve Resource Management: Utilize reporting features to identify underperforming trials and redistribute funding to improve resource management. 
• Accelerate Trial Recruitment: Equip nurse navigators and providers with a clear list of suitable trials at the point of care. Rather than manually assessing the availability of trials, they can quickly consult the data – making it far easier to match patients with clinical trials. 

Want to explore how the solution could help you design, fill, and complete more cutting-edge cancer research?