In his book, Applied Minds: How Engineers Think, Guru Madhavan explores the mental makeup of engineers. His framework is built around a flexible intellectual toolkit called modular systems thinking. He says that “systems thinking is more than just being systematic; rather, it’s about understanding that in the ebb and flow of life, nothing is fixed and everything is connected. The relationships between the modules of a system create a whole that cannot be understood only by analyzing its constituent elements.
In other words, the whole is greater than the sum of its parts.
Systems engineers are taught to think about all problems holistically and then design individual components accordingly. This mindset is missing from clinical trial design and is one of the main reasons why the clinical trial process is broken. Consider this: in the last decade there were 18 million cancer patients diagnosed in the US, but only 0.1% were offered clinical trials. At the same time, 66% of oncology clinical trials are closing prematurely because they can’t fill their trials with patients.
This makes no sense and deprives too many cancer patients of hope for a better outcome.g
The life sciences industry will be better equipped to address the inherent challenges prevalent in oncology clinical trials by using engineering principles that target individual components while accounting for their ramifications over the entire trial from the outset. Nowhere is this clearer than in the matching, recruitment, and enrollment of cancer trial patients. Today, this process is like finding a needle in a haystack.
Finding a needle in a haystack
Identifying patients for oncology trials seems like an intractable problem for clinical researchers, but that’s because they don’t think holistically about all the processes needed to identify, engage, and guide patients through enrollment and participation. Just as engineers do not design for just one process without considering the entire system—i.e. building NASA’s Orion cockpit without thinking about how it affects the entire spacecraft—clinical researchers must consider how patient involvement affects the entire value chain from recruitment to retention of results.
To solve problems, engineers dive deep into all failure possibilities, considering every potential outcome for every solution. It is also critical to success in clinical trials, where there are many possible points of failure. Companies will make a transformational change in clinical research when they apply an engineer’s mindset, thinking both horizontally throughout the trial process and vertically to thoroughly analyze all potential points of failure.
New thinking + new technology = scalable solution
As science advances cancer treatment, clinical trials are increasingly designed around very small, genetically defined subsets of cancers, making finding suitable patients difficult. Additionally, oncology trials typically require patients to be relapsed/refractory after standard cancer treatment or to have relapsed at least twice before being considered candidates. If the patient clears these first hurdles, he faces a rigorous pre-screening. Cancer trials are notoriously rigorous; in fact, 40% of patients with cancer trials available to them are ineligible for inclusion due to eligibility requirements, according to industry report.
Actually a recent study found that approximately 80% of patients with advanced non-small cell lung cancer did not meet the criteria for the studies included in the study. As a result, 86% of these attempts failed to complete the recruitment within the target time. Clinical investigators are also tasked with enrolling patient populations that reflect the diversity of cancer demographics, further complicating patient identification.
Combined, these hurdles make identifying and enrolling patients one of the biggest obstacles to oncology clinical research. Trial sponsors are grappling with this challenge despite investing in a variety of solutions, including many new and unproven approaches.
Some sponsors, for example, employ digital patient recruitment specialists who work to identify potential trial participants using widespread social media advertising to reach a larger pool of applicants. It’s effective… to a point. It addresses only part of the problem and does not take into account what happens after the patient is identified.
Other researchers are trying to use advanced technologies, such as data science and artificial intelligence (AI), to mine patient databases and medical records based on trial eligibility criteria. Again, these technologies are powerful, but they do not take into account what happens to patients once they are identified.
By thinking about this problem like an engineer, we can develop a more holistic solution that not only addresses patient identification, but also considers how best to guide patients through the many pre-screening requirements for participation. These requirements, such as gathering medical records and obtaining various lab tests, can be complex to navigate and burdensome, especially for the sickest cancer patients we are trying to help.
Next, there is the challenge of keeping patients actively engaged during trial enrollment so that they do not drop out before they have even completed screening. Engineers analyze and solve these potential problems that others don’t think about, while clinical researchers are focused on trying to prove a hypothesis. The engineer-minded researcher does both – addressing all pain points in patient enrollment, including:
- Patient identification – analyzing all direct and indirect patient acquisition channels in real-time and routing to a centralized location for further evaluation. Direct patient acquisition channels typically include referrals from call centers, patient advocacy groups, leads identified through digital advertising, leads from mobile apps, and public awareness events such as webinars and educational sessions. Indirect patient acquisition channels include referrals from providers, payers, next-generation sequencing providers, and specialty pharmacies.
- Patient file management – identifying specific trial eligibility requirements and ensuring that patient data are accurately extracted from medical records to meet these criteria. AI can make this process faster and more accurate.
- Full trial identification – considering all available trials in pre-screening cancer patients in case they are rejected from their first option. AI also plays a role here by automating searches across multiple sample databases that are challenging to manually navigate.
- Capture feedback – understanding why a patient was accepted or rejected can inform future patient recruitment efforts. New technologies provide transparency, allowing patients to be re-evaluated for a trial if they can meet criteria later, and lead to long-term improvements in overall population health as this transparency is applied across patient cohorts.
- Last mile patient support. – providing highly sensitive care to patients who are often overwhelmed by trials yet exhausted by the side effects of their treatment and illness. In this ‘last mile’, individual patient care can also serve to sensitively identify and remove any barriers to participation, such as transport logistics and costs, and maintain their active engagement until the last dose of their investigational treatment.
- Monitoring and feedback – understanding the success of clinical trial enrollment and continuing to obtain patient feedback on disease progression, the clinical trial process, and the implications of clinical trial participation, such as side effects.
Engineers see everything as a system, know how to design within constraints, and recognize the need for trade-offs. Adopting an engineering mindset in oncology research can fix any broken constituent processes such as patient enrollment to systematize clinical trials. Combined with the ingenuity of scientifically minded clinicians, this new approach could help more patients get better drugs faster.
Photo: Warchi, Getty Images