The US Food and Drug Administration (FDA) and similar agencies worldwide exist to ensure the safety, efficacy and security of drugs and medical devices. For each drug or device they approve, they have to weigh the benefits to public health against the risks of side effects or complications. Evaluating drugs and devices has grown more challenging — for regulators and drug and device manufacturers — in an environment overwhelmed with data.
To improve efficiency, regulators are urging clinical trial sponsors to rethink the way they design and run clinical trials. The recommended approach: risk-based quality management (RBQM), a holistic strategy that ensures sponsors plan for, prepare for and protect against harmful risk from planning to submission.
Led by recent amendments to International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Good Clinical Practice (GCP) guidelines, the FDA and other regulators have provided extensive recommendations for how to make better use of technology to build an RBQM strategy.
It's a lot to absorb. The guidelines, both comprehensive and vague, have left many sponsors wondering how exactly to prove safety and effectiveness using RBQM.
What the FDA Wants
To understand what regulators want, it helps to understand what they don't want — more data without related context.
According to a 2018 Tufts Center for the Study of Drug Development report, the number of endpoints used in clinical trials rose 86% between 2001-2005 and 2011-2015. The number of sites increased 63% over the same time period. This increased complexity means more data to collect and learn from but also more patient burden.
"Researchers want to collect more and more information, but this data needs to be relevant and useful," said Remarque Systems product specialist Erika Horan. "With advances in technology, whether it's DNA testing or wearables, there's a large volume of complex data out there coming from an increasing number of sources."
To reduce complexity, ICH E6 (R2) suggests that sponsors "identify those processes and data that are critical to assure human subject protection and the reliability of study results." RBQM requires that sponsors focus only on the data needed to determine safety and efficacy.
"RBQM focuses on your highest-priority data and processes so you can efficiently use your resources," said Amanda Coogan, Remarque Systems risk-based monitoring senior product manager. "Instead of 100% SDV — or the status quo — the intent is to manage and monitor smarter and more efficiently."
What Regulators Look For
As with any regulatory submission, sponsors should extensively document evidence that shows a product is safe for patients and effective in a targeted population. Using a risk-based approach includes documentation of a study's defined risks and prioritization of risks as well as mitigation and contingency plans that address those risks.
Regulators want to know how sponsors are executing RBQM. "You should show the metrics you track and thresholds set on those metrics," Coogan said. "For example, when adverse event frequency is higher than a predetermined expectation, you plan to get an alert from a system that calls attention to the issue and provides guidance for resolution and mitigation." Essentially, Coogan said, sponsors have to tell a detailed story about their data and RBQM process oversight.
About Quality Tolerance Limits
Quality Tolerance Limits (QTLs) are now an expectation under ICH GCP guidelines. When exceeded, QTLs trigger an evaluation to determine if a systemic issue exists. In other words, when the trial exceeds those limits, patient safety and study integrity are at risk. Examples include protocol violations, missed assessments that contribute to primary/secondary endpoints, and adverse events of special interest.
Sponsors define QTLs when planning a trial as part of the initial risk assessment. According to ICH E6 (R2), researchers define QTLs by "taking into consideration the medical and statistical characteristics of the variables as well as the statistical design of the trial." Key performance, quality and risk indicators also come into play.
How to Prioritize Risk
In a reflection paper on RBQM, the European Medicines Agency said the first step in prioritizing risk was to understand the processes and outcomes that really matter to achieve study objectives.
Determining the likelihood of occurrence, effect and detectability of risks helps researchers prioritize them. The highest risks — or rather, the highest-priority data — will vary from study to study, depending on the endpoints, the disease, and the drug or device being studied. As Horan explained, oncology drug researchers might have different expectations around frequency and types of adverse events than researchers studying a treatment on otherwise healthy patients.
Technology's Role in Safety and Effectiveness
Artificial intelligence (AI) and machine learning have already helped some sponsors improve patient recruitment and engagement and generate real-world evidence. RBQM can also benefit.
Because of RBQM's holistic nature, sponsors need to continually monitor data in real time. AI helps make that happen. It also provides insights that help sponsors, CROs and clinical monitors make strategic decisions to mitigate risk.
With the help of machine learning, advanced data platforms generate information sponsors can use to show the FDA they have maintained excellent documentation and oversight of their clinical trials. Technology can help make detailed logs, alert systems, validating results and documentation of updates all possible.
A platform that harnesses the power of AI allows sponsors to view disparate forms of patient, site and trial data all in one place. Actions such as measuring drug response to monitoring and alleviating risk are two of the many ways a comprehensive, data-agnostic platform helps facilitate RBQM.
When implemented in accordance with agency guidelines, RBQM helps lower the risk to patient safety because it protects against risk from the outset. Technology, combined with the internal process changes necessary to implement that technology properly, makes RBQM work for pharmaceutical manufacturers and the industry at large.
"RBQM isn't just about stripping out monitoring," Coogan said. "If you implement RBQM effectively, you should be able to focus your resources on data and processes that are most critical to proving safety and efficacy. Ideally, this results in catching errors and issues early, ultimately leading to more reliable data from which decisions can confidently be made. If that means you get a drug to market a year sooner, you've potentially saved millions."
References