Ensuring that a drug is safe is of the upmost importance during a clinical trial. Pharmaceutical companies must know that their product is of the highest quality, and that it won’t harm patients who have been prescribed it.
Traditionally, when a drug is undergoing a clinical trial, frequent visits to the trial sites to gather data – physical exam results, or adverse event reports, for example – are necessary. But this process of collecting and verifying data, known as source data verification (SDV), is costly in terms of time and money. Instead, pharmaceutical companies are moving towards risk-based monitoring (RBM), where a combination of activities, including centralised data collection and monitoring, are employed to direct resources to where they are most needed. It also enables better data and protection of subjects, without increasing cost.
RBM is an effective means of ensuring the quality of clinical trials by identifying, assessing, monitoring and mitigating any risk that might affect the quality or safety of a clinical trial. It can enable a company to quickly identify and track actions for high-risk sites, and reduce complexity by supporting multiple trial designs and data sources. It can also optimise monitoring using adaptive risk models applied to historical study data.
On target
“RBM is about monitoring the right data at the right time at the right site, enabling an even more focused collaboration with clinical trial sites,” explains Hanne Cloetta Lang, vice-president at Novo Nordisk Clinical Operations. “It starts with the inception of a development plan, where risks are proactively identified and mitigating actions predetermined.
It paves the way for a collaborative data evaluation with other colleagues, allowing clinical monitors to have targeted on-site discussions – based on data – to support the site in their activities related to patient safety and data compilation.
“ICH E6 (GCP) was updated effective November 2016; the revised wording includes a requirement for all sponsors conducting interventional clinical trials to adopt a risk-based approach to monitoring. This means using monitoring methods that are proportional to the risks associated with the trial. For example, the risks to patient safety, patient rights or data integrity.”
RBM can support and enhance current practices that are designed to ensure patient safety and quality data, and help allocate resources more efficiently. Resources are assigned based on the risk and need, with the aim of increasing monitoring without compromising patient safety or data quality. Handling all risks in the same manner is inefficient and prior experience with similar trials should help a company identify the greatest areas of risk and develop metrics to quantify any potential issues.
Pharmaceutical companies are turning to RBM because the traditional approach of SDV is not completely effective, and is incredibly costly: up to a third of a study’s cost can be attributed to the on-site review of trial data. The process is resource intensive and limited in its ability to identify and prevent issues.
It is generally accepted in the pharmaceutical industry that the process for clinical trial monitoring needs to change, now that a more centralised, risk-based approach is the preferred method for ensuring that trials are efficiently monitored.
“One of the methods used to do efficient risk-based monitoring is increased use of centralised monitoring,” says Lang. “This is enabled by a central review of large volumes of data; hence, tools and techniques to enable this centralised monitoring are essential.”
“Most tools to enable centralised monitoring of data need to source data from current electronic data capture tools or central data repositories. To enable seamless, or at least simpler, use of data, the standardisation is an important element. For many years, an effort has been made to develop and support global data standards. These standards enable information from clinical trials to be shared, analysed and combined independently of IT platforms. The new processes also open new opportunities to identify data outliers during the trial conduct or to test new hypotheses, potentially leading to new patient segmentations and better target the treatment.”
Dashboard view
Centralised monitoring techniques mean staff can perform data checks, review for inconsistencies, and complete regulatory reviews and updates remotely, without the need for visiting a study site. It also offers many other benefits over SDV.
It uses more automated reviews to determine the need for manual intervention and is more likely to uncover errors than on-site monitoring, which is limited in scope and prone to error. All data flows into a central risk dashboard, which can make statistical and graphical checks to determine the presence of outliers and unusual patterns in the data easier.
A centralised approach also offers the opportunity to compare data between sites to assess performance, identify potentially fraudulent data, or locate faulty or poorly calibrated equipment. It is significantly cheaper – activities like on-site monitoring can be limited to study sites where problems are most likely to be occurring. Results can be monitored while the trial is under way, meaning issues can be identified and resolved more quickly.
While the specifics will vary according to the trial, a risk-oriented monitoring programme will typically collect and submit data through a ‘dashboard’ built for that particular study. This is designed to provide at-a-glance information about the status of the study at each site relative to the specific risk factors of the trial.
A centralised approach requires a steady, reliable flow of data from each study site to the central monitoring system, through either manually entered and transferred data, or an automated connection between the data entry system and central dashboard. If a site shows a high risk level, a monitoring plan should help determine whether further investigation is necessary. The dashboard might also perform statistical analysis to help identify problems between sites and – along with further analysis – will help determine whether an in-person investigation is necessary at a particular site.
RBM can help combat the continually rising cost of conducting effective research trials. The limited resources and budget can be better used to ensure data quality. It may also help speed up the trials timeline.
“The duration of clinical trials is first and foremost driven by the treatment and recruitment time, as well as the internal time to process data. While treatment time is fixed upon protocol design, the remaining parameters can be optimised to achieve faster time to submission, potential market approval and, ultimately, delivery to patients,” comments Lang.
Slow on the uptake
The industry has been slow at adopting RBM, however, as it requires organisational change, and throws up many other challenges. “One of the main challenges we have in data management in relation to RBM is to make data available earlier for decision-making,” says Lang.
“That is to say, being able to provide data for medical review shortly after patients have been included in the trial. Other challenges include the introduction of risk-based data cleaning to ensure the highest attention to the most critical data points, as well as alignment of SDV and data-cleaning activities.”
RBM is evolving as it becomes more widely adopted. Novo Nordisk is supporting TransCelerate, a non-profit organisation started in 2012 to improve the health of people around the world by simplifying and accelerating research and development of innovative therapies.
As part of this, TransCelerate members – which include many high-profile pharmaceutical companies – have developed a standard RBM approach to the planning and conduct of clinical trials, with an aim to provide the greatest oversight and efficiency. This will improve the outlook for RBM.
“RBM is somewhat immature across the industry,” notes Lang. “As organisations’ RBM set-ups mature and we obtain insights from regulatory agencies (after more inspections of RBM trials, and evaluation of regulatory submissions based on RBM trials) the process framework, IT tools and approach to RBM will undergo optimisation in most organisations,” Lang states.
“Presumably, it won’t be a case of ‘one size fits all’, as every organisation will adapt implementation approach in accordance with ways of working within each company. There is, however, no doubt that anyone involved in clinical trials will have to learn RBM as part of their ABC.”
