UPDATE: This blog was originally titled “Taking on More Risk by Doing Less Clinical Data Monitoring.” TransCelerate BioPharma, Inc reached out to BioClinica to clarify some points in their methodology. What is important to note is that the TransCelerate approach does not advocate “less monitoring” but rather a shift from an excessive concentration on source data verification to comprehensive risk-driven monitoring.
Targeted clinical data monitoring is a hot topic. The FDA recently released its final guidance document, "Oversight of Clinical Investigations - A Risk-Based Approach to Monitoring" and the industry is buzzing with the implications.
Here at BioClinica, we're seeing hesitation from some study sponsors to embrace the new approach. The response is not surprising as clinical research tends to favor the "tried and true" over new methods. Take for example the switch from paper-based data collection to electronic data collection (EDC) — the change was slow despite the clear benefits.
Likewise, the move toward a modified monitoring strategy is likely to take some time — especially since it feels counter to ensuring clinical data accuracy. Can we really conduct clinical trials with reduced source data verification? What if the FDA questions the data? In this scenario, how do we show everything possible was done to ensure accuracy?
As a clinical trials technology company, BioClinica offers powerful software tools to successfully manage a risk-based monitoring approach. Express, a "tried and true" EDC tool, makes it easy. The real challenge is change itself — doing monitoring differently than how it's always been done.
So, how do you determine which study sites pose greatest risk in data quality?
Like so many difficult questions, there is no one-size-fits-all answer. It depends on your clinical study, your investigational sites, and your methodology. The major challenge is defining risk indicators. The key to success is to create a Risk Management plan.
Below are some of the leading industry approaches to help determine your risk.
TransCelerate's approach is outlined in the position paper "Risk-Based Monitoring Methodology" and advocates a philosophical shift in monitoring processes. The program employs centralized and off-site mechanisms to monitor important study parameters holistically and uses adaptive on-site monitoring to further support site processes, subject safety, and data quality.
After defining critical data, which should always be 100 percent source data verified, TransCelerate recommends taking the following into account when determining risk:
- Safety: Sites with Serious Unexpected SAEs, sites that are outliers with regard to non-serious AEs, and SAEs.
- Investigational Product: Compliance with regard to study drug receipt, drug accountability, dispensing errors, and temperature excursions.
- Subject Recruitment: Outliers in screen failure rate or discontinuation rate.
- Protocol Compliance: Protocol deviations (major and minor)
- Trends in Data: Abnormal data trends, for example.
- CRF Completion Statistics: Time to enter and approve data, for example.
- Query Issues: Time to answer queries, number of re-queried queries, for example.
- Regulatory Documents: Number of documents not reviewed or received.
- Site Staffing: Site turnover rate
- Site Management: Facilities review of adequate storage of supplies and adherence with procedures
TransCelerate Pharma Risk-Based Monitoring findings show 7.8 percent of all queries in clinical studies are related to SDV. Additional findings show that 2.4 percent of queries in clinical studies are related to critical variables.
Metrics Champion Consortium (MCC)
Volunteers from a variety of pharmaceutical companies and CROs are working together to identify the best metrics for industry use. MCC has identified the following considerations for Risk-Based Monitoring:
- Patient consenting process
- Patient enrollment / retention
- Protocol compliance
- Investigational product supply management
- Data quality
- Patient safety
- Sample quality problems (Lab, ECG, imaging, etc.)
- Site staff issues
MCC assembled this list after analyzing nine sample studies from six member companies. Despite variability in the way those companies manage data review activities, all were similar in the low rate of SDV-generated queries, with the average percentage of SDV queries generated 7.8 percent. The average percentage of SDV queries generated in critical data as represented as a part of the total number of queries was 2.4 percent.
The rate of SDV-only discrepancies in critical data (2.4 percent) suggests that SDV has a negligible effect on data quality. These data support TransCelerate methodology, which recommends shifting from onsite monitoring (including 100 percent SDV) to a risk-driven monitoring approach, and is consistent with FDA regulatory guidance.5,6,7
How will your company determine what indicators to use when determining risk?