Overcomplicated Clinical Trial Protocols Lead to Data Overload and Reduced ROI, Phesi Analysis Shows
核心洞察
Phesi's analysis of 2,401 Phase III clinical trials reveals that many trials suffer from overcomplicated protocol designs, leading to delays and increased patient burden.
The study found that trials with more outcome measures reported a lower percentage of those outcomes, with over a third of outcome measures going unreported.
Type 2 diabetes (搜索) trials exemplify the issue, where sponsors using fewer outcome measures achieved better site enrollment and higher enrollment rates.
Phesi's recent analysis indicates that an excessive collection of data, driven by a 'fear of missing out' (FOMO) mentality, is undermining the return on investment (ROI) in many Phase III clinical trials. The firm's report highlights that overly complicated protocol designs are creating avoidable delays in drug development and increasing the burden on patients.
The Problem of Redundant Outcome Measures
According to Dr. Gen Li, founder and CEO of Phesi, while collecting a sufficient amount of data is crucial, it is equally important to ensure that the data is relevant and collected at the appropriate time. The analysis revealed that many trials include redundant outcome measures, such as using multiple physical performance status measures on the same patients, which adds unnecessary burden without providing additional value.
Analysis of Phase III Clinical Trials
Phesi's analysis utilized proprietary data from its Trial Accelerator platform, examining 2,401 industry-sponsored Phase III clinical trials that had reached their primary endpoint since January 2020. Of these, 1,574 had reported patient data. The study identified five top disease indications and found that 146 protocols were designed to capture 1,821 outcome measures, ranging from 1 to 102 measures per protocol, including both primary and secondary outcomes.
The findings indicated an inverse relationship between the number of outcome measures included in a trial protocol and the percentage of outcomes reported in the results. On average, more than a third (35%) of outcome measures were not reported. Trials with fewer than the median number of outcome measures reported 94% of those measures, while those collecting more than the median reported only 56%.
Impact on Trial Efficiency
While the impact of complex trial designs on cost, timelines, and patient burden has been previously acknowledged, objectively measuring this impact has been challenging. Phesi's Trial Accelerator platform has enabled the firm to assign a 'complexity score' to trial protocols, modeling the true impact of the number of outcome measures.
For instance, a comparison of type 2 diabetes (搜索) trials from two different sponsors revealed that the sponsor using a median of 25 outcome measures had lower site enrollment performance (10.2 patients per site) and a lower enrollment rate (0.46 patients per site per month) compared to the other sponsor with a median of only 10 outcome measures (11.2 patients per site and 0.53 patients per site per month, respectively).
Recommendations for Improvement
Dr. Li advises investigators to avoid 'data FOMO' and focus on collecting only the data they truly need. By being more precise with outcome measures, investigators can improve site selection, patient recruitment, and data quality, ultimately leading to a better return on investment. This is particularly important in the current economic environment for the pharmaceutical industry.