Data FOMO may undermine clinical trials’ ROI, Phesi warns 

The firm’s report warns a ‘significant number’ of trials are responsible for overcomplicated protocol designs, that can create otherwise avoidable delays in drug development times, leading to increased burden on patients. 

“Collecting a good amount of data is important for clinical trials, but it has to be the right data at the right time,” cautioned Phesi founder and CEO Dr Gen Li.

“Many of the clinical trials we analysed had redundant outcome measures. For example, trials often use several different physical performance status measures on the same patients – which is a huge added burden for that patient even though each scale is measuring the same thing. This also puts undue pressure on investigator sites.”

Using proprietary data from its Trial Accelerator platform the firm analysed 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.

Phesi identified five top disease indications, including 146 protocols designed to capture 1,821 outcome measures, ranging between 1 and 102 measures per protocol, and including both primary and secondary outcomes.

The company said evidence revealed that the more outcome measures included in a trial protocol, the lower was the percentage of those outcomes reported in results. On average, more than a third (35%) of outcome measures were not reported.

Trials with fewer than the median number of outcome measures in a protocol reported 94% of those measures in their results; meanwhile, those that collected more than the median number of outcomes typically reported just 56% in their results.

While the complexity of trial design has been known previously to affect the cost, timelines and patient burden of clinical development, this had been difficult previously to measure objectively, explained Li. Advances in Phesi’s Trial Accelerator platform allowed the firm to analyse its proprietary data and assign a ‘complexity score’ to trial protocols, modelling the true impact of the number of outcome measures.

In one example, the report compared type 2 diabetes trials from two different sponsors; using a median of 25 outcomes measures for their trials had lower site enrolment performance (10.2 patients per site vs. to 11.2) and a lower enrolment rate (0.46 patients per site per month vs 0.53) compared to the other sponsor with a median of only 10 outcome measures.

“Investigators need to avoid data FOMO and make sure they are only collecting the data they truly need,” advised Li.

“Being more precise with outcome measures makes it easier to select trial sites, recruit more patients in a shorter period… and collect better quality data from those patients. All these compounded benefits lead to better return on investment, an increasingly pressing issue in the current economic environment for the pharmaceutical industry.” 

Pic: Hush Naidoo/Jade Photography