In a previous weeknote, we outlined our one-pager that described our recommendations in a quick way. We are now happy to provide more detail to one of our recommendations, and will consistently be updating these weeknotes with the other ones as they become available. Please click here for the Reduce, Recalibrate, and Redesign brochure that we will be taking to stakeholders in order to get partners on board to implement this recommendation.
We all know that too many clinical trials fail at different stages of the treatment assessment process.
These failures lead to both financial and human cost, among a number of other issues including delayed treatments, clogging of the regulatory pipeline, and an uncertain market.
So we built a machine learning model to help identify the items in trial protocol associated with this failure, so we could identify early which trials would fail. It turned out that the number of endpoints and amount of eligibility criteria were among the strongest associated with trial failure.
Both of these findings reflect many conversations we have had with stakeholders and experts- more complex trial design often lead to failure. There is a vicious cycle in treatment development: when a disease or molecule is less well-understood, the clinical trial protocol grows in complexity. In other words when there is more uncertainty, there are more eligibility criteria and more endpoints to manage that lack of understanding. But our analysis, as well as the findings of other researchers, indicates that this strategy leads to failure. Identifying these trials earlier in the treatment development cycle by using models like ours during the facilitated review process can provide more information to sponsors facing difficult decisions about whether these higher risk trials should be modified or halted. We will now take this to the pharmacuetical companies and government regulators and devise a pathway that this recommendation can be implemented.