Cutting Waste, Boosting Efficiency: Digital Solutions for Clinical Trial Supply Challenges

For anyone working in the clinical supply chain space, I highly recommend reading Kristel Rens’ article Digitalized Drug Forecasting Minimizes Waste In Clinical Trial Supply Chain. I found it incredibly validating—dare I say cathartic? It perfectly captures many of the challenges we face and aligns with the industry-wide changes we're striving for at DataCompass.

Clinical trial supply forecasting can often feel more like an "art" than a science. If you gave the same trial protocol to multiple clinical trial supply planners, you’d likely get just as many different supply plans in return. Kristel points out a key issue: the level of granularity that can be obtained in commonly used forecasting technologies (i.e. Excel) often doesn't match the complexity of modern trial designs. 

This is especially true in oncology, where trials frequently offer "investigator preference" for chemotherapy regimens, paired with intricate BSA/AUC dosing schemes. In this scenario, how does one accurately plan for an oncologist's choice to use either carboplatin vs. cisplatin, as well as an average AUC of 5 vs. 6 for carboplatin on a NSCLC study? This is where control and scalability of scenario planning is key.

What is an important concept to internalize is that the plan will change and evolve over time, and it is important to show this through effective archive and version control over change in forecasts and study design, and how it relates to different supply quantities and labeling campaigns over time. DataCompass's clinical supply forecasting tool captures just this, along with BI tooling integration to show how updated enrollment (as the actuals from past and current months "replace" the forecast from future months) is trending toward the current supply plan, so that timely adjustments may be triggered and escalated to leadership. 

This is an exciting time to be in the clinical trial supply chain, and with the continued development and adoption of AI-driven modeling, the need to stay present in these technological advancements is ever more important.