The STREAM trial included complex study regimens – recruited participants could be assigned to any of the 13 treatment combinations with eight IMPs typically prescribed during the intensive phase and four IMPs during the continuation phase. In total, 13 IMPs had to be available at each site to allow recruitment and uninterrupted treatment for participants.
In order to avoid treatment interruptions, there was a need for sufficient stocks of IMPs to be available while minimizing costs of inventory, storage, and waste due to expiry. Additionally, long supplier lead times and relatively short IMP expiry (typically less than one year for one of the IMPs) had to be taken into consideration. Moreover, delays in recruitment of participants due to strict inclusion and exclusion criteria made desired stock level calculations challenging.
These complex circumstances required the development of a robust forecasting system to continuously monitor risks of supply chain disruptions including timing of recruitment at each site; the estimated number of participants to be recruited monthly; typical participant’s weight; and the estimated number of participants who may be lost to follow up.
The supply chain strategy implemented for the trial relied on a number of forward planning elements including pre–positioning medicines at regional depots for onward shipping to the sites, regular monitoring of stock levels, and continuously reforecasting IMP needs based on the latest recruitment data and trends to mitigate the risk of supply chain disruptions.
Our initial forecasting was based on optimistic scenarios of rapid recruitment of up to 50 participants per month per site. This resulted in an oversupply of IMPs and subsequent expiry and need for disposal and replenishment. However, once recruitment started, monthly recruitment reports were generated which were used to calculate expected recruitment for the upcoming period and enabled realistic forecast calculations.
The STREAM trial was implemented at 15 trial sites, across eight countries with site pharmacists with various levels of experience in forecasting. In order to facilitate their tasks, a simple quantification tool was developed to help with calculating needed quantities of IMPs. This was disseminated to each site pharmacist along with training by the Sponsor pharmacists.