Of all the tools used by clinical teams to manage the clinical study supply chain, Interactive Response Technology (IRT) solutions are amongst the most ubiquitous. I’ve personally been involved in the implementation of hundreds of study IRT solutions over the past 20 years. Many facets of the solution implementation process have the potential to introduce risk to the seamless execution of the trial. These include ambiguous or poorly translated requirements, poor programming, configuration and validation and disruptions to systems caused by mid-study changes. One facet consistently introduces risk: the overall availability of the appropriate amount of medication to service the treatment needs for all participating subjects in the trial. The IRT solution may be well designed and implemented, but it will only be as good as the amount of drug made available to fuel it. It is this point of risk that often creates negative perceptions of IRT solutions, with the overall value of these solutions for clinical teams eroded as a result of critical stock out situations.
Stock out situations are painful for virtually every constituent of clinical trial set up – clinical supply and operations teams, IRT suppliers, drug distributors, sites and above all subjects. So how do you determine how much drug is needed to treat all the subjects in your study? Clinical teams will develop a supplies strategy at the outset of a study. Many critical ingredients are typically considered, including the study randomization/dosing design, treatment ratios, randomization balancing/stratification factors, dispensing/visit schedule, titrations and other dosing variability, optimal packaging configuration, countries involved, participating depots, supply lead times, expiry and labelling constraints, and predicted recruitment rates.
Based on our experience and feedback, the most widely used method to consider these complex factors is spreadsheets, which try to absorb as many of these considerations as possible. Alternatively, more sophisticated formal modelling tools are employed to simulate and forecast supply needs. It is important to note that many of these factors are assumptions that vary in their accuracy, and these inaccuracies have the potential to compromise the validity of any conclusions drawn from the forecasting activity. At worst, this could become a “best guess” as the number of inaccurate assumptions increases.
In a recent webinar, I described a real-life study scenario in which what we might call a perfect storm of events contributed to significant stock outs in a large, global, Phase III study for a neurological indication. In brief, initial forecasts were made using traditional spreadsheet-based tools, but many significant study assumptions were not reflected in the actual execution of the study, particularly in connection with subject recruitment rates and stock expiry constraints. Stock out situations started to occur. Analysis and remediation efforts using downloaded IRT reports proved to be ineffective. The study team elected to adopt Bioclinica’s unified IRT and Trial Supply Optimization (TSO) solution, which holistically integrates real IRT study data with the ability to model and forecast future study behavior. The modelling activity highlighted points of risk in the study supply network and easily highlighted some immediate actions needed to address some of the more urgent upcoming issues.
The supply chain consultants, IRT project management team, blinded sponsor supplies team and packaging vendor collaborated to help ride the storm. In addition, the clinical team elected to perform more of these re-optimization exercises throughout the course of the study to help avoid any further stock out situations.
The unified IRT/TSO methodology had not previously been considered by the sponsor and study team but enabled a solution to a very deep problem. These more sophisticated tools are appropriate for many of today’s complex, large-scale studies, helping to proactively identify potential issues that could compromise the study’s drug supply. We believe that it is best practice to use one of these tools both at study start to build a model and perform up-front planning and forecasting and then throughout the study for regular review to support proactive planning and adjustment. To learn more about the case study above and the solutions that were implemented, download and listen to the webinar recording or contact me at Ehsan.Ramezani@bioclinica.com.