Supply chain management is often an exercise ground for clichés. However, ‘navigating the obstacle course’, ‘overcoming the hurdles’ and ‘facing the Herculean challenges’ of a successful clinical supply chain barely do justice to the minefield, for want of a better word, that sponsors must cross to ensure materials reach participants on schedule.

‘Running the gauntlet’ must surely come closest, with the timely delivery of trial materials facing body blows from insufficient supply, cross-border regulation, cold chain breaches, delayed patient enrolment and date expiration. Then there’s the problem of oversupply: wasted drugs being stored or transported to clinical sites for participant numbers that simply haven’t been secured.

So, how can such idiom-strewn problems be overcome? The answer is, of course, by forecasting, but the large-scale, complex and cross-border nature of many trials requires a sophisticated approach indeed. 

Not just the means but also the methods of forecasting have evolved considerably in the past few years. With huge budgets at stake, and the desire to get safe and effective drugs to market as soon as possible, there is growing consensus that the time-consuming and occasionally error-prone spreadsheet simply won’t cut it anymore.

Changing forecast

A volume of industry-specific software and web-based solutions is now available to assist sponsors in their forecasting endeavours. While huge savings can be made, many companies are yet to make the move, fearful of the investment costs.

Zaher El-Assi, president of Merge eClinical, makes this point well. Outlining the benefits of financial forecasting software and electronic-data-capture (EDC) in an article for the Nice Insight website, he writes: “Despite the advantages of EDC systems, and cloud-based systems in particular, many sponsors and CROs, particularly smaller companies, continue to use paper-based systems or rely on myriad Excel worksheets to manage their clinical trial data.

Cloud-based platforms designed for clinical trial management are now available as modular systems that can be customised to meet the needs of any given clinical trial.

“Most often, the reason is an expectation that the required upfront investment for advanced software solutions is too great, and the need to sign long-term contracts is too burdensome. While that scenario may have been true when EDC systems were first introduced, it need no longer be the case today. Cloud-based platforms designed for clinical trial management are now available as modular systems that can be customised to meet the needs of any given clinical trial, regardless of the size and complexity.”

Controlled risks

Choosing the right solution for your business or specific project, and maximising its use, relies upon first recognising the elements of the trial that can actually be controlled – and, even more important, those that cannot.

Forecasting via the most efficient means can then be used to put the necessary precautions in place from the outset, and create a trial timeline that allows the ongoing risks to be closely monitored, projected and managed.

Take study design, to which changes after a trial has started can cause huge increases in outlay for vendors. Instead, this needs to be thoroughly interrogated during the planning stage, and cross-compared with available predictions. For example, if the intended design is financially feasible within the overall budget only if enrolment reaches 90% in each of the clinical sites within an allotted time, how have you guaranteed these conditions? Have you forecast that this percentage is achievable based upon previous knowledge of enrolment rates in the proposed sites, and current knowledge of the patient population?

With such tight margins, forecasting may reveal that creating a less demanding study design – or securing additional space in the budget – before the trial begins may well be advisable.

Data-driven approaches are key to the burgeoning models of forecasting technology, not simply in the way that analysis is carried out, but also in how data is recorded, shared and processed in the first place.

This is being driven by two factors: a growing availability of near real-time data for many moments in the supply chain, such as monitoring the temperature of drugs in transit in the cold chain; and a move towards a cloud-based approach that allows this data to be shared at source, aggregated and accessed immediately by clinical trial managers wherever they are in the world.

At the earliest point of forecasting, an assumption-based approach is, to a certain degree, inevitable, relying upon the wisdom and experience of the professionals involved. However, this doesn’t rule out helpful comparison with previous trials, and a wealth of insight can be gained from comparative analysis of multiple historical trials, which can throw up problematic areas and enable sometimes unidentified risks to be highlighted and mitigated.

Applying such an approach to the choice of vendors, especially when designing a new cross-border trial, can help ensure that expectations will be met. Particularly for smaller sponsors and CROs taking on a large-scale trial for the first time – one that could involve several countries and thousands of patients – requesting information from vendors and investigational sites regarding historical regulatory clearance times, enrolment rates and patient adherence can help distinguish between a reliable, good-value partner and a likely false economy. Enrolment protocols should also be considered in the light of previous performances, and adjusted accordingly.

Keep it current

Once a trial is up and running, assumption-based models should swiftly be supplemented and replaced by data-driven, evidence-based approaches. The best clinical supply management solutions are able to facilitate this steady drip of information as and when it arises, keeping forecasts accurate and current.

While outsourcing tasks such as cold-chain logistics can bring financial and performance advantages, and has supported growth in multinational trials, these advantages are reliant upon all stakeholders in the trial being engaged to communicate information and take part in timely data sharing. Only then can broad and accurate forecasting of the trial as a whole allow problems to be caught early and the effects mitigated.

Analytical dashboards, such as IBM’s Cognos-based solution, allow managers to drill down on data such as forecasting the under-enrolment of patients at each active site, and the anticipated costs of sourcing, setting-up and supplying additional sites if they need to be added. This can even include the advertising spend required per patient to publicise the trial, as well as other associated costs.

Consider an instance in which delays to patient recruitment mean additional costs for storing dispatched drugs, or adding a new market means that by the time regulatory clearance has been obtained, the current batch of drug will reach expiration point before the trial period is complete. Being able to visualise an overview of all of these considerations and the related costs and lead times allows clinical trial managers to act with agility and speed.

Stream team

Of course, any advanced software is only as strong as the professionals setting up, using and interpreting it. With individual elements such as the design of packages and patient kits taking on huge significance as a result of the scale of today’s late-stage trials, a multidisciplinary project team is essential from the start.

You need a very well-structured project team at the start, one that is cross-functional, because I think this is going to pull all the different functions together more effectively.

David Gilliland, Daiichi Sankyo’s director of clinical supply operations, says: “You need a very well-structured project team at the start, one that is cross-functional, because I think this is going to pull all the different functions together more effectively.

“The team approach is going to become more critical downstream, and the technologies we’re looking at need to work on a cross-functional basis – not just on one area.”

This, he adds, is particularly important in the changing climate of trials: “Nowadays, it’s become much more critical that the team works collaboratively. The timeline still needs to be hit, so we’re trying to do more complex studies but in a shorter timeframe.”

Take your time

While the race to market is compelling, taking an overly optimistic view of lead times will result in goals being missed, and greatly increased expense. Continuous forecasting can help tackle this, combined with in-built buffer periods for the kinds of delays likely to be encountered in, for example, soliciting comparator drugs from competitors or sourcing difficult-to-obtain measurement devices in the patient kit.

With more small and large CROs and sponsors recognising the importance of comprehensive forecasting, it’s a capability undergoing rapid development and expansion, which will ultimately benefit the whole industry.

As an industry with scientific innovation as its lifeblood, big pharma has nonetheless been seen as a slow adopter of some technological advances, such as continuing manufacturing, that have driven productivity in consumer industries. 

In supply forecasting solutions, there is an opportunity to contain the escalating costs of ambitious trials and to take control of a process at the heart of what pharma does: demonstrate the efficacy of treatments that can change patients’ lives for the better.  