The Biggest Early-Stage Clinical Supply Mistake? Building Too Much, Too Soon.
By Rachel Grabenhofer, Chief Editor, Clinical Supply Leader
Gobsmacking is how one visitor to the U.S. for the FIFA World Cup described his trip to New York, in an NPR interview. “From the size of the food, to the size of the buildings, to the skyscrapers.”
For those who don’t know (I had to look it up), gobsmacked refers to being overwhelmed with wonder, surprise, or shock (good or bad). According to NPR, other fan reactions flooding social media echo that same sentiment: everything in America is huge.
As consumers, we’re often led to believe more is better, and in some cases it can be. Especially in biopharma – where life-saving treatments are scaled up to expand patient access.
Perhaps counter-intuitively, though, in early-stage biotech development, when investing in design processes, systems, and infrastructure, more isn’t necessarily better – at least in the beginning. Alicia Collins at Tune Therapeutics explains, “You really need to resist the urge to overcomplicate. In an emerging company, there are many things you don’t know yet.”
Collins is the vice president of technical operations for Tune Therapeutics, a clinical-stage biotech company specialized in epigenome editing – or epi-editing – medicines. She also previously served as the vice president of supply chain and external manufacturing for the company, and before Tune, she directed patient operations and supply planning for cell therapies at Bristol Myers Squibb.
So, what does Collins mean by overcomplicate? Development teams at emerging biotech companies are under high pressure to prepare for scale up, and that pressure often leads to a familiar question: Are we going to need a system?
Her response? “No. Let's be honest. Excel is the planner's best friend." Collins argues that for early-stage biotech companies, one of the biggest clinical supply mistakes is over-building at the beginning – solving for problems that don't exist yet.
"I’ve had non-supply chain experts ask if we need a system, like for enterprise resource planning (ERP), and whether they should budget for it,” Collins says. “I’m not even sure if they themselves fully understood what they meant by system.”
In a recent discussion on clinical supply in emerging biotech companies, Collins circled back to the fundamental of keeping it simple at the start for a few reasons – early assumptions can change, and complexities are often underestimated. Our chat also revealed insights into planning (rather than building) for future growth and partner selection, as part 2 in this series explores.
Operational Ecosystem: Don’t Overbuild In The Beginning
For emerging biotech companies, uncertainty is a constant. "There are many things you don't know – and some of the things you do will 100% change," Collins says. Because future needs are difficult to predict, she cautions against building processes and systems for every possible scenario.
"Trying to make something fit all the phases and all the places you're going as a company is a waste of energy and resources," she explains. Instead, Collins recommends a disciplined approach: build only what is needed today and expand as needs become clear.
"Build the basics, and that is all," she says. "You only add on when you need to."
Accept That Early Supply Assumptions Are Evolving
The fact that early assumptions are still evolving is why Collins cautions against overbuilding from the start. A supply plan based on them may look good on paper but during execution, it may fail.
The biggest assumption for Collins is time. "In a supply chain, you have many timelines coming together to create a supply plan – especially because early-stage companies often depend on external suppliers and manufacturing partners.”
Misalignments between timelines can happen if it’s the first time you’re making your product at scale, producing a GMP batch, or running your process – or “if you don’t have a lot of data on how your process performs,” she notes.
Many of these challenges are amplified with advanced therapies, where manufacturing processes and supply networks are still forming alongside the science; for example, when companies are working with new modalities.
“Processes for cell and gene therapy are relatively new – especially when scaling them up for the first time or making the first GMP batch," she says, adding that sponsors and partners sometimes develop the processes together. "Many external manufacturers are also new in this area and still learning.”
Without established experience, predicting timing is especially difficult. "When a manufacturer gives you a schedule and an estimated lead time, you have to scrutinize that. But also know that you’re scrutinizing it through the lens of not knowing, yourself, exactly how long it should take," Collins explains.
Besides potentially delaying manufacturing, evolving assumptions can shift critical coordination points for complex processes – both across and within organizations. “You can see in the schedule where, ‘Oops, this didn’t finish in time,’” Collins illustrates, “but we needed it to start something else, and we had those processes running almost in parallel to get the timeline you wanted.
“That sequencing can be lost simply due to the amount of work you’re trying to have that manufacturer complete,” Collins continues. “That's why it's important to always be very closely in lockstep with your CDMO partners,” Collins emphasizes.
Forecasting Is An Art, Not A Science
Besides manufacturing and coordination, assumptions can break forecasting – which for Collins, is a clear example of the uncertainties supply teams must manage. “Forecasting is an art and not a science,” she says – teasing that supply professionals often think less about whether a forecast is wrong and more about how wrong it might be.
"Forecasting is wrong,” she states. “Everyone in supply chain has the saying: ‘It’s not if the forecast is right or wrong – it’s how wrong is it?’ Is it 5% wrong? Is it 95% wrong?" Those questions matter when it comes to inventory and demand – which rarely are perfectly aligned.
Early-stage companies are often conservative, sometimes making just one batch if they think that’s all they need. “You don't want to spend money on multiple batches if you don't have to. You don't want to risk expiry," Collins notes.
The challenge, of course, comes when the forecast changes. If the company made one batch based off the original projection and that increases, obviously they are going to need more supply. “But that supply could take nine months,” Collins highlights. “People need to understand the impact of those changes” – which doesn’t happen if supply teams are siloed, as is often the case.
Managing these situations depends heavily on communication across functions. This means helping teams understand not only what is changing, but what those changes mean for lead times, inventory, and future supply. "It's making sure that the people who are doing the forecast really understand the implications of changing it, and that they realize the lead times for a resupply," Collins stresses.
Ultimately, supply leaders are challenged to understand how evolving assumptions interact to anticipate the consequences when one of them changes. "There are multiple areas where assumptions can break down, so you're really trying to align all the uncertainties and understand the impact of one thing happening when it shouldn’t – or not happening when it should,” Collins summarizes.
Where Clinical Supply Complexities Are Underestimated
Compounding the challenge, the more interconnected the system, the more difficult it is to understand how assumptions interact – embedding complexity so deeply into the supply model that it’s underestimated until it becomes visible during execution. This takes different forms depending on the therapy.
“For example, with autologous therapies, there's no finished goods inventory buffer,” Collins explains. “If one thing on your bill of materials isn't available, you're shutting down everything." This complicates planning, as organizations must closely monitor supply risks and establish mitigation plans as early as possible.
For make-to-stock therapies, Collins sees different risks. "Here, cold chain and materials can be an issue. If materials go out of temperature when you're shipping them, hopefully you have data to support excursions – because if you don’t, you can’t use them.” This complicates the process because in order to determine what a product can withstand, stability studies must be planned in advance across all supply functions.
Yet another complication can be the reliability of supplier partners. "When you’re early phase or small, and you're sourcing materials for your manufacturing process, it is important to make sure your suppliers are capable,” Collins advises.
For instance, before Tune, Collins experienced development programs progressing faster than suppliers could scale. "The timelines were moving so quickly, the supplier either couldn’t support the amount required as it increased or they didn’t have the capability to change when needed.”
Examples like these reinforce Collins' broader point: companies shouldn’t overbuild in the beginning because there are too many underestimated complexities and untested assumptions. This is not to say they should ignore the future, however.
"Supply chain is obligated to future thinking – it is a big planning function,” Collins says. And as programs advance, supply leaders must determine whether today's supply model and partners can grow alongside them, as part 2 in this series explains.
Collins has 29 years of experience in the pharmaceutical and biotech industry at Merck, Genentech, BMS, and Tune in both domestic and international roles. Her experience encompasses manufacturing, supply chain, external manufacturing, strategic sourcing, and materials management. She holds a B.S. in Integrated Science & Technology with a double concentration in biotechnology and environmental science from James Madison University.