Planning for Scale in Complex Bioprocess Development

Whitepaper - Where speed meets complexity in bioprocess development Bioprocessing White paper | Bioprocess services Natraj Ram, PhD Vice President of Innovation, Thermo Fisher Scientific Natraj Ram serves as the Vice President of Innovation at Thermo Fisher Scientific, where he is responsible for developing and implementing strategic roadmaps for product, technology, and service innovations. With extensive experience in biologics, he has a track record in process and analytical development, verification, and commercial manufacturing. Natraj holds a PhD in Chemical and Biochemical Engineering from Rutgers University. Adam Goldstein, MSc Senior R&D Director, Thermo Fisher Scientific Adam Goldstein is an accomplished biotechnology executive and global leader in biomanufacturing innovation, operations excellence, and technical development. As Senior Director of R&D for Single-Use Technologies and Innovation at Thermo Fisher Scientific, Adam spearheads strategy and execution for next-generation manufacturing solutions that enable process intensification, sustainability, and scalability across the company’s bioproduction network. As biopharmaceutical development continues to evolve, biotech companies are facing increasing pressure to balance speed with growing process complexity. Platform approaches have streamlined development for established modalities like monoclonal antibodies, but those same expectations for speed and scalability are now being applied to more complex, variable biologics. At the same time, efforts to accelerate timelines and improve productivity through higher-titer manufacturing and process intensification are also introducing new layers of complexity. Decisions made early in development can significantly impact scalability, manufacturability, and regulatory readiness, particularly when technologies and infrastructure have not fully kept pace. In this Q&A, in-house specialists from Thermo Fisher Scientific explore how biotech companies are navigating these competing pressures, where challenges most often emerge, and how more integrated approaches to process design and development can help balance speed, complexity, and risk. Where speed meets complexity in bioprocess development 2 thermofisher.com/bioprocessforbiotech Q: Bioprocessing is becoming increasingly complex. What are the biggest challenges biotech companies face today in both upstream and downstream process development? Ram: In terms of the traditional mAb (monoclonal antibodies) modality, it’s really become platformized. So, then the question is less of complexity and more about speed. For example, what used to be a one-year gene-to-IND timeline is now, “Can you do it in six months?” That pressure has become more pronounced as people assume it can be done easily. When you look at bispecifics, gene therapies, and other complex modalities, complexity is still a real issue. The bar is set by what we expect for mAbs, both in terms of operational platforming as well as product quality and the speed of development. Additionally, the underlying technologies aren’t always there yet for these newer modalities. Goldstein: As we move into a new era, there’s a lot of mixed-mode complexity happening. Teams are trying to compress their timelines, reduce resources, and still push for speed and efficiency. Often, increasing titers center on adopting technologies like perfusion, which frequently serves to increase process complexity alongside driving higher titers. So, the game is really about how to achieve higher titers and get a better cost per gram, while also giving teams the tools, automation, and platforms to actually run those processes effectively. Q: Where can process intensification create unintended risks for more complex therapeutics? Ram: Today, process intensification is primarily a mAb play and not one for complex therapeutics. However, people are using perfusion for complex molecules—not for intensification, but for feasibility. If we look at the process of intensification in the context of improving efficiency and productivity, today that primarily applies to mAbs. The most crucial question then becomes this: at what point do you introduce process? Even some experienced companies are still not introducing it in Phase 1. They tend to bring it in later, as these approaches don’t always meet early-phase speed requirements. However, that’s where that unintended risk shows up. If I’m running fed-batch for Phase 1 and shift to process intensification for Phases 2 and 3, what does that comparability look like? The bottom line is, if we can operationalize the process intensification and provide a service that combines product knowledge and process knowledge, then it becomes easier for biotech companies to adopt these approaches much earlier in their lifecycle. Goldstein: We know intensification can deliver higher productivity, smaller footprints, and more agile manufacturing. But it also comes with a much tighter operating window. So now the focus shifts to managing that risk — getting real-time data, real-time feedback, and making sure performance can stay consistent at scale. That’s where AI platforms may play a bigger role. Q: What are the most common challenges biotechs encounter when planning for scale-up from development to manufacturing? Goldstein: A big part of it is designing the process with scale in mind from the beginning. What works at the bench scale doesn’t always translate to GMP manufacturing. For example, you can get very high titers during development using large volumes of media, but that may not be operationally practical or cost-effective at manufacturing scale. Another constraint can be the equipment. Tools available at the small scale may not exist — or may not be practical — at full scale. Even if they are sucess depends on whether teams have the training and experience to run it effectively and interpret what the data is telling them. The game is really looking at how to drive higher titers and get a better cost per gram, while also giving teams the tools, automation, and platforms to actually run those processes effectively. thermofisher.com/bioprocessforbiotech 3 Ram: Ideally scientists understand a drug’s critical quality attributes at 5,000 liters while they’re still working at small scale. It’s almost like reverse engineering. In reality, the knowledge gap is still there today. Once a process is developed, teams also need the right tools and support to translate it to large scale without having to do significant rework. Q: What can be some of the biggest pitfalls biotech companies face when applying traditional biologics manufacturing assumptions to more novel modalities? Goldstein: The industry has built strong, standardized platforms, and they work well for established modalities. But as you move into intensified processes and more complex molecules, you can’t always apply those same assumptions. You need a much deeper level of process understanding, stronger control strategies, and the right automation and PAT (Process Analytical Technology) tools in place to reduce additional risk. Ram: When you try to fit complex biologics into traditional facilities, things can get tricky. A scientist might design a process that works from a feasibility standpoint, but it doesn’t align with what a CDMO (Contract Development and Manufacturing Organization) can execute based on their platform or equipment. What is needed, then, is a deliberate translation layer that systematically converts theoretical process design into an operationally viable, execution-ready strategy aligned with real-world manufacturing constraints. Q: How can early consideration of downstream purification strategies prevent delays later in development? Ram: With mAbs, you’re usually optimizing for efficiency and cost, and with more complex biologics, you’re often optimizing for feasibility and quality. Both approaches can introduce delays if you’re not thinking ahead. That’s where early modeling can help. Even simple process economics or modeling can give you a sense of whether something will be scalable and what the cost implications might look like before you get too far down the path. Goldstein: As the industry moves toward intensification and more capable lean processing, a lot of the existing infrastructure is not designed for the types of processes we’re seeing emerge — higher titers, higher concentrations, faster turnaround. Retrofitting those facilities, especially if they’re stainless steel, can be expensive and time-consuming. That’s where single-use technology really can come into play. Over several years, single-Use technology has supported many customers in making rapid changes. In many cases, retrofitting the right single-Use technologies can be a more strategic and efficient path than retrofitting stainless steel systems. Q: How does early process design and data strategy influence regulatory readiness and what should biotechs prioritize to avoid delays during later stage submissions? Goldstein: Process intensification is not always practical to introduce in later development programs such as Phase 2 or Phase 3. While it is technically feasible, doing so can jeopardize regulatory strategies, including BLA (Biologics License Application) submissions, by introducing additional risk to validation and acceptance. These considerations should be addressed early in development and scale-up, where changes can be implemented with far less regulatory and operational risk. Ram: Data and regulatory readiness ultimately come down to awareness. Digital tools can be extremely effective. If you reverse-engineer a system based on requirements, you gain visibility into what’s required for regulatory filings and change management. That’s where digital tools could make a big difference by helping guide how experiments are designed, how data is captured, and how everything is documented so that it’s usable later in the process. For biotech companies, integrated process design, combining early modeling, scalable technologies, and connected upstream and downstream strategies, is important to reduce risk and accelerate the path to manufacturing. 4 thermofisher.com/bioprocessforbiotech For research use or further manufacturing. Not for diagnostic use or direct administration into humans or animals. © 2026 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries unless otherwise specified. WTP-13333850 Q: How can integrated upstream and downstream services help improve scalability and overall process robustness? Goldstein: Bringing these capabilities together allows us to help translate that technical knowledge directly to biotech customers, helping them understand how our technologies work, how they can help to improve their processes, and where we can apply strategies to help intensify operations. We can support this work internally at our global Thermo Fisher Bioprocess Design Centers (BDC) or through on-site support, which can make it easier to implement processes and create more efficient, streamlined workflows by utilyzing hands-on development and scale up with the customer at the points of use. Ram: In many biotech companies, it can be common for upstream and downstream teams to work in siloes. The same is true in manufacturing facilities. Most of the challenges we see aren’t upstream or downstream — they’re integration problems. You can look at this at two or three levels. At the facility level, you want to design an integrated operation that can flex across different titers, volumes, and process needs. Downstream must be able to adapt to that as well, so the ideal facility is one that combines flexibility with efficiency. The reality, though, is that no one has a complete view of that connectivity, especially not at the Phase 1 stage.When upstream development is carried out without clear visibility into how increases in volume and titer will impact downstream processing and manufacturing, delays become inevitable because critical interdependencies were not addressed early. If you can model those interactions and show teams what those downstream implications look like, you can make better decisions much earlier in the process.Q: What does true service integration look like and where do external partners often fall short in achieving it? Ram: CDMOs are generally set up to serve the majority use case. When an application aligns with that operational and service design, it can be well supported through established workflows. However, more novel or less conventional biologics often fall outside these assumptions and require additional support.That’s where there’s an opportunity for more flexible, integrated service models — ones that are tied to a specific platform and more focused on helping customers move their programs forward while also supporting mainstream needs.Ultimately, the evolution of biopharmaceutical development has reached a critical juncture where the pursuit of speed must be balanced against the inherent technical risks of increasingly complex modalities. Success in this new landscape may be supported by moving beyond fragmented development cycles toward a model of integrated process design. By prioritizing early-stage modeling, adopting flexible single-use technologies, and leveraging digital tools for real-time data strategy, biotech companies can help mitigate the unintended risks of process intensification and help to ensure that a molecule’s path to the clinic is defined not by unforeseen bottlenecks, but across a proactive strategy that accounts for the interdependencies of the entire production lifecycle.The companies that succeed in the next generation of biologics will not be those that move in isolation, but those that design for speed, scalability and regulatory readiness simultaneously will likely be well positioned for success. Learn more at thermofisher.com/bioprocessforbiotech Delaying process intensification or scalability decisions can introduce significant comparability and regulatory risks later in development.