Biologics are set to make up 57 percent of the pharma market by 2030, yet the volume behind each product is shrinking. Fewer than 20 percent of commercial products now need more than a ton of antibody a year, with 80 percent needing under 500 kg, and 47 percent under 100 kg, by Sartorius's own analysis. bioTRAK database services at BioProcess Technology Group, meanwhile,puts the median marketed antibody at around 48 kg a year, with roughly 60 percent of the pipeline projected below 100 kg.

Such a shifting landscape, incorporating more products at smaller scale, is ripe for flexible, intensified manufacturing processes. Sharpening that pressure is a patent cliff of roughly $236 billion in global sales due to lose exclusivity between 2026 and 2029, with every expiry opening the door to biosimilar rivals that erode the originator's price. AbbVie's Humira (adalimumab), the best-selling drug in history at more than $200 billion in lifetime sales, lost its US patent protection in 2023 and saw its net price drop more than 40 percent inside the first year. Its market is now being shared across a plethora of biosimilar makers, each manufacturing smaller volumes, which is a pattern intensification is built to serve.

Regulators have pushed the same way for over a decade. In fact, back in 2011, Janet Woodcock, then head of the US Food and Drug Administration's drug evaluation center, was already pressing the industry away from stainless-steel plants having cast them as decades out of date (this shift she later set out in print). Since then, the FDA has run workshops on speeding continuous manufacturing, ISPE has issued a good-practice guide on continuous biomanufacturing, and the ICH's Q13 framework was finalized in 2022. Thus, the regulators are with us and they want to move in this direction.

A single-use intensified process can cut cleanroom footprint by around 70 percent, by Sartorius's data, above the 30 to 50 percent independent facility-design analyses cite for single-use alone, since intensification consolidates unit operations further. That feeds straight through to the HVAC system, the single largest energy draw in a biomanufacturing plant, which shrinks by a similar margin.

Some drugmakers are already adapting to the new ecosystem and embracing more flexible technologies and platforms. More than 80 manufacturing facilities worldwide have implemented some form of PI, and one in every three bioreactors Sartorius sells is now ordered with a perfusion option.

Contract development and manufacturing organizations (CDMOs), in part because of their own multi-client flexible needs, have historically been viewed as leading the process intensification charge, but there are increasing signs large pharma is following. At the Sartorius Process Intensification Forum in Hydrabad, India, in June – where I also spoke about these issues – MSD (known as Merck & Co. in North America) described building its own continuous capacity from scratch, having skipped the legacy stainless-steel plant entirely as a relatively late entrant into biologics. Sanofi took a similar route with the intensified, continuous biologics facility it opened at Framingham, Massachusetts, in 2019. Meanwhile, AstraZeneca has conceived its concept of a “plant of the future” built on connected sensors and real-time process visibility, digital twins, AI that flags process drift, end-to-end digital orchestration across development, manufacturing, and quality, and humanoid robots on the floor for the routine work.

But much of the industry is still moving the other way. North Carolina alone secured more than $10 billion in life-sciences investment in the last year or two, much of it in large, fixed capacity. In the town of Holly Springs three Fortune 500 companies have together committed $6.7 billion within two square miles. CDMO Fujifilm Biotechnologies's site, which opened in 2025, runs 8 x 20,000 L bioreactors with plans to double that capacity, while innovators are building large stainless-steel installations of their own on the same doorstep: Amgen is adding a $1 billion drug-substance expansion and Genentech/Roche has more than doubled its commitment to around $2 billion.

These investments are not a misjudgment, per se, as much of the new capacity is aimed at genuinely high-volume products and at the surging obesity-drug market, and at US reshoring under a tariff regime pushing manufacturing back onshore. However, they are also a product of long-term planning decisions and legacy mindsets. Drugmakers sit on stainless-steel plants they cannot write off overnight, and the pressure to keep expensive capacity full pulls against building something smaller. Even the longstanding bottleneck of integrating the downstream is largely a legacy problem, with results of a Sartorius survey taken through the BioPhorum industry consortium finding only 14 percent of companies had achieved real connectivity between unit operations, the obstacle being control systems in older plants that are unable to talk to one another.

The question is whether all that fixed capacity, committed on long horizons, fits a pipeline moving the other way. The biggest new plants are being built around 20,000-litre reactors, even as 47 percent of commercial products now need under 100 kg a year.

The biologics facility of the future should be designed for adaptability, rather than utilization. It must combine modular single-use infrastructure, process intensification, automation, PAT, and AI-enabled decision-making to adjust continuously to changing product portfolios and market demand. The goal is no longer simply to build the largest facility possible, but the most productive, flexible, and intelligent one – without investing in fixed capacity that may never be fully used.

Newsletters

Receive the latest pharmaceutical news, personalities, education, and career development – weekly to your inbox.

I have read and understand the Privacy Notice*

Subscribe