Every Approved Biologic Was Grown in Media. Most of That Media Is Decades Out of Date.

May 28, 2026

|

ARTICLES

Cell culture media is not a glamorous subject. It does not appear in press releases. It does not feature in pipeline announcements. It is not discussed at investor days. But every biologic medicine that has ever reached a patient was grown in it.

  • Trastuzumab changed the prognosis for HER2-positive breast cancer.
  • Adalimumab became the world's best-selling drug. 
  • Kymriah and Yescarta were the first CAR-T therapies approved by the FDA. 

Each one of these medicines required cells to be grown at scale, and cells cannot be grown without media. Media is the infrastructure. This makes the gap between where cell biology is today and where media development is today a serious problem.

Why most cell culture media aren’t designed for the therapies being delivered today

Most of the cell culture media in use today were developed for a different era of biomanufacturing - one where CHO cells producing monoclonal antibodies were the dominant model, and the manufacturing demands of engineered T cells, NK cells, or viral vector-producing cell lines were not yet on the horizon.

When a process development team picks up a CAR-T construct today, they are often starting from T cell expansion media built for earlier-generation immunotherapy protocols - formulations not designed for the metabolic demands of engineered T cells at manufacturing scale. T cells are sensitive to osmolality shifts, oxygen tension, and nutrient availability in ways that generic formulations do not account for. This can lead to variable expansion, inconsistent activation, and yield that plateaus before it should.

The underlying cause is that media has never been treated as infrastructure. Bioreactor design attracts engineering investment. Downstream processing attracts automation investment. Cell line development attracts genomics investment. Media gets a catalog selection and a hope that incremental adjustment will close the gap. 

Why Design of Experiments can’t keep pace

The standard approach to media development is iterative experimentation guided by scientific expertise. A team of scientists designs experiments, runs them sequentially, interprets the results, and adjusts the formulation. This approach does work - it has produced most of the approved biologics on the market. But it has a ceiling.

A typical cell culture media formulation contains 60 or more interacting ingredients. Testing all possible combinations at just three concentration levels per ingredient produces more potential formulations than there are grains of sand on the earth. Conventional Design of Experiments (DoE) methodology can handle 10 to 12 variables before statistical coverage of the formulation space becomes too thin to be meaningful.

For legacy cell types running mature processes, that ceiling is rarely the constraint. The biology is understood, the formulation space has been partially mapped by decades of cumulative work, and incremental optimization produces incremental gains.

For novel cell types, that ceiling is hit almost immediately. But the consequence of getting it wrong - a failed batch, a regulatory hold-up, a scale-up failure that delays trials. The result is a delay in a patient accessing a therapy, which may be their only option. 

What optimized media development delivers downstream

The downstream effects of a well-optimized media formulation propagate through the entire manufacturing process. 

More consistent scale-up

The process variability that makes small-scale results hard to predict at the manufacturing scale is often a media problem. A formulation that is not well-optimized for the specific cell type behaves differently under different dissolved oxygen levels, different agitation regimes, and different bioreactor geometries. A formulation that is properly designed for the cell is more robust across those conditions.

Fewer batch failures

Media is implicated in a significant proportion of batch failures in biomanufacturing - through nutrient limitation, accumulation of inhibitory byproducts, or variability in undefined components. Optimized, defined formulations reduce the biological variability that makes batch-to-batch performance unpredictable.

Cleaner regulatory submissions

A media formulation built from defined, qualified components is easier to defend to a regulatory agency than one assembled iteratively from undefined starting points. The quality of the media development work feeds directly into that.

Media is manufacturing infrastructure

Bringing a CAR-T therapy to commercial scale requires cell engineering, viral vector manufacturing, apheresis and cell processing logistics, cold chain, and more. Each element of that system has seen significant investment and innovation in the last few decades.

Media, the substance in which the cells are actually grown, has not received the same attention. You can engineer a cell perfectly, optimize a bioreactor, nail your downstream process, and still watch a batch fail because the media was not designed for that cell at that scale.

That is not because cell culture media is less important. It is because the tools to develop it intelligently did not previously exist. Sequential, expert-driven experimentation was the only available method, and it moves at a pace set by human experimental throughput.

Those tools now exist. MediOP™ combines high-throughput robotic automation, Bayesian optimization, and millions of proprietary data points on cell culture media performance to optimize cell culture media.

What that delivers is the ability to invest in media the way teams already invest in their bioreactors, their downstream process, and their cell line development. This means a designed formulation, built for the specific cell, optimized across the objectives that matter to the program - titer, cost, consistency, scalability - before the process reaches a scale where changing it becomes expensive.

Talk to one of our scientists about media designed for the cell types you're working with.

Sign up for email updates:

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form. If the error persists, please use the backup form.