Concentration Feasibility in High Concentration Formulation: A Q&A with Eva Keilhauer

More biologics are moving from intravenous infusion to subcutaneous injection, and that shift almost always means pushing the protein to a higher concentration. It sounds like a small step. In practice, high concentration formulation is where a lot of programs quietly lose time.

We sat down with Eva Keilhauer to talk through what happens on a high concentration project: what clients tend to underestimate, what “concentration feasibility” really involves, and where the most avoidable delays come from.

When a high concentration project comes to Coriolis, what happens first?

We start by understanding the project in detail, rather than reaching for a default recipe. What phase is the molecule in: are we developing for Phase 1, or for Phase 3 and commercial supply? What do we already know about the molecule and its challenges? What is the planned primary packaging? From there we suggest a strategy tailored to the molecule and the situation.For classic monoclonal antibodies we do also have a standardized, faster approach, because so much is already understood about how those behave. But a standard mAb and a difficult molecule are very different starting points, and the strategy has to reflect that.

What is the most common thing clients get wrong before they reach you?

Many arrive believing upconcentration is easy, and that it can simply be done in their current formulation. That assumption is usually where the trouble starts.

Concentrating a protein changes how it behaves. Viscosity rises, often steeply. Self-association and aggregation become real risks. A formulation that worked well at a lower concentration frequently won’t carry over unchanged.

What does “concentration feasibility” mean in practice?

At Coriolis, two things, run together.

First, in-silico modelling, to predict whether a molecule is prone to the typical high concentration problems like self-association and high viscosity. We use an in-house pipeline built from in-silico descriptors, machine learning (for monoclonal antibodies) and molecular modelling. That already tells us a lot before we’ve consumed any physical material.

Second, an actual small-scale concentration procedure in the lab, to see how the molecule really behaves. How high can we go? Does the material aggregate? What does the viscosity-versus-concentration curve look like? We also collect early data on how the formulation matrix influences upconcentration behavior, which then informs how we prepare formulations later on.

Together, those two strands tell us whether a molecule is a realistic candidate for high concentration, and they do it early, before significant material or time has gone in.

How does in-silico modelling change your workflow? Does it save time?

In the beginning, it doesn’t save a huge amount of time, to be honest, because we still run the studies to confirm the predictions. The simulations don’t replace the experiments.

Its real value is elsewhere. It gives us a much more solid understanding of the molecule and its basic characteristics from the start, and it saves effort, and therefore costs for the client, because we can narrow the testing down to the most promising formulation conditions instead of screening broadly. Plus it mitigates risks, so while it may not save much time when everything goes right, it can save significant time by preventing wrong turns that lead to delays, repeat studies, and unnecessary rework. In the end, that is where the real time savings for the client come from.

Where in a typical timeline do you see the most time wasted?

When the high concentration strategy isn’t thought through from the start. Plenty of clients run the initial clinical phases at a lower concentration, which is completely fine, as long as you’re already doing the work to assess high concentration feasibility in parallel.

That’s exactly what our dedicated in-silico and lab-based assessments are for: predicting, very early, whether a molecule is suitable for high concentration. And if it isn’t, we can still choose a separate development path, for example delivering the required high dose through a larger volume. Choosing the right development path for each molecule is what makes high concentration/high dose development succeed.

How does primary packaging shape a high concentration project, and how early does it need to come up?

More than most teams expect. The primary packaging that comes with a standard high concentration drug product, a prefilled syringe being the obvious example, shapes a lot of the work. We develop the formulation, and we select the packaging too, then make sure it’s compatible with the drug and the formulation. That means looking at silicone oil and tungsten compatibility, testing usability, syringeability, and injectability, and assessing long-term stability in the intended prefilled syringe. By the end of a project, a surprising amount of it was about the container.

That’s also why it has to come up early. The formulation influences the container selection, and the container selection influences the formulation development, so the two can’t be handled in isolation. The sooner we understand where the client wants to go, and needs to go, with a solid grasp of the target product profile, the better we can set up the right strategy from the start.

We also have a strong network and collaborate with the relevant vendors, so we can bring them in on specific containers or device questions when those go beyond our own expertise.

Coriolis doesn’t own manufacturing slots. How does that change the advice you give?

We’re completely independent in the selection process. We’re looking for the best solution for the client and the product, with nothing else pulling on the decision.

At the same time, we stay very aware of what has to be true for that choice to work: the container and/or device must be available at sufficient quality and quantity, and it has to be manufacturable at the client’s CMO of choice. Independence is only useful when the recommendation is also practical.

Can you give an example where formulation work made a measurable difference?

On one project, we ran a systematic, iterative screening to reduce the viscosity of a specific high concentration product, and we brought it down by roughly 40 to 50%. A reduction like that can be the difference between a product that’s injectable through a standard needle and one that isn’t.

Another example was in a project where we needed to switch a monoclonal antibody from intravenous to subcutaneous delivery, with the concentration raised to 200 mg/mL. The client’s early efforts ran into viscosity and aggregation problems. Our scientists developed a self-buffering formulation using only a few amino acid additives, which stabilized the protein, reduced the viscosity, and met the client’s target product profile.

Partnering with Coriolis on High-Concentration Development

What ties this together, and what sets Coriolis apart on high concentration, is that the whole stretch can sit with one team: the high concentration formulation, the container selection and testing, non-GMP filling for all the relevant container types, and testing under real-life conditions like stability, stress, in-use and transportation studies.

For a lean biotech, that means one group and one connected set of data across the part of development where subcutaneous programs most often trip up, between the formulation and the finished product. If you’re mapping out a high concentration or subcutaneous program and want a second read on feasibility, that early conversation is usually the cheapest one you’ll have.