Toward Better Biotherapeutics

How did you get into analytical chemistry?

My first experience with analytical chemistry instrumentation was during my Master’s degree in Philip Jessop’s lab at Queen’s University, Canada. I knew straight away that it was an area I’d like to explore more. I joined the Nestlé Research Centre and spent two years developing and validating LC-MS/MS methods for determination of chemical contaminants in food materials. Next, I joined the pharmaceutical analysis lab in Geneva for my PhD, working with three well-known scientists – Jean-Luc Veuthey, Davy Guillarme and Szabolcs Fekete. While there, I used different liquid chromatography and capillary electrophoresis approaches for biopharmaceutical analysis.

What keeps you moving forwards?

It’s easy to be motivated about improving therapies for cancer patients! Of course, my research is a long way from the patient, which is one reason why I wanted to spend time in biopharmaceutical companies. There, I can see first-hand the impact of analytical chemistry on drug discovery, development and manufacture.

Your work has focused on antibody–drug conjugates (ADCs). Why is analysis of these biotherapeutics particularly challenging?

Antibody–drug conjugates (ADCs) combine a lipophilic drug with a monoclonal antibody, meaning that the hydrophobicity of an ADC is much more pronounced than that of an unconjugated antibody. In theory, SEC separates different species based on their size, but in practice the picture is complicated by a range of non-specific interactions, including hydrophobic interactions.

To reduce hydrophobic interactions, biopharma scientists typically include isopropanol in the mobile phase, but this introduces two drawbacks: i) you may no longer be working under native conditions and ii) isopropanol has been found to have deleterious effects on some antibodies.

How have recent innovations in SEC columns helped tackle the complexities of ADC analysis?

The ability of the previous generation of ultrahigh pressure SEC columns to limit non-specific hydrophobic interactions with little or no isopropanol has helped to establish the validity of these analyses.

In addition, most ultrahigh pressure columns are packed with sub-micron particles, which has allowed for smaller columns and faster separations. You can now do in a 4.6 mm ID, 150 mm length column what used to require 7.8 mm ID and 300 mm length, and a separation that used to take 45 mins can be completed in 10 mins. Such improvements start to look attractive in other areas of biopharma development, such as process control.

I would go as far to say that, when it comes to ADC analysis, the biggest advances over the last five years have been seen in SEC columns.

You published an article in the Journal of Chromatography A this year on “Extending the limits of SEC” (1) – what was your aim?

A critical quality attribute of ADC products is the amount of free payload in solution. Typically, the payload molecule is cytotoxic, so you don’t want it being released before the ADC reaches its target. Establishing the free payload content of a candidate drug in solution by SEC alone can be challenging. During SEC of an ADC, the high molecular weight species (whole ADC, monoclonal antibody, etc) will emerge first, followed by smaller molecules (payload, linker, etc). The smallest molecules all elute as a single band, making it impossible to accurately quantify the payload. Previously, researchers had tried a 2D-LC method – coupling SEC to reverse-phase chromatography to separate the smaller molecules – but we were able to find a faster, more streamlined method.

How did you develop the new method?

With colleagues at the University of Geneva and a major pharmaceutical company located in Basel, I was conducting experiments using an ultrahigh pressure SEC column (Tosoh). We found that the payload species were highly hydrophobic, and were therefore absorbed onto the stationary phase. It gave us an idea: why not try a two-part separation using the same SEC column? We allowed the largest proteins to elute, then applied an acetonitrile gradient to elute the smaller molecules. We were pleasantly surprised to find that, thanks to secondary hydrophobic interactions, the acetonitrile gradient allowed separation of the smaller molecules. For some ADCs, this meant we were able to quantify both the high molecular weight species and the free payload in a single run – in under 10 mins. By taking advantage of (usually undesirable) hydrophobic interactions, the streamlined method could allow R&D scientists to quickly rule out ADC candidates that release payload in solution. It’s a great example of how analytical chemistry can help in the development of new biotherapeutics. 

What technology is likely to have the biggest impact in biopharma – now and in the future?

Right now, the most important trend is towards multidimensional LC, as demonstrated by the large number of presentations on the topic at HPLC 2018 in Washington. Most vendors now sell at least some 2D-LC-specific solutions, although software development for these systems has lagged behind somewhat. I believe that over the next five years, most companies will be using multi-dimensional LC approaches, coupled to high-resolution MS instruments. 

Looking ahead 10 years, I think we will see separations going beyond two dimensions and into three, four or five dimensions. In parallel, I hope that suppliers will be able to improve column chemistry even further, in particular to reduce non-specific interactions in SEC and allow it to be coupled directly to MS.

Pharma is (rightly) a cautious industry and, before any new technology is adopted, companies and regulators must be sure it won’t cause unexpected issues. Nevertheless, despite the technical and regulatory challenges, all the company scientists I have worked with have been very open to new technology.

Keeping Apace with Biopharma Trends

Regina Roemling, Senior Marketing Manager, Separations at Tosoh Bioscience GmbH, Germany fills us in on the company’s latest innovations for the biopharma market.

What challenges do vendors face in developing products for biopharmaceutical applications?

The biopharmaceutical industry is highly regulated and established methods are not easily replaced. Consequently, products such as chromatographic resins used in manufacturing or columns applied in QC of an approved biologic have to maintain a consistent quality over a very long period. On the other hand, the development time for new analytical tools needs to be reduced to cope with the increasingly rapid development of new, complex biopharmaceuticals. Looking just at therapeutic antibodies, there is a huge range, from antibody–drug conjugates (ADCs) to small, single-chain variable fragments (scFv). Analytics need to keep pace with this variety.

What do biopharma customers want?

R&D labs dream of multidimensional analytical platforms allowing the thorough characterization of tiny amounts of candidate molecules overnight. In production, throughput is of higher importance, but robustness and reproducibility are essential too. 

What trends in biopharma analysis have you seen in recent years?

As Alexandre describes, hyphenation and multidimensional chromatography have been amongst the hottest topics at recent HPLC meetings. Another trend is the increasing use of affinity-based separations, not only in purification but also in HPLC analysis. A good example is Protein A affinity, which can be applied for fast analysis of cell titers or as a kind of sample pretreatment of a crude feedstock in process analytics when it’s coupled with techniques like size exclusion chromatography (SEC) for monitoring aggregate contents. In fact, we will soon launch a new affinity column, which is designed to analyze the antibody-dependent cell mediated cytotoxicity (ADCC) activity of antibodies with high selectivity and reproducibility.

Why develop a column for measuring ADCC activity? 

ADCC plays an important part in the mechanism of action of therapeutic antibodies, particularly cancer-targeting mAbs. When developing an antibody for cancer therapies it is extremely important to select a clone delivering the required ADCC activity, but current methods (in vitro cell-based assays or surface plasmon resonance techniques) are less reproducible than chromatography. In addition, there are many more potential applications for a fast and highly reproducible HPLC method; for example, monitoring lot-to-lot differences for antibody products or comparing the ADCC activity of an antibody biosimilar to the innovator. We are keen to hear further ideas from academia and industry for potential applications and are looking forward to releasing the new affinity column by the end of 2018, which we are confident will be a real asset to biopharma scientists (1). 

Reference:

1.  M Kiyoshi et al., Nature Scientific Reports, 8, 3955 (2018).