November 12, 2014

High-Throughput Process Development in an Historical Environment

Presentations, discussions, and information-sharing on modern bioprocess technology and high-throughput process development (HTPD) in an historical and culturally enriched environment is the essence of the HTPD meeting series, which now has been ongoing for more than six years.

After previous meetings in Krakow, Poland and Avignon, France, the 3rd international conference devoted to HTPD was most recently held October 6-9 in Siena, Italy, yet another beautiful city on the UNESCO world heritage list. About 100 delegates from 13 different countries joined the discussions on HTPD case studies in upstream, downstream, and formulations, as well as modeling, data analysis, and analytics. A new session this year was the HTPD Olympiad, for which five teams had signed up to develop a purification process for a domain antibody (Dab). The challenge was to use HTPD workflows for maximized Dab purity and yield.

A workshop held by Ashley Hesslein from Bayer HealthCare kicked off the meeting. The topic was “Lessons Learned in Building HTPD Capabilities” and comprised of a panel discussion and short presentations from five scientists, representing some of the leading companies in the HTPD area. The Q&A session was intense and the panel concluded that data management and analytical support are still remaining challenges. The requirement of having a critical mass of robotic users – at different levels of expertise from super-users to general users – and resources was also a common theme.

The workshop was followed by the plenary lecture given by Arne Staby of Novo Nordisk, in which he discussed mechanistic modelling versus – or complementing – Design of Experiments (DoE), and he also presented a case study and lessons learned from a regulatory submission that included mechanistic modelling.

In the Tuesday morning session on Modeling and Data analysis, the presentation from Georg Pollinger at Roche on internal software DAMAS (Data Acquisition Management and Analysis Software) garnered a lot of interest. Since the introduction of HTPD approaches, it has been a challenge with the generation of large data sets, and also the visualization and traceability of analytical results. As part of a strategic initiative, Roche developed the DAMAS software that merges an electronic lab journal with material and sample management also linked with analytical equipment.

HTPD approaches in downstream goes beyond chromatography
The approaches in downstream process development, with screening of chromatography resins and parallel screening of optimal process conditions, was, to a large extent, the origin of the new era in HTPD. This could be seen in the interest for the downstream session at the meeting (with more than 80% of participants ranking this session as the most interesting), as well as the maturity that HTPD has gained in downstream applications, especially in chromatography. As an example, John Welsh from Merck presented a collaboration with Cytiva on the development of the next- generation multimodal chromatography resins and how high throughput techniques and different formats were used to identify the effect of ligand density in monoclonal antibody (mAb) purifications. Two different novel multimodal polishing Capto™ ImpRes resins were evaluated and the optimal ligand density was identified to provide best aggregate clearance and largest operating space.

In addition, it is clear that HTPD now goes beyond chromatography in the downstream space, with several presentations on flocculation and refolding. Cornelia Walther from University of Natural Resources and Life Sciences in Vienna, Austria, presented work on refolding of inclusion bodies. Traditionally, optimization of the refolding steps has been considered a bottleneck during downstream process development. In the presented work, instead an automated platform using 96-well plates for parallel screening of inclusion body solubilization and protein refolding conditions was shown. With this approach, 24 solubilization conditions per plate could be studied, with quadruplets of each condition, and still requiring less than 90 mg protein per screening. For the fully automated screening of 96 refolding conditions, less than 4 mg of solubilized protein per screen was required. Time saving was impressive, with screening of all these solubilization and refolding conditions performed in less than five days.

Upstream and formulation are growing areas
HTPD activities in the upstream continue to grow rapidly. Several new equipment and innovations were introduced to the market, such as micro bioreactors in the 10-15 ml scale, controlled by an automated workstation. As an example of their utilization, Chris Daniels from Merck presented work with design and commissioning of a BSL-2 HTS platform for process development of viral vaccine production, where all liquid handling, plate set-up, incubation, assay set-up, and readings were contained within one robotic system. Essentially everything was automated to reduce variations as well as the workload while increasing experimental resolution. Some of the remaining challenges in the upstream area were reported to be high-throughput metabolic analytics.

On the subject of formulation, David Smithson from Genentech presented how the company began the process of implementing high throughput approaches in the area of formulation development several years ago. However, the conclusion was that Genentech did not anticipate much demand for high-throughput screening of formulation, at least not for non-research projects and based on the current product pipeline portfolio at the company. One reason was that Genentech already has an extensive in-house experience of formulation based on earlier work. Genentech is said to be working with less than 20 different formulations, developed mainly for mAbs, but these formulations have also been shown to work for other types of target molecules. Instead, the main utilization of its high-throughput technologies and equipment are mainly used for sample preparations for complex assays, which has proven to be very beneficial.

Conversely, Vanessa Jully from Université Catholique de Louvain in Brussels, Belgium, presented a different conclusion based on her collaboration with GlaxoSmithKline Vaccines, in which a high-throughput screening platform was developed to study adsorption capacity of aluminum-containing vaccines. Aluminum-containing salts are already established as important adjuvants in the formulations of several approved vaccines. However, for new vaccines it is important to develop a thorough understanding of the adsorption mechanisms of an antigen onto an aluminum salt. The value of HTPD methodology was helpful, with different aluminum salts and different types of vaccines (virus-like particles, polysaccharide conjugates, and proteins) evaluated.

Winning an Olympic gold medal in HTPD
Based on suggestions following the second HTPD conference in 2012, a new session – the HTPD Olympiad – was organized and results were presented in Siena. The purpose of the Olympiad is to allow process developers to benchmark their internal HTPD workflows against peers in the biopharmaceutical industry and academy. The challenge in the very first HTPD Olympiad 2014 was to develop a purification process for a Dab expressed in E. coli using high-throughput workflows for maximized Dab purity and yield. Six teams signed up for the challenge, and five teams completed it.

The Olympiad was a truly successful introduction and gave a new dimension to the meeting. The audience found it interesting to hear all the contestants presenting their work, their challenges, and the results. Or as Phil Lester, chairman in the Organizing committee, said in his concluding remarks: “One book – but five stories”.

The performed work and reports were evaluated by an independent scientific board. The winners in the Olympiad – winning two categories each – were Genentech (winning in Process performance and Data handling and modeling) and Merck (winning in Analytics and HTPD workflow efficiency). Sharing second place were Roche, Boehringer-Ingelheim, and the only academic contestant, University College London. But sometimes winning first price is not everything. Dominic Boeth, from Boehringer-Ingelheim, reported how the Olympiad can be used as a springboard to successfully improve the collaboration between two different sites within the company, and probably win something much more valuable than a gold medal.

In summary, the HTPD 2014 meeting must be considered another very successful event, with 67% of survey respondents rating it as excellent, and the remaining 33% as good. During the wrap-up of the meeting, there were several discussions about whether the HTPD meeting series should continue with a fourth event in 2016 or if HTPD today is such an established and mature area that the need for a specific conference no longer exists. The final decision has not yet been made, but based on the meeting survey, 96% answered that they were looking forward to another meeting, possibly with a slightly extended scope. If so, there are still several cities on the UNESCO world heritage list that could host the HTPD meeting. The organizing committee now also has to consider the new theme for the conference series: an impressive stairway leading up to the event.

Tags: biotechnology, process development, Sienna, conference, HTPD