Smarter DSP Design Based on Early Stage Development Data

Presented by Michael Box

Dr Michael Box is lead scientist of purification development at Lonza Biologics, UK.

Lonza Biologics is using its good manufacturing practice (GMP) production data, accumulated over many years, with the aim of informing its purification process development activities. The company will use such information to improve its high-throughput screening experiments that automate process development.

The company first conducts a developability assessment of its customers’ drug candidates in silico. This allows the identification and removal of undesirable product characteristics such as a propensity to aggregate or generate unwanted immune responses in patients. Lonza can use product generated early in the cell line selection process to perform additional laboratory testing to assess product stability under different pH and conductivity conditions.

Process development scientists perform high-throughput screening studies using Tecan robots that automate buffer preparation and protein purification. Lonza selects either RoboColumns, microtiter resin plates, or PhyTip® columns, depending on the objectives of the study. High-throughput and off-line analytical methods provide data on product quality and impurities, allowing the assessment of purification performance under different conditions.

Figure 1: Future state: knowledge flow into high-throughput screening (HTS)

This approach has helped Lonza identify protein A chromatography elution buffers that will minimize aggregation along with other parameters during antibody purification processes development. The impact of high- throughput techniques on the development of nonantibody recombinant proteins has been transformative. It allows the rapid evaluation of a large number of process conditions with little process material used. Although such experiments provide reasonably reliable information about yields and purities that the process can achieve, scientists are still required to make minor process modifications during scale-up from development to GMP production. In the near future, the company hopes to use these screening experiments to provide information about resin costs, plant capacities, scheduling, and costs of goods.

Furthermore, Lonza hopes to increase the effectiveness of high-throughput screening studies by informing their design with knowledge acquired from historical manufacturing batch and virus clearance study data. For example, reviewing data from historical virus validation studies has shown that higher protein-loading capacities on anionexchange resin can lead to a decrease in the resin’s viral clearance capabilities. Thus, attempts to improve process economics by maximizing the use of this resin chemistry capacity can have a negative impact on virus clearance. However, (according to Lonza’s past knowledge), early identification of purification schemes may not deliver sufficient virus clearance before any large-scale manufacturing batches commence.

Through better process understanding, Lonza can design highthroughout experiments that can identify conditions that maximize throughput while ensuring patient safety. By developing those concepts further, Lonza hopes to reduce the need for intermediate scale-runs, thereby lowering development costs further and increasing customers’ speed to clinic.