Purpose: With increasing pressure for drug developers to reach the clinic as quickly as possible, there is a growing demand for increased capacity and a shorter turnaround for analytical support. This requirement for shorter time frames and increased sample numbers can be overcome through high-throughput automation, which can decrease the time required to develop a product cell-line and purification process for more complex proteins. Product quality analysis during upstream and downstream process development is key to the production of safe and effective biopharmaceutical products. Therefore, an automated High Throughput Impurity Detection (HTID) system was designed to meet this demand and to efficiently test small volumes of hundreds of samples in only a few days.
Methods: Multiple vendors were approached to determine how commercially available Host Cell Protein (HCP), CHO DNA and Protein A (PrA) quantitation kits could be automated. Different chemistries were assessed and compared for HCP testing, alongside other requirements including sample volumes needed and assay sensitivity. After the vendor was selected the HTID System was installed.
Results: During vendor selection there were high levels of variability observed between the different kits supplied for HCP concentrations, see Table 1.
This could be due to the variability of the anti-HCP polyclonal antibodies used by the vendor. However, when testing samples throughout a purification process, clearance of HCP was always observed throughout the process, Figure 1.
Vendor 1 was selected due to the improved working range and ease of automation. The results generated were also the most comparable to the in-house ELISA methods although further optimisation of the method was required in order to ensure that the working range was maintained for the low sample volumes.
After final vendor selection, the HTID system was installed, Figure 2. The HCP method was then optimised using the HTID system to improve the working range which are now similar to the current in-house method. The throughput has also been greatly increased from 6 samples per assay using the current in-house method, to up to 190 samples per assay. This method is now being used in house for qualitative screening assays.
Residual CHO DNA testing was the first impurity assay to be established and Site Acceptance Testing (SAT) was completed to assess the inter- and intra-assay precision, accuracy, linearity, sample matrix interference, working range and method limit of quantification (LOQ). All testing requirements were met, including inter- and intra-assay precision (< 24% CV and < 14% CV respectively), accuracy (65-101%), impurity linearity (R2 =0.99) with a working range of 16.7pg/ml to 166,670pg/ml and no sample interference was detected. The CHO DNA method is now being used for commercial and internal projects. There were some difficulties with transferring the PrA method using the vendor’s kits onto the HTID system and therefore optimisation work on this method is still ongoing.
The fully automated and integrated platform allows the testing of up to 300 samples by one operator in just 3 days whereas the same number of samples tested using the manual method employed in-house would take up to 4 months by one operator. This allows for rapid analysis of samples to aid in pre-Biologic Licence Applications and time-critical process decisions. The system requires less than 150µl for all three methods, approximately 9 times less volume than required for the comparable current in-house validated methods making it compatible with TECAN small-scale purification optimisation processes allowing for rapid purification process development.
Conclusion: Selection and development of an automated impurity testing platform for rapid process development has allowed us to meet the increased analytical demand from process development and characterisation studies. We can now perform rapid screening of impurities such as CHO DNA and HCP to enable high numbers of purification chromatography steps to be screened alongside standard testing. Further development of the PrA method is required in order to provide additional impurity testing. The HTID system has also enabled the development of streamlined connections for sample and electronic data transfer.
Hannah Gibbins– Slough, England, United Kingdom
Benni Juckes– Slough, England, United Kingdom
Hemisha Patel– Slough, England, United Kingdom
Lewis Higgins– Slough, England, United Kingdom
Imtiaz Alam– Slough, England, United Kingdom