Recombinant antibody mixtures; optimization of cell line generation and single-batch manufacturing processes
© Rasmussen et al; licensee BioMed Central Ltd. 2011
Published: 22 November 2011
Recombinant antibody mixtures represent an important new class of antibody therapeutics as demonstrated by the increasing amount of literature showing that combinations of two or more antibodies show superiority compared to monoclonal antibodies (mAbs) for treatment of cancer and infectious diseases [1–5]. Sym004, composed of two antibodies targeting non-overlapping epitopes of the epidermal growth factor receptor (EGFR) act in a synergistic manner to induce an efficient internalization of EGFR leading to subsequent degradation and exhibit superior anticancer efficacy as demonstrated in several preclinical in vivo models .
At Symphogen A/S, we have developed an expression platform, Sympress™, for cost-efficient production of antibody mixtures. The antibody mixtures are produced using a single-batch manufacturing approach where a polyclonal working cell bank (pWCB) prepared by mixing the individual stable cell lines producing all the desired antibodies is used as seed material for a bioreactor process . By using a single-batch approach the CMC development costs of antibody mixtures are comparable to costs for monoclonal antibodies. However, the single-batch manufacturing approach raises questions with regard to control of composition ratios, compositional stability and robustness of the cell banking procedure.
Here, we present experimental data addressing these key questions and demonstrate that mixtures of recombinant antibodies can be produced under predictable, reproducible and stable conditions using the Sympress™ technology.
Material and methods
The second generation Sympress technology is based on expression in the ECHO cell line, a genetically modified version of the dihydrofolate reductase (DHFR) negative Chinese Hamster Ovary (CHO) cell line DG44 .
The expression plasmid used for stable transfection contained a bidirectional CMV promoter construct enabling co-expression of the IgG light and heavy chains from one plasmid. A DHFR selection marker was coupled directly to the heavy chain via an internal ribosome entry site (IRES).
ECHO parental cells were transfected separately with each of the individual antibody expression vectors using standard transfection technology, whereafter cells were subjected to a methotrexate (MTX) selection schedule. The selected stable pools were single-cell cloned by FACS and high-expressing clones were adapted to chemically defined cell culture medium, expanded and frozen, still as individual monoclonal cell lines.
The preparation of polyclonal cell banks followed a two-tiered cell banking approach. First, the relevant monoclonal cell lines were thawed and expanded. They were then mixed in predefined ratios and frozen as polyclonal master cell banks (pMCB). The pMCB was subsequently thawed, expanded and frozen as polyclonal working cell bank (pWCB).
Antibody purification was performed by capture on a MabSelect SuRe column and cation exchange chromatography (CIEX) was used to separate the different antibodies based on their charge differences. The characteristics of the chromatograms (peak area/peak height) were used to determine the relative distribution of the different antibodies in the mixtures.
We then examined the compositional stability in a more complex model composed of a mixture of six different antibodies. A seed train of approximately 3 weeks would be required to generate cells for inoculation of production reactors in a 10.000 L scaled-up manufacturing process. To examine the robustness of the relative cell compositions during this timeframe +/- one week the cells from pWCB were inoculated into a 14 days fed batch shaker after 14, 21 or 28 days of expansion, respectively. The antibody distributions at the three different time points were very constant as illustrated in Figure 1C. Additionally, the growth and production properties were also very constant over time (data not shown).
The reproducibility and robustness of the cell banking system was evaluated by comparing the antibody product composition produced from two different pWCBs generated from the same pMCB. A pMCB producing a mixture containing 6 antibodies was used for the study. We generated two polyclonal working cell banks (pWCB-1 and pWCB-2) from the same pMCB. Production of pWCB-2 was separated in time from production of pWCB-1 by several months. Four ampoules from pWCB-1 were expanded and used for inoculation of four 5 L bioreactors for fed batch production. Several months later, three ampoules from pWCB-2 were expanded and used to perform three new 5 L bioreactor runs. The antibodies were purified and antibody distribution determined by CIEX. The observed variation, both within the same pWCB and between different pWCB in regard to the antibody distributing was very limited (Figure 1D). This strongly indication that the pMCB/pWCB concept provides a reliable cell banking strategy. The reproducibility was also confirmed with respect to cell growth and productivity (data not shown).
Symphogen A/S has developed an expression platform, Sympress™, that can be used for predictable, reproducible, and stable production of antibody mixtures in a cost effective setting.
Here, we have shown that the relative antibody ratio in antibody mixtures can be effective controlled using appropriate mixing of the individual monoclonal cell lines before generation of the pMCB. Further, we have presented consistent data from fed batch productions with pWCB where the seed train lengths varied from 14 to 28 days. This strongly supports that the single-batch manufacturing concept is sufficiently robust for manufacturing, also at scales required for commercial manufacturing. The two-tiered cell bank approach composed of pMCB and pWCB is robust and reproducible, and showed very high consistency both within a given pWCB and between different polyclonal working cell banks generated from the same pMCB.
- Ben-Kasus T, Schechter B, Lavi S, Yarden Y, Sela M: Persistent elimination of ErbB-2/HER2-overexpressing tumors using combinations of monoclonal antibodies: Relevance of receptor endocytosis. Proc Natl Acad Sci U S A. 2009, 106: 3294-3299. 10.1073/pnas.0812059106.PubMed CentralView ArticlePubMedGoogle Scholar
- van der Horst EH, Chinn L, Wang M, Velilla T, Tran H, Madrona Y, Lam A, Ji M, Hoey TC, Sato AK: Discovery of fully human anti-MET monoclonal antibodies with antitumor activity against colon cancer tumor models in vivo. Neoplasia. 2009, 11: 355-364.PubMed CentralView ArticlePubMedGoogle Scholar
- Dong J, Demarest SJ, Sereno A, Tamraz S, Langley E, Doern A, Snipas T, Perron K, Joseph I, Glaser SM, Ho SN, Reff ME, Hariharan K: Combination of two insulin-like growth factor-I receptor inhibitory antibodies targeting distinct epitopes leads to an enhanced antitumor response. Mol Cancer Ther. 2010, 9: 2593-2604. 10.1158/1535-7163.MCT-09-1018.View ArticlePubMedGoogle Scholar
- Fuentes G, Scaltriti M, Baselga J, Verma CS: Synergy between trastuzumab and pertuzumab for human epidermal growth factor 2 (Her2) from colocalization: an in silico based mechanism. Breast Cancer Res. 2011Google Scholar
- Pedersen MW, Jacobsen HJ, Koefoed K, Hey A, Pyke C, Haurum JS, Kragh M: Sym004: Novel synergistic anti-epidermal growth factor receptor antibody mixture with superior anticancer efficacy. Cancer Res. 2010, 15: 588-597.View ArticleGoogle Scholar
- Frandsen TP, Naested H, Rasmussen SK, Hauptig P, Wiberg FC, Rasmussen LK, Jensen AM, Persson P, Wikén M, Engström A, Jiang Y, Thorpe SJ, Förberg C, Tolstrup AB: Consistent manufacturing and quality control of a highly complex recombinant polyclonal product for human therapeutic use. Biotech and Bioeng. 2011Google Scholar
- Nielsen LS, Baer A, Müller C, Gregersen K, Mønster NT, Rasmussen SK, Weilguny D, Tolstrup AB: Single-batch production of recombinant human polyclonal antibodies. Mol Biotechnol. 2010, 45: 257-266. 10.1007/s12033-010-9270-9.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.