Why are CHO cells the dominant host for monoclonal antibody production?

Why

A 2010 review of antibody cell culture estimated that about 70 percent of recombinant proteins were made in CHO cells. The reasons are practical. CHO cells adapt to grow indefinitely in suspension without serum, which makes stirred-tank scale-up and chemically defined media possible.

Their glycosylation is compatible with humans, while murine lines such as NS0 can add non-human sugar structures, like alpha-gal and NGNA, that may be immunogenic. Selection systems based on dihydrofolate reductase with methotrexate, or glutamine synthetase with methionine sulfoximine, let developers select and amplify high-producing clones. As a non-human cell, CHO is less likely to propagate human pathogenic viruses.

Rodent cells do contain endogenous retrovirus-like particles, which is why ICH Q5A(R2) expects processes for rodent-derived products to show clearance of retroviruses, using murine leukemia virus as a model. There are downsides. CHO cells can still add small amounts of non-human glycans, their productivity per cell is lower than microbial systems, and their genome is unstable enough that clones can drift over long culture, which ICH Q5D addresses through cell substrate stability testing to the limit of in vitro cell age.

For small, unglycosylated proteins, E. coli or yeast are often better choices.