Given the results discussed above, how could such a process be scaled, what technologies could be used, and where might it
fit in a standard MAb purification scheme? If the goal is to remove the majority of the impurities before the initial chromatography
step, then it must be located at the harvest step, where two major technologies, centrifugation or filtration, could be used
Figure 4. Potential purification schemes using a precipitation step
The type of centrifugation most commonly used in MAb processes is disc-stack centrifugation, which is ideal for separation
when the solids concentration is low, or where smaller particles are present.25 This technology and its adaptation to mammalian cell culture broth has been described extensively in the literature and
presented at conferences.26–29
For developing a centrifugation step to remove the precipitate, a scale-down model would need to be developed.30,31 Once this model was available, the parameters affecting separation could be studied and optimized. For centrifugation, such
parameters would include the sigma factor, the relative densities of the materials to be separated, angular velocity, viscosity,
feed flow rate, and temperature.32 In addition, solids removal from the bowl during continuous flow centrifugation could also play an important role in centrate
The other option is depth filtration, which is also used extensively to process MAbs and related molecules.34,35 Depth filtration has advantages over centrifugation such as the removal of DNA and HCP.36,37 However, it is no trivial matter to screen the wide array of depth filters currently available and determine the correct
one for the application, and then determine the optimal parameters for processing, which include the pH and ionic strength
of the feed fluid, feed flux, membrane area, and operating pressure.38
If a technique can be adapted for precipitate removal following treatment with an agent such as caprylic acid, the technology
could be very useful. Early removal of significant amounts of HCP and DNA could result in the adaptation of a shorter purification
process or better performance and longer life of the capture resin.
Alternatives to standard chromatography for the purification of MAbs have been discussed in detail in the literature. One
such alternative, precipitation, is used extensively in other applications for the purification of polysaccharides, DNA, and
viruses, and may be adaptable to antibody purification. The technology could be manipulated to precipitate out the antibody,
as is the case with ammonium sulfate, or to reduce impurity levels while leaving the antibody in solution.
Several potential precipitating agents that have been used in other applications were evaluated for their ability to purify
antibodies by reducing HCP and DNA. Two cationic detergents, CTAB and DB, were able to precipitate DNA much more effectively
than ammonium sulfate. In addition, the short-chain fatty acid caprylic acid was able to remove both DNA and HCP very efficiently
for CHO-produced antibodies, but less efficiently for NS0-produced antibodies. A precipitation step targeting the removal
of impurities could be developed using centrifugation or depth filtration technology during the harvest operation, but would
need to be optimized to ensure the cost of goods compared favorably with a standard chromatographic technique such as Protein
A. If these engineering challenges can be addressed, precipitation may prove to be a valuable tool for antibody purification.
I wish to thank Tara Carter for her excellent experimental capabilities that made this work possible, and to Paul Mensah for
his guidance and advice.
This is an excerpt from the chapter in the forthcoming John Wiley and Sons book Process Scale Purification of Antibodies edited by Uwe Gottschalk.
JUDY GLYNN is a senior principal scientist at Pfizer, Inc., Chesterfield, MO, 636.247.6519, firstname.lastname@example.org