System integration, product quality, and product safety impact factors (leachables, extractables, and biocompatibility) are
major concerns. It is beneficial for a company to use a family of SU products from the same manufacturer to save the time
and reduce the cost of safety evaluations and validation.
Before introducing an SU UF–DF system to our facility, several other SU systems made by Millipore, including buffer and media
mixing tanks and containers, were already evaluated and used in GMP production. The flowpath and associated retentate tank
liner for the Millipore SU Mobius FlexReady Solution for TFF are made of the same film and material as those used for the
associated containers for the Millipore mixing system, which are widely used in our facility. The risk of leachables and extractables
affecting product quality and the compatibility of buffers and protein were previously assessed,3, 4 saving both time and cost on those safety evaluations. Unlike SU UF–DF systems, the Mobius FlexReady Solution for TFF is
a fully integrated system for operation, process monitoring and control, and data management. In the preliminary evaluation,
two systems had the same assessment score based on operation specifications, process monitoring, and control. System integration
and product quality and safety impact were the differentiating factors. Ultimately, the Millipore Mobius FlexReady Solution
for TFF was chosen because of its better integrated system and the existence of previous safety evaluation assessment.3, 4
Summary and Conclusions
The operation of this single-use system is easier than our existing SS UF–DF systems and the performance is comparable. The
disposable flow path pieces can be quickly installed, resulting in short equipment turnaround times. The retentate diverter
plate and magnetically coupled mixing device helped avoid the concentration gradients in the retentate tank and improved the
UF–DF performance. Product yield and purity of the intermediate UF/DF step using the Mobius FlexReady Solution for TFF are
comparable to the results from previous runs (data not shown). Also, the low system hold-up volume (0.6 L) and working volume
(2.0 L) allowed for a wider operating range. The strong feed pump maintained a 20 L/m2 flow rate even at 30 psi back pressure. Because the whole flow path is disposable, the risk of cross-contamination during
product changeover was minimized, and the time and effort for CIP were reduced in our multiproduct facility.
The features of process monitoring, control, and data management enhance the process automation capability and reliability
of data recording. The Allen Bradley Controllogix system and the associated human–machine interface (HMI) software were user-friendly.
The PID-like display provided real-time information of recipe parameters and process data. The control system was interfaced
with our production information management system for data collection and archiving. This feature enables the future implementation
of electronic batch records in a GMP production environment. It also enables easy, batch-to-batch comparison for technology
The PCV and permeate flow meter increased process monitoring and control capabilities. We observed fluctuations of TMP while
using the PCV for constant TMP control. It was found that the fluctuations could be reduced to an acceptable level (Table
1) by initiating the operation with feed flow control by pump speed, then switching to constant TMP control by PCV once the
feed flow rate is within the ROP range. During the diafiltration, total diafiltration volume target can be preset and monitored
with the included permeate flow meter. Although there was a measurement difference of approximately 3% in the total diafiltration
volume versus the weight scale measurement, the result is acceptable because the ROP for total diafiltration volume was quite
wide and diafiltration completion was primarily based on the conductivity and pH of permeate filtrate.
Table 1. Summary of constant transmembrane pressure control (TMP) control
Implementing this single-use ultrafiltration-diafiltration system doubled the retentate capacity of the intermediate UF–DF
system, and therefore shortened the unit process time and improved productivity. By eliminating the six pre- and post-use
CIP steps, the usage of water for injection and caustic solutions for CIP also was reduced. Furthermore, the 16 corresponding
rinse samples and swab sampling testing for CIP steps were eliminated. The overall cost saving will depend on actual utilization.
We found the payback period to be 3.8 years when performing ten campaigns per year.
The authors would like to thank the Millipore team for technical and logistic support for introducing this SU UF–DF to our
facility. The authors also would like to acknowledge the Centocor pilot plant purification team for their support in implementing
the SU UF–DF system for GMP production.
Keqiang Shen is a senior scientist, Be Van Vu is an associate engineer, Nikunj Dani is a senior systems engineer, Bryan Fluke is a senior associate engineer, Lei Xue is a senior manager, and David W. Clark is the global head of supply execution, all in pharmaceutical development and manufacturing sciences at Johnson & Johnson,
Inc, Spring House, PA, 215.628.5953, firstname.lastname@example.org
1. Paul MS, Mytelka SD, Dunwiddie TC, Persinger CC, Bernard H, Stacy MR, et al. How to improve R&D productivity: the pharmaceutical
industry's grand challenge. Nature Rev Drug Discov. 2010;9:203–14.
2. Charles I, Lee J, Dasarathy Y. Single-use technologies—a contract biomanufacturer's perspective. BioPharm Int. Suppl. Guide
to Disposables. 2007 Nov; 31–6.
3. Millipore, Inc. PureFlex: Extractables, bioreactivity safety evaluation approach. Technical brief.
4. Millipore, Inc. Extractables bioreactivity safety evaluation of PureFlex film for Centocor. 2007.
5. Maigetter RZ, et al. Single-use (SU) systems. Encyclopedia of industrial biotechnology: bioprocess, bioseparation, and
cell technology. 2010: 1–39.