Diminutive Bacteria Useful for 0.2-μm Filter Studies, FDA's Brorson Says

October 7, 2009

Diminutive bacteria with the right geometry (i.e. very thin) may be useful for research and development work on sterilizing-grade and other 0.2 uM filters, said Kurt Brorson, PhD, a staff scientist for the FDA/CDER's Division of Monoclonal Antibodies.

Diminutive bacteria with the right geometry (i.e. very thin) may be useful for research and development work on sterilizing-grade and other 0.2-μm filters, said Kurt Brorson, PhD, a staff scientist for the FDA/CDER’s Division of Monoclonal Antibodies. Speaking at a Sept. 24 seminar hosted by Millipore Corporation at its facility in Jaffrey, New Hampshire, Brorson said that because very thin bacteria such as Hydrogenophaga pseudoflava can pass to variable extents through 0.2-μm filters, these organisms can be useful for quantitatively studying how different processing conditions affect a filter's’ capacity to remove bacteria.

Brorson shared results from studies conducted at CDER/FDA that replicated and built on results reported by Sundaram, et al. in the PDA Journal in 2001. In the FDA studies, Brorson and his colleagues purchased 0.2-μm membranes made of various materials from five different manufacturers, and challenged the filters with loads of H. pseudoflava ranging from 107 to 1010 cfu/cm2. They then tested the effects of three parameters of three model process solutions, including high osmolality (30% sucrose), high ionic strength (0.5 M NaCl), and high viscosity (50% glycerol). They found that H. pseudoflava penetrated 0.2-μm filters consistently, but that the level of penetration was affected by the chemistry of process fluids, albeit not as predicted, and varied somewhat among filter membrane chemistries, brands, and between runs.

The fact that the level of penetration can be quantified as a log10 reduction value is what makes H. pseudoflava more useful for research studies than Brevundimonas diminuta, the organism normally used to validate sterilizing filters. “With B. diminuta, you simply get a qualitative result—either some passed through the filter or none,” Brorson explained. “That is useful for validation, but not very helpful for R&D, because you can’t measure quantitatively how process conditions affect filter performance.”

This suggestion does not mean that manufacturers need to change their validation studies, Brorson emphasized. H. pseudoflava is an environmental bacterium found in well water and other environmental sources, and is not normally found in pharmaceutical manufacturing facilities. Therefore, B. diminuta is still considered suitable for drug and filter manufacturers for process validation of 0.2-μm sterilizing-grade filtration.

Brorson also noted that very high loads in spiking studies can lead to heavy caking on the filter surface, which appears to reduce passage of H. pseudoflava through the filter. “Our data shows that high loads can actually present a lower challenge, rather than a more difficult one,” he said. “Spiking at load of 107 cfu/cm2 seems to strike a good balance, confirming the suitability of the validation current approach.”

Filter users should also understand that the 0.2-μm rating is not standardized across the industry, Brorson said. “The rating is determined from bubble point data and a mathematical equation involving several parameters that are very difficult to measure,” he explained. He cited a comment by industry filter expert Ted Meltzer that filter pore size values should be seen as an individual filter manufacturer’s rating, rather than representing an industry-wide standard. “Maybe we should look for a better way to describe and rate these filters,” Brorson suggested.