June 1, 2006 (Vol. 26, No. 11)
Angelo DePalma Ph.D. Writer GEN
Firms Must Demonstrate to Both the FDA and the Public that Processes are Safe and Effective
Common wisdom holds that bacteria and viruses are everywhereexcept, of course, in bioprocesses. Keenly aware of this, biomanufacturers have numerous approaches to limiting biocontamination upstream and eliminating it to undetectable levels downstream. Processes stay clean mostly by assuring that microbes never enter the picture. Well-known techniques, such as clean-in-place, sterilizing filters, cleanrooms and other controlled environments, worker training, and strictly following validated protocols, minimize contact between products and the germ-filled world outside the manufacturing suite.
According to the FDA’s most recent aseptic-processing guidance, identifying the normal flora within the manufacturing environment, and those likely to contaminate a process, are keys to maintaining sterility. When microbes reach action levels for a particular organism, manufacturers attempt to identify them and identify the source to prevent future occurrences. From a regulatory standpoint, biotech companies must demonstrate to the FDA and, increasingly, to the public, that their processes are under control with respect to microbe detection.
Sterility failures are rare, but their occurrence carries disastrous consequences for biomanufacturers of sterile products. Since 2004, Chiron(www.chiron.com; recently acquired by
As closed systems, bioprocess equipment has built-in contamination control in the seals, valves, and physical walls of bioreactors. Microbial growth inside fermenters is rare unless a valve malfunctions or reactors were not properly cleaned.
Contamination doesn’t usually occur because there’s a hole in the reactor, says Michael Cox, senior director of process development at Goodwin Biotechnology (www.goodwinbio.com). More likely, bacteria enter the system through the inoculum or during transfer of process fluid during upstream processing. Downstream, filtration clears any pathogen about the size of a typical virus or smaller.
99% Prevention, 1% Detection
Contamination is almost never immediately detectable. Too often, it becomes apparent only after processors have expended considerable time on a batch. If you have one bacterium in a reactor, dividing every twenty minutes, versus cells that divide every 24 hours, which would you bet on? asks Cox.
In biotech, contamination control is 99% prevention, 1% detection. That percentage point is critical though, since companies rely on it to demonstrate that processes are clean. Because of the time lag between when a swab is taken and the resulting culture turns positive, contaminated product will not likely be picked up until late in manufacturing and sometimes only after release, notes Stacy Montgomery, Ph.D., director of global marketing for Biolog (www.biolog.com).
Biolog produces microtiter well-based microbiology tests that identify more than 2,000 unique organisms based on their carbon utilization. Panels consist of 95 wells containing various sugars and amino acids, plus one control well. Biolog offers panels for major organism typesGram positive, Gram negative, yeast, and filamentous fungi. As they use carbon sources for food, microorganisms generate NADH, a biological reducing agent, which transforms a tetrazolium redox dye from neutral color to purple.
Plates may be read manually, on a microplate reader, or through a completely automated system. Each assay uses two time points within one of two time regimes: 4𔃄 hours for fast-growing organisms like E. coli, and 16󈞄 hours for slow-growing Gram negative non-enteric bacteria. Organisms show unique fingerprints for carbon utilization based on their preferences for some carbon sources over others. Microorganisms are identified by comparing their utilization patterns against those stored in a database.
The fully automated OmniLogID contains an incubator, analyzer, computer, printer, software for reading Biolog’s MicroPlates, a turbidimeter for preparing inocula, colony-magnifying lamp, and an eight-channel electronic pipetter. Biolog also offers an semi-automated MicroLog analyzer and the MicroLog series manual systems.
BioVigilant Systems (www.biovigilant.com) distinguishes between standard microbiologic detection systems that are merely rapid and those that are instantaneous. Rapid systems can never be continuous since they employ inherently slow, labor-intensive chemical and biological methods. BioVigilant’s Instantaneous Microbial Detection (IMD) systems use optics to verify the presence (or absence) of microbes without growing cells or chemical staining.
BioVigilant offers two types of instruments for pharmaceutical-quality applications: IPD (Instantaneous Particle Detector) and IMD. Both may be used as part of a process analytical technology (PAT) program for microbial alerts or continuous monitoring/trending of air and process liquids.
IMD requires almost no sample preparation or human intervention. Since it operates nondestructively, IMD provides a high level of assurance during downstream processing of concentrated product streams, especially during fill/finish operations, according to the company. Its only limitations are the inability to detect viruses or identify microbes by species.
Water:an Obvious Source
Pharmaceutical and biotech manufacturers spend millions of dollars per process each year on highly purified water, for example WFI (water for injection). Because water is used immediately and has such broad impact on quality, the industry has been keenly interested in systems that provide rapid JMAR Technologies (www.jmar.com), which specializes in laser technology for imaging, analysis, and nanofabrication, is breaking into the water-monitoring business through a recently initiated distribution agreement with GMP Systems (www.gmpsystems.com).
JMAR’s BioSentry provides real-time, continuous, online monitoring and classification of bacteria and protozoa in critical water systems. The device uses laser-generated multi-angle light scattering that generates unique signatures for microorganisms based on the system’s pathogen-detection library. BioSentry integrates into existing information systems and is configurable for alert thresholds and notification depending on the pathogen. In April, GMP Systems became the exclusive pharmaceutical distributor for BioSentry in the northeastern U.S.
According to John Ricardi, vp for sensor products, BioSentry works at several points in process water purification systems.
Environmental air can be a significant source of bacterial contamination in controlled environments. Most monitoring systems have difficulty distinguishing externally introduced contaminants from those growing in the cleanroom, thus generating false positives that lead to process interruptions. Pall’s (www.pall.com) Ascotec environmental air-monitoring system claims to distinguish between the two types of contaminants through an innovative capture technique.
Asotec uses Pall’s low-velocity BioCapt impactor sampler, which provides slits that microorganisms travel through before landing on the assay media. Random contamination falling onto the media therefore lands outside that pattern and is counted separately.
Ascotec is available as a remote or portable system. The Ascotec AirCapt Multipoint MP8+ remote system features eight sampling lines configured for simultaneous and/or sequential sampling of up to 16 independent lines. The portable MiniCapt PS1 sampler operates at a constant flow of 25 or 50 L/min, and is monitored in real-time.
Pall is positioning Ascotec, which it made more broadly available to biotech markets in April, as a complimentary technology to the Pallchek Rapid Microbiology System, which according to the company fits into FDA’s PAT initiative. Through PAT’s real-time monitoring, FDA hopes that pharmaceutical and biotechnology companies can monitor process quality continuously versus post-hoc product testing.
A Skeptical View
As a chemical engineer, Jim Agalloco of Agalloco & Associates (www.agalloco.com) holds some rather original views on contamination control in bioprocessing. Most contamination problems, he says, result from improperly sterilized vessels.
Sterilization is not as straightforward as it seems, but not necessarily because of breakdowns in rigid steam-in-place operations. Rather, end-users of bioreactors fail to realize that they are in the business, indirectly, of designing sterilization systems.
Users modify reactors, adapting them with new piping that takes its capabilities beyond what it was originally designed for, Agalloco says. Modifications can easily introduce regions that do not drain condensed steam as well as they should, resulting in a breeding ground for microorganisms. Designing in cleanability is a responsibility that falls to whoever designs the bioreactor in its final format.
Agalloco makes some strong statements about using classified space for bioreactors. In a nutshell, he believes it is a waste of time. People have been successfully making wine and beer for thousands of years without classified environments. Why does biotech need classified space for bioreactors? Contamination at that stage is not a function of the environment, but of procedures used to introduce inoculum. If the inoculum is clean, and only contains one organism, you don’t need a classified environment for the reactor.
According to Agalloco, trying too hard to protect the bioreactor environment can adversely affect the ability to sterilize equipment. For example, a steam sterilizer normally requires an atmospheric break between its drain and the facility drain, but some biotech companies object to that layout because it compromises the controlled environment. Eliminating the atmospheric break introduces more piping and surfaces, which leads to more opportunities for microbes to grow. To protect the outside of the tank, they purposely risk contaminating the inside. Product is not formed outside the vessel, but inside.
Another pitfall is to apply the strict environmental classes for fill/finish to operations that are much further upstream, where the product is not sterile. Over-playing environmental restrictions causes manufacturers to chase after contamination problems in areas where microorganism or particle levels are not an issue.
If you slap a class 10,000 requirement on an upstream area, a certain microbial limit comes with it. Unfortunately, that limit was established for different circumstances, and in the end you have to deal with excursions that are meaningless.
Agalloco cites one manufacturer who had originally set different microbial limits for similar particle classifications but was advised by an ex-FDA official to apply fill/finish standards throughout. This firm subsequently suffered through numerous internal investigations for microbial levels that were out of specification, even though these excursions had no impact on product quality. Aseptic fill/finish is the only hard and fast standard, Agalloco says, so manufacturers automatically feel they must apply it everywhere.