Guide Will Clarify Use of Barrier Isolation
Unclear regulatory expectations hinder the use of barrier isolation technology, despite its potential. But FDA and industry are working to address issues.
by Erik Swain, Senior Editor
Aseptic processing of liquid pharmaceuticals is perhaps the most complicated task in healthcare packaging. A mastery of chemistry, biology, engineering, sterilization techniques, and personnel procedures is essential for the product to be packaged without contamination.
Executing barrier isolation—taking an aseptic filling line and enclosing it in an isolator that separates it from the external environment—is even more complicated. There are few isolators in operation to serve as models for new lines, and there are no official standards on how to implement the technology properly. Without a set of standards, pharmaceutical companies carry out barrier isolation as they see fit. Therefore, one company's method will differ from that of another.
"When one pharmaceutical company does barrier isolation one way, and another does it another way, that makes it very hard for a vendor," says Mark Zarembo, manager, barrier isolator division, The Baker Co. (Sanford, ME). "There is still a lot of custom stuff being made. My goal is to build the exact same unit twice."
The Baker Co.'s isolator protects products, the environment, and personnel.
There may be guidance for industry soon. The International Society of Pharmaceutical Engineers (ISPE; Tampa, FL) just published a guide on sterile manufacturing facilities, which contains a chapter on barrier isolation. While it is not an official regulation, standard, or guideline, FDA has provided comments on it.
Perhaps as early as this year, FDA plans to revise its 1987 aseptic guideline and add a section on barrier isolators, according to Richard Friedman, compliance officer with FDA's Center for Drug Evaluation and Research. "FDA for the first time will publish barrier isolation guidance," he says.
This will be a very positive development, as industry has been abuzz for years about what FDA will expect from this technology, especially in terms of validation. FDA, in turn, has eagerly sought input from industry about which barrier isolation methods and procedures have worked and which ones have not. With the publication of barrier isolation guides, pharmaceutical companies will have a resource to direct them as they add this technology to their capabilities.
If designed and maintained correctly, a barrier isolation line can provide a lesscontaminated environment, lower long-term facility and staffing costs, and better operator protection than conventional filling lines. Despite the current difficulty in setting up such a line, the technology is becoming more widely used.
Jack P. Lysfjord of TL Systems Corp. (Minneapolis) and Michael E. Porter of Merck & Company, Inc. (Whitehouse Station, NJ, and West Point, PA) released a study last year showing that there were 84 barrier isolator lines delivered, more than anyone had expected. The lines were used more widely in Europe, but were catching on in the United States. Last year, six systems had FDA approval, and there have been a few more since.
"The more biotech drugs there are, the more important [barrier isolators] will be," says William Arden, marketing manager for TL Systems, part of the Bosch Group. "The original reason for them was to keep humans outside the filling process. Now, with drugs that are potentially dangerous to operators, the question is how to keep the drug inside the isolator." This may explain the increasing need for barrier isolation systems.
Although becoming more widely used, barrier isolation still brings up questions and concerns. Friedman says one of the foremost is coming up with an accurate definition of barrier isolation.
"If it is not a sterilized unit, supplied with HEPA, ULPA, or sterile filtered air and providing uncompromised continuous isolation of its interior, then it isn't an isolator," Friedman says. "You must isolate the two environments so that they are exclusive of one another. If a firm does something less and stretches the definition, it will in fact undermine the exceptional efforts of its industry counterparts who do barrier isolation the right way. If one company tries to bring down the standard, that will be problematic for the rest of the industry."
The key attributes of a true isolator, he says, are that it is sterilized, it is uncompromised, and it provides continuous isolation. All of these are at the core of the debate over what exactly barrier isolation can be expected to do.
"It's not coming up with an isolator with no discernible leak rate that's important," Friedman says. "It's important only that there is an adequate design and overpressure that is continuously maintained and that precludes the ingress of any particulates from the environment surrounding the isolator."
But, he says, claiming that isolators can achieve sterility assurance is a mistake, because aseptic processing does not involve a terminal sterilization process of known lethality. Rather, media fills are performed to determine a contamination rate, which is the more accurate phrase. "Aseptic processing, no matter how it is done, is not the equivalent of terminal sterilization," he says. "A barrier isolator is still housing an aseptic filling line."
If the definition of an isolator is something that completely separates the inside from the outside, then the methods of getting the product in, such as mouseholes and transfer ports, will continue to be considered one of the technology's problem areas. This is where breaches of isolation often occur. To combat such breaches, IMA North America (Fairfield, CT) recently patented a method of using vaporous hydrogen peroxide in the cold zone of the sterilizing tunnel, says Warren Roman, vice president.
Another breach of isolation occurs with the gloves used by the operators. Because there is no way to test gloves for smaller leaks that permit the passage of microorganisms, "conservative, mandated glove replacement schedules" are essential, Friedman says.
There have been inconsistencies with how industry has dealt with breaches, from glove, valve, and power failures to out-of-specification pressures. But there is no inconsistency in Friedman's instructions. "Any breach of the isolator . . . should be investigated and the affected product rejected, and should lead to a mandatory sterilization cycle," he says.
Other areas that draw attention during an inspection of a breach are structural integrity, materials of construction, pressure differential, clean area classification, frequency of sterilization, and personnel practices, Friedman says.
Besides breaches, another concern is ensuring the isolators are designed so that all portions, especially the driving machinery, can be cleaned and sterilized properly.
"A firm should minimize sterilization challenges to the greatest extent possible," Friedman says. "There have been product failures due to cluttered design concepts that prevented surfaces from being well exposed during the surface sterilization cycle."
For that reason, FDA frowns on retrofitting—taking a line that has not been designed for use in an isolator and building an isolator around it. "A more narrow, sleek type of line is used in the isolator, with most moving drive parts outside the isolator," Friedman says. "This is not necessary for conventional operations. They are different animals. I have not seen any retrofitting, for good reason."
With that in mind, some companies are designing filling lines that they say can eventually be adapted for use in isolators, especially making sure all parts can be cleaned and sterilized.
Bausch + Stroebel Machine Company, Inc. (Clinton, CT) even offers an option to install certain brackets, holes, and other parts up front to make an eventual conversion easier. "We help the customer foresee what is in the future," says Paul R. Chimino, sales engineer. "It's best to get the hardware in place ahead of time." However, he says, the company is not trying to force customers into that option, and it is most concerned with selling the customer the most appropriate machines, with or without an isolator.
Whether FDA will approve of such conversions remains to be seen.
THE CONVENTIONAL WAY
Despite the attention given to barrier isolation, the vast majority of filling lines are produced for conventional aseptic processing. There are fewer up-front costs with conventional lines, which have a history of effectiveness. In addition, conventional lines are seeing their share of advancements. Increasing automation of filling lines is perhaps the most significant trend, because less operator involvement means less opportunity for contamination.
"The automated equipment must be gentle enough not to fracture vials or cause other unforeseen stresses to the product," Friedman says. "All in all, the trend toward increased automation has been very positive. Automated weight checks are easily adapted into current lines."
Filling machinery manufacturers say that they are applying some of what they have learned in barrier isolator design to their conventional filling lines.
"Reducing the overall footprint is important, as is making the changeover items toolless," says Joel Slazyk, vice president of sales and marketing, Chase-Logeman Corp. (Greensboro, NC). "The market is still not wide open for isolator systems, so we still spend a lot of time on standard filling systems. But when the market does open up, we'll be prepared."
One of the most important issues in designing a filling line, Chimino of Bausch + Stroebel says, is to determine which filling technique is best for the product in question—rotary piston pump, time pressure, or weight density.
Stopper technology has also evolved because of barrier isolation. Last year, West Pharmaceutical Services (Lionville, PA) introduced ready-to-use stoppers, already sterilized, eliminating the need for human contact. They also offered an autoclavable package.
Automated stopper transfer systems, such as the one West makes, remove the human factor even further. "To the extent that you remove an operator, even though in this case it is an operator outside an isolator, you are improving the process as a general rule," says Friedman. "When you have stoppers in plastic bags and rip them open, you are generating particles. Also, when you manually charge them into the stopper bowl, it is not ideal."
Barrier isolator technology, though showing great promise, will probably not be fully accepted in the United States until there are formal standards, regulations, or guidelines in place, and until more data are available to prove its superiority to conventional aseptic processing.
Yet, the technology will always require sound design, control, and maintenance—challenges that can be difficult to meet.
"If this technology is able to consistently reproduce what it is apparently capable of achieving, one of the only remaining impediments for a firm would be inattention to the stringent preventive maintenance requirements," Friedman says. "An ever-vigilant eye is of the utmost importance to this technology.
"Frankly, it may not be for everyone," he continues. "For a firm whose operating philosophy is to meet only minimal GMPs and operate on the cusp of CGMP compliance at all times, this technology is not the right one. But it is if the firm has a good quality philosophy and understands the importance of designing quality into machinery, and if the firm operates and maintains the line to exceedingly high standards on a routine basis."
Validation of an Isolator System Using Vaporous Hydrogen Peroxide
by Anne Booth, president, Booth & Associates (Barrington, IL)
While barrier isolator technology for liquid pharmaceuticals and medical devices is still evolving, it is very important that any isolator being put into use is validated. Isolators must meet certain predetermined performance criteria; perhaps the most important of which is a sterility assurance level of 10-3 to 10-6 depending on the exact use.
Anne Booth offered the following isolator validation advice for systems using vaporous hydrogen peroxide (VHP) at the Medical Design & Manufacturing West 99 conference, held in January in Anaheim, CA.
The documentation should include a detailed description of the physical system and a diagram of the unit layout with interfaces and transfer systems clearly marked. The following items should also be included: equipment description, manufacturers' specifications, construction materials, instruments (with calibration status), utility specifications, HEPA filter certifications, and computer software.
The isolators must be checked independently of each other and of the generator. Collect data by performing the following: a mock run to check cycle alarms and alerts, an integrity test for leaks, and a pressure test to ensure positive pressure can be maintained. Establish preventive maintenance and cleaning procedures, document proper air exchanges, and collect data on the piping system that connects the isolator, generator, and outside exhaust.
Each phase of the sterilization cycle must be evaluated, taking into consideration HEPA integrity, computer alarms and alerts, drier capacity and status, sterilization cycle verification, temperature distribution and mapping in the isolator, and uniform sterilant distribution using chemical indicators.
Process qualification determines appropriate cycle parameters given the isolator configuration, room temperature, and loading. The system need not be installed in a controlled environment but must be in a room maintained at a constant temperature and relative humidity.
Prior to starting, develop the maximum fixed load configuration to be used during routine operation. Then conduct temperature mapping to determine heat distribution. This analysis, using thermocouples placed throughout the isolator and load, is needed to determine the cold spot within the isolator in order to calculate the maximum safe concentration of hydrogen peroxide that can be used without causing condensation. A good figure is 20%, as anything lower can add considerable time to the sterilization cycle. For best results, the drier capacity should be greater than 12 hours.
Chemical indicators should be placed adjacent to the thermocouples (as well as in difficult locations such as gloves and half-suits) to make a general evaluation of gas distribution. The color should gradually change from white to gray violet within 25 minutes. For most uniform gas distribution, mount fans within the isolator.
Watch for condensation. If it occurs, it can indicate that the gas concentration is too high for the temperature within the isolation, or that the concentration is not uniform. Estimation of the concentration is determined from tables provided with the generator.
STERILIZATION CYCLE DEVELOPMENT
The isolator should be validated to a sterility assurance level of 10-6. The sterilization of all internal surfaces of the isolator and of all external surfaces of the items in the isolator is validated using a resistant biological indicator (BI)—the best to use in this case is Bacillus stearothermophilus with a spore population of 106. Spores are inoculated on carrier material that does not absorb the sterilant and are packaged in Tyvek pouches.
There are two ways to determine the appropriate half cycle. One is to gauge performance of sequential fractional cycles by increasing the gas exposure time for each, similar to the method used in EtO validation. After each fractional cycle, retrieve and sterility test the BIs. The other method is to test pairs of exposed BIs within media tubes in the isolator's worst-case location at intervals during a single gas exposure cycle. In both cases, the time that produces total kill becomes the half-cycle gas exposure time.
Once the half-cycle gas exposure time is determined, run three consecutive cycles with BIs and thermocouples. Sterility testing of BIs should yield all negatives if the cycle parameters are correct. The appropriate BI growth controls should all be negative, and the media growth promotion should test positive.
Then run a full cycle to determine the time needed to aerate the isolator completely. The hydrogen peroxide gas should be reduced to an acceptable level, usually 1.0 ppm. Aeration time depends on isolator volume, mass of adsorptive material, and the rate of outgassing from materials. Use a semi-quantitative gas detection tube to gauge aeration efficiency.
At several points, regeneration cycles may be required to remove the humidity from the drying agent. The typical time is 18 hours, but the lower the drier capacity, the longer the process will take.
OTHER VALIDATION ISSUES
Other tests are needed to determine that the gas is not penetrating product containers, supplies, and other objects.
Maintenance of isolator sterility over a period of time should be established and monitored. That means a schedule of routine sampling. Worst-case situations, such as loss of power and transfer of additional supplies, should be included during this period. Periodic inspection of gaskets, ports, and gloves to detect imperfections is essential.