Safe & Sterile


Dedicated equipment, minimal operator intervention, and single-use technologies are just a few keys to success in the aseptic processing of biopharmaceuticals.

By Bruce Flickinger

Click here to enlarge image

Although biological products and their potential contaminants are fickle entities, the manufacture of sterile biopharmaceuticals has matured into a well-characterized, industrialized endeavor. Much of the uncertainty is gone, the constraints and opportunities are largely understood, and incidents of compromised products reaching consumers are relatively rare. These facts speak volumes about the technologies, training, processes, and ongoing research involved in bringing aseptically processed drug products to market???often for use against the most serious ailments in the highest-risk patients.

“A standard, well-controlled aseptic process should not afford conscientious operators many opportunities for error,” says John Dobiecki, vice president of manufacturing operations, Microtest (Agawam, MA). “If you do what you’re supposed to do in terms of facility design, process flows, the education of people in the processes, good training programs, and segregation practices, you really shouldn’t have any problem performing a formulation and aseptic fill/finish of a biologic.”

Sterile fill/finish processes are used throughout the continuum of biopharmaceutical development and manufacture. They are the essential aseptic process???defined as ensuring the absence of microorganisms capable of causing infection or contamination???and entail separately sterilized drug product, container, and closure being brought together in a controlled environment. So while virtually any solution, powder, or suspension can be sterilized prior to fill/finish, there is no terminal sterilization step for the biological product in its final container.

Aseptic fill/finish is thus highly dependent on technique, detailed procedures, equipment, controls, and facility design. The more unusual or complicated the product or container system, the greater the technical or operational challenges that can ensue. Potential problems include stability issues in the final dosage form, or a requirement for special or abnormal manipulations that have to be done aseptically. Still, Dobiecki says, “As long as methods are developed, engineering studies performed, the manipulations validated for these aseptic steps, and workers are educated to carry them out correctly, processes remain well controlled.”

Furthermore, different sterilization processes are used in fill/finish operations: Glass containers are treated with dry heat; rubber closures are subjected to moist heat; and liquid dosage forms are filtered. Irradiation might also be used and ethylene oxide can be used on packaging. Solutions also might be subsequently lyophilized in a sterile dryer to further improve the drug’s stability profile. Each of these processes requires validation and control because each of them could introduce an error that ultimately could lead to a contaminated product.

A control issue

Well-controlled though these processes may be, all stakeholders in the aseptic space constantly seek to tighten the screws and reduce margins for error. Any manual or mechanical manipulation of a sterilized drug, components, containers, or closures poses the risk of contamination and thus necessitates careful control.

Recalls of sterile products are relatively unusual, and problems most often occur in the field, whether resulting from a process issue or mishandling by the user. To gain detailed insights into these problems, the U.S. Food and Drug Administration (FDA) recently conducted an in-depth analysis of 17 products involved in Class II recalls from 2006???2007 due to drug product sterility concerns. Of these, seven recalls were the result of “non-sterility” and 10 were the result of “lack of sterility assurance,” which can occur when a preponderance of current Good Manufacturing Practice (cGMP) violations cast doubt on product sterility or because of problems found with the protective packaging.

Results of the analysis were due to be presented at a joint PDA/FDA conference this September [to take place after this issue went to press]. Generally, however, the agency has indicated that cGMPs were the underlying reason in six of the 10 recalls for lack of sterility assurance???the most disconcerting from the process owner’s standpoint???and packaging component integrity the cause in four of them.

Figure 1. Plastic and stainless working together: sterile transfer operation using a Millipore Express™ SHR (Sterile High-Retention) filter and Lynx™ ST (Steam-To) connector on a stainless-steel tank. The Express devices provide sterilizing-grade performance and mycoplasma removal for cell culture media, media additives, process intermediates, and other biological fluids. Photo courtesy of Millipore Corp.
Click here to enlarge image

FDA’s 2004 guidance on “Sterile Drug Products Produced by Aseptic Processing???Current Good Manufacturing Practice” could also help mitigate sterility process issues. This document is significant in that it recommends building quality into products through science-based facility, equipment, process, and system design. Risk-based approaches are addressed in the guidance relating to personnel, design, environmental control, and media fills in an aseptic processing operation.

The guidance is part of FDA’s broader vision for a more agile, flexible approach to overall quality assessment that allows industry to make drugs to the same level of quality without extensive regulatory oversight. Where regulatory submissions for biologics and other drugs currently tend to be prescriptive, with less focus on critical analysis and scientific rationale, the guidance seeks to move companies toward demonstrating product and process understanding and emphasizing robustness instead of reproducibility. This allows for process change within a “bookended” design space.

Certain tenets of sterile manufacturing hold true regardless of the particulars of a process and its validation. One-way movement within the cleanroom, especially in the sterile fill/finish suite, is critical for all of the materials, enclosures, and components. The cleanroom design should eliminate two-way transfers from occurring concurrently, where sterile goods are physically passing “non-sterile” goods.

Physical or process barriers are also important in the environment. For example, after a stoppering process for solution drugs, sealing occurs immediately, usually with some kind of aluminum seal. The design of the cleanroom or equipment should include a barrier or space separation between the stoppering and sealing processes to minimize any potential particle contamination from the stoppering process.

An alternative to stoppering is the use of form/fill/seal. This fully automated process forms a plastic container system, filling the solution during the process and immediately sealing the containers. The form/fill/seal process minimizes environmental exposure and can provide microbial contamination results similar to an isolator process.

For lyophilized drugs, the filled and partially stoppered vials are transferred to a sterile lyophilizer. It is normal for the stoppers to be seated in the vials inside the sterile dryer at the end of the lyophilization cycle prior to opening the door. The stoppered vials then are removed from the dryer and immediately capped. This delay in sealing the container immediately after the filling process, during which time the drug is exposed to the environment, presents an additional risk that needs to be addressed in validation protocols.

Regarding the 2004 guidance, Microtest’s Dobiecki says, “As a contract manufacturer, we take the most conservative approach in incorporating new guidances or technologies. We have used risk-based approaches in all aspects of our manufacturing processes and they have been effective. Aseptically we still concentrate on putting in place appropriate systems and monitoring them on a continual basis.”

Equipment control is a key safeguard, especially in a multi-product facility where cross-contamination is a major concern. “We operate using a very robust equipment management system,” Dobiecki says. “We employ multiple check and balances in every process to make sure that the proper components and equipment are being used.

“Everything used is either disposable or dedicated to a single product,” he continues. “Even where one client has multiple products, each product will have a unique set of dedicated equipment and disposables.”

Single-use technology gains traction

On this point, there is intense interest currently in product-dedicated systems, and particularly in single-use technologies.

“There is a huge amount of interest in single-use technology in the field,” says Jerold Martin, senior vice president, scientific affairs with Pall Life Sciences (East Hills, NY). “As with any new idea, you have early adopters and those who prefer to wait until there is a greater body of data and experience. We’re seeing this second wave of users starting to jump in.”

“Single-use technology fits perfectly with final fill and finish where the end users’ key goals are GMP compliance, reliable retention assurance, and reduced process risk,” adds Laurelle Sciola, senior product manager with the Bioprocess Division of Millipore Corp. (Billerica, MA). “There is also the added convenience of a ready-to-use, self-contained, pre-sterilized filtration system, which reduces validation costs and contamination risk.”

The single-use concept essentially involves the use of disposable plastic process components instead of traditional stainless steel. Plastic components are nothing new, of course; filter capsules have been used for many years for sterilizing-grade final filling applications. Tubing, connectors, and biocontainers are becoming widely used as well in buffer and cell-culture media filtration applications.

But the shift entirely from stainless steel is something different altogether. As acceptance grows, single-use systems are becoming more complex, extending beyond a simple filter and tubing configuration on a bench to complex systems on a stainless-steel frame or skid, and plastics as the sole fluid pathway. “Certain steps in full-scale manufacturing still rely on processes that are not yet available in a single-use format, such as chromatography and centrifuging,” Sciola notes.

“The biggest question about using plastic systems relates to the potential impact on drug product quality,” Martin says. “It’s a big step to move from stainless, which, rightly or wrongly, is assumed to be inert, to plastics, which are assumed to have higher levels of leachables that could affect product quality and safety.” He points out that some early converters to plastic processing were working with protein products that were sensitive to metal ions.

“We just don’t know yet how much data has to be generated for FDA to feel comfortable,” Martin says. “The guidances say essentially that process equipment cannot affect the product, and this doesn’t change based on materials. There’s just more and different testing involved.”

Martin says his discussions with FDA indicate the agency “is keen” on companies incorporating single-use systems, particularly at the clinical stage. Indeed, most of the conversion efforts to date are in processes for clinical batches, which don’t often need FDA approval. Stakeholders are “eager to see what FDA will accept and approve in these processes,” he says.

The aversion to change in regulated industries notwithstanding, plastic single-use systems are attractive on several counts. First, dedicated equipment minimizes the risk of cross-contamination and single-use equipment “further minimizes the risk from lot to lot, not just across products,” Martin says. Furthermore, as the technology matures, manufacturers are offering pre-connected closed systems that offer better protection against external contamination compared with users assembling the components themselves.

Pall’s single-use product line, for example, includes filter capsules and sterile connectors, disposable membrane chromatography units, and a new line of biocontainers. These components can be assembled into its systems available in varying configurations. “We have a well-established engineering organization from our experience in stainless steel, and we’re able to leverage this infrastructure for our single-use systems,” Martin says.

Millipore also offers a range of single-use components, such as a line of filtration devices that find application in a range of biotech and pharmaceutical sterilizing-grade applications, including final sterile filling, bulk sterile hold, final formulation buffers, mycoplasma protection of cell culture media, vaccines, and chromatography column protection steps.

Figure 2. Allegro™ System components come pre-assembled and gamma-irradiated to enable quick startup and rapid scaleup. They include Allegro Biocontainers, Kleenpak™ capsule filters and sterile connectors, and Mustang® membrane chromatography capsules. Photo courtesy of Pall Corp.
Click here to enlarge image

Single-use systems also eliminate the need for cleaning, which not only closes another window for potential contamination but also is a huge cost saver. The significant time and costs involved in cleaning validation, as well as use of chemicals and water for injection required for cleaning and rinsing, can be sidestepped with single-use technologies.

A third advantage offered by single-use systems is speed: Martin says a single-use facility can be up and running in a matter of months compared with years with stainless steel.

Adequate testing, minimal intervention

Martin is chair of the Bio-Process Systems Alliance (BPSA) Guidelines and Standards Committee. The BPSA was formed three years ago to promote the acceptance and implementation of plastic single-use systems in pharmaceutical manufacturing, and to develop consensus recommendations for components and methods that safeguard the quality of drugs produced using these technologies.

The group, comprising about 40 member companies, has published four best practices guides since 2006; they cover component quality tests, extractables and leachables, irradiation and sterilization, and disposal. The BPSA technology committee currently is working on two additional guides that will present consensus methods for extractables and leachables separately.

“It’s important to have standardized methods so that end users can compare core data from their suppliers,” Martin says. “End users will always have to do more specific testing, mostly concerning leachables in their particular processes.” BPSA’s efforts have spurred other organizations into action; both ASME and PDA have established single-use task forces.

Test methods and strategies are critical throughout the aseptic processing lifecycle. Moving from purified bulk drug substance to the sterile final drug product is a multi-step process that calls for testing at several points to ensure product identity, composition and microbiological integrity (see table). There are several universal tests for sterile products, as well as testing specific to the particular product.

Table 1: Examples of testing performed during aseptic processing
Click here to enlarge image

Bulk biological drug products, as opposed to their small-molecule counterparts, typically arrive at the fill area either as a refrigerated liquid or frozen material, so they need to be sampled in a way that minimizes temperature increases of the bulk container. This sample then needs to be temperature controlled until time of analysis. These types of precautions need to be delineated in specification and SOP documents, Microtest’s Dobiecki says.

Frozen bulk drug substances pose an additional difficulty because thawing the material to obtain an identity sample should be avoided. Three approaches can be taken to address this, Dobiecki says. One is to perform sampling, testing, and release of the substance concurrently with the formulation process. Testing can also be performed on a small, representative sample that is prepared with the bulk substance but shipped separately. A third option is to receive the material solely against its certificate of analysis along with formal instructions from the client to proceed in this manner.

Dobiecki notes that beyond taking representative samples, additional samples might be desired to gather information on the effects of the thaw and processing time on product purity. “This can be valuable for early-stage clinical trial material where process history is not yet available,” he says.

During formulation, which usually occurs in at least an ISO Class 7 clean environment, Dobiecki says three principles need to be incorporated into the sampling/testing plan: Perform all sampling using aseptic techniques; test to assess only critical product/process parameters; and minimize process interventions, including sampling.

Generally, the more involved a formulation is???requiring the addition of diluents to achieve a desired concentration, isotonicity, or pH level, for example???the more additional testing might be necessary to supplement drug concentration data. Multiple sample points might also be required depending on the design of the formulation process.

Figure 3. Filling, capping, and crimping of sterile final drug products. The operation is semi-automated: Vials are manually loaded and removed from equipment but all filling, stoppering, and capping is done automatically. Photo courtesy of Microtest.
Click here to enlarge image

Once the material is formulated, the next step is sterilizing filtration. Testing associated with filtration minimally consists of pre-filtration bioburden, sterility testing of the bulk sterile drug product, and filter integrity testing of the sterilizing filters and any associated sterile vent filters.

Following this, fill/finish operations are performed in an ISO Class 5 area, during which several in-process tests are performed. The first involves the assurance of delivery volume; here, Dobiecki says, the filler is adjusted to a “best guess” setting and the fill head is purged to remove any air. A minimum of three sequential vials are filled and removed from the line, their contents gravimetrically verified, and volumes determined. Additional weight checks should then occur periodically throughout the fill.

“The fill suite operator’s other responsibilities include monitoring the fill process and assessing the general quality of the finished drug product coming off the fill line,” Dobiecki says. Maintenance of fill sequence is important when sampling for release testing; sterility samples must be taken from the beginning, middle, and end of the fill. Release tests will vary but typically will be developed to determine product concentration; activity (for biologics); purity (and impurities); endotoxin; USP particulate, color, and appearance; and deliverable volumes. It is important to note that “samples should not be pulled for release testing until after the finished drug product has left the fill/finish suite and gone through 100 percent vial inspection,” Dobiecki says.

Deciding on pre-use testing

One type of test that generates ongoing interest and debate is pre-use filter testing. Although integrity testing of sterilizing filters after use is a regulatory obligation, opinions vary as to whether drug manufacturers should conduct their own integrity testing, or if the assurances provided by their filter suppliers are sufficient to meet regulatory and process validation requirements.

Regulatory authorities tend to hedge on the topic. The EU guidance says post-sterilization, pre-use testing “should” be conducted, which generally is interpreted as “must,” while FDA says pre-use testing “can” be conducted, which carries more of an optional connotation than “should.”

Furthermore, “the FDA aseptic processing guidance does not distinguish between post- and pre-sterilization pre-use testing,” Martin notes. “These are risk-based decisions, and pre-use filter testing is primarily a decision based on the potential for lost product, not on the risk to the patient.”

Beyond deciding to conduct pre-use integrity testing, the processor has to do it correctly. “It’s important that operators are trained to do filter-integrity testing properly, because most negative results turn out to be false negatives,” Martin notes.

Filters used in sterile processes may be received from the vendor as pre-sterilized capsules or non-sterile, in which case they then are assembled by the user into the appropriate filter housing assembly configuration. The entire assembly then is sterilized using a validated autoclave or steam-in-place (SIP) cycle. Certainly, manufacturers of bioprocess filters conduct rigorous integrity testing of their products and supply the attendant documentation to their end users. The question becomes whether the handling and configuring of the filter by the user then requires final testing for integrity before use in the process.

Millipore, for example, uses a multi-step integrity testing process. First, membranes are bubble-point tested. “True correlations have been demonstrated between membrane BP values and bacterial retention,” Sciola says. “Only membrane that exceeds the minimum critical BP, as defined in the membrane validation, is used to fabricate cut disks and devices.” In addition, each lot of membrane is tested and certified to meet specifications, including complete retention of B. Diminuta at a minimum challenge level of 107 CFU/cm2 per ASTM 838-83 methodology.

After pleated filter devices are fabricated, they are tested using a low-diffusivity gas to minimize background diffusive flow for heightened test sensitivity. Each filter device lot is tested and certified to meet specifications, including those specified in ASTM 838-83 methodology.

Millipore also conducts a validation exercise in its ACCESS validation lab, in which bacterial retention of B. Diminuta at a minimum challenge level of 107 CFU/cm2 is assessed using the end user’s actual process fluid, processing conditions, and with membrane that is within 10 percent of Millipore’s minimum validated bubble-point specification.

“We perform a 100 percent in-process integrity test on all filters prior to release and provide validated integrity test specifications to end users,” Sciola says. “Whether to perform a pre-use, post-sterilization test of the filter’s integrity is at the user’s discretion and depends on a number of factors involving system design and regulatory compliance. The user must perform his/her own risk assessment based on the application requirements and validated claims.”

Dobiecki feels that, unless it is specifically requested, “pre-use testing generally is not advised because it exposes the filters to additional challenges and handling with questionable benefit.” Filter manufacturers, as well as contract manufacturers, have experience working with filters and will have associated data concerning extractables, the potential for product/membrane retention, and other factors, and can “assess that information against the formulated drug substance’s matrix for compatibility.” Coupled with the use of a redundant sterilizing filter, “this provides a very high level of assurance that the filters will perform as designed,” he says.

Conversely, “We support the EU regulatory guidance for post-sterilization, pre-use testing,” Pall’s Martin says. “The bias against doing this is based on the perception of risk of additional complexity to the system, but these concerns are not supported by science.”

Playing it safe

Sound science underpins the development and manufacture of biopharmaceuticals and other biological products, which include vaccines, small molecules, biological proteins, oligonucleotides, and various types of suspensions. The impact of these products on the improvement of human health is just beginning to be felt. But their promise imparts a momentous responsibility for ensuring their safety, a responsibility borne not only by health care providers but also the myriad manufacturing and technology companies working to put medicines in their hands.

Aseptic processes are indispensable links in this chain of manufacture and supply. Practitioners continue to move toward disposables and single-use technologies, product-dedicated systems, and less human involvement with the aim of better ensuring sterility and, ultimately, patient safety. Their efforts span early clinical-phase hand filling to small-volume, semi-automated filling to fully automated, high-volume production over multiple days and runs. Risk and the potential for error are inherent regardless of process volume or complexity.

“When I hear about a product in the market with sterility issues, it’s troublesome for all of us,” says Microtest’s Dobiecki. “Thankfully, it’s an extremely rare occurrence. With proper operator education of each process, strict adherence to cGMP practices, validated container/closure systems, and qualified supply chains, there shouldn’t be any issues with final drug product sterility.”