Friday, July 31, 2009

biopharmaceutical manufacturing

Given the financial realities in the pharmaceutical industry today, there is intensifying pressure to decrease time to market and raise productivity, while containing costs for both facilities and R&D staff. These factors, coupled with the FDA’s elimination of the Establishment Li-cense Application (ELA) for biologics conforming to the definition of “well-characterized” products1, are driving rapid expansion in the area of contract manufacture for biologics. All areas of manufacture are available, including preclinical test materials, clinical trial (Phase I-III) supplies and marketed product. The decision to outsource manufacturing services is most effectively driven by a comprehensive strategy within the sponsoring organization for at least the near-term (and preferably, long-term) overall development of the product.

Clearly, adjustments to the strategy will need to be made along the way as more is learned about the properties of the drug (e.g., its pharmacokinetic and pharmacodynamic properties, scalability and yield of the process, as well as regulatory strategies). Many decisions can be made early to facilitate the product development process later. Let us focus on the search for outside manufacturing services in each of the above phases of development and discuss how this search interfaces with other activities necessary in the manufacturing and development of a biopharmaceutical product.

Development Stage
The development stage of a product clearly plays a major role in selecting a contract group for its manufacture. The considerations in the different developmental stages are obviously quite different, especially with regard to regulatory aspects specific to manufacturing.

Preclinical Testing
At the earliest stages of biological product identification and testing, its DNA sequence(s) may or may not be known (in the case of a recombinant protein product). The particular type of protein will dictate the type(s) of expression systems (i.e., mammalian/microbial) that are likely to be successful in producing viable quantities for testing.

For instance, monoclonal antibody production is, at present, limited to mammalian systems (although some recent reports have suggested that it is possible to produce antibodies using yeast expression systems2). Technologies are also currently available to produce fully human antibodies to an antigen of choice using transgenic mice, thus eliminating the need for “humanizing” a murine (mouse-derived) monoclonal antibody. Does the protein have glycosylation that is necessary for its activity?

The sponsor should keep in mind that the expression system chosen will determine to a large degree how much protein will be available for testing. Therefore, careful and thorough screening at this point will pay off later in the manufacturing process (and probably in later clinical manufacturing as well). In addition, some expression systems carry with them licensing fees—another factor that must be considered in the selection process.

All of the above issues must be considered during the selection of a site for production of preclinical material. Other important activities that must be ad-dressed at this stage are robust analytical assay development (addressing how to identify and quantify the amount of protein yielded) and fermentation and downstream recovery development. These all play a key role in outsourcing selection. Obviously, all of the above will be greatly facilitated by in-house expertise at the sponsoring company.

When looking for a contract manufacturer for preclinical material, the sponsor must decide whether to go to a “full-service” provider, i.e. a contract organization that can supply all of the services needed, from molecular biology to fermentation/cell culture development, as well as downstream purification and analytical methods development, or alternately to go to different suppliers for these various activities. While it is obvious that there may be advantages and disadvantages associated with both approaches, some mixture of the two choices could be most effective from both development and cost perspectives. Certainly, there is quite a broad range of choices at this stage of development.

Early Stage Clinical Product (Phase I/II)
The experience gained in preclinical outsourcing should be quite informative in driving the selection process for outsourcing of clinical-grade material to support phase I/II clinical testing. Beginning with human trials in Phase I, product must be produced under current good manufacturing practices (cGMP). This serves to limit the selection of potential contract suppliers considerably, in comparison with the options available to supply material for preclinical testing. Specifically, the contractor must be able to demonstrate compliance with FDA guidelines (or those of another country, depending on where the trials will be conducted) with regard to facilities, raw materials handling and manufacturing control and their associated documentation.

Since the ultimate responsibility for product compliance rests with the sponsor, one of the first steps in the selection process for outsourcing clinical product must be a thorough audit of the facilities and procedures used in the proposed contract site. Someone with intimate knowledge of current FDA guidelines should conduct the audit. In addition, this individual should have prior experience in the area of biomanufacturing.

During the early stage of clinical product development/ manufacturing, the contract manufacturer can play a vital role in assisting the sponsor with all of the activities related to the manufacture of the product. When it comes to selection, it is important to assess the level of expertise in each of the critical areas related to the successful manufacture (and IND filing) for the product.

It is now quite common to find personnel at contract organizations with extensive industry experience. This experience can be invaluable during planning and execution of the project; the staff can also provide advice at key points in the project. Usually, a process coming out of preclinical development needs at least some refinement in order to translate into a fully documented manufacturing process ready for execution under cGMPs. The contract manufacturer should be able to supply experienced guidance in this area.

If the process (or parts thereof) is to be developed by a contracted group, the sponsor must make sure to ask, “Who owns the process?” In other words, are there potential areas in a process that could be claimed by the contractor as inventions and thus are subject to licensing by the sponsor? This becomes an issue if the sponsor decides to move to another contractor. This area must be addressed up front with any potential contract manufacturer.

Late Stage Clinical Material (Phase III)
Production of material for Phase III clinical trials is a much more complex activity than that encountered in the early phases of clinical trials. By this time, the sponsor’s process should be very clearly defined (with, of course, some modifications and optimizations still in the works) and there should be a general idea of how much product will be required both for ongoing and future clinical studies as well as initial market launch. Is the facility under consideration capable of producing enough material using the current process to support not only Phase III trials, but also (very probably) the initial product launch? The level of regulatory scrutiny directed at the product at this stage will be quite high, in view of the fact that the lots produced will be used to support the eventual Biologics License Application (BLA) filing.

The above considerations limit the choices for contract sites in Phase III to a handful and, in some senses, make the job of selection easier. The facility used at this stage for product manufacture should be either already-licensed (that is, already producing marketed product) or licensable since your first marketed product may well be produced here.

The downside of all of this, of course, is that the small number of contract sites that meet the above criteria have become flooded with requests from potential clients, all looking to use outsourcing as a strategy for manufacturing. Since the site of Phase III manufacture will likely be at or near the scale contemplated for commercial launch of the product, discussions should begin early in negotiations regarding the long-term outlook for facility availability, potential expansion and, in general, where the contract manufacturer sees its business going.

Again, the contract site can play a major role in assisting client companies with expertise in a variety of areas, ranging from process scale up to validation issues. It is quite important at this stage to maintain a very close (and effective) working relationship between the sponsor and contract group, since many issues will arise that will require close collaboration. The smoother the relationship is, the more effective the problem solving.

The Search
How does one go about effectively looking for information about contract manufacturing and development activities? A number of sources exist for identifying organizations that offer contract manufacturing. For development and manufacture of preclinical material, choices range from university-based programs to full service contract organizations that can go from preclinical material through late-phase clinical trial product and beyond. Increasingly, the internet is becoming a valuable source of information for contract sites. For instance, www.biospace. com (the website of BIO, the Biotechnology Industry Organiza-tion) offers links to organizations offering various contract manufacturing and development services. Detailed information about contract manufacturers can be found by accessing their individual web sites.

The Selection Process
Once the sponsor has made its preliminary selection of a group of contract sites, the next step is to screen them to see which ones fit the criteria most closely. As mentioned above, the number of candidate organizations decreases dramatically as the project moves from preclinical development through clinical trials to marketed product.

The process usually begins with a call to the contract organization to get a better idea of its capabilities in the area of interest. During this initial discussion the sponsor should be prepared to offer some (non-proprietary) information about the product, its development stage and what some of the expectations are. If the fit looks good, the next step is to sign a Confidential Disclosure Agreement (CDA) with the contractor. This will allow the conversation to enter into more detail about the product, the organizations, the expectations for product supply (if appropriate) and the near- and long-term availability of the contract manufacturer’s facility and personnel. These initial discussions can take place by telephone, fax or e-mail and will set the stage for a personal meeting. During these discussions, it is also appropriate to ask for references from past customers; these can be a very good source of information.

Visiting the Site
During the sponsor’s initial visit to the contract site, it is usually possible to get a good sense of the compatibility of the project with the facility and also of the sponsor personnel’s compatibility with the people. During the tour of the facility, representatives from the sponsor can talk with in-house experts in the fields applicable to the project. They can also meet with management and the project coordinator (there should be an individual assigned to handle all of the details associated with the project and to act as contact between the organizations). The sponsor should be prepared to discuss the project in intimate detail in these meetings; this will allow both sides to assess whether the site is appropriate for the project.

If the project involves process development/improvement activities on the part of the contract organization, how will any discoveries arising from this work be handled from the standpoint of intellectual property? These negotiations are better handled up-front than after the fact. Will the project require any specialized equipment that must be purchased and installed (and also validated) prior to manufacturing? The latter point is potentially rate limiting with respect to the overall timing of a project. It may be much easier to go to a site that already has the necessary pieces of equipment than to try to install such equipment elsewhere.

The sponsor must ask the contractor how other projects have been handled and how the two groups will work together to develop successful strategies for technology transfer and trouble-shooting between the organizations. Both sides must work to formulate a plan for decision-making during manufacturing. This usually involves having a representative of the sponsor present during the production process (preferably on the manufacturing floor) to help in critical decision-making. The value of this will depend upon the input of the sponsor to the development and scale up of the manufacturing process itself. Therefore even if the contractor develops the entire process, someone associated directly with the sponsor organization should possess detailed knowledge of the technical fine points and of its strengths and weaknesses.

As stated above, once the choice of manufacturers is narrowed down, a thorough audit of the facilities is in order. This will ensure that the chosen sites are in compliance with cGMP regulations. Remember: the sponsor bears ultimate responsibility for the regulatory compliance of the manufacturing environment and procedures, even when it is carried out at a contract site. Therefore, this critical due diligence should not be overlooked.

The Final Choice
After investing all of the work to explore the possibilities, the list of viable candidates will be short. If all of the indicators are positive at this point, the sponsor must get a clearer idea of the cost of the services. This comparison can be quite interesting since, obviously, there is no fixed formula for determining the cost of contract manufacturing and the associated activities. While this is a complex issue, given all of the above considerations, I can attest that cost for similar services can vary widely. It should be pointed out that the cost of goods at the early (and intermediate) stages of development is unlikely to impress the chief financial officer anyway, so cost should not be the driving force in a sponsor’s decision.

The final choice of the appropriate contract site for a project involves a complex interplay between technical capabilities, regulatory compliance, project management and cost. All of these must be factored into the decision of where to manufacture an important biopharmaceutical product at each stage of development. n

1. In 1996, the FDA eliminated the requirement for establishment licensure for certain biologics, such as recombinant DNA-derived therapeutic products, monoclonal antibodies for in vivo use, synthetic therapeutic peptide products of 40 or fewer amino acids, synthetic plasmid nucleic acid therapeutic products and therapeutic DNA plasmid products. The latter are products that can be extensively characterized from a physicochemical point of view. The FDA excluded blood products, gene therapy viral vectors and therapeutic cells and tissues from the list of potentially well-characterized products.

2. N. Motwani, personal communication, presentation at IBC “Antibody Engineering” meeting, San Diego, CA, December, 1999.

biopharmaceutical product

As patent protection expires for the first wave of biopharmaceutical products, the potential marketplace for generic substitutes looms large. Biogenerics offer a multi-billion dollar marketplace, one which has yet to be exploited. I would like to discuss future developments that may lie ahead for biogenerics, as well as a tentative perspective and a (necessarily incomplete) overview of the issues and hurdles to be overcome if the golden opportunities predicted by some observers associated with the future of biogenerics are ever to materialize. I must stress that my perspective is based on a personal interpretation of a very complex situation and that my views do not aim by any means to be comprehensive.

We should start by defining what a biogeneric is. Generics (or, more appropriately, multisource pharmaceuticals) are typically defined as pharmaceutical preparations that:

• are essentially similar to an original product;
• involve an active substance with expired patent protection;
• are approved through a simplified registration process; and
• sell under a common name typically with very little (if any) promotional activities.

Essentially, generic pharmaceutical companies offer consumers a cheaper version of a product for which patent protection has expired. Within this framework, a critical success factor is represented by the ability to develop the generic copy with relatively modest investments, in order to make the economics work.

By extension, the term biogenerics is used to designate pharmaceutical preparations involving a biologically active substance stemming from modern biotech tools. As with pharmaceutical generics, these biogenerics are essentially similar to an original biopharmaceutical whose substance patent has expired, are approved through a simplified abbreviated registration process, and are sold under the generic substance name as opposed to a brand name (e.g., EPO as opposed to Epogen).

Table 1: Top biopharmaceuticals and their biogeneric status
Brand Active Substances Marketer Year of Approval 1999 Sales
in $Millions
Generics Under Development
Epogen Epoetin alfa Amgen Inc. 1989 1760.0 Yes
Procrit Epoetin alfa Ortho Biotech Inc. 1990 1505.0 Yes
Neupogen Filgrastim Amgen Inc. 1991 1260.0 Yes
Humulin 50% human insulin isophane suspension,
50% human insulin (recombinant DNA origin)

Eli Lilly & Co.
1992 1088.0 Yes
Intron A Interferon alfa-2b, recombinant Schering Oncology 1986 650.0 Yes
Avonex Interferon beta-1a Biotech 1996 621.0 Yes
Engerix-B Hepatitis B vaccine, recombinant Biogen Inc. 1989 540.0 Yes
Rebetron Ribavarin and interferon
alfa-2b, recombinant
SmithKline Beecham 1998 530.0 No
Ceredase Alglucerase Schering Corp. 1991 500.0 No
Cerezyme Imiglucerase Genzyme Corp. 1991 479.0 No
Genotropin Somatropin Genzyme Corp. 1995 460.8 Yes
ReoPro Abcicimab Pharmacia Corp. 1994 447.3 No
Betaseron Interferon beta-1b Eli Lilly & Co. and 1993 413.0 Yes
Kogenate Antihemophilic factor, recombinant Centocor 1993 403.4 No
Enbrel Etanercapt Berlex Corp.
1998 366.9 No

Market Drivers
Having defined biogenerics, we should review the reasons behind the interest they have sparked. The biggest driver of this market is money, as should be expected. Presently, the world market for biopharmaceuticals is approximately $20 billion (source: IMS Healthcare). If biogenerics achieve the same 10-15% medium- to long-term penetration rate of the rest of the generic market, that means there’s a potential U.S. biogeneric market of $2 billion.

In addition, during the next five years, several of the top 15 biopharmaceuticals will come off patent, creating opportunities for generic formulators. Most of these biopharmaceuticals are currently under development by generic producers, including EPO, hepatitis B vaccine and colony stimulating factors. Also, some biopharmaceuticals, including insulin and human growth hormones (HGH), have already lost their patent protection—at least for the drug substance—and are therefore open to generic competition.

Not surprisingly, a number of companies are jockeying for position in what is perceived as a multibillion dollar opportunity. Several players are emerging either located outside the Triad (namely western Europe, Japan and North America) or working in partnership with non-Triad companies. These companies include:

LG Chemicals – a Korean biotechnology company producing EPO, insulin and interferons for the domestic market and reported to be cooperating with several Western companies.

GeneMedix – a UK company with strong ties to several institutes in China and expecting to launch its first product this year. Interestingly, GeneMedix has elected to focus on Asia, since the regulatory and patent situations have not been clarified in western countries.

Cangene – a Toronto-based biotech company with a "generic-like" approach. Cangene concentrates on the development of existing molecules but performs limited clinical trials to demonstrate equivalence./

Rhein Biotech – producing biogenerics through partnerships in India and Argentina.

In addition to these a few large generic players have expressed interest in biogenerics, including:

E. Merck – In the late 1980s, this company acquired rights for HGH and interferon from Tecnofarma (Argentina) and has subsequently worked with LG Chemicals for the production of six other proteins. E.Merck Group’s current commitment to biogenerics is unclear.

Teva – Through its alliance with Bio-Technology General in the U.S., this Israeli company plans to market generic HGH, which was a $350 million market in the U.S. in 1999

Stada – This large German generic producer has begun development through its minority-owned subsidiary Stada Biogenerics and is working in collaboration with DSM Biologics in Canada to produce bulk active substance. Main targeted molecules include erythropoietin, filgrastim, interferon alpha and interferon beta.

Yet, despite all this interest, the substantial potential market and the fact that some biopharmaceuticals have seen their substance patent already expire—if one excepts gray market like Latin America or the Far East—a market for biogenerics has yet to emerge. The originators of biopharmaceuticals appear to have been remarkably successful in preserving their franchises. We must ask why this is and whether the situation will change. Will the promises of substantial growth and financially rewarding opportunities associated with biogenerics ever emerge? To answer this question, we must examine the conditions typically required for generics to succeed, and see how they apply to this unique market.

At least on paper, several biopharmaceuticals coming off patent represent ideal candidates meeting most of the criteria for the successful development of generics.

These drugs satisfy the large market size qualification. The first biotech products expected to come off patent comprise several blockbusters still showing double digit growth. The top example of this is EPO, a molecule with sales in excess of $3 billion, and interferon alpha, which posts more than $600 million in sales.

In addition, they offer large profit margins, with relatively low impact of the cost of the bulk biopharmaceutical on the overall price. This would allow biogeneric producers to offer discounts compared to the nongeneric versions that are currently available.

These potential biogenerics also tend to have simple non-proprietary formulation systems. Most of the molecules are straight injectable forms such as solutions or lyophilizates.

However there are key issues are associated with biogenerics that differ from regular generics. The most intricate obstacle is the availability of bulk active biopharmaceuticals through non-patent-infringing routes. Patent coverage for biopharmaceuticals represents a thorny path for biogenerics. Also, affordable demonstration of bioequivalence has held back development of this market.

The most critical divergence from the pharmaceutical generic market is the lack of a clear regulatory framework. Without this element, it is impossible to envision an economically viable development of a generic product. Biogeneric developers need regulatory standards so that they may avoid undertaking massive investments to develop the generic file.

As of today no specific regulations exist for biogenerics. Traditionally, biologicals have been considered by regulatory authorities as a distinct category from synthesized drugs. This has resulted in a situation of regulatory near-vacuum (or at least vagueness).

For example, in Europe it is not clear whether a biogeneric should be registered through the abbreviated procedure applicable for well-defined traditional generics requiring "that the generic is essentially similar to a reference product; if not, the generic should undergo a full registration process logged with the European Medicines Agency" (EMEA Committee for proprietary medicinal products- Note for guidance on development pharmaceuticals January 1998, EU Directive 65/65/EEC Article 4.8, medicinal product marketing license directives). This is not an unlikely or impossible proposition, as this procedure is theoretically mandatory for all "high-tech products," including biopharmaceuticals.

The regulatory situation for biogenerics is even more complex in the U.S., where the regulatory paths for biologicals and chemicals differ. Biologicals are approved by the CBER (Center for Biologics Evaluation and Research) under the Public Health Act, while conventional drugs are regulated by the U.S. Food, Drug and Cosmetic Acts and are evaluated by the CDER (Center for Drug Evaluation and Research).

The schism is not absolute, we must note. Complicating the classical distinction between biologicals and small molecule pharmaceuticals is the fact that some well-characterized recombinant proteins (such as growth hormone and insulin) have been registered as pharmaceutical drugs through the CDER.

This has far-reaching implications, for biologicals approved under the CBER are specifically excluded from the generic abbreviated approval process that is applied for synthetics. The rationale for this is the viewpoint that current bioanalytical methods are not adequate to assess biopharmaceutical equivalence. However "biologics" approved under the CDER are theoretically subject to generic competition, an element of critical importance.

Even if biogenerics were made eligible for the abbreviated procedure, another hurdle immediately emerges: the need to demonstrate "essential similarity" to a reference product. Indeed, unlike low molecular weight synthetic chemicals, where equivalence can be easily demonstrated through full analytical characterization, biopharmaceuticals most often consist of complex substances that are difficult, if not impossible, to fully characterize from a physico-chemical perspective, given limitations in current analytical techniques.

Furthermore there is the widespread view that, for biopharmaceuticals, "the process makes the product." Even minor modifications in a bioprocess such as changes in agitation or aeration systems, reactor size, operating conditions or culture media, not to mention changes in the cell line or microbial system applied are viewed as potentially leading to possible variations in the quality or properties of the biopharmaceutical associated as an example with different postranslational modification patterns (such as glycosylation), variation in tridimensional structures or altered impurity profiles.

Such changes (at least as seen by regulators who are vested with the task of overseeing the safety of pharmaceuticals) could potentially result in an altered safety and efficacy profile. This would require extensive (and expensive) clinical trials to gain regulatory approval. The authorities’ perspective has been—up until now—that simple bio-equivalence studies are not enough for biogenerics.

In fact, this issue is not specific to biogenerics. Marketers en-gaged in the development of innovative biopharmaceuticals currently face the same hurdles when changes are brought to the production process during development or after market authorization.

Getting Over the Process
However, this situation is evolving. The dogma that the "process makes the product" has been increasingly challenged. It appears that—if not strictly identical—at least comparable biologics can be obtained through different processes. This is the case for most well-characterized proteins. In Table 2, we see how HGH is being produced in Europe in a variety of expression systems (standard E. coli, special strains of E. coli and transformed mouse cell lines), with identical results, a 191-amino acid sequence copying the human pituitary growth hormone. Thus, there are at least five HGH products currently on the European market, each obtained by different companies based on distinct processes and even involving different expression systems. All of these products appear to show the same profiles in terms of amino acid sequence, potency and safety and efficacy, undermining the belief that "the process makes the product."

Producer Expression System End Products
Pharmacia Standard 191 amino acid
Ferring E. coli sequence
“identical” to the
Lilly Special strains human
Novo of E. coli pituitary growth
Serono Transformed mouse
cell line

This is leading to a consensus that the concept of "product comparability" is more appropriate for bioproducts than "essential similarity." The regulatory authorities in both Europe and the U.S. have become increasingly aware of the almost Kafkaesque situation associated with this regulatory vacuum surrounding the comparability of complex molecules (including biopharmaceuticals). In Europe, the CPMP (Committee for Proprietary Medicinal Products) has started to develop specific guidelines within the frame of an initiative launched in 1998 by the Biotechnology Working Party.

These guidelines are designed to address the issue of comparability for what is referred to as "well characterized biotech-derived products, namely recombinant proteins and peptides" (Note for guidance on comparability of medicinal products containing biotechnology derived proteins as active substance EMEA May 25th, 2000). The guidelines include the physico-chemical and biological tests required to demonstrate structural equivalence of the two products, the assessment of the potential impact of process changes on the quality of the products, and the comparability in terms of toxicology and clinical efficacy. Interestingly, these guidelines were developed to facilitate changes in the processes applied by originators, but they have been extended to compare recombinant proteins developed by different manufactures such as biogenerics and multisource biopharmaceuticals.

Does this means that the regulatory hurdles facing biogenerics are all cleared and solved? Not really.

Although this step potentially opens the door for biogenerics, we must recognize that, as presently written, the proposed draft leaves substantial room for interpretation. No universally applicable rules are being proposed, so each product must be reviewed on a case by case basis. Also in the proposed regulation, relatively vague words and statements are used, with rather strict conditions proposed to demonstrate ‘comparability.’ This continues to favor the originators rather than potential biogeneric producers. Still, it’s a start.

Patent Wars
The complex patent environment surrounding biologicals represents another barrier to the development of biogenerics. The unique aspects of biotechnology patent law and its complexities certainly represent a fertile ground for biopharmaceutical innovators to hinder the development of biogenerics, providing a nightmare for potential generic producers. Issues include the overlap of process and product patents, in addition to such matters as exclusivity period and orphan drug status. A salient example of this complexity is the fierce battle raging between Amgen and Transkaryotic Therapeutics surrounding EPO.

As if this would not be enough, in the U.S.there is no Bolar-Roche (or Waxman-Hatch) provision that applies to biopharmaceuticals that seems to apply to biopharmaceuticals (at least to those registered under the frame of the CBER procedure). This represents an additional hurdle for the development of biogenerics whilst providing an additional period of exclusivity for the originators, as it means that the generic company would not be able to start any substantial development work on the biopharmaceutical before the effective patent expiry date, a situation in sharp contrast to what is prevailing for traditional pharmaceuticals based on small molecules.

The Future of Biogenerics
Is the situation completely doomed for biogenerics? Should generic marketers shelve and forget forever hopes of ever capitalizing on this potential multi-billion dollar opportunity? The answer to this question must be nuanced and hinges on a variety of factors.

First, the biogeneric market requires the development of a favorable regulatory and patent framework. This situation, although evolving, has not yet materialized.

The availability of bulk material and bulk biogenerics suppliers is also a must for the development of a biogeneric market. In this respect several reliable contract biopharmaceutical producers are starting to emerge (Boehringer-Ingelheim, Cambrex-BSCP, Diosynth-Covance, DSM-Biologics, Lonza). We should note, however, that their focus has been mainly on exclusive custom manufacturing of new biologicals under development by biotech companies or pharmaceutical industry majors active in the development of new biopharmaceuticals. At this stage it is far from clear what strategies contract biopharmaceutical producers will follow in respect to biogenerics. Will they eventually move into this market , risking possibly alienating their present customer base or will they instead stick to the innovator biopharmaceutical market? Given the current capacity shortage having emerged particularly for mammalian cell culture it is quite likely that most contract biopharmaceutical producers will elect the latter strategy, creating thereby an additional hurdle for biogenerics as it will be difficult to have access to the bulk substance.

The receptivity of the customer base, including patients and health care providers, to generic formulations also represents a potential barrier. Most of the biopharmaceuticals on the market target serious diseases in which the risk of switching patients to new formulations could be viewed by health providers as unwarranted and dangerous, given fears about biogeneric quality.

Another factor is the policies and approaches of present-day pharmaceutical generics suppliers. Many are still waiting on the sidelines, hoping the biogenerics situation will become clear. Undoubtedly, once the first biogeneric hits the market, many companies will step up their efforts in this field.

In addition, advances or breakthroughs in analytical technologies potentially have far-reaching implications. New technologies in this field may allow a more accurate characterization of macromolecules, opening new frontiers in terms of demonstration of equivalence or even essential similarity.

The Empire Strikes Back
We must also take into account the strategies followed by biopharmaceutical innovators, the originators of these products. They are expected to fiercely defend their franchise, engaging in a wide array of tactics, including proactive efforts to influence and shape the regulatory framework and stress to the medical community the value and safety of the original product, compared to biogenerics. Innovators stand the most to gain from the belief that "the process makes the product," and that even small process changes can impact the product equivalence and hence its safety and efficacy profile

Innovators are accustomed to the methods of blocking biogenerics, having already used these techniques to hinder traditional generics. These potential obstacles include:

• Adopting an aggressive legal stance and suing any patent infringement: examples include Amgen defending EPO’s position against Genetic Institutes and now against Transkaryotic Therapies;

• Repositioning the molecule through new enhanced formulations offering tangible benefits in terms of patient compliance, sustained release forms, lower adverse effects. Novartis did this with Neoral, its blockbuster transplantation drug;

• Accelerating the introduction of a second generation product, and possibly phasing out the original molecule: see Amgen’s strategy with NESP, the novel erythropoiesis stimulating protein currently under registration.

So, will the promise of major opportunities associated with biogenerics ever materialize? Undoubtedly, major hurdles are still hindering the development of biogenerics, but the overall field is rapidly evolving. Developments are not always predictable. Breakthroughs in the regulatory environment must not be ruled out. This may well have a catalytical effect on the onset of biogenerics. Perspective participants, eager to capitalize on this potential opportunity, must carefully monitor the field, keeping in mind that timing is of the essence.

Within this frame it is important to note that whilst it is beyond doubts that one day biogenerics will become a reality not only in gray markets but also in the U.S., Europe and Japan, a major uncertainty is represented by the precise timing of such developments. Given the critical impact of time on the ultimate financial return of investments, it is probably wise for perspective players on the biogeneric scene to carefully hedge their bets.

Manufacturing and laboratory efficiency

Manufacturing and laboratory efficiency issues have direct effects on cash flow, balance sheet, product quality and customer satisfaction. Performance behind the competition—or ahead of it—can dramatically affect shareholder value. Therefore, in the face of ever-challenging profitability and revenue goals, pharmaceutical companies and contract manufacturers alike are striving to improve and enhance their manufacturing and laboratory efficiency. A solid understanding of manufacturing costs and performance metrics among world-class companies is the first step in evaluating a company’s own current practices.

Best practice benchmarking is an ideal tool for such analysis. By engaging executives at comparable companies in a detailed survey, industry leaders can be identified; and by conducting interviews with executives at those companies identified as “best-in-class,” insights into best practices, performance gaps and opportunities for meaningful improvement can be identified.

The Best Practices LLC research team recently conducted a study of pharmaceutical and contract manufacturers dealing with a large number of products, but their production is primarily in small lots. This manufacturing and planning challenge is of key interest to contract manufacturers dealing with similar issues.

This particular engagement involved a survey of manufacturing executives at FDA-regulated plants in North America with small lot sizes, considerable lot variety, no API production, and no biological production. Of the participating facilities:

Seven of the 11 plants in the benchmark class manufacture solid dosage forms only;
Plants A and C manufacture both solid and injectable dosage forms [they are designated by a single asterisk (*) in all subsequent graphs];
Plants H and K manufacture injectable dosage forms only [these are designated with double asterisks (**) in all graphs]; and
Plants B, C and K are contract manufacturers [and are designated with a plus (+) in all graphs].

The participating companies were

• aaiPharma,
• AstraZeneca,
• Aventis,
• Bayer,
• DSM,
• Lilly,
• Patheon,
• Pfizer and
• Wyeth.

Participants in the survey were executives at manufacturing facilities highly regarded for their cost and performance management. The survey was designed to identify performance gaps, averages, and best-in-class measures. This gives executives a sense of magnitude concerning organizational differences, as well as identifying potential areas for needed improvement.

Product Breakdown
We asked participating plants to provide the following background metrics regarding their facilities:

Total number of different products (based on formulations, sizes, and doses) produced:
Solid dosage forms (before filling/packaging)
Injectables (filled vials before labeling)
Liquids and others (before filling/packaging)

For most participating facilities, focusing on one type of manufacturing process, or
a limited line of products, yields significant efficiencies, economies of scale, and a
fine-tuned operational excellence (see Figure 1). Such focus generally results in better facility performance as measured by other parameters in this study. Product diversity requires larger facility footprints (leading to more capital costs); potentially underutilized equipment and systems; diluted personnel expertise (such as maintenance, quality, testing, etc.); and management focus.

Figure 1
Figure 2

Converison Costs
We also looked into facility conversion costs as compared to total product sales attributable to the participating facilities. In this regard, conversion costs included those directly related to converting APIs to finished form, including employee-related costs (salaries, wages, benefits), depreciation (equipment costs) and other direct costs (including consumables, utilities, etc.).

Figure 2 explores the ratio between “Conversion Costs” and “Total Sales” (x 100%). Overall, higher conversion ratios are due to product complexity, capacity utilization, staffing decisions, degree of automation, and other factors. These responses demonstrate that contract manufacturers (C and B)—and those facilities that do some contract manufacturing or zero-profit product transfers—typically have considerably higher ratios than do more conventional facilities.

In understanding the conversion cost ratio, we must recognize the significant bearing market forces can have on such a calculation. Product pricing, number of competing products, market share, product availability, product types (animal health vs. human health, etc.) and other factors can greatly influence the ratio’s denominator (product sales), and therefore a facility’s relative standing. The higher ratio of conversion costs to total sales of other companies without such pricing issues are usually due to staffing issues, production productivity, degree of automation, newer equipment, or excess capacity.

We asked participants to share their headcount in areas related to Conversion Costs, including

Maintenance, engineering, utilities
Quality control (including chemical testing, inspection, microbiology, environmental monitoring and stability testing for marketed products)
Quality assurance/compliance (including documentation)

Figure 3
Figure 4

Figure 3 looks at relative headcount by function. Relative production headcount clearly is affected by the number of shifts in operation. Greater utilization via shift work reduces the conversion costs as a percent of total sales. For companies with similar shift strategies (as indicated on the capacity utilization slide), relative production headcount should be similar. Levels of automation also have a bearing on production staffing compared to the benchmark class.

QA and QC staffing levels are generally not dependent on the number of shifts run at facilities. Low staffing in these areas is often indicative of future quality problems and should be closely evaluated and acted upon by plant management.

Maintenance, engineering and utility (MEU) staffing typically is not dependent on the number of shifts employed at a facility, since most of these functions are staffed around the clock. High staffing levels in MEU usually increase the facility’s maintenance costs as a percent of replacement value (see Figure 7). Equipment reliability and on-site utility facilities can have significant bearing on such performance.

Validation figures reflect current efforts on 21 CFR 11 initiatives. A low percentage of people on validation issues generally indicates one of two scenarios. Either the facility could be ahead of the curve, with much of its validation efforts behind it; or it could be behind the curve, with more validation activity ahead of it. The number and type of systems to validate also have a bearing on such performance. Generally speaking, facilities on the far end of any measure warrant more detailed analysis to identify specific improvement and cost containment opportunities.

One such area for focus deals with the span of control, or how many workers are typically assigned to an individual manager or supervisor (on average).

Most facilities (9 of 11) reported very few directors and above at their plants compared to their relative number of supervisors or line workers. Contract manufacturers B, C and K, however, did report a noteworthy percentage of supervisors relative to production workers and presents an improvement opportunity for the industry. Interestingly, most of the facilities producing injectables reported a higher percentage of supervisors/higher than did the other benchmark partners. This is due, in part, to production complexity.

Facility span of control, defined as the number of workers assigned to managers (on average), was generally 10:1, which is consistent with other staffing studies performed by Best Practices LLC.

Figure 5
Figure 6

Capacity Utilization
We also asked participating plants to provide metrics regarding capacity utilization levels for areas producing Solid Dosage Forms (bulk product), Solid Dosage Forms (packaging), and Injectables (filling and packaging). Capacity utilization figures are based on a one-shift operation (average numbers in case of parallel equipment or lines). Those facilities with more than one shift are expected to have values in excess of 100% or 200%, as applicable (see Figure 5).

For solid dosage manufacturing, five plants in the benchmark class reported a significantly greater capacity utilization than did other participants. As expected, those plants reporting a greater capacity utilization a relatively low amount of production overtime expenditures.

Those facilities with comparatively low utilization have significant revenue-enhancing opportunities by consolidating company-wide production (bring in more products from other company facilities), performing additional contract manufacturing, or selling unneeded capacity (see Figure 6).

Participating facilities dedicated to a specific task usually enjoy a higher capacity utilization. For example, Company H’s facility is limited to injectable manufacturing, so its equipment and production commitments are focused to the task. Contract manufacturers (such as C and K) generally reported lower utilization, but this is typically driven by market forces rather than production train reliability, maintenance, or other issues.

Figure 7

Maintenance & Automation
We asked participating plants to provide the ratio between maintenance cost and replacement value (x 100%). Best-in-class maintenance cost performance (exhibited by a low ratio to replacement value) is due to very low maintenance costs, comparatively higher replacement values, or a combination of both (see Figure 7). Depreciation (a logical proxy for capital equipment and facility expenditures) is the biggest driver in such a calculation, with a low ratio (good performance) significantly impacted by high depreciation. For those companies with comparable depreciation figures, relative performance on the maintenance cost to replacement value ratio are more illustrative—especially with respect to cost controls. Large cost drivers in this area include MEU personnel, equipment reliability, equipment age, degree of preventive maintenance performed, consumable utilization, and corrective maintenance effectiveness.

Key Findings
In general, significant differences exist between contract manufacturing facility performance and that of their more conventional brethren. The following key findings emerged from the research:

Capacity utilization is a key driver in cost management and production efficiency. Those facilities getting the most out of their equipment tend to perform better in most categories measured in this study. Contract manufacturers tended to have low utilizations. Forecasting, scheduling, production design, and current business levels all have a significant impact and must be addressed in getting the most out of production facilities
Too much variety in product types can lead to under-utilized equipment and diluted management, maintenance and quality focus. Contract manufacturers that focus production facilities on one product form generally enjoy better utilization, lower costs and greater profitability.
Contract manufacturers often have lower product margins and resulting higher conversion costs as a percentage of total sales. Facilities that do some contract manufacturing, or production of product for transfer overseas with no profit recognition, experience similar conversion cost to sales ratios. Equipment utilization is a key issue for such facilities, with over-capacity built in to maintain a high degree of production flexibility, but with understandable cost penalties.
Most survey participants reported only average degrees of automation in their facilities. Regulatory compliance pressures are typically blamed for slow adoption. Those facilities that aggressively embrace automation, however, enjoy substantial benefits. One company with a high degree of automation reported considerably better maintenance cost, headcount and overtime performance when compared to other benchmark partners.

Best practice benchmarking is an excellent tool for facility and corporate management to compare structural organizations, staffing levels, roles and responsibilities, and performance to critical measures. While benchmarking metrics illuminate the magnitude of performance gaps, more in-depth best practice surveys and interviews should be employed to identify specific best practices, lessons learned, mutual challenges, and prioritized improvement opportunities based on available resources and company culture.

pharmaceutical labelling

In 2003, the FDA announced a mass recall of the cholesterol-lowering drug Lipitor. The recall, triggered
by the discovery of counterfeits, affected more than 130,000 bottles.

While drug counterfeiting in the U.S. is still rare, the number of investigations are on the rise. According to the FDA, the number of counterfeit drug investigations has risen from an average of five per year in the 1990s to more than 20 per year since 2000.

Growing use of the Internet to purchase prescription drugs also complicates matters. “The biggest threat now to every pharmaceutical manufacturer is online buying of prescription products. How do you control that, make sure that the authentic product is being purchased, plus at the correct price?” asked Neil Sellars, director of product development and marketing for National Label Company in Lafayette Hill, PA. “Online purchasing brings everything to a whole new threshold.”

“Counterfeiting is a huge problem for the pharmaceutical industry as it relates to direct liability. The drug companies annually lose billions of dollars to very clever counterfeiters. From a cost standpoint, if they could reduce the amount of counterfeit drugs in the marketplace, that will help them with their bottom line,” said Bob Piefke, business development manager for Appleton in Dayton, OH.

Widely publicized cases such as the U.S. Lipitor recall of 2003 have catapulted the issue of counterfeited drugs into public view. All this attention has trickled down into the label industry, where many industry pundits report a rise in interest of tamper-evident and security packaging.

Return to top

Track and Trace
There’s no doubt security measures are being talked about heavily within the pharmaceutical supply chain. According to a report released in February 2004 by the FDA, the Agency is recommending a multi-pronged approach to combat Pharma counterfeiting.

Among the suggestions, the report states that using one or more authentication technologies such as color shifting inks, holograms and taggants is an important anti-counterfeiting tool. Furthermore, it reports that radio frequency identification (RFID) “tagging of products by manufacturers, wholesalers and retailers appears to be the most promising approach to reliable product tracking and tracing.”

RFID, a technology that allows for multiple chips to be scanned simultaneously, has made headlines for its usefulness in the logistics industry. But it is also gaining notoriety for the role it can play in the pharmaceutical supply chain. That is, tracing pallets or individual pharmaceuticals to make sure they come from—and stay with—legitimate sources.

While track-and-trace is a popular catchphrase in RFID circles, some question whether or not it is feasible today. “The infrastructure globally doesn’t exist for them to take advantage of the track and trace benefits of this technology, which is really what the FDA is highlighting,” said Sellars.

“It’s really not feasible because you have to have the infrastructure around the globe to read the technology and download it. Right now, as you prepare the product you can take that information and uplink it, but uplink it to where? Who’s going to take that information? How’s it going to go globally?” asked Sellars.

Despite the barriers, Sellars and others say RFID will have a growing presence within Pharma. “At the present time, RFID is in its infancy in the pharmaceutical industry,” says Piefke. “I would say that probably in the next two years it could have a fairly major effect on the direction that the industry goes regarding variable information labeling, and certainly in the long term it will have a major effect.”

Other Security Features
Within the industry, some pharmaceutical companies are choosing to adopt a security triangle, or a three-tiered approach to security measures in packaging.

Level one includes overt features such as holograms. “It’s still probably the most important [level] because many of the pharmaceutical products that have been counterfeited were caught by the consumer, so you have to communicate with the consumer,” said Sellars.

Levels two and three utilize covert features. For example, Sellars said, level two measures might include products also seen on currency—pen-reactive inks, for instance. And level three would employ forensic technology.

“There’s a ton of different types of technologies out there. The most common are IR taggants that are added to the ink and they basically are tuned to a specific reader that a manufacturer would own.”

So what does the spotlight on counterfeiting, brand protection and security features mean for the label converter? At the moment, some would argue not very much. One issue is cost. Talk is cheap. Packaging security features and RFID implementation is not.

“We’ve looked at other technologies, taggants and that sort of thing. We’ve offered it to some of our clients, and most of the technologies we’ve offered are still under debate,” said Andrew Vale, East Coast sales manager for Ampersand Label, headquartered in Garden Grove, CA.

“Security measures are of increased importance, but as far as getting involved with the dollars and cents, it is one of those things where [our clients] would like to adopt the technology, but they really don’t want to pay for it. At this point, the cost is still enough of a roadblock to stop the technology from being adopted,” Vale added. “There’s a lot of talk about counterfeiting, but no real action.”

Others are more hopeful about the growth of security features. St. Paul, MN-based label converter Tursso Companies has placed high confidence on the growth of security features within pharmaceuticals. The company has recently focused in on three features: taggant encoded inks, label materials with fluorescent fibers, and optical watermarks.

So far, the response has been encouraging. “We’ll be gathering our market data over the next two or three months to find out what the receptivity is, but I think it will be strong,” said David Gray, vice president of sales and marketing for the company.

Return to top

IIf RFID and complex features are out of the price range of some pharmaceutical customers, less expensive alternatives that add security to the pharmaceutical supply chain are appearing as well. Verify Brand is a relatively new company that works with pharmaceutical companies and their label vendors. The label printer would generally print a code on the pharmaceutical label, along with a URL or phone number. Consumers and suppliers can use that contact information to verify that the product is genuine.

“We don’t cause the manufacturers to change their production or distribution processes. They just continue to order labels the way they do today. Once we find a place to put a code on the label, it causes very little disruption, and very little, if any, cost or equipment to the [label] manufacturer,” said Kevin Erdman, president of the Minneapolis, MN-based company.

When it comes to anti-counterfeiting and security initiatives, cost is not the only barrier. Pharmaceutical manufacturers are wary of another problem related to counterfeiting: the ease of duplication.

“A major issue today is the broad range of ‘solutions’ being offered to the pharmaceutical companies and the converters. It has become increasingly difficult for the brand owner and the converter to differentiate between truly effective solutions, and those that are easily counterfeited or imitated,” said John Keane, vice president of sales and marketing for Kurz Transfer Products, in Charlotte, NC.

Outside pressures are not only fueling interest in anti-counterfeiting measures. They have also affected the amount of content found on the label, leading to a rise in constructions that accommodate extended text.

“You need extended text now because of the drug fact requirements, minimum type sizes and the amount of content that needs to be on the label. You need more real estate on the label these days for any type of drug,” said Shev Okumus, president of Star Label Products, Fairless Hills, PA.

Photo courtesy of Ampersand Label

“Expanded content labels are growing in the pharmaceutical industry, since they are available in so many different formats and sizes,” said Rob Ryckman, director of marketing and product development, RFID and security products for CCL Label located in Hightstown, NJ. “It is the only way that the vast amount of required information can be applied directly to the item in multiple languages.”

Also driving a trend toward extended text is the convenience of a booklet label. “It makes it a more self-contained package. Historically, there have been inserts, on-serts, top-serts and products that get affixed to the top of the package or somewhere in the carton. If you can eliminate the carton with an insert and have a label with a booklet affixed, you’re saving a significant amount of money,” said Mr. Gray of Tursso Companies.

While booklet labels can be a money-saver if it eliminates the need for cartons and instruction sheets, they are more expensive than using a regular label. Fortunately, there are other ways to get more information without ordering a booklet label.

“We’re trying to come up with different ways, even using paper with different release coatings, things like that, without having to make a multi-layered label. We’re trying to keep the cost down for some of our generic brand customers,” said Mr. Okumus.

Using specialty release coatings allow a converter to print on both sides of a label. The ability to print on both sides creates more room for text without adding another layer of material. “Through the printer’s sophistication, the release coating can be spot applied or pattern printed to some extent. They would be able to get a particular label to come off completely, or it could be part of the label design, where it can be pulled back and reapplied,” said John Donaleski, senior chemist for Craig Adhesives and Coatings located in Newark, NJ.

Return to top

Paperless Labels, White Labels
While currently there is a trend toward extended text labels, there is also a move in the industry for the opposite. The Pharmaceutical Research and Manufacturers of America (PhRMA) will soon be launching a trial of its paperless labeling initiative. After successfully testing it at 10 pharmacies, PhRMA now plans to bring the trial to 265 pharmacies for a larger test.

In this system, the drug information “would be available on the Internet. And you can access a web site and download or print out the information,” said Mr. Gray. The information would be in portable document format (PDF). Although it is only in testing stages, and there are definite opponents to the initiative, the association has publicly stated that full electronic dissemination of labeling information is the ultimate goal.

Is this initiative a concern for label converters? “It’s a concern, but I don’t see it as an immediate concern,” said Mr. Gray. “People aren’t ready yet in the 45-and-up age group to get their drug information from a web site . . . I still believe there’s a generational shift that needs to occur before it’s more readily available.”

And even if the paperless labeling initiative came to pass on a wide scale, labels won’t be eliminated all together. “There may be labels with less information, but in no case would I conceive of bottles without some identification information on them,” said Lowell Matthews, chief executive officer of Ampersand Label.
Internationally, another trend has been noted, coined as the “white label initiative”.

“It’s very popular in Europe because they run with smaller SKUs and a great number of languages,” said Mr. Sellars. The white label initiative is not a mandate, but a trend. Pharmaceutical companies purchase blank labels and print the labels themselves using a digital printer. Or they use a generic preprinted label and have a variable information section that is printed.

The chance that this new trend will reach the U.S. is unlikely. “A lot of our customers have looked at it, but it’s just not feasible right now,” said Mr. Sellars. “Digital serves two great markets: on demand and variable information. If you don’t have either one of those requirements, then it is an expensive luxury.”

Return to top

Special Considerations
As with any market, trends come and go. But one thing that remains a constant within the pharmaceutical industry is the need for quality. If a label is not manufactured correctly, it could mean huge implications in the form of human and financial losses. Consequently, the converters who print for pharmaceuticals must adhere to strict guidelines.

The procedure to get a pharmaceutical account can be laborious as compared with other label markets. “With the Pharma industry, there’s a whole process that has to be done in order to get that client’s business, a whole number of steps starting with them auditing your facility and setting up procedures with you,” said Mr. Matthews.

Once the business is there, it’s a whole different process to keep them. Attention to detail is critical. Many converters offer clean rooms and double inspections. Some must destroy their trash before it goes outside.

“We set up our new building so that there is an indoor trash compactor. When you get rid of their waste, their labels, if you have a dumpster outside, it means anyone can get in there and pull out labels. We have an indoor trash compactor that all of our trash goes into, but it was put there because of our pharmaceutical label customers,” says Okumus.

Another way converters are helping their customers is by offering to consecutively number the label liner. Pharmaceuti-cal customers must account for the number of labels they have in their facilities at all times. If the Pharma company only uses half a roll of labels, the consecutive numbers save employees from hand counting the leftovers.

While there are certain challenges associated with servicing pharmaceutical clients, some things are the same across all label markets. “At the heart of all it’s a printing business. The three key factors will never change: price, quality and service. All of the innovations are a natural evolution, but at the heart of it is the basics,” said Mr. Gray.

Pharmaceutical Sterility Testing

Essential things to know

By Steven Richter

Sterility testing of pharmaceutical articles is required during the sterilization validation process as well as for routine release testing. USP1 requirements employ sterility testing as an official test to determine suitability of a lot. An understanding of sterility testing is beneficial in terms of designing a validation process. The need to provide adequate and reliable sterility test data is an important quality assurance issue. Sterility testing is a very tedious and artful process that must be performed by trained and qualified laboratory personnel. The investigation of sterility test failures is a process that requires attention to environmental data as well as many other factors including training and sample difficulty.

This paper presents the general concepts and problems associated with sterility testing as well as the various testing methodologies. Most USP <71> sections are harmonized with the EP/JP.

Sterility testing is an essential part of every sterilization validation. Sterility testing is an extremely difficult process that must be designed and executed so as to eliminate false positive results. False positive results are generally due to laboratory contamination from the testing environment or technician error. The testing environment must be designed to meet the requirements of the United States Pharmacopeia (USP) in terms of viable microbial air and surface counts. Growth media used in sterility testing must be meticulously prepared and tested to ensure its ability to support microbial growth. Procedures for sampling, testing, and follow-up must be defined in the validation procedures.

Sampling Plans

The official test, the USP (Volume 30) recommends testing 40 units per production lot. A reprint of Table 2 "Minimum Quantity to be Used for Each Medium2" is on the next page. Some of the quantities are not harmonized with the EP/JP volumes.3

For combination products, the ISO 11137/111354 standards recommend various sterilization validation sampling plans based on lot size and validation method. In cases where small lots (>1000) are manufactured, the sampling size depends on lot size.

Environmental Concerns Related to Sterility Testing

The sterility test environment is described in USP General Informational Chapter <1211>. The environment should be as stringently controlled as an aseptic processing environment. An aseptic processing environment (clean room) is used to dispense sterile pharmaceuticals into presterilized containers. A clean room is generally a room that delivers laminar flow air which has been filtered through microbial retentive High Efficiency Particulate Air (HEPA) filters. The room is maintained under positive pressure and has specifications for room air changes per hour. An environment used for sterility testing should be similar in design to an aseptic processing environment; there should be an anteroom for gowning and a separate area for the actual sterility testing. The testing area should meet ISO Class 5 particulate control requirements (specified in USP chapter (1116)). Sterility testing should not be carried out under a laminar flow hood located within a room that is not maintained as ISO Class 5. Along with particulate testing in the environment, the laboratory must test for viable bacterial and fungal organisms ubiquitous to it. The sterility test technician must be suitably gowned in sterile garments that prevent microbial shedding into the room. The room should be validated in terms of particulate and microbial levels. The laboratory must have a validation and training program for gowning and sterility testing.

Our validation programs require that technicians consecutively test 40 simulated samples for both membrane filtration and direct immersion methods without a false positive test result under less than ideal environmental conditions. Isolator technology is utilized to create a sterile environment for one to test pharmaceutical articles. The validation required to qualify an isolator is extensive. The isolators are generally sterilized using chemical sterilization.

Many issues surround the robustness of the sterilization process. Qualifying and maintaining an isolator system for sterility testing may require extensive work. In testing pharmaceutical articles in a closed system such as SteritestTM, an isolator may not be the best cost approach to the environmental concerns. Most environmental concerns can be obviated by standard aseptic processing GMP's.5


The United States Pharmacopeia is a compilation of validated methods and official monographs for pharmaceuticals and medical devices. IT is broken down into the following sections: Monographs, General Informational Chapters, and General Requirements. General Informational Chapters <1000> series are not legal requirements. The Sterility Test (USP Section <71>) is categorized under General Requirements and is therefore a legal requirement.

For combination products, the ISO radiation sterilization microbial methods (11737-2 1998)6 describes a sterility test which is a modification for the USP method. This test is specific for the detection of aerobic organisms that have been exposed to sub-lethal sterilization cycles. This ISO sterility test method is recommended for the validation of both gamma and electron beam sterilization processes.

The method of choice for EO7 sterilized products is the official USP <71> procedure.


Prior to actual sterility testing, it is prudent to send an example sample to the testing laboratory so the laboratory can determine the appropriate testing procedure. Each product should have a unique procedural specification for testing. The procedure should be very specific in terms of which items (or vials/syringes) to test. The procedure must indicate the Sample Item Portion (SIP). The Sample Item Portion is the percentage of the complete product tested. Since medical devices come in all shapes and sizes, it is very difficult to test large and cumbersome medical devices in their entirety. Therefore, the test laboratory will determine a Sample Item Portion which is a portion of the sample expressed in fractional terms (i.e. 0.1 for 10% of the sample).

This number is used in gamma and electron beam dose setting methods. The SIP portion should be validated by sterility testing.

Combination products have unique challenges. A combination product is defined as one that has a drug component with medical device. For example, a drug coated stent. The agency's Office of Combination Products (OCP) would determine which regulatory branch (CDRH, CDER or CBER) is officiating the product. Official USP sterility testing of combination products is required for all sterile drug products. The drug product component applied aseptically creates the largest challenge to laboratory personnel. Biologics must be aseptically processed and cannot be terminally sterilized. In the near future, we will see more biologics that are combination products. Combination products sterilized by radiation are generally handled as medical devices following the ISO 11137 standard. For the most part, pharmaceutical GMPs would take precedent over 820 QSR8 requirements with all combination products. The more robust GMP9 requirement would assure reduced bioburden counts and consistent microbial populations during manufacturing.

The USP <71> Sterility Test contains two qualifying assays which must be performed prior to sterility testing. They are the "Suitability Test" (Growth Promotion Test) and the "Validation Test" (Bacteriostasis and Fungistasis Test).

The Suitability Test is used to confirm that each lot of growth media used in the sterility test procedure will support the growth of fewer than 100 viable microorganisms. If the media cannot support the growth of the indicator organisms, then the test fails. Secondly, a portion of each media lot must be incubated and assessed for sterility according to the incubation parameters (time, temperature) established by the method. If the media is found to be non-sterile, then the test fails.

The Validation Test is used to determine if the test sample will inhibit the growth of microorganisms in the test media. Stasis, in terms of microbiology, is defined as the inability of a microorganism to grow and proliferate in microbiological media. Media that is bacteriostatic does not necessarily kill bacteria; it simply may retard bacterial growth and proliferation. The Validation Test must be performed on each product prior to and/or during sterility testing. This test determines if the media volumes are valid for the particular product. Some medical products contain bacteriostatic and fungistatic compounds that may require special procedures and special media for testing. This test is similar to the Suitability Test described above, however, the product sample is placed in the media along with the microorganisms. Microbial growth in the presence of the test samples is compared to controls without test samples. If microbial growth is present in the sample and control containers, then the test is valid. The next step is to proceed to actual sterility testing. Suitability, validation and sterility tests can be performed simultaneously.

The USP describes three general methods for sterility testing: 1) Membrane Filtration, 2) Direct Transfer (Product Immersion); and 3) Product Flush.

Membrane Filtration Sterility Testing

The Membrane Filtration Sterility Test is the method of choice for pharmaceutical products. It is not the method of choice for medical devices; the FDA may question the rationale behind using the membrane filtration test over the direct transfer test for devices. An appropriate use of this test is for devices that contain a preservative and are bacteriostatic and/or fungistatic under the direct transfer method. With membrane filtration, the concept is that the microorganisms will collect onto the surface of a 0.45 micron pore size filter. This filter is segmented and transferred to appropriate media. The test media are fluid thioglycollate medium (FTM) and soybean casein digest medium (SCDM). FTM is selected based upon its ability to support the growth of anaerobic and aerobic microorganisms. SCDM is selected based upon its ability to support a wide range of aerobic bacteria and fungi (i.e. yeasts and molds). The incubation time is 14 days. Since there are many manipulations required for membrane filtration medical device sterility testing, the propensity for laboratory contamination is high. Therefore, in an open system, more sterility failures are expected when using this method. A closed system is recommended for drugs and small devices or combination products. Most pharmaceutical articles are tested using a closed system. In closed systems, the propensity for extrinsic contamination is very low.

Direct Transfer Sterility Testing

Combination products: This method is the method of choice for medical devices because the device is in direct contact with test media throughout the incubation period. Viable microorganisms that may be in or on a product after faulty/inadequate sterilization have an ideal environment within which to grow and proliferate. This is especially true with damaged microorganisms where the damage is due to a sub-lethal sterilization process. All microorganisms have biological repair mechanisms that can take advantage of environmental conditions conducive to growth. The direct transfer method benefits these damaged microorganisms. The entire product should be immersed in test fluid. With large devices, patient contact areas should be immersed. Large catheters can be syringe filled with test media prior to immersion. Cutting catheter samples to allow for complete immersion is the method of choice.

The USP authors understand that appropriate modifications are required due to the size and shape of the test samples. The method requires that the product be transferred to separate containers of both FTM and SCDM. The product is aseptically cut, or transferred whole, into the media containers. The test article should be completely immersed in the test media. The USP limits the media volume to 2500 ml. After transferring, the samples are incubated for 14 days.

Product Flush Sterility Testing

Combination products: The product flush sterility test is reserved for products that have hollow tubes such as transfusion and infusion assemblies where immersion is impractical and where the fluid pathway is labeled as sterile. This method is easy to perform and requires a modification of the FTM media for small lumen devices. The products are flushed with fluid D and the eluate is membrane filtered and placed into FTM and SCDM. This method is not generally used.

Bulk Drug Products / Biologics and Pharmaceuticals

Bulk Pharmaceuticals (APIs) are tested for sterility per USP 71 prior to release to the manufacturing processes.

Bulk Biologics are tested according to 21 CFR 610.12 for sterility testing. This method requires one media (FTM). The sample test sizes are listed in the document. Volumes are no less than 10 ml.10

Interpretation of Sterility Test Results

The technician must be trained in the method of detecting growth during the incubation period. Growth is determined by viewing the media, which is generally clear and transparent, against a light source. Turbid (cloudy) areas in the media are indicative of microbial growth. Once growth is detected, the suspect vessel is tested to confirm that the turbidity present is due to microorganisms and not due to disintegration of the sample; sometimes samples produce turbidity because of particulate shedding or chemical reactions with the media. Once a suspect container has been tested, it should be returned to the incubator for the remainder of the incubation period. Samples that render the media turbid are transferred on Day 14 of the test and incubated for four days. Growth positive samples require further processing such as identification and storage.

Sterility Test Failure Investigation

For every positive sterility test (OOS), the laboratory should perform an OOS investigation to determine the validity of the positive growth. This investigation encompasses the following items:

  1. clean room environmental test (EER) data;
  2. media sterilization records;
  3. technician training records;
  4. the relative difficulty of the test procedure;
  5. control data (open and closed media controls);
  6. technician sampling data (microbial counts on gloves and/or garments post testing).

The USP allows for a re-test of the product if persuasive evidence exists to show that the cause of the initial sterility failure was induced by the laboratory. Identification and speciation of the isolate(s) is a significant contributing factor to the final decision. If the First Stage sterility test can be invalidated by the laboratory, then the USP allows for Second Stage sterility testing. Second Stage sterility testing requires double the original number of samples tested. The Second Stage test can be repeated if evidence exists invalidating the test due to a laboratory error as above.

A detailed investigation may uncover circumstantial evidence to support a final decision. It is recommended that sterilization cycle data, environmental data, and bioburden data be reviewed prior to making any decision to release product.

It is recommended that medical device manufacturers qualify the test procedure with non-sterile samples.

The probability of a false positive can be calculated using John Lee's formula.11 The formula is based upon sample container diameter, amount of time container is left open and the room particulate count.

Sterility testing requires high levels of control with regards to GMPs, Good Laboratory Practices12, environment (aseptic clean room ISO class 5 or better), and employee practices. It is essential that meticulous technique be employed in the practice of sterility testing. Sterility testing is an integral part of sterilization validation as well as a routine quality control. Generally, false positive results are uncommon in testing drug products using a closed system. Combination products have challenges that should be planned into a robust QA program.


  1. The United States Pharmacopeia, 30th Revision, The United States Pharmacopeial Convention: 2008
  2. USP 30 Table 2 Minimum Quantity to be Used for Each Medium
  3. USP 30 Table 3: Minimum Number of Articles to be Tested in Relation to the Number of Articles in the Batch
  4. ISO 11137 Sterilization of health care products – Radiation – Part 2 2006: Establishing the sterilization dose
  5. FDA Guidelines 2004 "Guidance for Industry Sterile Drug Products by Aseptic Processing, Current Good Manufacturing Practices," September, 2004
  6. ISO 11737 ANSI/AAMI/ISO 11737-2 1998 – Sterilization of Medical Devices – Microbiological Methods – Part 2, Tests of Sterility Performed in the Validation of a Sterilization Process
  7. ISO 11135 1994 Medical Devices Validation and Routine Control of Ethylene Oxide Sterilization
  8. Code of Federal Regulations Title 21/Chapter I/Part 820, "Quality Systems Requirements: General," 2006
  9. GMPs CFR 201 Title 21 2006
  10. 21 CFR Part 610.12 Bulk Biologics
  11. Lee, John Y. "Investigation Sterility Test Failures" Pharmaceutical Technology, February 1990
  12. Code of Federal Regulations Title 21/Chapter I/Part 58, "Good Laboratory Practice for Nonclinical Laboratory Studies," 2006