The Akers–Agalloco Method

Published on: 
,
Pharmaceutical Technology, Pharmaceutical Technology-11-02-2005, Volume 29, Issue 11

The authors present a new approach to risk assessment for aseptic processing that emphasizes the contributions of personnel.

Many in the pharmaceutical industry consider aseptic processing to be among the more difficult activities to execute properly. A substantial number of variables affects the safety of sterile products manufactured aseptically. The design and operation of an aseptic production facility must minimize the risk to the patient. This article presents a new approach to risk assessment for aseptic processing that emphasizes the contributions of personnel.

In September 2004, the US Food and Drug Administration issued the long awaited guidance on sterile drug products produced by aseptic processing (1). That month, FDA also published Pharmaceutical CGMPs for the Twenty-First Century—A Risk-Based Approach. (2) Each of these documents refers to the other, but there is no substantive information offered in either to assist the practitioner in establishing a risk-based approach for the design and control of aseptic processing operations.

Numerous pre-existing risk-analysis methods have been adapted for the pharmaceutical industry, including:

  • Fault Tree Analysis (FTA);

  • Failure Mode and Effect Analysis (FMEA);

  • Hazard Analysis and Critical Control Point (HACCP);

  • Hazard and Operational Studies (HAZOP);

  • Failure Mode, Effect, and Criticality Analysis (FMECA).

These commonly used methods have proven successful in a variety of applications. None, however, was ever specifically intended for application to aseptic processing, which relies on a large number of variables and has the lowest tolerance for failure of any process in the entire industry. Simply put, failure in aseptic processing is unacceptable, and a suitable risk-analysis method must incorporate all of the factors affecting product sterility and patient safety. The risk-analysis approach used must incorporate the recognition that there is no acceptable level of risk associated with sterile products, regardless of the manufacturing method. The goal is, and must always be, perfection in all elements of sterile product manufacturing, especially when products are made using an aseptic process (3). With this in mind, the authors embarked on an evaluation of previously established risk-evaluation methods as applied to aseptic processing.

Risk and risk-evaluation methodologies

The risk-analysis method that is chosen must focus clearly on the manufacturing activities that are essential for success in aseptic processing. This normally is manifested by a concern for criticality. Not all of the constituent activities of a process being evaluated have an equal effect on the end result. Risk, as defined within FMECA, is a multiplication of the criticality of an occurrence by its frequency. This definition seemingly can be applied directly to aseptic processing. Determining what constitutes an "occurrence," however, is generally difficult in aseptic processing. One could argue (perhaps incorrectly) that environmental monitoring results can provide a measurement of occurrences, but this assumption is theoretical at best. Environmental monitoring, especially in the ISO 5 environments commonly used for aseptic processing activities, hardly can be considered a quantitative or fully effective tool. In the typical ISO 5 environment, microbial contamination is detected only rarely, and even the most aggressive sampling plans only sample a limited amount of air or surfaces* (see endnotes). Another drawback of viable environmental monitoring is that results are unavailable in real time when using common growth and recovery methods. Nonviable particle monitoring, which is available in near-real time, has never been correlated sufficiently to viable data. The authors consider it of limited value in aseptic risk assessment.

The authors believe that there is a straightforward way to analyze risk in aseptic processing. Risk is by general agreement a function of the release of human-derived contamination into the operating environment. The full extent of human contamination risk is substantial: a gowned operator may release as many as 10,000 colony-forming units (cfu) or more per hour (4, 5). The estimated value is derived from operators performing controlled and defined movements immediately after donning sterile gowns. These methods merely provide a way to estimate the intensity of the contamination source from activities substantially different from those performed during aseptic processing. Direct application of these data to aseptic processing unfortunately is not possible, given the differing environments and activities.

Whyte developed a risk-assessment model based on his research on microbial deposition, which was an extension of his earlier efforts with settle plates (5–8). The premise behind this approach is the recognition that personnel are the primary source of contamination. It assumes the mechanism for dispersion of that contamination is airborne deposition. Whyte's model is the following:

Number of microbes deposited on product = C × S × Pd × Pa × T

in which C is the concentration of microbes in the source (people), S is the quantity of air or material dispersed from a source over time (usually cfu/m3/s), Pd is the proportion of organisms effectively transferred, Pa is the proportion of organisms that arrive into the product area, and T is the time during which microbes could be transferred.

Whyte also has proposed a slightly simpler deposition model for risk evaluation in which the risk from microbial contamination is defined as:

Risk = A × B × C × D

in which A is the microbial contamination on or arising from a source (e.g., glove or air), B is the ease of dispersion or transfer, C is the proximity of the source to the critical area (one can assume the contamination decreases as the inverse of the square of the distance), and D is the effectiveness of the control method (e.g., isolator, restricted-access barrier systems [RABS], automation, sealed container, intervention frequency).

Whyte chose five levels of risk for each of the terms in the equation: 0 indicates no risk, 0.5 indicates very low risk, 1 indicates low risk, 1.5 indicates medium risk, and 2 indicates high risk. In the case of factor D (effectiveness of control), he suggests 0 for "full barrier control." This means that, should a truly full barrier exist, the overall risk effectively would be zero. Products such as sealed vials that mimic the capabilities of a closed isolator system logically would fall into the "full barrier control" category (9, 10).

The authors' main criticism of this risk model is that it may underestimate process risk because it implies that an exclusionary process such as aseptic processing could produce a level of contamination control effectively equivalent to terminal sterilization. Certainly aseptic processing, in its present state, cannot be considered equivalent to terminal sterilization. Although it may be possible in the future for extremely advanced aseptic processing to achieve levels of actual product safety equivalent to those attained using terminal sterilization processes, there are no aseptic systems currently in operation that can attain process capability comparable with that of a terminal sterilization process. Furthermore, the destructive control afforded by terminal processes enables scientists and engineers to assess risk far more reliably than is possible in aseptic processing.

The deposition models for contamination risk analysis have their advantages and disadvantages. They take into account the following technical conditions that have been included in informal risk assessment for years:

  • the size of container opening;

  • the exposure time for the container;

  • the estimated microbial content of the surrounding air.

The authors believe that a major disadvantage of these models is their inherent assumption that the deposition of microorganisms from the air represents the only vector for contamination. Having observed numerous aseptic processes over the years, the authors find that approach inadequate. In many aseptic processes, operators must approach sterile materials with tools and protective measures. Focusing on personnel seems more appropriate. Although deposition may be the mechanism of dispersion, one must not be so focused on the containers that one ignores the potential presence of microorganisms in other stages of the aseptic process. The authors have developed an approach that is much broader and is not limited to the effect on the container alone.

In addition, the authors believe that assumptions regarding the microbial content of air are potentially misleading. It is tempting to consider that the worst-case content of airborne microbial contamination is that given in the common international recommendations (i.e., <1 cfu/m3).

The authors take a broader view of the contamination potential in aseptic processing and recognize that contamination comes from various sources. Table I presents a vision of risk assessment that has been used for many years (11). This vision draws upon that broader view and incorporates concerns that affect each of these tasks and the others that affect sterility assurance for aseptically produced sterile materials.

Table I: Validation of aseptic processing.

There are several basic concepts that result from this conception of aseptic processing. The effect of personnel on the aseptic process is all-important, and one must define risk to sterility predominantly in terms of the human activities required to execute the process. Other factors can be included, but focus on the contribution from personnel must be paramount. Risk is directly related to the number of human interventions required during the process: the fewer interventions required, the lower the risk of contamination. Interventions can be scaled relative to criticality. The type of interventional activity performed and the proximity of that activity to sterile materials suggest that some form of weighting of the intervention's influence is appropriate.

Other factors in contamination exposure include:

  • the container-opening diameter (smaller diameters are preferred, following the deposition models described previously);

  • the length of time the container is exposed to the environment, from its initial entry until its closure;

  • the length of time the closure is exposed to the environment before the container is sealed;

  • lyophilization, which can increase the exposure risk as a result of extended exposure time of the unsealed container and a need for additional interventions;

  • container type, (e.g., ampuls versus. vials, sealed versus. unsealed);

Advertisement
  • automation that significantly reduces the need for operator interaction with sterile equipment and materials;

  • complex assembly activities, which depend to a large extent on operator skill and thus increase risk;

  • novelty of personnel, equipment, or procedures, which increases risk.

One of the authors' fundamental premises is that all ISO 5 environments are not equivalent in their ability to support successful aseptic processing. Limitations to microbial detection in these clean environments make distinctions among them subjective. Nevertheless, the authors acknowledge that some technologies are superior to others in their ability to limit microbial intrusion and contamination. The environmental conditions in which the aseptic process is performed were included as a major factor in risk assessment. Environmental conditions are an overall factor in the present model. The application of these general principles and others of similar intent form the basis for the authors' risk-assessment model.

The pharmaceutical industry seems to give little consideration to risk mitigation during equipment selection. Only slightly more consideration of risk mitigation is given during the choice of container or delivery system. Ease of assembly and operation can make a substantial difference in risk exposure. Reducing the number of required aseptic connections is beneficial. Automated component handling is helpful in reducing risk. Low in-process adjustment and maintenance requirements should be key specifications. Equipment that can operate with minimum accumulation time also is desirable because it reduces exposure.

Akers–Agalloco method for risk assessment

The objective of this effort is a method that is:

  • easy to use;

  • based on real-world risk factors rather than more theoretical concepts;

  • based on an occurrence and criticality model;

  • focused on the effect of personnel;

  • inclusive of fatigue as a factor;

  • prone to reward processes in which interventions are reduced to a minimum.

Occurrence in the present model includes the quantity and criticality of interventions as well as other elements of process risk that are indirectly related to the core activity of aseptic filling. It includes:

  • background environmental risk, which is distinct from the filling environment;

  • aseptic compounding risk, which is heavily dependent on interventions;

  • aseptic setup risk, which is heavily dependent on interventions;

  • aseptic-filling risk, which is heavily dependent upon interventions and calculated separately for manual and machine processes;

  • lyophilization risk, which is included where present.

The risk to contamination associated with personnel is such that the contributions for these factors must be weighted more heavily than those derived from the contributing factors that are not human-related. An extra weighting of 10× has been assigned to those process steps in which personnel presence and close interaction with sterilized products, materials, and equipment is mandated: aseptic compounding, aseptic setup, and aseptic filling.

The risk for the overall process is the sum of the contributions from all of the relevant process factors, weighted according to the effects of various practices. Determination of the individual factors varies and includes simple multiplications that reflect the combined influence of variables within an individual process step. Weighting factors are included in the lyophilization and aseptic filling parameters to provide a more balanced consideration of these process steps. Addition is used as the primary means of calculation for the following reasons:

  • It substantially simplifies the calculations once the individual contributions have been determined.

  • It is consistent with the sense that adding an activity (e.g., lyophilization) to a process adds risk to that process (e.g., solution filling). This allows the elements of a larger activity to be addressed individually and directly and optimized from a risk perspective more easily.

As in the other risk-assessment models, a low score in the present model suggests a low potential for contamination. The risk contributions for the various process elements of the overall aseptic process are described below and quantified in the tables that accompany this text.

The risk assessment relies heavily on the environmental technology used in the aseptic process. The authors have applied a corrective factor that is based on the individual contributions of the various process steps to the final score. If aseptic compounding and aseptic filling are not performed using the same environmental technology, then the contribution risks for each should be calculated separately, on the basis of the technology used.

Facility contributions

Facility contributions depend on the types of environments provided for the supporting tasks associated with aseptic processing, including sterilization unloading and personnel gowning, and the background environment for the fill zone. Additional contributions from the facility derive from the sanitization method used and the frequency of treatment (see Table IIa).

Table IIa: Facility considerations for cleanrooms.

Similar parameters relate to using isolators, but the factors and weighting are adjusted to suit the differences in technology (see Table IIb). The authors have included RABS within the cleanroom category because this is the technology that it most resembles. The model for facility contributions can be expressed as:

cleanroom contribution = sanitization interval factor × sanitization process factor × background classification factor

or:

isolator contribution = decontamination interval factor × decontamination factor × background classification factor

The intervals between sanitization and the type of process applied are multiplied and added to the contribution from the background environments. Smaller facilities where one room serves multiple purposes (e.g., filling and unloading) receive a lower score because of their simplicity.

Table IIb: Facility considerations for isolators.

Aseptic compounding contribution

Processes that require substantial amounts of aseptic processing during formulation have an increased risk that must be factored separately from the aseptic filling process. The number of personnel interacting with sterile materials; the duration of the process, including setup and execution (exclusive of hold times where the sterile material is secured and there is no personnel activity in the environment); and the technology used for the process are factors in this area (see Table III). Because these processes generally involve frequent human intervention, interventions are not considered separately. This contribution is present in the production of even the simplest products in which the only postfiltration activity involving sterile materials is associated with sampling or verifying filter integrity.

Table III: Aseptic compounding.

The process time, including any required aseptic setup for compounding, is multiplied by a novelty factor—based on experience of personnel, equipment, and process—to determine the aseptic compounding contribution to process risk, which can be expressed as follows:

aseptic compounding risk contribution = process duration × novelty factor × aseptic personnel factor

Aseptic setup contribution

The assembly and setup of filling equipment require direct human manipulation of sterilized equipment and tools within the critical environment. In some firms, only the most experienced personnel, whose proficiency in the required task has been confirmed by a successful media fill, are allowed to perform this task. The hands-on nature of this activity often requires evaluation separate from that of the fill process. The duration of the setup, which reflects the sophistication and complexity of the filling equipment and the uniformity of the components, should be factored into this activity. Because this activity is almost entirely accomplished by humans, related process interventions are not considered separately.

Typically included in this activity are product connection and introduction, initial weight and closure systems checks, and any other relevant activities, including environmental monitoring during this assembly and setup process. The additional setup concerns related to suspensions, creams, ointments, and powders are factored into the setup duration time, as are other elements such as stopper and vial uniformity, inert gassing system installation, and the more demanding requirements for double-chamber filling and other more complex package designs (see Table IV). Product-formulation factors also are incorporated into contributions from the aseptic execution. The location of sterilizing-grade filters in the product delivery system affects the risk of contamination of a filled container. The aseptic setup risk is expressed as follows:

aseptic setup risk contribution = setup duration × complexity factor × product delivery factor × novelty factor × aseptic personnel factor

Table IV: Aseptic filling setup.

Aseptic-filling process contribution

The perfect intervention is the one you don't have to perform. When evaluating aseptic processing, one must focus on the need to avoid interventions and, when they are unavoidable, to minimize their influence. Intervention management during aseptic processing has received increased attention with an increase in the importance of interventions (1). A recent article about intervention management provides the following definitions and examples related to aseptic filling activities (as opposed to setup or changeover):

Routine interventions are activities that are inherent parts of the aseptic process and integral parts of every batch. Typical routine interventions include:

  • periodic component replenishment;

  • periodic fill weight or volume checking and verification;

  • fill weight or volume adjustment;

  • environmental monitoring;

  • product sampling;

  • filter integrity testing;

  • product container replacement;

  • any other interventional activity that is an integral part of the process (12).

Nonroutine interventions are a significantly greater concern than routine interventions because their frequency is substantially lower than that of routine interventions. In contrast to routine interventions, the execution of nonroutine interventions cannot be as narrowly scripted. In the article cited previously, the user is encouraged to define how to perform interventions of all types, train personnel in those practices, and adhere to them during process simulation and routine production. That is sound advice, but there are simply no means to orchestrate nonroutine interventions so that they conform to the predefined practices. During an actual process, nonroutine interventions may vary somewhat from expectations, and a company may be forced to rely on adaptations by the operator to execute them successfully. For this reason, the authors place greater emphasis on any nonroutine interventions that must be performed. The following definition and list come from the same reference:

Nonroutine interventions are activities that are predominantly corrective and may not be a part of every batch. Although in theory nonroutine interventions may not be necessary during the aseptic process, in practice such interventions are almost always required to correct some anomaly. Some common nonroutine interventions involve:

  • stopper misfeeds or clumping;

  • fallen, broken, or jammed containers;

  • defective seals on containers;

  • product spillage or leakage;

  • product filter change;

  • sensor adjustments or replacement;

  • filling-needle replacement;

  • fill-pump replacement;

  • stopper-bowl changes;

  • timing adjustments;

  • conveyor or guide-rail adjustments;

  • any other line malfunction requiring manual correction (12).

The aseptic-filling risk contribution can be determined using an approach designed strictly for manual fills and a more general method for machine fills in which personnel fulfill only a supporting role.

Manual fills

The intervention risk is the number of times the individual parts of the package (e.g., vial or stopper) are handled to prepare a filled container, and is expressed as follows:

intervention risk for manual filling (IR) = touches per unit

The result is combined with the other relevant factors for aseptic filling found in the tables to define the overall risk for manual aseptic filling.

The intervention risk for any container is defined as the number of times the container must be handled during the aseptic process. In manual filling, that number is usually greater than 1. Machine filling has an intervention risk of less than 1 (see the "Machine fills" section).

Manual filling must be considered the most risky of all aseptic processes because the minimum number of required interventions to fill and seal a container is greater than one.

Machine fills

Interventions during machine filling are substantially fewer, and thus of lesser impact, relative to manual fills. Intervention risk IR, with respect to criticality factors considered, is the distance from the exposed product contact parts and components. Critical interventions (e.g., replacement of fill pumps or other critical dosing equipment) are scored as 5, as are any aseptic connections made or remade after the initial setup. All other interventions within one foot of exposed product contact parts or components parts are scored as 3 on every occurrence. Interventions within two feet are scored as 2. Interventions outside two feet are scored as 1. Routine interventions that are an inherent part of every process are weighted as 1, and nonroutine (or corrective) interventions are weighted as 3.

Number of interventions. Calculate or visually confirm for a period of not less than one hour during the process all of the interventions, routine and nonroutine, required during the process. Multiply each by the appropriate proximity and type score. Determine the weighted number of interventions per hour by adding these values. For example:

4 routine interventions within 1 foot: 4 × 1 × 3 = 12

2 routine interventions within 2 feet: 2 × 1 × 2 = 4

1 nonroutine intervention within 3 feet: 1 × 3 × 3 = 9

2 nonroutine interventions within 2 feet: 2 × 3 × 2 = 12

1 critical intervention: 1 × 5 × 3 = 15

Weighted interventions/h: 12 + 4 + 9 + 12 + 15 = 52

The score should be based on the number of interventions observed or the maximum allowed. The ideal number of required interventions is always zero.

The weighted number of interventions (normalized for criticality and proximity) per hour should be determined first. A longer evaluation period provides a more accurate assessment, as does averaging the number of observed interventions over the entire batch or multiple batches. The role of container and closure consistency in determining handling requirements is automatically included when determining intervention risk in this manner.

Containers/hour. The number of containers per hour is the actual number of units produced during a one-hour period, not the theoretical line speed per minute multiplied by 60. Determination of this value over a longer period of time is preferable for the sake of accuracy. Do not include periods when filling is intentionally stopped for activities such as lunch, breaks, or shift change. Include in the calculation those times when the fill is interrupted by interventions of any type. Divide these values by each other to determine the number of interventions per container. Again, a lower number is desirable. This value is the intervention risk (IR), expressed as follows:

(normalized interventions/h) ÷ (containers/h) = interventions/container or intervention risk (IR)

Adjusted product-filling risk

Estimate the total risk from filling (for either manual or machine fills) by incorporating the remaining variables associated with the filling process: container size, complexity, container introduction method, closure-handling technology factor, and process duration (see Table V).

Multiply the intervention risk by the process duration in hours, the container design factor, container feed factor, closure feed, aseptic personnel factor, and the technology factor.

Table V: Aseptic filling.

IR × fill duration × container factor × container feed factor × closure feed factor × novelty factor × product factor = contribution to risk from aseptic filling

The longer the process duration, the greater the possibility that at least one unit will be contaminated because of the increased number of interventions. Multiplying the intervention risk by the length of the process emphasizes the effect of filling speed and includes consideration of fatigue as a factor in causing contamination.

Multiplying the resulting number by the product and technology factors adjusts for the varying levels of contamination potential associated with various aseptic technologies and product formulation types.

Table VI: Lyophilization process (if present).

Lyophilization risk. Lyophilization risk is associated with the time that filled components are exposed to the environment, between first exposure and closure, as well as with the handling practices, lyophilizer sanitization** (see endnotes) and sterilization practices, and environmental technology (see Table VI). The equation is as follows:

lyophilization risk contribution = loading time × lyophilizer sterilization factor × load factor × transfer factor × tray factor × technology factor

Environmental technology risk

Once the risk contributions from the components of an overall aseptic process have been determined, the authors recommend an overall adjustment according to the environmental technology providing the aseptic conditions. If aseptic compounding is a substantial part of the process, and is not performed using the same technology, the environmental factor should be applied separately and then added. The authors have placed an arbitrary weighting on the various technologies throughout this article (see Table VII). As stated earlier, consistency in the evaluation of an aseptic process will be manifest in the overall risk assessment, regardless of the specific values chosen.

Table VII: Environmental technology.

Conclusion

In the application of the present method, a glaring difference in risk analysis "score" is apparent between conventional staffed aseptic processing and advanced technologies. Manual processes fare even more poorly. The authors believe that the distinctions this method reveals among different aseptic technologies are real and accurately represent the risk of contamination. The reader may disagree with the numerical weighting factors selected, but the rationale for them is sound. It is possible to debate the present recommendations in a strictly numerical sense, but such a debate would result only in subtle changes and would not materially affect the objective assessment of the differences among the various practices evaluated. This method should not be used to determine what is acceptable or unacceptable in absolute terms, but should serve as a means of identifying opportunities for process improvement, regardless of the practices and technologies used. There may come a time when a system similar to this one can be used to define the acceptability of practices for products, but that time has not yet arrived. The authors also see potential for the application of this method in the selection of technologies to be used.

Perhaps even more important, the authors believe that the present risk analysis model can facilitate the determination of an appropriate level of validation and revalidation suitable to the technology used. Aseptic technologies that either mitigate the risk of human intervention or eliminate such activities entirely through automation or separative technologies are inherently safer than technologies that allow direct human intervention. It follows that the more reliant a technology is on human intervention, the more risky it must be. Logically then, such traditional tests as the media fill are far less likely to provide value in assessing the performance of advanced technologies. Thus, media-fill test frequency could be reduced, as could sample size, particularly in systems that allow direct real time control and evaluation of the environment and the process.

In fact, the authors propose that it is possible to draw a clear line of demarcation between "advanced" aseptic technologies and more conventional systems using risk analysis. For a technology to be considered advanced, there must be no direct human intervention at any time. For example, a RABS system would be considered advanced only if there were never direct human interventions through temporarily opened barriers. Similarly, a blow–fill–seal system would be considered advanced only if no containers were put at risk after an intervention and the fluid path were subjected to sterilization-in-place before the resumption of manufacturing. In the authors' view, firms that invest in technologies that have been demonstrated objectively to mitigate risk and in which key control parameters can be measured and controlled in real time should benefit from reduced validation, revalidation, and traditional microbiological-monitoring requirements. Imposing the same validation requirements for advanced technologies as for more risk-intensive alternatives not only provides no safety benefit, it does actual harm through implementation delays caused by excessive requirements.

This article may be considered a positive step in an effort to assess risk in aseptic processing. The authors believe that a broadened perspective of risk relative to aseptic processing increases awareness that risk can vary substantially in what are perceived by many to be equivalent—and thus equally acceptable—practices and technologies. The authors welcome constructive criticism of this effort in the hope that it will lead to a clearer definition of risk issues and evaluation approaches as they relate to aseptic processing.

Postscript

As a further guide to the reader, the authors have applied this method to various facilities used for aseptic processing (see Appendix 1, posted at www.pharmtech.com). The facilities discussed are models based on real installations, though the authors have altered some of the characteristics to prevent identification. The risk assessment of these facilities was used to fine-tune the model. The authors believe the relative (and, of course, subjective) capabilities of the facilities are consistent with the values obtained and demonstrate the potential utility of the present model for application in aseptic risk assessment. Should readers visit each of the facilities the examples are based on, they might develop a similar perspective.

Endnotes

* The authors do not support increases in environmental monitoring in these already very clean environments in a misguided effort to find what should not be present. Increasing monitoring scrutiny typically increases the number of interventions and thus increases the risk of contamination. In any case, there are no means to prove the absence of microorganisms from an environment; additional samples provide no benefit.

** Sanitization of the lyophilizer is not considered in compliance with CGMP regulations, but some laboratory-scale lyophilizers cannot be sterilized.

James Akers is the president of Akers Kennedy & Associates, Kansas City, MO. James Agalloco* is the president of Agalloco & Associates, 856 US Highway 206, Suite B-11, Hillsborough, NJ 08844, tel. 908.874.7558, jagalloco@aol.com He also is a member of Pharmaceutical Technology's Editorial Advisory Board.

*To whom all correspondence should be addressed.

References

1. US Food and Drug Administration, Guidance for Industry: Sterile Drug Products Produced by Aseptic Processing—Current Good Manufacturing Practices (FDA, Rockville, MD, Sept. 2004)

2. FDA, Pharmaceutical CGMPs for the Twenty-First Century—A Risk-Based Approach, (FDA, Rockville, MD, Sept. 2004)

3. Parenteral Drug Association, "Process Simulation Testing for Aseptically Filled Products, Technical Report No. 22," J. Pharm. Sci. Technol. 50 (6 suppl.), 1996.

4. B. Reinmuller, "Dispersion and Risk Assessment of Airborne Contaminants in Pharmaceutical Cleanrooms," Royal Institute of Technology, Building Services Engineering 56 (2001).

5. W. Whyte, "Reduction of Microbial Dispersion by Clothing," J. Parenter. Sci. Technol. 39 (1), 51–60 (1985).

6. W. Whyte, "A Cleanroom Contamination Control System," Eur. J. Parenter. Sci. 7 (2), 55–61 (2002).

7. W. Whyte and T. Eaton, "Microbial Risk Assessment in Pharmaceutical Cleanrooms," Eur. J. Parenter. Pharm. Sci. 9 (1), 16–23 (2004).

8. W. Whyte and T. Eaton, "Microbiological Contamination Models for Use in Risk Assessment During Pharmaceutical Production," Eur. J. Parenter. Pharm. Sci. 9 (1), (2004).

9. www.medInstill.com

10. www.aseptictech.com

11. J. Agalloco, PDA Course Notes on Aseptic Processing, 1988 to date.

12. J. Agalloco. "Management of Aseptic Interventions," Pharm. Technol. 29 (3), 56–66 (2005).

Appendix I: Application of the method

In this appendix, the proposed methodology evaluates five different aseptic processing systems. Execution of the methodology describes the facilities better than a written summary, but a brief description of each is provided by way of introduction.

In each of the facilities, a freeze-dried formulation is used for the evaluation. The authors have chosen to simplify the process by providing the weighted interventions, line speeds, process duration, and thus the intervention risk for each system. The authors recommend using the method described in the text, but in the interest of brevity we have eliminated that step in these examples. The intervention risk for each of these is included in the listing below.

Facility A. An older facility producing a range of small-volume parenterals of various formulations and configurations. Weighted interventions per hour: 90; fill speed: 120 vials/min; process duration: 6 h, and intervention risk (IR ): 0.0125 interventions per container.

Facility B. A heavily automated facility built in the late 1980s and dedicated to the production of a single freeze-dried product in multiple containers and strengths. Weighted interventions per hour: 5; fill speed: 300 vials/min; process duration-5 h; and intervention risk (IR ): 0.00027 interventions per container.

Facility C. An early-generation isolator-based facility intended for a variety of products and formulations. Weighted interventions per hour: 60; fill speed: 80 vials/min; process duration: 4 h; and intervention risk (IR ): 0.0125 interventions per container.

Facility D. A small-volume suite producing clinical materials. Weighted interventions per hour: 60; fill speed: 30 vials/min; process duration: 2 h; and intervention risk (IR ): 0.033.

Facility E. A low-volume clinical suite relying on manual filling. Interventions required per container: 4, thus the intervention risk (IR ): 4. Process duration: 4 h.

The latest media fills at each of these facilities were free of microbial contamination, which reveals the relative inability of process simulations to evaluate relative risk.