Sterile Fill Facilities: Problems and Resolutions

September 1, 2003
Majda Imensek

BioPharm International, BioPharm International-09-01-2003, Volume 16, Issue 9

The author shares a case example that applies an eight-step structured approach to the front-end engineering of a vaccine R&D fill-and-finish facility project. Assumptions, issues, action items, owner interface, and interdisciplinary coordination are covered to meet the challenges of timing, technology, compliance, and cost.

Designing a modern vaccine research and development fill-and-finish facility is a complex endeavor. The project team must be able to meet the increasing challenges of technology, compliance, and cost with skill and imagination in a timely manner.

Most project teams agree that the two most important factors that contribute to a successful project are a well-defined scope and early and extensive planning.

The ability to influence project cost over the life of a project is greatest during the conceptual planning stage of the project life cycle (Figure 1). The curve indicates that the first 5 to 15% of project effort expended during scope development affects 75 to 90% of the project's final cost and operating expenses, following Pareto's Principle, which states that a minority of effort drives the majority of results.

Figure 1. The ability to influence project cost over the life of the project is greatest during the conceptual planning stage of the project life cycle. The curve indicates that the first 5 to 15% of project effort expended during scope development will affect 75 to 90% of the project final cost and operating expenses, following Pareto's Principle.

The more information that an owner makes available to the design team at the start of the project, the more likely the design effort will be a success. Effective communication and information exchange between the design team and client will produce a sound budget and a well-defined scope, minimizing changes to budget and schedule alike.

A Front-End Engineering Approach

In the vaccine research and development fill-and-finish facility project example described here, we applied a front-end engineering approach. This process ultimately results in a conceptual design report (sometimes called a conceptual design review), a well-defined basis of design (BOD), and a ±30% appropriation grade estimate. The sidebar summarizes the methodology applied to the front-end engineering of this particular vaccine R&D fill-and-finish facility (after project award and team mobilization).

Each step of the methodology is described below from a process design perspective and includes starting assumptions, issues, action items, owner interface, and interdisciplinary coordination. In the sections that follow, this case example shows how a well-defined scoping methodology provides structure to the design process and ensures that a client's needs are met.

Step 1: Client Goal-Setting Session

The first step in defining the scope of the project provides a forum for an interactive exchange of ideas and expectations between the client's project team, the technology teams, and the project design team. Many of the starting premises of the project are defined or verified during this stage.

During this goal-setting session, the designer team's expectations are to

  • obtain a statement of objectives that communicates the client's vision and needs

  • understand the markets into which the products will be introduced (the United States, European Union, or global markets).

  • establish team member roles and responsibilities.

The starting premises of the project as defined by the client are

  • The R&D fill-and-finish facility shall be designed to handle live and subunit vaccines.

  • Two vial lines - one for live vaccines and one for subunit vaccines - shall support vial production of 15,000 2-mL vials per batch.

  • The vial line for live vaccines shall meet BL-2 containment requirements, as defined by NIH guidelines.

  • The vial line dedicated to subunits shall meet BL-1 containment requirements, as defined by NIH guidelines.

  • Well-established technology shall be used.

  • The just-in-time receiving warehouse shall support formulating, filling, and packaging operations.

Step 2: Technology Transfer Session

The purpose of this step is to highlight product specifics and operating areas, with specific consideration of product and equipment interface and operational preferences. In addition, user groups are interviewed to learn about potential product features and to discuss technologies envisioned for the facility formulation suite and filling lines.

During this information gathering stage, emphasis is placed on collecting spatial and functional requirements rather than on finding solutions. The following product features or specifics are determined:

  • sensitivity, including temperature, pH, O2, and light

  • nonfoaming or foaming characteristics

  • water or solvent-based formulation

  • operator safety, such as biological safety levels and toxicity levels (operator exposure level)

  • product corrosiveness

  • definition of the "model" process using design parameters.

Formulation. The next step is to define the formulation process. In our example, the formulation area will be located on the same floor as the filling equipment. Maximum portable formulation-tank working-volume will be 400 L. Tanks above 400 L will be stationary.

It is also necessary at this stage to address the process design issues, including the following:

  • Is a segregated formulation room for vial lines required?

  • Is a weighing and dispensing subarea required? If the answer is yes, then required equipment - hoods, scales, peristaltic pumps, stirrers, isolators - needs to be defined.

  • Provide a clear definition of buffer prep and excipients. Will preparation be in situ, in a pilot plant then delivered to the area, or will buffers be prepurchased and delivered to the area?

  • Define formulation and solution. Filter sterilization is used where the component is soluble and likely to be adversely affected by heat. Sterilizing filters should be routinely discarded after process of a single batch. Industry practice is to employ a second sterile filtration at point of fill.

  • Consider suspension and solution issues. For example, products containing aluminum adjuvant are formulated aseptically because, once they are alum absorbed, they cannot be sterile filtered. We determined that no filtration was required.

  • Consider formulation tanks: What is the maximum acceptable working volume for portable tanks? How many are required? Will formulation be prepared in portable containers?

  • Consider floor scale resolution issues. Higher scale resolution creates higher equipment costs.

Component prep area and central tank wash. Our starting assumptions for the component preparation area are that the ultrasonic sink, pass-through washer and dryer, pass-through CGMP sterilizer, and pass-through dry heat oven will be used to support manual or semi-automated fill and filling line component preparations for vial line operations.

Presterilization preparation of glass containers typically involves a series of wash and rinse cycles. Containers and closures should be sterilized and, for parenteral drug products, free of pyrogens. Glass containers are typically subjected to dry heat for sterilization and depyrogenation. Rubber closures, including stoppers, are cleaned by multiple wash and rinse cycles before final steam sterilization. Silicone used to prepare rubber stoppers should be steam sterilized.

Front-End Definition Methodology

Process design considerations at this stage are

  • To maintain sterility, equipment surfaces that contact sterilized drug products or sterilized container/closure surfaces must be sterile so they do not alter the purity or potency of the drug.

  • Is manual or semiautomated fill required for vial lines?

  • What is the fill batch size? What vial sizes will be used? This information is needed to properly size the batch vial washer.

  • How is the formulated intermediate product received? If the product is received in glass bottles, what is the largest bottle volume and size? The dry heat oven volume can be sized based on batch and glass bottle size.

  • Is it preferable to use one CGMP sterilizer for product contact parts and utensils or to segregate parts and utensils into separate sterilizing equipment?

Our starting assumptions for the central tank wash area are that the formulation and product tank issues are fully resolved and that the central wash area is designed to clean all facility portable tanks (from BL-1 to BL-2) and supported with a clean-in-place (CIP) skid and steaming station to clean and sterilize all portable tanks. Tanks exiting the BL-2 area will pass through the decontamination autoclave.

Finally, the process design issues for the central tank wash area are whether the starting assumptions are acceptable and if segregation for BL-1 and BL-2 wash areas should be considered.

Filling line and lyophilization. Figure 2 shows multiple options that can be considered during the technology transfer session. Filling lines and lyophilizers are complex, expensive, and long-lead items. The technology supplier's timely input is necessary for determining space allocation, utilities, budget, and schedule.

Vial filling. Vial filling activities include multiple manipulations of sterile materials (aseptic connections and sterile ingredient additions, for example) before and during filling and capping operations. To maintain product sterility, the environment in which aseptic operations are conducted should be of appropriate quality throughout operations. Measurements to confirm air cleanliness in aseptic processing zones should be taken with the particle counting probe oriented in the direction of oncoming airflow and at specified sites where sterilized product and multiple containers are exposed.

Our starting assumption for the vial hand fill is simply that the manual fill operation is required. Process design issues include the following:

  • batch size

  • location of manual fill location (inside the formulation room or in a separate room)

  • whether to use manual fill or batch fill with tabletop or stand-alone equipment under a unidirectional airflow hood

  • whether the manual fill operation will support liquid only or liquid and lyophilized product.

Vial filling line. Our starting assumption for the vial filling line is that the client specifies the maximum batch size as part of the "starting premises" of the project. Process design issues include the following:

  • batch size for Phase 1, 2, and 3 clinical trials

  • container closure systems (size, type, and compatibility with equipment)

  • vial sizes and fill volumes

  • lyophilization, if applicable

  • range of line speeds and configurations

  • frequency and number of runs

  • product campaigning or line change.

Also important is determining whether the process will be batch or continuous. If the batch process is used, process design issues include

  • filler type (peristaltic pumps, time pressure, piston type, and so on). Ensure sterility of product contact surfaces by sterilizing the entire sterile-liquid stream-path

  • fill accuracy (±0.5%, ±1.05%, and so on)

  • whether stoppers will be prepared in-house or prepurchased sterile

  • if a diluent is required; if so, will the diluent be prepared online or prepurchased?

  • for batch-filling and stopper or capping equipment enclosures, determine whether product protection will be achieved with rigid plastic shields, restricted access barriers (RABs), or isolators. The sterile product and container closure should also be protected from activities occurring adjacent to the production line to segregate the aseptic processing line.

Lyophilization processes include the transfer of aseptically filled product in partially sealed containers. Facility design should ensure that the area between a filling line and the lyophilizer - and the transport and loading procedures - provide Class 100 protection.

For lyophilized products, determine whether prefreezing vials using a blast freezer is required before lyophilization loading. In addition, determine whether the lyophilizer is intended to process only online product, or if it is to be shared between online product and lyophilization development studies. If a lyophilizer will be shared by online product and development studies, more complex lyophilization features will be required, and the lyophilization equipment will need to be placed in a separate room.

Typically, transfer carts are used for batch fillers to move the aseptically filled vials, with partially seated stoppers, from the filler to the lyophilizer. The cart should maintain class 100 environmental conditions during the storage and transfer of the vials to the lyophilizer. The transfer cart environment is HEPA-filtered with unidirectional airflow and provides protection for both product and operator. If a continuous process is used, process design issues should define the filling line enclosure and determine whether product protection will be achieved with rigid plastic shields, RABs, or isolators.

Figure 2. Multiple options can be considered during the technology transfer session. The technology supplier's timely input is critical at this stage for determining space allocation, utilities, budget, and schedule.

An emerging aseptic processing technology uses isolation systems to minimize the extent of personnel involvement and to separate the external cleanroom environment from the aseptic processing line. A well-designed positive pressure isolator, supported by adequate maintenance, monitoring, and control procedures, appears to offer an advantage over classical aseptic processing. The design of an aseptic processing isolator employs unidirectional airflow that sweeps over and away from exposed sterile materials, avoiding any turbulence or stagnant airflow in the area of exposed sterile materials, product, containers, and closures.

Transfers are performed using a direct interface with a decontaminating transfer isolator or dry heat depyrogeneration tunnel. Properly operated rapid transfer ports (RTPs) are also effective transfer mechanisms. The room surrounding the isolator should be class 100,000 active.

If isolation technology is used, determine the decontamination agent to be used, the cycle time, and whether cleaning will be performed manually or using CIP technology.

A decontamination method should be developed that renders the inner surfaces of the isolator free of viable microorganisms.

Other issues related to vial filling include

  • whether to use a depyrogenation tunnel with cooling zone sterilization

  • filler type (peristaltic pumps, time pressure, or piston-type rolling diaphragm)

  • whether to clean product contact parts in situ or manually using CIP or steam-in-place (SIP) methods

  • fill accuracy (±0.5%, ±1.05%, and so on)

  • whether inert atmosphere is required during filling and, if so, should gaseous nitrogen or argon be used

  • what the room temperature and relative humidity control requirements are during the filling operation

  • product recirculation requirements during the filling operation

  • whether check weighing should be 1%, 5%, 100%, or statistical

  • whether stoppers should be sterilized in house or prepurchased sterile.

If stoppers are prepurchased, decide whether delivery to the line will be in bags alone or in bags with rapid transfer ports? Likewise, if stoppers are prepared in house, will delivery to the stopper bowls come from the same level or from above?

If the client prefers using isolation technology, then the delivery of stoppers to the line should be discussed. Stopper hoppers should be maintained using batch-by-batch sterilization. In addition, determine if a diluent is required and whether diluent preparation should occur online or prepurchased. If the diluent is prepared online, an additional in-line sterilizer is required.

Lyophilization. Our starting assumption regarding lyophilization is that the client's user group should provide process parameters for lyophilization based on model processes (batch size and vial sizes) and lyophilizer cycle data (such as lyophilizer loading temperature, lyophilizer cycle duration, operating temperature, and vacuum profile). The client's input is required to size the lyophilizer production surface area.

Process design issues related to lyophilization include whether manual loading will be performed using HEPA-filtered carts or whether automatic loading and unloading will be performed using HEPA-protected loading and unloading carts. If the client prefers automatic loading and unloading, consider using a conveyor pusher system and an automatic loading and unloading system (ALUS).

For the isolated line, loading and unloading carts should have a vapor-phase hydrogen peroxide (VHP) sanitization feature. For R&D facilities, a typical lyophilizer surface area is approximately 150 ft2 and offers single-door installations with a compact design and CIP and SIP features. A pusher conveyor system is less expensive and well suited for R&D fill-and-finish lyophilizer shelf sizes.

Capping and crimping. Our starting assumption is that the vial capping and crimping machine will be positioned in a separate room because of particulate generation, therefore the only process design issue is delivery of the flip-off caps to the isolated capping machine.

In batch operation, lyophilized vials are delivered in HEPA-filtered transfer carts and unloaded onto the capping and crimping station conveyor. In continuous line operation, closed liquid and closed lyophilized vials are delivered by conveyor or ALUS carts to the capping and crimping station.

External washing and drying. The starting assumption is that external washing and drying are required only if product specifics indicate the need to wash the exterior of the vials. Process design issues include determining whether an external washing is required. If yes, decide whether a water rinse is adequate or if chemical decontamination is necessary to inactivate the outer vial surface. In addition, determine whether the laser and UV coding will be applied to vials after the vial exterior wash or after capping and crimping.

Vial inspection. Our starting assumption is that vials will be inspected before the packaging operation. Any damaged or defective units should be detected and removed during visual or automated inspection of the final sealed product. The process design issue is whether liquid and lyophilization "nude" (that is, unlabeled) vials will be inspected manually, semi-automatically, or automatically. For an R&D pilot plant, a semi-automatic inspection machine is most frequently the client's choice.

Tray loading of inspected vials. In this example, we assume that a tray loader will be used to load nude vials into trays before staging to nude vial storage and waiting for quality control release for final packaging. The only process design issue is to determine what methods will be used to label, palletize, and shrink wrap finished product.

Packaging lines for vials. Our starting assumption is that packaging for nude vials will be manual with semi-automatic labeling.

Labeling operation and label inspection design issues include the number of labeling and packaging rooms, and whether labeling machines, and the label inspection machine, will be manual or semiautomatic.

In R&D facilities, packaging is typically accomplished manually using a linear flow operation. The product pack is designed to be distributed for clinical use.

From the technology transfer session, the design team expects to learn to

  • incorporate the client's operational preferences into the final design and define the fill line for batch or continuous operation

  • prepare block flow diagrams

  • define equipment options, equipment, and lead time

  • prepare preliminary equipment list adequate for pricing

  • establish contact with critical vendor resources to obtain information related to pricing, utility requirements, and equipment space needs.

Client action items from the technology transfer session include the following white papers to narrow the options and facilitate the decision-making process

  • restricted access barriers (RABs) compared with isolation technology advantages and disadvantages

  • lyophilizer batch loading (using transfer carts) compared with automatic pusher loading

  • liquid nitrogen or additional compressor (mechanical refrigeration system) as back-up for lyophilizer compressor failure

  • assessment of any equipment technology alternatives with cost comparison for filling lines and lyophilization.

The design team's action items include the following:

  • prepare white papers and submit to client for review (material will be subsequently discussed and any unresolved issues settled during the next technology transfer session.)

  • generate block flow diagrams

  • size process equipment

  • develop preliminary equipment list.

Step 2a: Technology Transfer

The purposes of this step are to improve and solidify the definition and requirements for the vaccine R&D fill-and-finish facility, review white papers, and agree upon solutions and the steps to take.

Session outcome. Issues resolved during this step become facts and directives. In our example, they include

  • Formulation must be disengaged from filling because most of the formulated products are held for three weeks in the intermediates storage, during sterility testing before release.

  • A continuous vial filling line will be enclosed in the isolator with a line speed of 100 vials per minute for 2-mL vials.

  • The isolator will be decontaminated using vapor-phase hydrogen peroxide (VHP).

  • The vial sizes are specified.

  • The vial filling line will be designed for 5% check weighing and will have the capability to purge vials with inert gas before and after fill of oxygen-sensitive products.

  • Peristaltic pumps will be used as the vial filling system.

  • An automatic pusher system will be used to load and unload the lyophilizer automatically and deliver vials to the capping machine.

  • The lyophilizer will be designed with a redundant compressor (mechanical refrigeration system) as back-up in the event of a lyophilizer compressor failure.

  • A decontamination washer and dryer will be incorporated into the BL-2 vial filling line.

  • Vials with liquid content will be inspected inline with semiautomatic inspection machines.

  • Lyophilized vials will be inspected manually.

  • Nude vials (lyophilized and liquid) will be automatically loaded into trays and manually palletized.

  • Palletized vials will be moved to the nude vials cold staging area (maintained at 4°C, –20°C, and –70°C)

  • An automatic label inspection machine is required.

Remaining process design issues include

  • determining changeover between campaigns

  • determining changeover time requirements

  • determining whether same-product production lines will be used for diluent and placebo batches

  • reviewing the fill-and-finish design approach to determine that it is compliant with EU and U.S. CGMP regulations and with NIH guidelines for separation and concurrent use of live and subunit filling lines.

The design team's action items include incorporating newly determined facts and

  • updating equipment list

  • updating block flow diagrams

  • developing process flow diagrams with material balance from client's "model process" batch records

  • developing process utility summary and determining the size of utility generating equipment

  • providing all data to other project disciplines

  • providing vendor space allocation data to the facility integration group

  • sizing process support areas to be included in architectural room design criteria sheets.

  • working with the facility integration group and HVAC engineers to develop functional diagrams and adjacency matrices to show the relative proximity of the various process spaces.

The diagram must be carefully constructed from sound data because the success of this type of facility is largely measured by its ability to smoothly integrate the movement of people, equipment, raw and intermediate materials, and finished product. Floor plan development will not proceed until these critical relationships are understood, and their broad design implications (dedicated HVAC systems, airlocks, unidirectional flow, and so on) are clearly defined.

Step 3: Regulatory Review

The purpose of this step is to review the proposed design approach from a regulatory perspective. The focus at this stage is compliance with EU, U.S., CGMP, and NIH guidelines on the separation and concurrent use of live and subunit cell lines.

Outcomes at this stage include

  • Formulation rooms should be dedicated for live and subunit products.

  • A formulation room to formulate live vaccine product will be supported with an exiting decontamination autoclave.

  • Sterilization or VHP decontamination of the depyrogenation tunnel-cooling zone is required for live vaccine continuous vial filling line.

  • Filling lines will be separated and dedicated for live and subunit products.

  • Define all storage areas for live and subunit product in process staging. Address the need for segregation on a case by case basis.

  • Develop HVAC cleanliness classification requirements for various areas.

  • Classify the room surrounding the isolator as class 100,000 active (class C).

  • Develop gowning procedures as a function of progression from uncontrolled areas to controlled areas to classified areas.

  • Achieve unidirectional flow of people, equipment, material, and product as much as possible.

  • Define the waste collection and waste decontamination systems.

  • Determine applicable codes and standards.

At this point, the design team should work with the facility integration group and HVAC engineers to update and modify functional diagrams (also called bubble diagrams or adjacency matrices) to reflect changes and decisions from the regulatory review.

A critical concern for the client is achieving process and functional segregation between live and subunit activities and areas. The functional diagram should stress segregation strategies to support concurrent production of live- and subunit-derived vaccines.

Once the functional process diagram and all support areas are defined, sized, reviewed, and approved, construction of the vaccine R&D fill-and-finish facility can begin. A facility integration group develops overall building systems that form the infrastructure of the facility. Issues such as material, product, personnel, and equipment are raised and resolved. A layout drawing is presented to the client team during Step 4.

Step 4: Workshop

During Step 4, a complete review of design progress to date is conducted. The workshop brings together the client company and the design team to review, challenge, iterate, and, ultimately, confirm the function, cost, and schedule parameters that define the project. Key activities include

  • reviewing all conceptual documents to date (For process discipline, this includes process description, process and utility flow diagrams, priced process and utility equipment list, and utility demand schedule and summary.)

  • reconciling all meeting minutes

  • reviewing snapshot estimate

  • reviewing project schedule

  • reviewing conceptual site plan

  • developing a list of outstanding issues

  • outlining conceptual design review table of contents.

Step 5: Scope Definition

The conceptual design should be well documented with drawings, narratives, and reports. This documentation includes integrated perspectives in each of the primary areas of consideration: process and equipment, facilities, clean utilities, and infrastructure and plant utilities.

Key activities in this step include development of facility plans and finalization of scope documents, project schedules, and the ±30% estimate.

Step 6: Conceptual Design Report

This step's objective is to publish the first draft of the conceptual design report (also called a conceptual design review) for the client's review. This document includes the discipline scope definition, the cost estimate, and workshop documentation.

When the client has read, commented, and annotated the conceptual design review draft, a formal facilitated review of the document is performed.

Step 7: Formal Review by Client

This facilitated interactive review allows the client to examine the conceptual design review. Comments resulting from this review are discussed and scheduled for incorporation into the final conceptual design report, as mutually agreed to by the the client and the design team.

Step 8: Design Delivery/Acceptance

When the client signs off on the conceptual design review document, the front-end definition process is complete. This milestone is considered the starting point of the multidisciplinary, preliminary design effort.

The process and documents described here are meant to illustrate the need for a systematic, well-established approach to the conceptual design of a vaccine R&D fill-and-finish facility. The careful collection of the data and criteria that forms the scope of the design is especially important in the conceptual stage of the design.

Prudent allocation of resources and inclusion of technology suppliers in the process is of utmost importance. Establishing the ground rules through a formalized process significantly improves the prospects of achieving both the design team's and the client's objectives. BPI