Of late, there has been considerableinterest and intrigue in the pharma-ceutical industry with regard to therecently approved FDA guideline on
PAT. In September 2004, the FDA issued theirfinal guidance for the industry, “PAT – A Frame-work for Innovative Pharmaceutical Develop-ment, Manufacturing, and Quality Assurance.”The guidance describes a framework for the“implementation of innovative pharmaceuti-
cal development, manufacturing, and qualityassurance.” The guidance extends beyond mereinstallation of process analyzers – it encour-ages the application of process control, continu-ous improvement, and knowledge managementtools along with the vision of an exciting, newapproach to pharmaceutical manufacturing andregulatory efficiency. PAT promises to deliver a“culture-change” in the industry, which has toooften treated innovation and productivity asstep-children to regulation and
compliance. With PAT guidance in place,manufacturing companies now have theFDA’s encouragement to adopt a new, risk-based regulatory framework that has itsbasis in scientific and engineering principles.Though as with any new initiative, ques-tions are rife as to “how” to implement PAT,what is the best approach, and how do weoptimally enforce IT and automation strat-egies to support the PAT framework.
We were able to incorporate the ideas ofthe PAT initiative into a new biological pro-cess that our company developed and whichis currently deployed at our facility inClayton, North Carolina. Utilizing thisproject as a case study, this article presentsthe approach to PAT that we adopted at oursite, along with an automation strategy thatwe believe enhanced our PAT effort. We willapproach this article by first explaining ourinterpretation and the framework for PAT.We will then move on to discuss the role ofautomation and explain the concept of “scal-able” automation. After a brief introductionto the case study, we will compare automa-
tion concepts and explain the importance of having a suffi-cient automation infrastructure in place to support PAT. Wewill then present two examples from our case study todemonstrate the benefits of deploying PAT. The first exampleis centered on the operation and control of our Water forInjection (WFI) systems for our processes. The second ex-ample describes a reporting application that utilizes theadvantages of the automation infrastructure to provide areal-time comparison of parameters over multiple batches,also referred to as the fingerprinting of “golden batches.”
The PAT ApproachThe FDA guideline1 presents a very broad interpretation ofwhat the Agency considers PAT to be. As a site, we are in theprocess of developing a well-documented “master-plan” thatdefines our interpretation of the guideline and presents aroadmap for our company to follow while implementing PAT.The master plan defines the mission, vision, strategy, andframework of our PAT effort. It also presents a blueprint-document that all PAT projects follow to ensure consistencyacross various PAT initiatives. Within the master plan, thePAT framework is detailed, which explains how a PATopportunity is initiated, executed, and evaluated. The frame-work is presented in Figure 1, and a detailed explanation ofthe proposed steps to a PAT implementation are provided inSidebar 1.
The process of identifying, monitoring, analyzing, control-ling, and reporting combined from the PAT approach. Oncethese discrete steps have been deployed, we expect to achievethe final goal of most PAT initiatives – process understand-ing. Understanding the process well, with critical pointsidentified and controlled, and all sources of variability undercheck, we can then reap the benefits of the PAT model. Thesebenefits include, but are not limited to:
• real-time quality assurance• right first-time and enhanced root cause analysis tools• reduced cycle times• yield improvement opportunities
• faster time to market potential• decreased burden of final product testing• reduced manual testing
Role of AutomationThe model in Figure 1 underlines the importance of havingautomation in place, which can support the monitoring,modeling, controlling, and reporting of the Critical QualityAttributes (CQAs) for any PAT effort. To implement PATeffectively, automation needs to support the effort of continu-ously and automatically collecting data not just from thesensors directly associated with the process, but from all ofthe other factors that could influence the results. In otherwords, close integration with the control system becomesvery desirable if decisions are being made on the basis ofonline measurements, while deviation handling capabilitiesare necessary to ensure process control. The control conceptis based on the model of the so-called automation pyramid -Figure 2. The pyramid describes the layers and functions ofautomation beginning with the plant floor or field instrumen-tation and actuators. It can extend all the way up to theManufacturing Execution System, (MES) and EnterpriseResource Planning, (ERP) level.
The idea behind the applied concept of “scalable” automa-tion is to put the basic infrastructure and functions forautomation in place with every new project. This allows forhigher levels of automation, including PAT functions, to beadded later on in the project or even during the operationalphase of the facility. The application of today’s “New Genera-tion” control system technology makes this control conceptflexible, affordable, and it allows for quick implementation.The “New Generation” control system technology is based onopen interfaces, modular configuration, qualification, as wellas scalability. So, automating the process would start at thebasic control level and then the described PAT principleswould be applied - Figure 1, Sidebar 1. The more we under-stand our process, the more we can increase the level ofautomation to support the increased understanding. Auto-mating a full batch process to start the project, it was feltwould result in a lot of re-work as the project progressed andprocess idiosyncrasies were better understood.
The Case StudyFigure 3, shows at a very high-level, the process flow for anewly implemented process at our facility. This process is thebasis of our case study on PAT and scalable automation. Theprocess flow is typical of biotherapeutics facilities and con-sists of numerous discrete steps, including dissolution, filtra-tion, chromatography, ultrafiltration/diafiltration (UF/DF),nanofiltration, followed by formulation, filling, and freezedrying (not shown). The chromatography and UF/DF skidscomprise the heart of our process. In addition, there arenumerous vessels, tanks, pH adjustment carts, and Tem-perature Control Modules (TCMs) that we combined in the“balance of process” environment. Two other vital compo-nents in the process are the CIP system and the WFI system.The CIP system, like chromatography and the UF/DF, is
typically a skid-based system. The WFI system is probablythe most crucial ancillary system in the process. WFI isutilized approximately 50 percent of the time in the overallproduction process.
During normal production, data is collected at each step inthe process, and the results of each step are critical for normalsequential processing of the product. The chromatographyand UF/DF skids (as with the CIP skids) are all typicallyautomated in their operation and are controlled locally withlocal operator interfaces. The conventional approach untilnow, in the industry, has been to buy chromatography, UF/DF, and CIP skids from different vendors usually with propri-etary and differently configured/documented control systemsinstalled on each skid. Besides lending itself to arduous datacollection, communication (also referred to as “handshak-ing”) between the skids for optimal operation and processcontrol is a huge and often costly challenge. Furthermore,understanding and reducing variability in the overall processbecame very difficult with so many different control plat-forms controlling the same process. In addition, the installa-tion of skids from different vendors with individual controlsolutions would result in multiple, different operator inter-faces (graphics design, color codes, alarm handling/mes-sages, commands, and logins). We consider this a seriouschallenge for our production operators who have to operatemultiple skids, control the non-skid related “balance of pro-cess” environment, and interact with the utilities systems.Different operator interfaces not only cause inefficiencies,but can develop into a main source of operator errors.
Comparison of Automation ConceptsWhen we were tasked to develop the automation concept forour case study, we used the model of the Automation Pyramidto assess the different approaches available to us. For thisarticle the two most extreme approaches to bioprocess auto-mation will be discussed. As described before, pharmaceuti-cal and bioprocess facilities usually consist of an assembly ofskids from different vendors set in a process environmentthat supplies utilities, process aids, storage, and cleaning. InFigure 4, automation concepts for the same process utilizing
skids from different manufacturers that are set in a processenvironment are compared. In the “islands of automation”concept on the left of Figure 4, the option of having indi-vidual automation solutions for the skids and the environ-ment is analyzed. The skids are not interfaced with eachother and have individual operator interfaces. The scalableautomation concept on the right of Figure 4, presents thesolution we selected for our case study. We worked with ourskid manufacturers and system integrators to implementtheir automation and expertise on the same “new generation”control system platform. Thanks to inherent scalability,modular approach, and ease of configuration, these controlsolutions are also attractive to the skid manufacturers. Inthis concept, we have individual skids that can be developedand tested stand-alone. Later on they can be applied to theprocess and play in concert with each other (imagine the “plugand play concept” from your home and office computer world).Skid manufactures and integrators are required to utilizestandardized and pre-qualified configuration modules, aswell as to adhere to the same configuration rules (such asgraphic standards).
Figure 4 uses the automation pyramid to compare bothsolutions layer by layer.
Automation Concepts and PATBased on the comparison in Figure 4, the advantages ofautomating the entire process on one scalable control systemplatform became apparent to us. Not only for one skid, butacross the entire process, scalable automation:
• fulfills the need to effectively access all sources of variabil-ity
• provides a method to monitor the CQA’s in a commonformat
• provides a clear relation of data to the batch and processstep information
• allows for handshaking and interlocking between skids aswell as common process and utilities systems
• These characteristics are imperative to enhance the PATeffort.
Figure 3. Process flow for new biological process.
The “islands of automation” method of control systeminfrastructure (as depicted on the left in Figure 4) makes eachone of the characteristics listed above complex on manyaccounts. It is difficult to collect and exchange data from fiveskids and supporting infrastructure systems. Once available,correlation of data becomes a huge challenge. Batch data areusually unavailable for the entire process, and the bottomline is that control is still being done with individual and non-interfaced control systems. So the final element of PAT, beingcontrol, based on real-time measurements to provide realtime quality assurance, becomes difficult to implement.
Trending across an entire batch becomes possible if acentralized historian for data collection is added, but philoso-phies like model predictive control are difficult to realizesince the effects of one process step can rarely be correlatedwith another process step down the line.
After all these considerations including a total cost ofownership analysis, it was decided to implement the central-ized and scalable automation approach. This lends itself veryeffectively to the PAT concept. Having installed all skids, theproduction equipment, and the critical utilities systems onthe same control system platform, we are able to:
• obtain all the process relevant data via one commoninterface
• collect multiple pieces of real-time data to develop reportsand even models
• report on multiple batches using one common reportingengine
• handle deviations from the process specifications in atimely manner
The last facet is achieved due to the fact that skids across theprocess have all the relevant information with regard to theentire process, as opposed to just operating with informationregarding their own skid. With all systems on one commonautomation platform, we were truly able to lay the frameworkfor deploying the entire breadth of PAT and begin to under-stand our process from a holistic standpoint.
Implementing the concept of scalable automation enabledus to install the framework for automation and PAT. Thisframework best provides information and tools to increaseour process understanding. Since the expectation was to usedata from the control system for process decisions and releaseof product, we ensured that the control system met allrequisite Part 11 requirements.
With the scalable automation concept in place, we parsedour process into smaller more manageable sub-systems, anddeployed PAT principles on sub-systems when the opportu-nity presented itself. One such sub-system which benefitedfrom the centralized automation concept and PAT principleswas the WFI sub-system.
WFI ExampleOur site uses WFI as a solvent in processing its products, asa solvent in equipment and system cleaning processes, and asa significant component of many of its products (the productbeing considered for our case study is roughly 90 percent
Figure 4. Comparison of automation concepts for new process.
water). A WFI system at our site is structured so that thereis a centralized WFI generation with main distribution tanks,which then feed intermediate production, sub-distributiontanks. The intermediate production, sub-distribution tanksstore WFI until production actually requires it. At that time,the main distribution loop starts re-feeding the intermediatesub-distribution tanks.
The requirements for WFI systems are set forth in themajor pharmacopoeias: United States Pharmacopeia (USP),European Pharmacopoeia (EP), and Japanese Pharmaco-poeia (JP). Figure 5 shows a still and a main distributiontank, which feed a utilities sub-distribution tank, which inturn feeds a production intermediate, sub-distribution tank.Production at our site uses water by opening and closing thevalves into the main process. For the sake of illustration, wehave just concentrated on one chromatography system and aUF/DF system, but the same process applies to most of theskids in the process.
WFI Example – “Non-PAT” ApproachBefore we illustrate PAT and scalable automation applied tothe WFI systems using centralized automation, we will takea look at the operation of the WFI systems without theapplication of PAT utilizing the “islands of automation”concept (WFI control system different from the skid controlsystem). When production wants WFI, the production opera-tor would have to confirm with the utilities operator whetherWFI is available for use or not. Electronic requests via seriallinks or hardwired signals back and forth are options; how-ever, both are susceptible to failure and costly in a validatedenvironment.
Upon receiving a request for WFI, the utilities operatorwould ensure that the critical parameters were within speci-fications, and return production’s request, allowing them touse WFI. Production would then open the valve and use WFIfor processing.
Among the major problems with this approach, a few inparticular standout – (1) The major problem is that if anyCritical Quality Attributes go out of specification duringproduction usage, there is no way of the production controlsystem taking action, until the operator manually stops theWFI flow into the process. In essence, there is no real timequality assurance. (2) Secondly, since the other skids areoblivious to the interactions of this skid with the utilities,there is no room for any sort of predictive control to occur.
WFI Example with PAT and AutomationConcept in PlaceRevisiting the WFI example from a PAT perspective, westarted the deployment of the bandwidth of PAT at the firstlevel, the identification of the CQAs. In the case of WFI, weutilized USP standards for chemistry, and determined TOC,conductivity, and return temperature as critical attributesthat we needed to monitor. The next step was identifyingwhere these measurements needed to be made to ensuresufficient WFI monitoring. Considering line sizes, systemdynamics, etc.; we placed online analyzers (TOC and conduc-
tivity) at the returns of all tanks and loops. Figure 6 shows theplacement of the analyzers.
The next step was to monitor the analyzers. Signals fromthe TOC and conductivity analyzers were sent back to theutilities controller (note that the utilities and productionDCS are now on the same platform). The signals in thisexample were 4-20 mA signals; however, the control systemis compatible with all of the latest bus (i.e., fieldbus, profibus,etc.) technologies that could have been deployed. Alarmlimits were set in the control system at values at or below the“out of tolerance” limits. The alarms were identified as “GMPcritical alarms,” and any out of tolerance alarm was reportedin an integrated reporting package. Here, it is important tonote that we relied on at least two years of operating experi-ence with these online analyzers (particularly TOC), beforewe began to completely rely on them for process decisions(e.g., automated interlocks and reduced sampling).
When production needed WFI, a recipe parameter re-quests it from the utilities controller. Once the controllerdetermines that conductivity, TOC, and return temperature(the CQAs for the utilities systems) are within acceptablelimits, it returns another recipe parameter to the productioncontroller allowing it to open the valve and the productionvalve opens.
For example, if TOC exceeds the acceptable limits, inter-locks automatically close the production valves and preventproduction from using WFI. Real-time quality assurance ofWFI is maintained throughout the entire production process.In addition, the other skids can take mitigative action toaccount for any deviations from the standard values.
The centralized automation concept also supports thegeneration of QA relevant reports, ad-hoc reports and offlineanalyses for the entire WFI system. This achieves the finalaspect of the PAT idea – wherein we identify, monitor,control, and report the CQAs.
The benefits of this approach of the WFI example in ourcase include:
• real-time quality assurance with the constant monitoringof the CQAs.
• WFI management is electronic. Online totalizers allowutilities to send WFI to its most critical users with ad-vanced predictive information on how much will be neededfor that phase of the recipe.
• Strong reporting components allow for multivariate analy-sis.
• Reduced manual testing. TOC and conductivity needs tono longer be tested daily due to constant monitoring of thecentralized control system. This also eliminates the possi-bility of manual errors that occur during the processing oflaboratory samples.
The described online WFI monitoring and control provide thehighest degree of assurance that WFI contamination does notimpact production. In addition, we were able to cut our QCcosts by replacing conductivity and TOC related manualchemistry testing.
The centralized and scalable automation concept providesthe required infrastructure for the PAT based paradigm shiftof our WFI utilization philosophy. All the I/O in the central-ized automation approach is wired directly to the new genera-tion control systems with an integrated historian and report-ing packages. All systems work in a virtual, shared networkin such a way that data from all skids is available on acommon platform. The centralized control system controlsthe entire process, and facilitates the comparison of criticaldata from all skid-based systems on one platform. So now, allthe skid information is available, and if a critical parameterfor WFI would trend toward an unacceptable limit, valvescould be proactively shut off to mitigate the risk of contami-nation.
Figure 6. WFI example with PAT in place.
Another benefit of this approach was that using thisexample, we were able to implement PAT into the site in aphased manner with a quick-win, and thereby build manage-ment confidence. Other opportunities like rapid micro andreal-time endotoxin analyzers are under evaluation thatwould further improve the quality assurance of WFI. Theimportant point here is that as analyzer technology provesitself on site, we add it onto the centralized control system –thereby quality critical data is available on one controlsystem, which facilitates easier deployment of the “control”aspect of PAT.
Fingerprinting of Golden BatchesWithin the scope of our case study, we were able to deploy anonline reporting tool that provides a real-time comparison ofparameters over multiple batches, also referred to as thefingerprinting of “golden batches.” It allows us to compare datawith respect to time, process step or batch-ID over multiplebatches, which greatly facilitates process monitoring andunderstanding. This reporting application fully utilizes thecentralized automation infrastructure we discussed before.
Our interpretation of a “golden batch” is no different thanthe one shared by the rest of the industry – we considered agolden batch to be one that performs ideally with respect tocycle times, yield, and quality. Our goal is to identify charac-teristics of this golden batch, and control the parameters ofinterest so they fit the values of the golden batch. To trulyunderstand all the dynamics of our process, we need tocompare parameters over time, from batch to batch, and thencorrelate these parameters with other parameters to analyzethe various interactions in our process. Having a centralized
ConclusionPAT promises to deliver a new culture to the pharmaceuticalindustry – a culture of innovation. Automation deploymentneeds to support this FDA encouraged effort by providing aninfrastructure that fosters integrated data management anddata analysis abilities across the entire process. This leads toimproved process understanding, which in return allows forthe implementation of additional automation and PAT func-tions where it is determined to be most beneficial for productsafety and yield. The presented concept of scalable automa-tion enables an efficient addition of these automation andPAT functions throughout the entire lifecycle of the facility.
The presented WFI example shows that the application ofPAT and automation leads to increased product safety andefficiency. The “fingerprinting of golden batches” examplegives an exciting outlook on what the potential of today’s PATand automation tools can provide us with.
This article not only focuses on the need to have a suffi-cient automation structure in place, it also emphasizes theimportance of defining and following a structured PAT ap-proach. The FDA guideline on PAT describes a broad frame-work that allows pharmaceutical and biotech companies todefine their approach to PAT considering their product and
automation concept gives us a strategic advantage in achiev-ing this goal; in that we have information regarding all theparameters in one repository, the centralized historian. Whatwe needed to do was find an effective way to utilize the datato help us understand the causes of variability in our process.
The tool we deployed sits on top of the historian and hasaccess to all the batch and continuous data. The tool is a user-configurable tool that allows a user with appropriate privilegesto select the points or the CQAs that the user wants to trendand compare. After navigating a few setup screens, it allowsthe user to select the batches of interest for comparison.
The user can drill down to the phase and unit level granu-larity if a particular CQA needs to be compared from one phaseto the next phase - Figure 7. The net result is that a user cannow obtain trends for that particular CQA across variousbatches/phases/units. The user can then compare this CQAwith other CQAs to determine statistical correlations andobtain a greater understanding of the process. Once the pro-cess is better understood, the process can be automated to becontrolled at the characteristics of the golden or ideal batch.Using this tool and the fact that all the data can be overlaid viaone reporting engine, sources of variability can be understoodand PAT implementation is further enhanced.
process characteristics, as well as their infrastructure andorganization. This article is written as a case study describ-ing approaches and examples that have worked best for ourneeds at Talecris. The intent of this article is to share ourexperiences and to contribute to an industry wide discussion
on PAT. As it is true for the entire pharmaceutical and biotechindustry, PAT at our company is still in the process of beingdefined and structured with some very encouraging and realvalue-adding results.
AbbreviationsCIP Clean-In-PlaceCQA Critical Quality AttributeEP European PharmacopoeiaERP Enterprise Resource PlanningJP Japanese PharmacopoeiaMES Manufacturing Execution SystemPAT Process Analytical TechnologyTOC Total Organic CarbonUF/DF Ultrafiltration/DiafiltrationUSP United States PharmacopoeiaWFI Water For Injection
References1. Talecris Biotherapeutics operates one of the largest sites
dedicated to products that are derived from human bloodplasma.
2. FDA, “Guidance for Industry: PAT – A Framework forInnovation Pharmaceutical Development, Manufacturingand Quality Assurance,” Center of Drug Evaluation andResearch, Food and Drug Administration.
3. FDA, “Pharmaceutical cGMPs for the 21st Century: ARisk-Based Approach,” http://www.fda.gov/cder/gmp/index.htm.
4. Cohen, Nissan, “Validation of a Water System for the 21st
About the AuthorsJoydeep Ganguly is the PAT Group Leadat Talecris Biotherapeutics (formerly BayerCorporation, Plasma Division) in Clayton,North Carolina. He holds an MS in electricalengineering from the University of NotreDame, a BS in electrical engineering fromIndia, and is currently pursuing an MBAfrom North Carolina State University. Prior
to leading the PAT effort for Talecris, Ganguly was the leadcontrol systems engineer for numerous projects at BayerBiological Products Corporation. His areas of expertise in-clude statistical process control, automation, data analysismethodologies for biological processes and analytical instru-mentation. Among other things, he has developed standardsfor his site in the areas of PAT, control system design andimplementation, and advanced reporting packages for dis-tributed control systems. He has previously presented andpublished papers in the areas of fault tolerant control sys-tems, dependable networks, and automation concepts forchromatography systems. He is a member of the Eta KappaNu Electrical and Computer Engineering Honor Society,ISPE, and ISA. He can be contacted via email at:[email protected]
Proposed Stepsto a PAT Implementation
1. Identify: this step includes the process of identifyingan opportunity that would benefit from the PAT ap-proach, as well as identifying the critical quality at-tributes that need to be monitored and controlled in theprocess. For a WFI system, that may be Total OrganicCarbon (TOC) and conductivity; for a chromatographyprocess, it may be pH. Essentially, this process detailsthe critical control points that directly impact productefficacy or quality.
2. Monitor: the next step after identifying the criticalquality attributes would be to monitor them. Monitor-ing is usually achieved using on-line instruments. Re-cent advances in on-line analytical instrumentationhave encouraged more online monitoring of parametersof interest. The simple premise is that we cannotcontrol something we cannot monitor. The monitoringstep allows us to collect data for the CQA of interestand evaluate the effect of adjusting the CQA on theoverall process efficacy.
3. Analyze: the analysis step ensures that once wehave identified our critical quality points and monitoredthem, we employ statistical analysis to determine howthe critical quality attribute is related to the overallprocess efficacy. This step includes the development,verification, and validation of any statistical modelsthat could define the process. Experimental studies,engineering test plans, and retrospective data analysisare methods that we employ to analyze the CQArelationship to the overall process.
4. Control: after we have analyzed the relationshipbetween the CQA and overall process efficacy anddeveloped any statistical models, the next step in thePAT effort would be to control the process to ensure thatthe CQA is within specified limits at all times. This is themost critical step of the PAT roadmap that essentiallyensures that “real-time” quality assurance is met.
5. Report: the last step in the PAT implementationframework in our model is a reporting element. Thereporting element encompasses any tools that aid inassuring that the process was in fact in control through-out the processing period. Reporting tools serve twopurposes – they allow for data to be reported in afashion that aids in developing process understanding,and they allow for any exceptions from the “ideal state”to be documented in the final release records.
Gerrit Vogel is a Senior Engineering Man-ager at Talecris Biotherapeutics (formerlyBayer Corporation, Plasma Division) inClayton, North Carolina. In his current posi-tion, he manages and oversees projects andoperations functions for all control systems,electrical equipment, and instrumentationon site. He also has championed the intro-
duction of centralized automation systems within variousBayer sites to optimize operations and enhance processmonitoring and control. He has more than 15 years of expe-rience in plant and project engineering, coupled with eightyears of experience in engineering management. He is cur-rently the Chair of the PAT Core Team at Talecris, which isresponsible for developing the overall strategic vision for PATwithin the site. His areas of expertise include process auto-mation, computer validation, and analytical technology. Heis a very active member of the Pharmaceutical AutomationRoundtable and ISA. He can be contacted via email at:gerrit.vogel@ talecris.com
Talecris Biotherapeutics, 8368 U.S. 70 West, Clayton,North Carolina 27520.
This articleinvestigates theoptions forupgrading thecontainmentperformance oftraditionalpharmaceuticalfacilities usingalternative (lowcost)containmenttechnologies.
Introduction
Make no mistake, pharmaceutical sci-ences are passing through an excit-ing period of discovery and progress.New drug molecules, often with in-
creased pharmacological potency, are beingdeveloped along with novel delivery systems,all in the hope of finding new cures for humanailments. Even though we are a few stepscloser to the route to the market place, we have
to accept that these new Active PharmaceuticalIngredients (APIs) are going to pose seriouschallenges to our existing manufacturing fa-cilities. Do we look to provide new facilities toaddress these challenges or can we refurbishexisting plants? The ISPE Zurich Seminar (Oc-tober 2003) discussed these issues in detail.
This article highlights the process contain-ment aspects and provides advice on how alter-native containment technologies may be used
to provide additional levels of Safety Health and Environ-ment (SHE) and regulatory compliance (GMP) in instanceswhere new molecules and mature facilities are combined, atleast for the short term.
As with any refurbishment project, the primary issues aresummarized by the four Ps:
• Product (the SHE & GMP implications of the new mol-ecule)
• Process (identify critical activities needing containment)• Procedures (define new SOPs and emergency procedures
to ensure staff safety is not compromised by the newmolecule)
• People (seek “buy in” of the operating teams to developeffective ways of handling the new molecule and its asso-ciated containment devices).
This approach is based on my “from experience” empiricalformulae that attaining GMP/SHE goals split down into:
“device” + “correct operation” + “maintenance”= Attainment of target
The Containment IssuesProduct: Recognizing Risk and HazardA detailed review of the manufacturing process is needed toclearly identify every potent or API material handling taskthat places the operator at risk of exposure. Simplistically,the quantities of material being handled in “open transfer”and how dusty the compound is at this stage of the processmust be understood. The pyramid chart permits risk ofexposure – or – exposure potential to be evaluated accordingto the quantity of powder and dustiness of the powder - Figure1.
The hazard associated with the new API needs the fullinvolvement of the industrial hygiene specialists. The Opera-tor Exposure Limit (OEL) needs to be identified. Also, is therean acceptable Short Term Exposure Limit (STEL) that per-mits occasional tasks to transgress the normal 8-hour timeweighted average exposure?
Several pharmaceutical manufacturers are now adoptinga more flexible, statistical analysis of API handling taskexposure. This method uses nine or more of pumped environ-mental measurement filter heads to evaluate a specific task
and its surroundings in detail. Often referred to as the “3x5rule,” whereby the normal Engineering Control Limit (ECL)may be transgressed by up to three times during a 30-minutetask duration. The exposure of any one of the multiplesamples around the task site can be exceeded by up to fivetimes the ECL for the task duration. Any application withexposure to the API exceeding the 3x5 rule is considered “outof control” and therefore requiring further containment at-tention.
There are other over riding constraints, including whetheror not the compound is a dermal or respiratory sensitizer.These factors may have significant implications to the riskposed to our process operator and indeed may point to a morecontained approach as to how the material is handled in thefacility.
A risk and hazard profiling of every open API transfer taskideally needs to be carried out – this also should consider “nonproduction” aspects such as QA sampling of the compound inadjacent laboratories through to the risk posed to mainte-nance operatives.
Figure 2. Attachment of flexible encloser to a stainless base rail.
Table A. Selecting the containment device according to scale of operation and process frequency.
Small Scale Operation Medium Scale Large Scale
Consider disposable laboratory Consider machine top only enclosure. Consider full M/C enclosure tent.glove bag. Zipper airlocks. Bonded to facility floor.
Zipper airlocks.
Very small applications. Consider machine top only enclosure. Consider full M/C enclosure tent.Consider acrylic mini isolator. Rigid airlock with gasket seals. Bonded to facility floor.PVC flexible isolator with rigid RTP entry/exit.
work platform.Zipper airlocks.
PVC flexible isolator with rigid Consider machine top only enclosure. Cleanability needs suggest conventionalwork platform. RTP material entry/exit. rigid isolator may be necessary.
This procedure, although lengthy, permits the correctlevel of process containment equipment. The next thing toconsider is the equipment to process these new APIs and thetasks the operating teams are expected to perform in normalproduction.
ProcessWe have a facility to refurbish to allow new APIs to behandled. Let’s ask what scale of production are we looking tocontain? Smaller operations at laboratory or pilot plant scaleare easier to enclose in flexible isolators than operationsinvolving the handling of 50Kg containers. A good rule ofthumb is 10Kg batch size maximum for flexible isolation –this can be stretched to 25Kg where special drum supportplatforms are developed as part of the design. Containersmuch above this size really need drum lifters/tippers and asolid interface such as a pouring hopper to ensure a repeat-able and safe powder transfer system. Also at this time, whilereviewing the inherited equipment, we can ascertain the bestmethod of isolation. Some large floor standing machines suchas tablet compression presses or granulators can only beeffectively contained by dropping a full enclosure over themachine. Here an open base enclosure that is sealed to thefloor is a good solution. For processes such as weighing anddispensing or QA sampling where operator interface is inten-sive, an isolated work bench type containment is probably abetter solution. This planning period is the time to considerthe best methodology of entry and exit of materials from thecontained enclosure without loss of containment and nega-tive pressure. Table A lists recommendations on the type ofenclosure together with entry/exit methods.
With regard to full machine enclosures; these can either beequipped with an integral fully sealed PVC ground sheet floor(ideal for where the machine is light and portable), or theenclosure can be open at the base to permit its constructionof the enclosure around the machine or equipment requiringcontainment. The latter method requires careful consider-ation of how to achieve an airtight/liquid tight seal to thefacility floor (or machine top) without creating a permanentanchorage and damaging the machine or facility floor. Onemethod of attachment is illustrated in Figure 2.
Figure 4. Machined acrylic airlock – easier operation than zipperdoors.
Figure 3. Flexible glove bag systems.
Procedures and PeopleThese final two Ps are just as critical as understanding theproduct (new API) and the process. Indeed, because we are ina refurbishment project mode where the containment solu-tions are a retrofit, we need to verify that the operators willwork with our plans. Development of suitable SOPs andoperator training are going to be vital in attaining satisfac-tory levels of containment and operator safety. These topicsare covered later.
The Refurbishment Project:Accommodating a Potent API in an
existing Non-Potent FacilityIn order to get a feel for the key aspects, let’s take a look at atheoretical solid dosage form facility that has been in use for20 years handling “non potent compounds” and now mustadapt to handling a new API with an Operator ExposureLimit (OEL) of around 1.0 microgram/M3. Let’s also be clearthat in this scenario, the facility will be dedicated to handlingthe new compound on a continuous basis.
Key GMP/SHE parameters to address are:
• process specific containment• improved isolation of areas handling the new API• improvements to HVAC system• additional protection to service crews carrying out main-
Figure 6. Lab scale fluid bed in ventilated glove bag.
weigh scales, a support frame and an entry airlock. On thesubject of airlocks, the author has abandoned the use ofzippered access chambers for frequent applications. Experi-ence shows that zips may jam and or be subject to operatorabuse, pointing to a disappointing short operational life cycle.Instead of the zipper door approach, look at the use of rigidPlexiglas/Lexan or even Trespa (phenolic resin) airlock cham-bers with proper hinged doors and gasket seals. These pass inchambers provide a reliable and easy to use airlock thatpermits easy entry of API containers into the handlingchamber, without breech of containment. Built into thisfabrication can be the inlet/outlet HEPA filter housings andfan unit. This permits validatable and replaceable HEPAfilters to be used in glove bag applications. Figure 4 shows thetype of airlock housing being considered.
For API drum and weighed material exit from the han-dling chamber, there are several well established methodolo-gies:
• bag out with continuous liner• Rapid Transfer Port (RTP)• split butterfly valve
The most suitable method will depend on your process driver;however, focussing on lowest cost, we generally look to over-bag empty drum containers exiting the powder handlingchamber. The task is not frequent and therefore not a majorinconvenience to the operators.
Containing Powder Transfer TasksIn our theoretical facility, we need to look at typical openpowder transfer tasks forming essential processing stages,these could include the following:
• QA sampling of the incoming API• powder weighing – sub division• addition of the API to the batch
The facility has for many years operated a number of downflowdispensing booths that have provided satisfactory contain-ment for non-potent products. Operator exposure monitoringwithin these booths suggests exposure levels of around 100+µg/m3 (dependant on operator practice). This renders thebooths totally unsuitable for the new incoming API (1.0µ/m3)that will arrive in a 10Kg fiber drum. The small quantity ofthe new API is a key factor in selecting our upgrade route.
The frequency of weighing the new API at just a few kilosevery batch cycle suggests, at least at this stage, that asophisticated rigid stainless steel isolator may be overkill.Containment needs:
• barrier isolation between operator and new API• negative pressure enclosure to alleviate API migration• ability to pass in API drums and pass out weighed without
loss of containment• mobile to permit storage when not in use
These features can be brought together in a relatively low costglove bag type system as illustrated in Figure 3.
Of course the absolute lowest cost containment is the allPVC or Urethane envelope. The light plastic glove ports/sleeves allow for easy manipulation. However, to provide aworkable and easy to use device, I have tended to differenti-ate between “hardware” and “software.” The plastic envelopeshould be designed for the lowest cost of manufacture so itsdisposal on an occasional (campaign or batch basis) is notprohibitive.
To support the minimalist “software” with durable “hard-ware,” this could be a rigid stainless work surface to carry
Figure 7. Microwave granulator in ventilated glove bag.
This combination of lowest cost “software” and durablefunctional “hardware” is seen as providing the optimumrelationship between ease of use, ease of decontamination,and low price. The design can be adapted to work for QAsampling, dispensing, or batch additions – all by varying thedesign of the base plate and envelope.
Will this type of design attain the 1.0 microgram targetmaximum operator exposure? Early user feedback suggestsit will.
Operating at a negative pressure of 50pa the devices havebeen tested using placebo materials with the following proce-dural constraints:
• clean uncontaminated API drums entered via pass throughdoor
• API bag out using continuous liner is carefully set up toavoid gross contamination of API powder in sacrificialliner bag
Difficult Containment Upgrade ApplicationsWith small scale manual operations such as powder dispens-ing or sampling, we have the freedom to tailor the solutionaround the operator and process. Yet, other areas of themanufacturing process are often not as easy to contain via lowcost methods.
Take for example the IBCs used for intermediate batchtransfer of process powders and granulation lots.
Here, it is easy to foresee the escape of small quantities ofdust at the IBC loading and docking stations. As our theoreti-cal plant wishes to introduce APIs, these once acceptableemission levels (pre potent materials) could now easily breechthe maximum Operator Exposure Target (OEL) when consid-ering the newly introduced API.
How do we move forward? Should we consider the “engi-neered solution” and look at the wholesale conversion of 100sof existing IBCs to split butterfly valve operation? This is anestablished yet, costly solution.
Many plant managers would ask, can a containmentupgrade be attained without this big investment step?
Some split butterfly valve manufacturers are lookingtoward a disposable docking solution in the form of a flexibledocking flange system. The flexible docking flange takes theengineered split butterfly valve forward into a mass producedlow cost disposable product. The smaller flexible dockingflange is aimed at small scale powder handling (up to 20Kg),and may when fully developed, provide the ideal low costdocking system for glove bag type isolators. At the larger endof the product handling scale, is the flexible docking flangethat is a collapsible flange that can either lie folded up andeffectively flat (powder port closed) – or unfold – to form a350mm Sq powder transfer opening (powder port open). Justas with precision engineered RTPs, the flexible dockingflange (snap) dock together to form a contamination freemake/break connection surface. When the large powder portof the flexible docking flange is open conventional IBC dis-charge spigots can be located inside the flexible dockingflange. This means that our “marginally leaky” IBC dis-
charge can now be contained by an outer flexible sleeve withflexible docking flanges. With this concept, docking onto thedischarge station is possible – with containment provided bythe clean make/break joint flange as illustrated in Figure 5.This developing technology should see the inclusion of lowcost containment upgrades on an IBC/FIBC transfer system.
Large Scale Equipment InterfacesWith the introduction of APIs into the manufacturing pro-cess, some of the greatest containment problems are seen inthe granulation suite where high shear mixers and fluid beddryers are in use. Many of these devices may have beenoperated with open powder transfer procedures that cannotbe tolerated when more potent APIs are introduced. Depend-ing on the scale of the operation and the process validationthat exists, we can provide a “containment upgrade” byenclosure methods or alternatively larger scale processesmay elect to by-pass containment enclosures in favor of aclosed vacuum transfer system. The latter may have thedisadvantage of necessitating engineering changes to thegranulation equipment, but if this can be accommodated, theprocess will benefit significantly from automated powdertransfer. Several experienced vendors exist in the US andEurope, hence, in this article I will look at the smaller scale“enclosable” operations. High shear mixers and “one-pot”processors have been successfully isolated within glove bagtype enclosures, as shown in Figures 6 and 7.
Larger scale equipment such as fluid bed dryers can becontained utilizing polythene liner “bag tricks” to containremoval of the dryer bowl from the main dryer body; however,the considerable bag manipulation needed by the operatorsmay render this method risky for long term operation.
The key point to consider when looking to contain such
Table B. Validating your containment equipment selection within your organization.
InputChemical Industrial Safety and Operations Maintenance Quality EquipmentSupplier Hygiene Health Group Group Assurance System
Elements Validation Designer
Hazard Group
Scale of Operation
Exposure Potential
Frequency and Task Duration
Operability of Device
Cost of Device
Figure 8. Typical operator training slide.
large items of process plant is operability and benchmarkedcontainment levels. An engineering study with user groupinput to the development of any soft-wall containment solu-tion is a sensible approach.
Operator Training For Flexible Isolation SystemsThe cost saving and ease of installation benefits gained byusing flexible containment solutions in a refurbishmentproject must be considered against the risks and hazards thatcould be created by incorrect operation of a 0.3mm thick PVCbarrier. There is a strong case for extensive operator/usergroup involvement with the flexible containment equipmentfrom the early design stage right through to installation.After this milestone, operator training needs to emphasizesome of the golden rules of this technology as illustrated inFigure 8, and defined as the following:
• Always pre-load the flexible isolator with the cleaning anddecontamination materials required to make it safe at theend of the campaign.
• Always clean the flexible isolator as soon as the operationends.
• Never use the flexible isolator unless design negativepressure is attained.
• Develop safe material entry and particularly safe mate-rial/waste exit routes and procedures.
Other Refurbishment Project Considerations:Segregation of Production Areas (EnvironmentContainment)As we have seen, an increasing number of “low cost” andconventional containment devices now exist to create aneffective process facility upgrade to ensure safe containmentat critical powder handling stages during manufacture. Anycontainment device must be seen as a first line of defense, sowhen considering plant safety and environment separation,how do we create improved isolation between API handlingareas and general work areas?
In our theoretical plant, we are considering the permanentintroduction of an API compound for a dedicated manufactur-ing process. I think it’s worth considering basic solutions forplants where APIs may be introduced on an occasional basis.Here, temporary segregation between work environments
Figure 10. Plate glass doors a facility upgrade option.
Figure 9. Temporary flexible enclosures: (a) Temporarycontainment enclosure for Comil (b) Walk in enclosure for activepowder campaign.
may be required and the use of operator PPE may be anacceptable short term “fix” to get the required batch pro-cessed.
I say “fix” as in reality no respectable pharmaceuticalmanufacturer should be relying on PPE when so manycontainment solutions are available.
Nevertheless, at the most basic level, the use of weldedPVC or Urethane material may be used to form a temporarywork chamber with airlocks to segregate material entry/exitroutes from personnel entry/exit routes. Fogging or mistingshowers can be accommodated where emergency decontami-nation is viewed as a possible requirement. As with theirsmaller glove bag siblings, these low cost processing enclo-sures cannot be considered as a permanent durable solution.However, for the facility considering the occasional process-ing of API materials, their use may be acceptable. Twoexamples of the use of such enclosures are illustrated inFigure 9.
Environmental SegregationWith the permanent handling of APIs within the maturefacility, thought should be given to improvements to air lockpressure cascades, personnel, and materials flow to avoidcross contamination and to alleviate migration of the API into“safe” areas of the plant. In this area, facility door designshave seen major advances in the last 10 years. Doors are nowavailable designed specifically for pharmaceutical facilitieswhere cleanability and effective door seals are highly desir-able. Key features to look for are flush and crevice freesurfaces, flush glazed vision panels where needed, and someof the better designs feature door closer actuators concealedin the doorframe. While stainless steel doors have long beenconsidered the norm, there is a growing acceptance of plateglass doors in pharmaceutical facilities. A typical example ofa glass door in a pharmaceutical facility is shown in Figure10. Glass doors are often significantly cheaper than stainlesssteel items, enhance the visual appearance of the upgradedfacility, and add a touch of bold design flare. The downside ofplate glass doors is obviously breakages, but it seems that theplant operators treat glass doors with much more respect.
Where migration of API between areas is totally unaccept-able, the use of gas tight door sets with inflating seals, asshown in Figure 11, is becoming more popular.
Gas tight doors are now being applied to potent handlingfacilities – in the same way that leak proof vision panels arespecified on isolator systems.
House Exhaust – Lev SystemsIn this age of high containment solutions, the house exhaustsystem that extracts dust-laden air from compression ma-chines, packing lines, and a myriad of other operations is stillan object of contempt. It is very surprising to see all mannerof “low tech” filtration systems placed in either the roof topplant space or in segregated exhaust filter rooms for serviceby technicians protected by PPE.
Not only are we creating a potential “time bomb” of APIcontaminated ductwork and dust filter housings, we areputting the service crew at risk.
Believe it or not, the unfashionable technology of localexhaust dust filtration has moved on. Not only has thefiltration efficiency of the best devices moved up to nanogramlevel (guaranteed by some vendors), the entire service workroutine has been re-designed to be totally contained. Thesenew generation dust collectors permit designers to challengeconvention regarding the traditional location for dust filtra-tion technology. Tradition means “plant room” and all col-lected material “waste.” The latest generation of pharmaceu-tical dust collectors may be set up to permit hygienic dust(process waste) recovery. These exhaust filter devices willutilise crevice free and highly polished powder contact parts.The collection hoppers are designed for vacuum transfer ofcollected powder recovery. Hence, what was traditionallyconsidered waste may now be considered for either re-intro-duction to the process where regulatory authorities permit.Alternatively, the collected powder may be easily and cleanly
Figure 11. Gas tight door with inflatable seal technology.
weighed for accurate calculation of in processing losses.Some pharmaceutical companies are taking advantage of
these new generation high efficiency dust filtration systemsand are locating the filtration devices directly into the pro-duction area. The benefit of placing the filtration plant in theproduction zone is much shorter (often demountable forcleaning) exhaust duct lines. Powder recovery or waste mea-surement is performed directly by the production staff. Manyengineers will be appalled at the thought of a “dust collector”handling potentially explosive dusts being located in a pro-duction zone. Surely, these devices create a lethal explosionhazard. Not so, just as we see large scale fluid bed driers nowfabricated to 10 or 12Bar pressure rating – so too are thesenew dust filtration devices built to these standards.
• dust free filter service and dust recovery – no risk to workroom environment or service crew
• wash in place clean down technology available• can be located in a production zone – filter housings fully
pressure rated and smooth cGMP fascia for architecturalintegration
The Way ForwardBefore any decisions are made about the containment add onequipment, the issues must be clarified and the operatorsmust support the decision. Down stream of this in what wemay call the detail design or engineering phase you will seegreat benefit in having prototype or mock-up devices deliv-ered for the user group to critique. A well designed prototypewill often permit preliminary performance testing, wherebythe ability to attain the required levels of operator protectioncan be benchmarked.
While nobody can feel comfortable developing serial num-ber 001 for any device in a production facility, what we arehoping to achieve in a facility containment upgrade is notrocket science and often relies on a variation of proventechnology. As a safeguard to ensure the selected equipment/designs are up to the job, the selection criteria must be
validated. The data in Table B suggests interface with thevarious groups responsible over the life cycle of the project.
ConclusionIn this article, some of the new approaches to upgrade a“standard” facility to a level capable of handling potentcompounds and new APIs safely were presented.
• Product (the SHE and GMP implications of the newmolecule)
• Process (identify critical activities needing containment)• Procedures (define new SOPs and emergency procedures
to ensure staff safety is not compromised by the newmolecule)
• People (seek “buy in” of the operating teams to developeffective ways of handling the new molecule and its asso-ciated containment devices.
In conclusion, my optimistic view is that the growing need forfacility upgrades to standards capable of handling APIs neednot be prohibitively expensive.
However, the key issue when introducing alternativecontainment technologies is to get the buy in from the opera-tors and user group. They then understand the risks posed bythe new compounds, they appreciate the efforts being madeto assure their safety, and they work with us rather thanagainst us in attaining reliable and safe working conditions.
References1. Turner, K.: AstraZeneca UK – Refurbishment as a catalyst
for improvement. ISPE Zurich seminar October 2003.2. Dayre, S.W.: AstraZeneca France – Renovation of phar-
maceutical facilities – maintaining compliant operations.ISPE Zurich seminar October 2003.
3. Hirst, Brocklebank, and Ryder. Institution of ChemicalEngineers UK – Containment Technology a Design Guide.2001.
4. Andrasfalvy, K.: Fette-Absolut gmbh. Fine 04 SymposiumStockholm January 2004.
About the AuthorMartyn Ryder was educated at Leeds Uni-versity in HVAC and Building Services. Ryderwas a founder of Extract Technology in 1981,now part of the CPS Pharma Group of compa-nies with operations in the US, Europe, andAsia. CPS Extract has grown to be one of theworld’s leading suppliers of pharmaceuticalpowder handling related containment sys-
tems, currently exporting more than 70% of its productioncapacity to major manufacturing sites around the world.Ryder now provides support to the CPS Pharma Group on anexclusive consultancy basis and shares his time between hisCPS duties and farming.
On-line Particle Counters ProvideDetection and Control in Water forInjection (WFI) and Purified Water(PW) Systemsby Dr. Hans-Walter Motzkus and Joe Gecsey
This case studypresents earlycontaminationdetection in anestablished WFIand PW systemthrough the useof on-lineparticlecounters.
Creating an effective water system toprovide Purified Water (PW) or Waterfor Injection (WFI) in a life scienceapplication is a careful adjustment of
design, materials, monitoring, and mainte-nance. By convention, certain parameters mustbe monitored in a WFI system: endotoxin lev-els, conductivity, TOC, and microbial (CFU)values.1,2 In the EU, other parameters thatmust be monitored are nitrates and heavymetals.3 It has been common in the past toobtain grab samples of the water at a pre-defined frequency and to later perform a labo-ratory analysis for the key parameters. In re-cent times, some of the instrumentation forthese parameters, for example, TOC, has al-lowed continuous on-line measurement to beaccomplished. Determination and monitoring
of particulate levels in WFI systems are notrequired by regulation, but can provide anenhanced degree of control, as described in thisarticle, resulting in early detection of potentialbreaches in the integrity of the water system.
The water system in Building S166 at theSchering facility in Berlin provides PW to anoral dosage forms manufacturing area and PWand WFI to a parenteral area for manufactur-ing clinical trial supplies. In 2000, an on-lineparticle counter was installed on the WFI sys-tem - Figure 1. The instrument was connectedto monitoring software that provides watersystem information from three different build-ings - Figure 2. The software includes opera-tional details of various parameters such as thesystem temperature and pressure in the vari-ous piping systems on the campus. Based on
the performance of the ini-tial analytical monitoringsystem on the WFI system, asecond particle counting sys-tem was installed in 2005 onthe PW system in BuildingS166.
As with most WFI sys-tems, the water is circulatedin a continuous flow loop atan elevated temperature ofat least 80°C (176°F) to ex-clude bacterial growth. Nei-ther the pH sensor nor theparticle counter are designedto operate in this tempera-ture range so a heat exchangesystem is positioned up-
stream from the counter to reduce the temperature of thesampled water to approximately 35°C (96°F) - Figure 3. Thecooled water flows through the particle counter with the flowcontrolled to 60 milliliters per minute (0.95 gallon/hour) andthen through the pH meter. It is then sent to drain. Sendingthe sampled water to drain avoids potential introduction intothe WFI system of chemical contaminants such as the re-agents used in the pH sensor.
At this time, alarm indications in the monitoring softwareare provided only for conductivity and TOC pending estab-lishment of appropriate limits for the pH and particlesthrough further monitoring.
Conductivity, TOC, and temperature are measured di-rectly on the heated loop. Separate temperature transducersare used to monitor the temperature in the main loop and inthe sample line following the heat exchange unit. Down-stream from the particle sensor is a flow controller, followedby the pH sensor. The output of the pH sensor goes to drain.
Particle DetectionThe sensor provides detection of particles starting at approxi-mately 2 µm through 400 µm. The counter updates theparticle count data each minute in each of four size channels:
≥5, ≥10, ≥25, and ≥50 µm. The data is shown on the monitor-ing system screen in terms of “counts per 60 ml” in each of thefour size ranges (channels). During steady-state operation,some counts are observed in the first two size channels (≥5and ≥10 µm), usually less than 100 counts per 60 millilitersat 5 microns. Counts are rarely seen in the upper two sizeranges (≥25 and ≥50 µm) during steady-state operation -Figure 4.
Observed count values of particles with this system arewell below 1% of the USP limits [25 counts per mL at ≥10microns and 3 counts per ml at ≥25 microns] for Large VolumeInjectables (LVI). Care has been taken to monitor not onlypeaks in particle count values, but also changes in thebaseline readings. Peak readings convey information onsystem performance that previously were never recordedwith grab sample methods at Schering due to their once-per-week schedule. Today, the consistent and minute-by-minutedata readings of the on-line system permit relevant data to begathered for trend analysis and even SPC methods in thefuture.
Because the system is still regarded as being in theinvestigative stage regarding particle counts, alert and ac-tion levels have not been set. Particle count changes are
Figure 2. Overview of DigiPlan FMC-DAS software architecture.
monitored by the Quality Group, evaluated, and then re-ported and discussed with both the Production Group fromClinical Supplies and Engineering. Elevated readings arenow used to initiate maintenance and repair of the systemsin cooperation with the Engineering Department.
Detected Particulate Matterin the WFI System
One of the major reasons for investing in an on-line particu-late monitoring system was the significant variance in thedata obtained from the hand-drawn grab samples Scheringhad previously employed. Often, these variances were falsepositives due to sample handling. In one incident, Ramanmicroscopy was successfully used to determine that theprobable source of high readings of particles larger than 5microns in the grab samples was particles generated from apolyethylene component of the screw-on caps of the jars usedfor the grab samples.
The on-line system avoids false positives caused by manualsample gathering technique and materials. Note that samplepoint valves are another notorious generator of particles andpotential false positives; grab samples should be drawn onlyafter waiting long enough for the particles generated by theopening of the valve to be flushed out of the sample port.
A further impetus for the on-line system installation wasthe reduction of manpower needed to obtain samples from thewater system. Previously grab sampling was carried out on aweekly basis and the results of the sampling were delayed bythe time needed for the laboratory analysis. The on-linemonitoring system now provides current data minute-by-minute at an estimated annual savings of at least one man-month of labor.
Within Schering’s water system, high-purity membranevalves are used to control flows. The membranes of thesevalves are exposed to the circulating WFI. Particle sheddingfrom compromised membranes will therefore also show up asbaseline rise in the detected particle levels. A significantimprovement in the valve system was triggered by the detec-tion of black particles in the WFI; it was discovered that theinitial EDPM membrane material was deteriorating more
Figure 3. Instrumentation panel with HIAC HRLD sensor.
Figure 4. Chart with typical steady-state or “normal” values.
quickly than expected in the hot WFI and clean steamsystems. Changing the contact surface to a Teflon-coatedEDPM material and establishing scheduled annual or bi-annual replacement [depending on use and application] ofthese membranes eliminated this particle source.
In another incident, particles recovered from the systemfollowing detection of increased counts were determinedthrough Atomic Absorption Spectroscopy (AAS) to be metallicparticles from the piping system itself. Recent constructionand repairs to the piping system were identified as theprobable cause. After the data was brought to the attention ofthe maintenance group and contractors, procedures wereinitiated both to reduce the amount of particulate mattergenerated during maintenance and to more effectively cleanrepaired sections before they were re-connected to the sys-tem.
High Flow Demand TriggersParticulate Release
An on-line system can show events that grab sampling wouldbe unable to detect; on-line systems also permit the possibil-ity of correlation to other parameters of the system in orderto help establish the root cause of the abnormal readings.
Figure 4 shows the normal, steady-state values. Comparethis to the screen capture Figure 5 that recorded episodes ofelevated counts following both a sudden reduction of the levelof water in the WFI holding tank by high rapid demand andalso two successive events of rapid refilling action by the still.
The red line in the graphic display represents the outputof the level detector for the WFI tank. The supply system(WFI still) works to keep the tank from dropping belowapproximately 70% of 3000 liters (793 gallons). Upon reach-ing the trigger value, the supply system adds water from thestill and brings the tank level back to a nominal “full” valueof approximately 74%.
The green trace shows the particle counts obtained in thefirst channel (≥5 µm) of the particle counter. Although somecounts are present during normal circulation, the rapid dropin level (marked as 1) caused the detected counts to jumpsignificantly.
Figure 6. Comparison of outputs from particle counter, TOC, pH,and conductivity during a particle spike event.
Figure 5. Particle surge as result of rapid drop in level of WFIholding tank.
The large drop in the tank level could have been caused bya cleaning operation in the main facility or, for instance, theinitial high demand of a vial washing machine.
Note that there also are significant increases in particlecounts when there are rapid, but lower volume demands on theWFI system, causing a rapid activation of the refilling system.In these cases (marked as 2 and 3), the refilling system had thecapacity to return the holding tank to its “full” state, but thedemand for WFI was maintained at a slow, but steady rate sothat that the refilling cycle occurred at a high frequency. Thisactivity also was accompanied by high counts. It has beendetermined that these spikes of counts are the result of theturbulence in the holding tank caused by the refilling action.Trapped particles or sediment on the bottom of the tank couldbe released into the WFI circulation - Figure 6.
This tank and the general piping system go through ascheduled maintenance including annual cleaning. If thisannual cleaning event is delayed, the spikes of particlestriggered by the high demand/refilling cycle increase inamplitude and continue until the cleaning is undertaken.Cleaning of the system causes a definite reduction of thebaseline counts and of the spikes triggered by the refillingaction. By analyzing the debris material removed from thetank during cleaning, it is expected that further improve-ments to the materials and the operation of the water sys-tems will be made.
By comparison, the responses of the other instruments onthe system during a refilling spike are not at all as strong asthe response of the particle counter. In Figure 6, the greentrace again represents the first size channel of the particlecounter. The red trace is the tank level, the yellow one is theoutput of the conductivity sensor, the magenta one is the TOCdata, and the blue is the pH value. The reaction of the othersensors is very muted compared to the very observableresponse of the particle counter clearly detecting the in-creased particle concentration.
The reaction of the particle counter correlates preciselywith the rapid depletion of the holding tank contents.
The particle counter has become an excellent indicator of
the resident particulate content of the tank and has been usedto determine the frequency of the cleaning cycle necessary forthis part of the WFI system, as well as a superior indicator ofunexpected increases in baseline particle count levels.
Because of the positive experiences with the automatedWFI analysis system, Schering AG has now equipped a PWloop with a similar system of sensors including particlemeasurement.
Evolution of the WFI Monitoring SystemAt this time, daily review of the on-line monitored param-eters is conducted as a joint session attended by Qualitypersonnel, the Production personnel for Clinical Supplies,and Engineering personnel during which any observationsare reported and discussed. This investigative phase willcontinue through 2006, after which it is likely that automaticalarms will be initiated; SPC limits also may be imple-mented. Alert and action levels will be decided based on theongoing observations of the baseline values and uniqueevents that occur. Analysis of any sediment collected duringannual tank maintenance cycles will lead to further improve-ments in the system and its operation.
On-Line Monitoring and theFDA’s PAT Initiative
The increasing movement toward on-line sensing technologyand automation in the pharmaceutical industry has thesupport of the main regulatory agencies worldwide. Thegrowing interest in Process Analytical Technology (PAT) isdue to the potential for improved control of the process,improved product quality, and shortened analysis times forIn-Process Controls (IPCs). Minute-to-minute monitoring ofthe key utilities in support of production – such as WFI – canaid in the assurance of product quality and rapid release ofproduct because there is more complete data to back up adecision for product release. Particle counters can assistplant operators in maintaining and assuring an appropriatequality of water for use as a rinsing agent and as an ingredi-ent in a product.
The FDA has provided encouragement and support for theconversion of existing processes into PAT processes. Datacollected can at first be regarded as “research data” until theuser has decided that the additional method is sufficientlyrobust.4 This allows a PAT process to be developed by theaddition of suitable analytical and control processes of whichthis might be one.
ConclusionReal-time sampling helps avoid errors common in manualsampling, such as compromised cleanliness of glassware,contamination generated when sample point valves areopened, and contamination contributed by the screw caps ofthe sample jars.
If a product contamination event were to occur, the moni-toring details from the WFI system would allow investigatorsto quickly determine if the WFI was a contributing elementto the contamination. Generally, the root cause for a contami-nation event will be found to be something other than thequality of the WFI system, but with the continuous monitor-ing data available, investigators can be on solid scientificground when they eliminate the WFI system as a suspect.And there is an economic benefit due to the reduction of laborrequired to obtain and analyze grab samples, in this case,estimated to be more than one man-month annually.
Particle counters can be placed in key positions of criticalsystems such as PW and WFI systems to assist in developingmaintenance cycles and as “watchdog” instrumentation tomonitor the continuing stability of the system. This technol-ogy, based on stable, field-proven light-extinction sensors,can be an affordable adjunct to existing monitoring systemsby providing a highly sensitive and real-time reaction toperturbations in water systems.
References1. United States Pharmacopoeia Twenty-eighth Revision
(2005), U.S. Pharmacopoeia Convention, Inc. Rockville,MD.
2. European Pharmacopoeia Fifth Edition 2005. Council ofEurope, Strasbourg.
3. Santoro, M. and Maini C. Which water for pharmaceuticaluse? Eur J Parenteral Sciences 2003; 8 (1):15-20.
4. Heinze, C.L., and Hansen, J.R., “Implementing PAT Stepby Step as a Process Optimization Tool,” PharmaceuticalEngineering, May/June 2005, Vol. 25, No. 3, pp. 8-16.
About the AuthorsDr. Hans-Walter Motzkus studied phar-macy and received his PhD in pharmaceuti-cal chemistry from the technical universityin Braunscheig, Germany. After five years inpharmaceutical technology with a mid sizepharmaceutical company in Salzgitter, Ger-many, he moved in 1989 to Schering AG,Germany, where he is responsible for the
GMP monitoring of the pharmaceutical development areas.He has recently co-authored articles on the use of ramanscattering for microbial identification. He can be contacted bytelephone: +49-3046815827 or by e-mail at:HansWalter.Motzkus@ Schering.de
Joe Gecsey is the Life Sciences ApplicationManager at Hach Ultra in Grants Pass, Or-egon, US. He is responsible for tracking regu-latory changes regarding particulate count-ing in the life science industry. He has con-ducted seminars throughout the world onparticle counter design and applications. Hereceived a BS from the University of Califor-
nia in 1974 and has been employed as an engineer andtechnical advisor by Hach Ultra (previously Met One) since1984. He can be reached by telephone at: +1-541/472-6526 orby e-mail: [email protected].
Hach Ultra, 481 California Ave., Grants Pass, Oregon97526.
Disclaimer:The viewspresented in thisinterview arethose of theinterviewee andshould not beunderstood orquoted as beingmade on behalfof the EMEAand/or itsscientificcommittees. Personal
Background
Q How long haveyou been in
your current posi-tion? What experi-ences prepared youfor this and what isyour educationalbackground?
A I have been in my current position since1 July 2002. Prior to this, I spent nearly
four years working in the pharmaceutical unitin the European Commission covering interna-tional issues (ICH, enlargement, mutual rec-ognition agreements on GMP), pharmacovigi-lance, orphan medicines, and GMP and qualitymatters. From 1992 until 1998, I worked forthe European pharmaceutical industry asso-ciation (EFPIA) as manager of scientific andregulatory affairs, and before this, I spent sometime working in Ireland as an R&D managerfor a pharmaceutical development companyand as a pharmaceutical assessor in the Irishdrug regulatory authority. I qualified as a phar-macist, did a post-graduate degree in pharma-ceutical chemistry, and subsequently a mas-ters degree in business administration. I be-lieve that the combination of my formal educa-tion and my varied work experiences have beenexcellent preparation for my current position,as they have provided me with a combination oftechnical, legal, and regulatory skills within amulticultural European environment.
Q What has been your most fulfilling roleso far in your career?
A I have been very fulfilled in all of myEuropean positions to date, but for differ-
ent reasons. I love the challenge of “helping to
make Europe work,” and achieving solutionsthrough working together with people fromdifferent nationalities.
Q What kind of activities do you enjoy inyour free time?
A In my spare time, I try to spend as muchtime as possible with my 11-year-old
twins, and also enjoy swimming, reading andwalking.
Agency Background
Q How, why, and when was EMEA founded?
A European Medicines Agency (EMEA) is adecentralized body of the European Union
with headquarters in London.Its main responsibility is the protection and
promotion of public and animal health throughthe evaluation and supervision of medicines forhuman and veterinary use. The EMEA coordi-nates the evaluation and supervision of medici-nal products throughout the European Union.The Agency brings together the scientific re-sources of the 25 EU Member States in a net-work of 42 national competent authorities. Itcooperates closely with international partners,reinforcing the EU contribution to global har-monization.
The EMEA is headed by the Executive Di-rector and has a secretariat of about 400 staffmembers in 2005. The Management Board isthe supervisory body of the EMEA, responsible,in particular, for budgetary matters.
The EMEA began its activities in 1995 whenthe European system for authorizing medicinalproducts was introduced, providing for a cen-tralized and a mutual recognition procedure.The EMEA has a role in both, but is primarilyinvolved in the centralized procedure. Where
PHARMACEUTICAL ENGINEERING InterviewsEmer Cooke, Head of Sector,Inspections, Veterinary Medicines andInspections, EMEA
the centralized procedure is used, com-panies submit one single marketingauthorization application to the EMEA.A single evaluation is carried outthrough the Committee for MedicinalProducts for Human Use (CHMP) orthe Committee for Medicinal Productsfor Veterinary Use (CVMP). If the rel-evant Committee concludes that qual-ity, safety, and efficacy of the medici-nal product is sufficiently proven, itadopts a positive opinion. This is sentto the Commission to be transformedinto a single market authorization validfor the whole of the European Union.
In 2001, the Committee on OrphanMedicinal Products (COMP) was estab-lished, charged with reviewing desig-nation applications from persons or com-panies who intend to develop medicinesfor rare diseases (orphan drugs). TheCommittee on Herbal Medicinal Prod-ucts (HMPC) was established in 2004and provides scientific opinions on tra-ditional herbal medicines.
A network of some 3,500 Europeanexperts underpins the scientific workof the EMEA and its committees.
Q Could you please give us an overview of the responsibilities and
activities in your organization? Howlarge is your staff and what are theirqualifications?
A As head of the inspections sector,I am responsible for a team of
about 18. Although some colleagueswere “inspectors” in previous positions,EMEA staff does not actually performinspections; this is the responsibility ofthe European inspectorates in each ofthe 25 member states. Of the 18 staff inthe sector, eight are support staff, onewith a financial degree. The remainderof the staff have degrees in the lifesciences or pharmacy (four) with arange of post-graduate experience from6 to 19 years.
My primary responsibility is to man-age my staff to ensure that the objec-tives and performance measures out-lined in EMEA’s work program areachieved effectively, and that work onimplementing provisions outlined inEMEA’s roadmap to 2010 is begun.
The tasks of EMEA’s inspection sec-tor include the following:
• coordinating any product defects andassociated follow-up and/or recalls
• chairing and providing technical,scientific, and administrative sup-port to quarterly meetings of in-spectors (GMP, GCP) from all EUmember states
• providing technical, scientific, andadministrative support to the jointCHMP/CVMP quality working party
• establishing and providing secre-tariat to the EMEA Process Ana-lytical Technology Team
• development/improvement of theGMP Guide and the Compilation ofCommunity Procedures on Inspec-tions and Exchange of Informationin partnership with the EuropeanCommission
• implementation of Mutual Recogni-tion Agreements (MRAs) on GMP
• coordinating the sampling and test-ing of centrally authorized products
• producing certificates of medicinalproducts in line with WHO recom-mendations
• The sector is also, in close collabora-tion with other EMEA sectors, re-sponsible for the EudraCT databaseon clinical trials and the establish-ment of a European database onmanufacturing authorizations andGMP certificates (EudraGMP).
Q What are your current key priori-ties?
A Our current key priorities reflectour core business and are out-
lined in the inspections’ chapter of theEMEA work program for 2005.
The major priority for 2005 was toprepare for the implementation of thepharmaceutical legislative review, inparticular the new requirements forGMP for starting materials and thesetting-up of a database on manufac-turing authorizations and GMP certifi-cates.
Other 2005 objectives included:
• Support the implementation activi-ties relating to GCP inspectionsunder the Clinical Trials Directive
2001/20/EC for human medicinesand the directive on GCP, in par-ticular the implementation of thesecond phase of the EudraCT data-base.
• Support the European contributionto international discussions onGMP/quality systems in coopera-tion with the FDA and within theICH and VICH framework.
• Coordinate activities in the contextof the joint audit program for GMPinspectorates to ensure mainte-nance of consistent quality stan-dards and harmonized approaches.
• Work on implementation of mutualrecognition agreements is expectedto move toward consolidation as allagreements with the exception ofthat with the US, become fully op-erational.
• Completion of the internal evalua-tion work with new Member Statesin the context of the Canadian mu-tual recognition agreement.
• Coordinate and manage effectivelythe requests for GMP, GCP,pharmacovigilance, and GLP inspec-tions relating to applications forproducts through the centralizedprocedure within the timeframe laiddown in Community law and to thestandards required by the Agency’squality management system.
• Implement an action plan for revi-sion of the sampling and testingprogram for centrally authorizedproducts in cooperation with EDQMto streamline activities and focusresources taking a risk-based ap-proach. Improve general transpar-ency and communication betweenall stakeholders.
• Provide support to all 25 MemberStates to optimize compliance withCommunity requirements in rela-tion to GMP and GCP andPharmacovigilance, and cooperateon planning initiatives to secure theallocation of sufficient resources forthe conduct of inspections through-out the EU and in third countries.
Priorities for 2006 are due to be pub-lished shortly.
Q How is your organization funded?How are the funds allocated?
A EMEA is funded by a combina-tion of fees for scientific services
(marketing authorization applications,variations, inspections, annual fees,etc.), and a contribution from the Com-munity budget. The fee income is usedto cover payments to Member Statesfor scientific services and the coordina-tion work done by the EMEA on prod-uct related matters. The contributionfrom the community budget coversmuch of the coordination work of theEMEA on matters such aspharmacovigilance, harmonization ac-tivities, telematics projects etc. A spe-cial fund is provided to support work onorphan medicinal products, i.e., in-tended for the treatment of rare dis-eases.
Q What is the mission for EMEA?What do you see as the chal-
lenges or barriers to achieving EMEA’sgoals?
A Perhaps the best way to answerthis question is to quote EMEA’s
mission statement:
“The EMEA’s Mission Statement is, inthe context of a continuing globaliza-tion, to protect and promote public andanimal health by developing efficientand transparent procedures to allowrapid access by users to safe and effec-tive innovative medicines and to ge-neric and non-prescription medicinesthrough a single European marketingauthorization, controlling the safety ofmedicines for humans and animals, inparticular through a pharmacovigi-lance network and the establishment ofsafe limits for residues in food-produc-ing animals, facilitating innovation andstimulating research, hence contribut-ing to the competitiveness of EU-basedpharmaceutical industry, and mobiliz-ing and coordinating scientific resourcesfrom throughout the EU to provide high-quality evaluation of medicinal prod-ucts, to advise on research and develop-ment programs, to perform inspectionsfor ensuring fundamental GXP* provi-sions are consistently achieved, and toprovide useful and clear information tousers and healthcare professionals.”
* GXP means Good Clinical Practice(GCP), Good Manufacturing practice(GMP), and Good Laboratory Practice(GLP) collectively.
Q What is your role and involve-ment in international harmoni-
zation? Why do you think that this isimportant?
A Europe has, of course, a longtradition of international harmo-
nization between the EU memberstates, formally dating from the firstpharmaceutical directives in 1965. Inaddition, the EMEA contributes to anumber of international harmoniza-tion activities, particularly within theICH and VICH frameworks. In thearea of GMP, GCP, and quality relatedmatters, international harmonizationis essential as the pharmaceutical in-dustry is a global industry, and shouldapply the same scientific standardsirrespective of where the product isdeveloped, tested, marketed, or pro-duced. Due to historical reasons anddifferent regulatory frameworks, thisgoal may not always be immediatelyachievable. International harmoniza-tion activities are important becausethey help to identify what the currentbarriers are and to develop mecha-nisms to resolve differences that mayexist. Technical and scientific differ-ences are generally easier to solve thanregulatory differences. When it comesto international harmonization, I cur-rently have both a personal involve-ment in and a role derived from myresponsibilities within the EMEA. Ihave personally been involved with thework of the International Conferenceon Harmonization (ICH) since 1992.When I worked for EFPIA, I started offas Steering Committee member andICH coordinator, and then retainedthe role of ICH coordinator for the bestpart of six years. When I moved to theEuropean Commission, pharmaceuti-cals unit, I became ICH coordinator forthe European Commission. I startedworking again within the ICH frame-work just after the FDA launched itsGMP for the 21st Century Initiativeand am currently one of the EU topic
leaders for the Q9 topic on risk man-agement.
At an organizational level, work oninternational harmonization within thequality and GMP area comes withinthe scope of the ad hoc GMP inspectorsmeetings, and the work of the QualityWorking Party respectively and in thoserespects, I am responsible for coordi-nating the input into the various dis-cussions from the side of the EU. Vet-erinary harmonization activities in thequality areas within the VICH Frame-work are also addressed in the QualityWorking Party.
Currently, the Quality WorkingParty has produced a draft guidelineon inhalation products jointly withHealth Canada, and a number of an-nexes to the EU GMP guide were pre-pared jointly by the ad hoc GMP in-spectors meetings and the Pharmaceu-tical Inspection Cooperation Scheme(PIC/S), which, apart from the EUmember states, involves Switzerland,Canada, Australia, New Zealand,Singapore, and Malaysia.
Q There have been several MutualRecognition Agreements (MRAs)
signed over the past few years – whatbenefits have these achieved? Do youenvisage that there will ever be suchan agreement between Europe and theUS?
A EU has signed MRAs in the GMParea with Canada, Australia,
New Zealand, US, Switzerland, andJapan. Apart from the MRA with theUS, which is not operational, all ofthese MRAs have been extremely ben-eficial for the EU. The MRAs are de-pendent on the recognition of equiva-lent GMP standards in the regionsconcerned, and mean that the resultsof GMP inspections by an MRA partnercan be accepted within the EU frame-work and vice verse and that no re-testing of each batch that enters theEU is required. The benefits for EUregulators include the saving of re-sources as additional inspections neednot be performed and the focus of for-eign inspection programs can be onareas of higher risk. From the industryside, less regulatory inspections needto be hosted and the expense of re-
testing can be avoided. e.g., Switzer-land figures.
In addition to the immediate re-source benefits for both industry andregulators, the MRAs include provi-sions for exchange of information onquality defects, alerts, and recalls whichhelp to ensure that coordinated ap-proaches to responses are taken andthat patients can be protected at aglobal level. They also provide for an-nual maintenance programs so that allpartners are aware of regulatory ororganizational changes that may havean impact on the agreements.
While there is a common basicframework, all the MRAs differ slightlywith respect to the implementationdetails. Contrary to popular belief, aMutual Recognition Agreement be-tween the US and the EU was actuallysigned in May 1998. However, in theGMP area, this provided for a threeyear transitional or confidence build-ing phase, which was never actuallycompleted, therefore rendering theagreement non-operational from thepoint of view of recognition of inspec-tion outcomes. However, exchange ofinformation on defects and alerts con-tinues to take place.
Business Strategies/Vision
Q What do you think the majorbarriers are for you and other
regulators?
A Availability of resources, effec-tive implementation of risk man-
agement principles in the interests ofpatients, and developing communica-tion networks so that issues such asfraud and counterfeiting can be tack-led in a global framework.
Q What is the long term vision forEMEA?
A The long term vision of the EMEAis outlined in EMEA’s Roadmap
to 2010 and key aspects of this arereproduced below:
The main challenge for the EMEA overthe next few years will be its ability tomeet the increasing expectations of itsstakeholders. The Agency will particu-larly focus on the needs and expecta-
tions of patients and users of medi-cines. The EMEA will have to find theright balance in terms of expectationssuch as applying high scientific knowl-edge for the timely delivery of sciencebased opinions, increased involvementin the protection and promotion of pub-lic and animal health, regulatory andscientific consistency, predictability,greater transparency, better informa-tion, and enhanced communication.
In addition, the EMEA will have toaddress issues stemming from theLisbon strategy for economic, social,and environmental renewal since theAgency’s role in enabling the pharma-ceutical industry to achieve the objec-tive of industrial competitiveness iscrucial. The EMEA has an essentialrole in bringing safe and effective inno-vative medicines as quickly as possibleto patients and users of medicines.Apart from economic competitiveness,the EMEA also contributes to the EUcitizens’ quality of life. In respondingto the above challenges, the Agencywill have to adequately address:
1. additional tasks allocated to theEMEA in accordance with new Com-munity legislation
2. new developments such as the per-ception of the safety of medicinesand the environmental impact ofthe use of medicines
3. the assessment of new types of prod-ucts (such as gene therapy,pharmacogenomics, proteomics,xenotransplants)
4. bi/multilateral scientific coopera-tions
In addition, specific segments of thepharmaceutical market deserve spe-cial attention, such as Small and Me-dium-sized Enterprises (SMEs).
The EU Regulatory System conceptrequires the EMEA to find adequateanswers to the above challenges inclose cooperation with its Member Statepartners. Therefore, the continuationand adaptation of the Agency’s net-working model also will require thatMSs are able to adequately respond tothe changing environment, which willresult from the political, institutional,legislative, and scientific develop-
ments. In order for the EU RegulatorySystem to position itself successfullyin the international environment asone of the world’s foremost regulatorysystems, NCAs should carefully exam-ine how they can best contribute to thefuture system since this will be key forthe overall success. It should be em-phasized that the network between theEMEA and NCAs can only be opti-mized if there is a stronger cohesionbetween all parties concerned, lookingat complementing the achievementsalready obtained by introducing fur-ther actions aiming at reinforcing thenetworking model. In order to achievesuch aims, a common understandingon the architecture of the future EURegulatory System is paramount. Oncesuch common understanding has beenobtained, in a next step, importantissues such as roles and responsibili-ties (in different fields such as regula-tory, scientific, organizational, andtechnical), of all involved parties needto be addressed in order to reach com-plete transparency on the accountabil-ity for the different activities to beundertaken in the context of the EURegulatory System.
Leadership Style
Q What type of strategy works bestfor the management of a regula-
tory agency?
A First let’s be clear, I don’t man-age a regulatory agency, I man-
age a small sector within a regulatoryagency and I can only comment frommy own experience. I think it is veryimportant to realize that, in general,the tasks of a regulatory agency arelaid down by legislation and the fund-ing of activities is very tightly con-trolled. In the case of the EMEA, thismeans through oversight of its man-agement board, the European Com-mission, and the European parliament.This means that the planning and bud-geting process tends to become part ofa negotiation process rather than asmuch of a management tool as wouldbe the case in an operations manage-ment company.
I also think working in a Europeanmulticultural environment is a uniquechallenge that requires specific adap-
tation of management skills. Out of my18 co-workers, I have representativesfrom nine different European coun-tries with three countries being domi-nant in terms of numbers. I am verylucky to have a great team working forme, all highly motivated and with ex-cellent experience. I try to be as non-interventionist as possible, as long asresults are achieved and performancestandards are met, I would leave it upto the individuals to decide how tomeet the goals of the organization. Ihave regular bilateral meetings withall of my professional staff and if thereare issues or difficulties, I may suggestthat focused brainstorming meetingson specific topics be organized. Per-sonal feedback on performance mat-ters also is very important.
I am passionate about efficiency,avoidance of as much bureaucracy andduplication as possible, “not reinvent-ing the wheel,” and transparency so Itry to encourage integrated thinkingand learning from others’ experiences.
I think I have learned a lot aboutmanagement styles in every aspect ofmy career and I try to use my trainingand experience in my current position.The first priority is to listen to whatother people have to say; there arealways new ideas that may be useful.People with different backgrounds anddifferent cultures have different per-spectives so that there can be no as-sumption that there is “one right way”of doing things. It is impossible to im-pose solutions without “buy in.” Thisapplies equally to my own staff, otherstaff of the Agency, as well as in thecontext of inspection coordination, andharmonization activities involving EUmember states. It also is extremelyimportant to be transparent – I haveseen too much resentment and dis-trust develop from lack of communica-tion.
Finally, I think one of the keys toeffective management and gettingthings done is never ceasing to believethat mountains can be moved if this iswhat is necessary to make things bet-ter. And no matter how slow or hard itis, never to give up!
Manufacturing/Operations Inspections
Q The US FDA believes thatdevelopers and manufacturers
need to increase efficiencies and re-cently issued a guideline on ProcessAnalytical Technology (PAT) whichaligns with their Initiative for the 21stCentury. Is EMEA working on some-thing similar?
Part of the PAT initiative involvesusing new technologies (i.e., on-linesensors, advanced controls, etc.) withthe ultimate goal of improving processefficiencies and product quality. Howdo you see new technologies impactingthe future? If so, what types and how?
A Let me tell you a bit about thework of the EMEA PAT team.
Over the last two years, the EMEA hasdone a lot of work to support and de-velop the PAT concept within the EUregulatory framework. This has in-cluded the organization of focused com-bined meeting of assessors and inspec-tors from all 25 member states, invit-ing a number of companies to presentto this group on the status of theiractivities, and the subsequent estab-lishment of an EMEA PAT team madeup of assessors and inspectors and thechairs of the QWP and Ad Hoc GMPinspectors groups respectively. We or-ganized a specific and dedicated PATtraining session for inspector and as-sessors in October 2004 and a secondtraining session is planned shortly.
In order to support the PAT activi-ties in EU, an EMEA PAT team wascreated in November 2003. It is a fo-rum for dialogue and understandingbetween the Quality Working Partyand the Ad Hoc Group of GMP Inspec-tion Services with the aim to reviewthe implications of PAT and to ensurethat the European regulatory frame-work and the authorities are preparedfor and adequately equipped to con-duct thorough and effective evalua-tions of PAT-based submissions. Theteam’s mandate provides further in-formation on the make-up and aims ofthe team. The meeting is chaired byDr. Keith Pugh of the UK regulatoryauthority and nine meetings have takenplace since the end of 2003.
Part of the work of the PAT teamhas involved an open invitation to com-panies to provide us with mock PATsubmissions. In association with these,we have organized two specific sitevisits.
The team has examined the FDAguidance document, and indicated thatthe EU is in general agreement withthe principles outlined. For this rea-son, the value of a specific EU guidancedocument, which would have outlinedmuch of the same philosophy, was ques-tionable. However, the PAT team hasdeveloped some simple questions andanswers that may be more specificallyapplicable in a EU context and contin-ues to work on expanding these. Allthese are published on the ProcessAnalytical Technology section of theEMEA’s Web site.
The key message we would like tostress is that the current regulatoryframework in Europe is open to theimplementation of PAT in marketingauthorization applications, and thatall efforts to facilitate these applica-tions will be made.
Regulatory, Quality, andPolitical Concerns
Q What involvement does EMEAhave in anti-counterfeiting?
A The national competent authori-ties of the EU member states are
engaged in the prevention and detec-tion of counterfeit medicines into thelegitimate supply chain. EMEA has nolegislative role in this area, but wefully support the work of the memberstates. The European Heads of Medi-cines Agencies have established a bi-annual meeting of European MedicinesEnforcement Officers (EMEO). EMEAattends this meeting and provides aliaison between the EMEO and theGMP inspectors meetings.
The establishment of this group pro-vides a formal method of developingfurther the close liaison and coopera-tion between the EU agencies, as wellas a mechanism for dissemination ofinformation, the establishment of com-mon objectives, and the developmentof training programs in the area ofenforcement. The EMEO are currently
surveying the extent of the problem ofcounterfeit medicines in the EU legiti-mate supply chain and have been askedby the Heads of Medicines Agencies todevelop an EU wide anti-counterfeit-ing strategy.
Industry, Government,and ISPE
Q What is your involvment withISPE? When did you first en-
counter ISPE?
A The inspections sector of theEMEA has been involved in in-
teractions with ISPE since shortly af-ter its establishement in 1995. Since Ijoined as head of sector, I have contin-ued this collaboration and committedto provide EMEA speakers to key ISPEevents, as well as ensuring the voice ofISPE is heard in consultation on guide-lines and relevant interested partiesmeetings. I think my first personalencounter with ISPE was in Brusselswhen I worked for EFPIA.
The purpose of this document is to pro-vide guidance on how to achieve anappropriate level of traceability between requirements, design, and test-
ing documents for regulated GxP applications.Although the expectation for traceability byregulatory authorities has been clearly stated,1,2
there is little definitive guidance on thepracticalities of achieving and sustaining trace-ability.3
This guidance addresses this gap and shouldbe treated as a supplement to GAMP® 4, GAMP®
Guide for Validation of Automation Systems.4
PrinciplesProcesses and supporting documentationshould be established and maintained to linkrequirements, design, and testing. In addition,it should be possible to trace back from testingto both design and requirements - Figure 1.This traceability provides a means to ensurethat all elements of design, as well as allrequirements, have been tested. It also enablesthe identification and flow of documentation inthe event of requests during an audit.
The linkage between requirements, design,and testing is not necessarily limited to a 1:1:1relationship:
• Multiple requirements may be covered by asingle design specification and tested by asingle test.
• Multiple design specifications may be linkedto a single requirement.
• Multiple tests may be required to addressone requirement or one design specification.
Whatever process is used to achieve traceabil-ity, it should be:
• appropriate to the system size, complexity,impact, and risk
• documented and approved in the validationplanning stage
• an integrated part of the overall life cycle ofthe project and beyond into the support andmaintenance of the system
Benefits of TraceabilityGood traceability yields a number of tangibleand intangible benefits. Examples include:
• Traceability will assist risk management.Focus should be placed on any critical re-quirements as part of the risk assessment.Traceability will help to identify critical de-sign elements and necessary testing. Thereshould be increased testing rigor applied tothe critical aspects of a system, compared tothe non-critical aspects of the system.
• Traceability will improve test coverage.Traceability should make it possible to dem-onstrate which requirements and design el-ements are tested. Therefore, duplicate orredundant testing may be avoided.
• Traceability can help demonstrate that vali-dation is complete. All requirements shouldbe functionally tested, covered by an audit,handled through a user operating proce-dure, or accepted as not requiring testing,and monitored in the live environment.
This granularity will influence the needfor the traceability matrix and its con-tents; the greater the granularity thelarger the matrix, and therefore, thegreater the need for a tool to maintainthe matrix.
An example Requirements Trace-ability Matrix (RTM) is shown in -Figure 3. Each reference within thetraceability matrix, e.g., U1.1.2, F3.1,D1.2, T8.2, could be a reference to asection or subsection within the rel-evant document, or to a totally sepa-rate document. The method used andthe process should have been declaredand approved within the validationplan.
Level of Detail PracticalitiesIt can be difficult to determine the levelof detail required for traceability. Thefollowing information is intended tohelp pitch the detail at a level whichsatisfies regulatory expectations fortraceability, while remaining practicalto maintain.
A strategy for traceability should beestablished during validation planning.User requirements should be devel-oped with traceability in mind.
The level of traceability could stopwith a reference to vendor documenta-tion; if documentation needs are metby the vendor documentation whensupported by in-depth supplier assess-ment and a vendor management plan.
The supplier should have their owntraceability for the documentation andtesting under their control. This shouldbe verified during Supplier Assess-ments, where appropriate.
Requirements need not trace to tech-nical controls in all circumstances.Requirements may trace to proceduralcontrols, in which case cross-referencesto identified SOPs is appropriate.
• For simple systems, an RTM is notrecommended, as sufficient trace-
• Traceability will improve changemanagement. When a change con-trol is raised, traceability enablesan accurate assessment of its im-pact by identifying related require-ments, design elements, and testscripts. Regression testing canthereby be clearly scoped.
• Traceability will help root causeanalysis of software malfunctions.It should be possible to more easilytrack and trace design element in-terdependencies when conductingroot cause analysis of incidents at-tributable to software malfunction.
• Traceability will help audits andinspections. It should be relativelyeasy to identify any and all support-ing documentation for any givenoperation. It should be much easierto provide timely responses to re-quests for information.
Methods of AchievingTraceability
Traceability may be achieved in a num-ber of ways, including:
• a Requirements Traceability Ma-trix (RTM)
• automated software tools
• excel spreadsheets
• embedding references directlywithin documents
If an RTM is chosen, it may be gener-ated as a separate deliverable or aspart of an existing deliverable, such asthe requirement document: the choice
will be dependant upon maintainabil-ity of the deliverable.
Traceability for simpler systemsmay be achieved through common orconsistent numbering of requirements,design statements, and testing - Fig-ure 2. The numbering for “temperaturerecording” in this example is the samein the requirements, design, and testdocumentation; thereby enabling trace-ability without creating a separatetraceability matrix. This approachworks well with smaller systems in lowrisk situations.
For purely Commercial Off-the-Shelf (COTS) software products, thetraceability may be reduced to that ofrequirements to testing (or qualifica-tion) only. However, this will dependupon the user’s knowledge of the sup-plier and their processes, the systemusage within the company, and thelevel of acceptable risk. In most cases,the design column in Figure 2 could bereplaced with a link to configurationitems, providing traceability betweenrequirements, configuration, and test-ing (or qualification).
The user of the COTS software prod-ucts will need to be able to demon-strate an intimate knowledge of thesupplier’s quality process, as a mitiga-tion of risk to the user processes whenusing the COTS system. This may ne-cessitate multiple visits to the supplierduring the project phase, as risks areidentified at points throughout theproject, and during the ongoing contactof the support and maintenance phasesof the system life cycle.
The depth or granularity of the re-quirements will be influenced by thesize and complexity of the system, alongwith its potential to impact drug prod-uct:
Figure 2. Example of embedded traceability for simple systems.
ability may be incorporated withindocument cross-references.
For global systems, planning for trace-ability in the validation plan is im-perative since the control of local andglobal requirements needs to be re-solved at this point for tracking thecombination of local and global require-ments.
Extending RTMsThere are other features that may beadded to a basic traceability matrixthat can assist with the overall effec-tiveness and efficiency of validationactivities. Examples include:
• A column to include a brief writtendescription of each requirement,which may assist in the verificationthat matrix contents are referencedcorrectly.
• A column to include change controlnumbers to enable tracking the sys-tem history and change impact.Reference also may be made to otherdocumentation and processes whichimpact the system, such as devia-tions or SOP changes.
• A column to indicate the criticalityof the requirements to assist levelsof testing applied to any given re-quirement. High criticality require-ments may have greater testingapplied; therefore, may referencemultiple tests, whereas low criticalrequirements may have a referenceto a single test. There may be a needto reference the executed tests andthe supporting test result documen-tation along with any failed tests.
• A column to indicate where a re-quirement has been met by proce-dural controls, along with the refer-ence to the procedure and its ver-sion number. In this case, the re-quirement and design columnsshould be blank, but the testingcolumn may not, as the use of theprocedure may be tested at the sys-tem test level.
• The test column may be expandedto indicate at what level the testingoccurs: unit, integration, acceptance(hardware or system) or where andwhen the testing occurs, develop-ment, qualification, production, orglobal, local. In this case, the level ofeffort in testing should relate to thecriticality of the requirement andthe level of acceptable risk. For ex-ample, a high-risk requirement maybe tested many times and at manylevels, whereas a medium risk re-quirement may be tested just once,and a low risk may not be tested atall, only verified through systemuse.
• A column linking a test to a mainte-nance or calibration record for theinstrument required for a test andrequirement. For process automa-tion documents such as installationrecords, loop checks and tuning,cable integrity checks may be linked,enabling traceability from the cali-bration certification on an instru-ment all the way through to the useof that measurement in the busi-ness process and system testing.
The above additions increase the diffi-culty of navigating and maintainingthe traceability matrix. Therefore thereneeds to be a balance between what isexpected of the traceability matrix andthe maintainability of the method cho-sen. Where large projects are beinginstalled, such as ERP/MRP II, LIMS/CDS, or large process control systems,it may be prudent to seek out a docu-ment management system which hasthe capability to both maintain thelinks between documents within thedocument management system andreferences to documents generated andstored outside.
Documentation andMaintenance of TraceabilityThe chosen process and method whichany given system will use for traceabil-ity should be documented and under-stood. It is recommended that this beachieved within the validation plan forsmaller systems, or perhaps proce-duralized for larger and more complexsystems. All members of a system de-velopment team should be acquaintedwith the process and method to ensurethat it is adopted and maintainedthroughout the system developmentlife cycle.
Once a system has been acceptedinto use, the maintenance of traceabil-ity is required to preserve its useful-ness. The method of maintenance willalways be linked to whatever process isused to maintain the requirements,design, and test documentation, all ofwhich must be updated to reflect thecurrent system. In addition, the versioncontrol may be linked to enable systemconfiguration controls, e.g., version 1.0of the requirements, design, test, andtraceability are in use at go-live of anysystem, then all versions may be in-creased at the same time to maintainthis configuration control periodicallythroughout the system life cycle.
Whatever changes are made to thedocumentation they will be controlledthrough the control of a change pro-cess. Within this process, the methodby which the documents will be up-dated should be documented, for ex-ample:
• at every change
• with a number of changes batchedtogether
• on a chronological basis
Figure 3. Example Requirements Traceability Matrix (RTM).
The method should be justified withinthe process and based upon documentedand reasoned risk.
ConclusionAlthough traceability is a valuable toolfor any system, its scope, depth, granu-larity, and level of detail should becommensurate with the criticality andrisk associated with the business pro-cess being controlled by the system. Iftraceability is sized correctly it may bethe one tool which can influence thesuccess of the project, support andmaintenance, and ‘auditability.’ How-ever, like any other tool it can achievethis success only if it is maintainedthroughout the system life cycle.
References1. PIC/S Guidance on Good Practices
for Computerised Systems in Regu-lated “GxP” Environments (PI011-2) (available at www.picscheme.org).
2. FDA, ‘General Principles of Soft-ware Validation: Final Guidance toIndustry and FDA Staff,’ publishedin 2002 by the Food and DrugAdministration’s Center for Devicesand Radiological Health (CDRH),(www.fda.gov)
3. Wingate, G.A.S. (Editor), ‘ComputerSystems Validation: Quality Assur-ance, Risk Management, and Regu-latory Compliance for Pharmaceu-tical and Healthcare Companies,’published in 2003 by Taylor andFrancis (www.crcpress.com), ISBN0-8493-1817-8.
4. GAMP® 4, Good Automated Manu-facturing Practice (GAMP®) Guidefor Validation of Automated Sys-tems, International Society for Phar-maceutical Engineering (ISPE),Fourth Edition, December 2001,www.ispe.org.
AcknowledgementsThe GAMP Forum would like to thankScott Lewis (Eli Lilly), Guy Wingate(GlaxoSmithKline), and Mark Cherry(AstraZeneca) for leading the develop-ment of this guidance.
Look for theCountry Profileon Argentina inthe March/April
issue ofPharmaceutical
Engineering.
Dear ISPE Members,
It is my pleasure to present the Thailand Country Profile on behalfof the ISPE Thailand Affiliate for this issue of PharmaceuticalEngineering. Thailand is an active participant within the rapidlychanging and developing ASEAN region and is ideally placed to takeadvantage of the exciting and ever increasing opportunities of theregion, its neighbors, and the world.
Thailand has both a rich cultural heritage full of proud achievementsand an ambitious outlook to further development that will beaccomplished using the resourceful and entrepreneurial nature ofher people. Thailand has successfully rebounded after the financialcrisis of the late ’90s and external investment is steadily increasing.Consistent financial growth, social stability, her close relations withsurrounding countries, and an experienced and well educated laborpool are all hallmarks of Thailand’s potency.
I hope you will be interested enough by thisCountry Profile to consider Thailand as yourbase for investment in manufacturing, re-search, and development.
A Look at Thailand: History and Financial ProfileIntroduction
“T he land of smiles” as Thai-land is often referred, is fa-mous for her variety of
beautiful nature and rich culture.It’s also known for its ancient ru-ins, fine arts and handicrafts, deli-cious food, and the contagiouswarmth of the Thai people.
A unified Thai kingdom wasestablished in the mid-14th cen-tury. Known as Siam until 1939,Thailand is the only SoutheastAsian country never to have beentaken over by a European power.A bloodless revolution in 1932 ledto a constitutional monarchy andsince 1946 King BhumibolAdulyadej, also known as RamaIX, is the Chief of State.
Thailand has an area of 513,254sq km. It is situated in the heart ofSoutheast Asia, and shares bor-ders with Myanmar in the West,Laos in the Northeast, Cambodiain the Southeast, and Malaysia inthe South. Bangkok is the capitalcity and center of political, com-mercial, industrial, and culture ac-tivities. It is the largest city withapproximately one sixth of the Thaipopulation of roughly 64.5 million.Thailand is home to various ethnicgroups with the largest being Thaiand Chinese with 75% and 14%respectively.
Thailand is a participant in in-ternational organizations like theAssociation of Southeast AsianNations (ASEAN), the United Na-
tions (UN), the United Na-tions Educational, Scientific,
and Cultural Organiza-tion (UNESCO), theWorld Health Organiza-
tion (WHO), and the World TradeOrganization (WTO).
In the past three decades, theoverall physical health indicatorsfor Thai people have been improv-ing. For instance, life expectancyat birth has increased from 59years in 1964 to 71 years in 2004(Source: National Statistical Of-
fice). Significant im-provements in the qual-ity and standard of theThai healthcare systemhave contributed im-mensely to this success.The country’s healthcaresystem has evolved froma system dependent andbuilt on local wisdom toone that relies heavily ontechnology and collabo-rative efforts of healthcare profes-sionals from multiple disciplines.
Thailand ranks as the world’sfourth most attractive nation forforeign investment in a survey bythe UN Commission for Trade andDevelopment in 2004. Thailandenjoys a strategic location right atthe heart of Asia – home to what isregarded today as the largest grow-ing economic market. It serves asa gateway to Southeast Asia andthe Greater Mekong sub-region,where newly emerging marketsoffer great business potential.
FinancialThailand has a well developed in-frastructure and a free-enterprise
economy. She has fully recoveredfrom the 1997-98 Asian FinancialCrisis and was one of East Asia’sbest performers in 2002-04. In-creased consumption, investmentspending, and strong export growthpushed the Gross Domestic Prod-uct (GDP) growth up to 6.9% in2003 and 6.1% in 2004. The growthoutlook for 2005 is set to remainimpressive, despite a sluggish glo-bal economy and the tragic 2004tsunami that took 8,500 lives inThailand and caused massive de-struction in the southern prov-inces.
The highly popular governmentlead by Taksin Chinnawat has pur-sued preferential trade agree-
Figure 1. Real GDP Growth %. (Source: NationalStatistical Office, 2004)
Figure 2. Thailand fact sheet. (Source: CIA factbook, 2005)Continued on page 10.
Thailand Pharmaceutical Industry OverviewMarket Size and Growth
The Thai pharmaceutical market has a current valueof $1.32 billion. IMS Health has projected that
Thailand will soon join China as one of the fastestgrowing areas for pharmaceuticals. Figures for 2004show that the market has grown by 6% from 2003.
The pharmaceutical spend per capita in Asia is fore-cast to continue its strong growth, and Thailand hasone of the highest growth rates which was 19 millionUS$ in 2004 and projected to be 36 million US$ by 2009(Source: IMS Health, 2005). The largest share of theThai market for pharmaceuticals is occupied by locallymade products according to IMS Health, 2005.
Of the total market share, locally produced prod-ucts amount to 65% with imported goods back uptoward pre-financial crisis levels at 35%. This localgrowth also has been reflected in the local manufac-tures with strong increases at 14% in 2002 comparedagainst 11% for foreign based companies (Source:Diethelm, 2003).
HealthcareThailand has a universal healthcare scheme that is inplace since 2001 allowing Thai citizen’s greater accessto medical services. The popular scheme setup by thegovernment is expected to be further funded with taxeson cigarettes and alcohol sales. Currently, more than95% of the population has health security in Thailandwith an increase of 2.85% in 2004 (Source: IMS Health,2005). However, there have been some negative im-pacts attributed to the scheme on the finances andnumber of medical staff leaving state hospitals.
Pharmaceutical Manufacturingin Thailand
In Thailand, there are three categories of drug manu-facturers:1. Multinational corporations: manufacture active in-
gredients and pharmaceutical formulations in theirown manufacturing facilities
2. 171 privately-owned Thai companies: primary focusis on producing pharmaceutical formulations and toa smaller extent, manufacturing active ingredients
3. One Government-owned Thai company: theGovernment Pharmaceutical Organization(GPO), which primarily prepares pharma-ceutical formulations for public medical estab-lishments
Market leaders include: Pfizer Inter. Corp., SiamBhaesaj Co., GSK, GPO, Biolab, Aventis Pharma,AstraZeneca, Novartis, Berlin Pharma, and Roche(Source: Diethelm 2003).
DistributionAccess to pharmaceutical products for the consumer ismainly through the general and specialist hospitals inThailand. The distribution of manufactured drugs isthrough independent distributors or self distributedby the manufacturers themselves.
Key Data Value Year WorldRanking
Pharmaceutical Market 1,320 2004 33(US$ millions)
Pharmaceutical Market 21 2004 54per capita (US$)
Market Growth (%) 6 2004 -----
Table A. Market overview. (Source: Espicom BusinessIntelligence, 2004)
Figure 1. Share of market by manufacturers. (Source:International Trade Centre UNCTAD/WTO, 1999)
Figure 2. Pharmaceutical distribution at the point ofconsumption. (Source: Diethelm, 2002)
Thailand: Moving Forward in Biotechnologyby Dr.Thanit (on behalf of Prof. Dr Morakot Tunticharoen, Director, NationalCenter for Genetic Engineering and Biotechnology)
necessary support for investmentin biotechnology.
Key Factors forPharmaceutical Success:Infrastructure and Skilled
Research PersonnelRecognizing that research and de-velopment is a driving force forpharmaceutical success, Thailandhas developed a variety of scien-tific infrastructures. The 323,748.5Sq Metres Thailand Science Park(TSP) is a landmark governmentinitiative. It was built with an ini-tial investment of $175 million.TSP provides main laboratories,incubator units, pilot plants, green-houses, and accommodations, aswell as financial, management,and legal support for TSP custom-ers. The TSP also offers long termleases of land for construction andready made wet-lab space for rent.
Skilled research personnel is an-
Like other countries in South-east Asia, Thailand’s biotech-nology has been rapidly de-
veloped in the last two decades asa result of government backing.Biotechnology is a priority sectorfor the country; therefore, it is re-ceiving soaring financial support.Biotechnological activities can befound in many research institutesand universities throughout thecountry. One of the institutes sup-porting biotechnology developmentin Thailand is the National Centerfor Genetic Engineering and Bio-technology (BIOTEC). BIOTECprovides resources for the countryto support Thailand’s developmentof biotechnology. This can beachieved through conducting R&Dprojects, facilitating the transferof advanced technologies fromoverseas, developing human re-sources at all levels, providing in-formation services, developing col-laboration with world class insti-tutes, and promoting public un-derstanding of the benefits of bio-technology.
Strong Political Supportfor the Promising Future
of BiotechnologyReinforcing biotechnological devel-opment, Thailand has formulatedthe National Biotechnology PolicyFramework in line with the
government’s policy to pro-mote sufficiency of living and
enhancement of competi-tiveness for the country,toward a proper balance
and direction. One of the six goalsfor biotechnological developmentin Thailand is that Thailand rep-resents a Healthy Community andHealthcare Center of Asia. Withthis political back-up, the countryis anticipated to drive biotechnol-ogy forward with a speedy pace.And this also will guarantee any
other key workforce of biotechno-logical development. Most univer-sities in Thailand have educationalprograms in biotechnology at alllevels, ranging from bachelor todoctoral degrees. The developmentof the qualified human resourcesystem is one goal under the Na-tional Biotechnology Policy Frame-work. This aims that in the year2011 Thailand will produce no lessthan 5,000 professional biotechnol-ogy researchers, no less than 500biotechnology managers, and noless than 10,000 students. On aver-age, Thailand can produce 400Bachelor’s degrees, 150 Master’sdegrees, and 10 Doctoral degreeswith the growth of 10% per year.Also, the Thai government has sentThai students to study biotechnol-ogy overseas. The first two phases(1990 -1995) aim to produce ap-proximately 330 biotechnologists,whereas the third phase anticipatesto see 370 graduates.
World Class Research andDevelopment Initiatives
Thailand has initiated many worldclass R&D projects in biotechnol-ogy:
1. Thailand SNP DiscoveryProgramBIOTEC and Centre Nationale de
Figure 1. National Center for GeneticEngineering and Biotechnology.
Figure 2. Thailand Science Park.Continued on page 6.
Genotypage (CNG), in cooperationwith the national collaborators,namely Ramatibodi hospital,Rajanukul Institute, andChulalongkorn University haveinitiated a collaborative projectwith an aim to analyze candidateDNA samples from 32 healthy Thaivolunteers. The resulting informa-tion is curate in Thailand SNPdatabase (ThaiSNP) that couldserve as a reference for variousSNP spin-off projects. For example,there are several research at-tempts to study multi-factorial ge-netic influenced diseases endemicgreatly in Thailand, including theinvestigation of genetic suscepti-bility to clinical malaria or thesearch for biomarkers in the de-velopment of genetic tests for theprevention of osteoporosis.
To assist these spin-offs, theSNP discovery process coversmainly SNPs inside a certain groupof genes believed to be associatedwith important diseases such ascancer, cardiovascular, SLE, andothers. Exonic regions are the maincoverage inside these genes. TheThai SNP database also hosts SNPdata from public domains such asdbSNP (NCBI) and JSNP (JapanSNP). This allows us to compareSNP properties across differentpopulations. This database alsowill serve as a basis for Thailand’sfuture research programs in sys-tematic genome screening,pharmaco-genomics, and anthro-pology. The Thai SNP databasealso will be used for the Asian SNPconsortium as a contribution fromThailand. The preliminary analy-sis of this data is being conductedby providing bioinformatic Web-based applications for interestedresearchers. Examples includetools for designing SNP-free prim-ers and tools for inferring a con-sensus haplotype of SNP from vari-ous haplotype inference algo-rithms. With these current fea-
tures, ThaiSNP database projectcan host more SNP informationfrom submitters, which will thenbetter ThaiSNP allelotype map-ping. In parallel to this databaseproject, an automatic SNP discov-ery program is being developed.Based on a direct SNP discoverymethod, this tool can reduce thetime researchers spend on correctlyidentifying SNPs or mutationsfrom input Chromatogram traces.Furthermore, if successful, thisSNP discovery tool has good po-tential in being commercialized.
2. From Biodiversity toDrug Discovery ProgramThais have a long tradition of us-ing nature as a healing tool. Me-dicinal plants and other remediesfrom nature have played a vitalrole and are still important eventoday. With the advance of scien-tific methods, Thai scientistsstarted searching for biologicalactive ingredients that confer ben-eficial activities more than 40 yearsago. At that time, the main objec-tive and most activities were con-fined to identifying new chemicalcompounds from medicinal plantsand reporting the results in scien-tific publications. Many new com-pounds were reported, but nonehave been further developed intomodern drugs even though therewere many scientists working inthis field. The major obstacle wasthat these ‘newly discovered’chemicals did not show beneficialbiological activities like thosefound in medicinal plants or natu-ral remedies. In general, biologi-cal assays that should guide everystep of the fractionation processhave not been used as part of theisolation and identification. Al-though some scientists have con-ducted biological assays which of-fer a better chance of finding po-tentially useful compounds, mostof the assays are low throughput,
and thus greatly reducesthe probability of success.In 1997, the National Cen-ter for Genetic Engineeringand Biotechnology (BIOTEC)established the Bioassaylaboratory to systematicallyscreen natural products for dif-ferent biological activities ina rapid and cost effectivemanner. The service wasfirst offered to BIOTEC’s in-house research group, whichfocuses on identifying compoundsfrom microorganisms as well asplants using bioassay guided iso-lation techniques. The number ofassays has been subsequently ex-panded and the service has beenoffered to scientists within thecountry and even to foreign re-searchers. Currently, the discov-ery of lead compounds from natu-ral resources in Thailand has beenconducted more systematically anda number of compounds with rel-evant biological activities havebeen discovered which has allowedthe private sector or internationalagencies such as WHO with drugdevelopment experience to haveaccess and evaluate them for com-mercial potential or for the good ofhumanity. In addition, in the lastseven to eight years, the focus ofbiological resources has shiftedfrom plants to other organisms,such as fungi, bacteria, and ma-rine organisms. This trend hasgreatly increased opportunities forfinding new active compounds be-yond those from plants alone. Cur-rently, Thailand is utilizing re-cent advances in biotechnologythat have accelerated the discov-ery of new drug targets that can beincorporated into biological assays,as well as new techniques to makethe existing assays more sensi-tive, less time consuming, andmore cost efficient. With these im-provements in the lead discoveryprocess, we aim to be more inter-nationally competitive employingour existing wealth of biologicalresources.
Thailand: Moving Forward in BiotechnologyContinued from page 5.
Pharmaceutical Regulations in Thailandby Dr. Nithima (on behalf of Wilai Bundittanukul, Director of Drug Div., Thai FDA)
The FDA is not a stand-alone Agency, but worksclosely with the Drug Committee, which is appointedby the Minister of Public Health every two years toadvise him/her on both regulatory and technical as-pects concerning the administration of pharmaceuti-cal control. The committee also is authorized to ap-prove or withdraw pharmaceutical registration, stan-
dard specifications, criteria and guide-lines, including suspending or with-drawal of licenses to manufacture, im-port, distribute, or sell. There are 14regular members on the Drug Commit-tee: five of them being ex-officio mem-bers who are appointed based on theirpositions in pharmaceutical-related or-ganizations and the others being ap-pointed from among pharmaceutical andmedical experts. The Committee can then
appoint subcommittees to assist them with certaintasks. Presently, 19 subcommittees have been ap-pointed.
Pharmaceutical Control SystemThe pharmaceutical control system is divided into twophases: pre-marketing and post-marketing. Thepre-marketing phase involves licensing regulations(regarding manufacturing, importing, or selling phar-maceutical products), drug registration, and drug ad-vertising regulation. The post-marketing phase fo-cuses on surveillance activities (e.g., inspection ofGMP compliance at manufacturing sites, adverse drugreactions, monitoring the use of marketed drugs forunexpected health risks), responding to consumer com-plaints, and reevaluation of pharmaceutical products.
Pre-Marketing PhaseLicensingThe Drug Act requires that any person who wishes tosell, manufacture or import drugs into the Kingdommust obtain a licence from the licensing authorities.The Drug Control Division is the licensing and regis-tration authority for manufacturing, import, and saleof drugs within Bangkok metropolis and its territories.Provincial Public Health Offices are the licensingauthorities for manufacture and import of traditionaldrugs and sale of drugs in other provinces.
Applications for licenses must be submitted to thelicensing authority. Their buildings and facilities willthen be inspected. A license will be issued after theinspection has confirmed that the applicant has ad-equate capabilities of doing such business, and he/shecan secure appropriate facilities and personnel for thatpurpose. Licences are issued, according to the business
Organizational structure of the Thai FDA.
OverviewThailand’s Food and Drug Administration (FDA) isone of the departments under the Ministry of PublicHealth (MOPH). It is a national agency responsible forsix health products, i.e., foods, drugs, cosmetics, nar-cotic/psychotropic substances, toxic and hazardous/volatile substances, and medical devices.
In relation to pharmaceutical products, the ThaiFDA has consulted or cooperated with experts in sci-ence, medicine, pharmacy and public health, consum-ers, manufacturers, importers, distributors and retail-ers of drugs. It works closely with several other orga-nizations (e.g., universities, industries, hospitals,healthcare professional groups, consumer groups, otherrelevant agencies, and foreign governments) in thedrug development and review processes.
Its mission continues to be protecting the publichealth by assuring the safety, efficacy, and quality ofpharmaceutical and biological products. It also is re-sponsible for advancing the public health by helpingtechnological development and researches to makemedicines more effective, safer, and more affordable;and helping the public get the accurate, science-basedinformation they need to use medicines to improvetheir health.
Drug Laws and CommitteesTo achieve the mission in consumer health protec-
tion, the FDA functions under the Drug Act BE2510 (1967). The 1967 Drug Act has been em-
ployed for almost two decades and it has quitesubstantially improved all aspects of drug con-trol. However, four more revisions subsequently
emerged in order to cope with the growth of the pharma-ceutical industry and the global situation. In the future,a new Drug Act will be promulgated to supersede the1967 Drug Act. When the new Act becomes effective,many features will be changed accordingly, for example:reclassification of medicines, renewal of product licenses,establishment of product liability, revision process ofGood Manufacturing Practice (GMP), and practices ofpharmacists and prescribers. Continued on page 8.
of the applicant, in the following nine categories:• license to manufacture modern medicines• license to import modern medicines• license to sell modern medicines• license as a wholesaler of modern medicines• license to sell modern medicines in sealed packages
which are classified as neither dangerous nor spe-cially-controlled medicines
• license to sell modern veterinary medicines in sealedpackages
• license to manufacture traditional medicines• license to sell traditional medicines• license to import traditional medicines
Good Manufacturing PracticeThe Thai FDA has begun campaigning on GMP compli-ance since 1984. Projects on development of the localpharmaceutical industry up to internationally accept-able standards were part of the Sixth National Economicand Social Development Plan (1987–1991) and also of theSeventh Plan (1992–1996). The projects aimed to pro-mote and support local drug manufacturers in imple-menting good manufacturing practices. The first guide-lines of Thai Good Manufacturing Practices were pub-lished in 1987. Since then, numerous workshops, semi-nars, and conferences, as well as consultative visits havebeen held or carried out to promote the guidelines adop-tion. Currently, the GMP is a mandatory requirement forall manufacturers of modern medicines.
A current GMP standard used in Thailand is theWorld Health Organization’s Good ManufacturingPractices. However, the Pharmaceutical InspectionCooperation Scheme (PIC/S) is planned to replace thiscurrent one to ensure that medicinal products manu-factured in Thailand will be in line with internationaldrug market requirements. Additionally, Thailand isspeeding up its standard upgrading as ASEAN plansto allow free trading in healthcare products in 2010and is likely to require all its 10 member countries toadopt the PIC/S standard. Therefore, Thailand is go-ing to apply for PIC/S membership in 2006, expectingto be endorsed in 2008.
Drug RegistrationThe registration process is necessary to ensure qual-ity, safety, and efficacy of the drugs being marketed inThailand. Only authorized licensees are qualified toapply for product registration. Manufacturing plants,in which drug products are manufactured, are subjectto inspection for GMP compliance. For the purpose ofregistration, drugs are categorized into three groups:• generics or pharmaceutical products with the same
active ingredients and the same dosage forms asthose of the original products, but manufactured by
different manufacturers• new drugs include pharmaceutical and
biological products of new chemicals, newindications, new combinations, new deliverysystems, and new dosage forms
• new generics are pharmaceutical and bio-logical products with the same active ingredi-ents as new drugs, which need to prove for theirtherapeutic equivalent by conductingbioequivalent studies on the same doses, anddosage forms as those of the new compoundsregistered after 1992
The amended registration procedure for new drugproducts, adopted in August 1989, involves a two-yearperiod of safety monitoring program. This means thatnew drug products will be firstly approved for use onlyin hospitals or clinics for at least two years. Thensafety reports must be submitted for consideration asto whether general marketing should be allowed. Mean-while, new generic products have to pass bioequivalencestudies to assure comparatively therapeutic outcomes.The bioequivalence data must be submitted to theauthorities as proof of the product bioavailability alongwith product information and quality dossiers.
Quality assurance of drug safety and efficacy beforemarketing can undoubtedly be achieved through GMP.Inspection of drug manufacturers and sampling ofdrug samples from manufacturers, importers, or retailpharmacies for analyses by the regulatory authoritiescannot effectively solve the problems encountered.Drug manufacturers, importers, and distributors mustestablish their quality assurance systems according tothe GMP guidelines to ensure that the drug productshave and continue to have the quality as claimed.
Drug AdvertisingDrug information available to healthcare profession-als and consumers is as important as drug quality forthe safe use of drugs. Drug advertisements and otherpromotional materials need to ensure truthfulness,non-misleading, and non-exaggeration.
Advertisements through any means must be ap-proved by the authorities before actually being dis-seminated. Advertisements of prescription or phar-macy-dispensed medicines are permitted only to pro-fessionals, but prohibited to the general public. Drugsin the household remedy category may be advertiseddirectly to the general public.
Post-Marketing PhaseTo further ensure quality, safety, and efficacy of theapproved drug products, the marketed products areregularly sampled for testing at the drug analysislaboratory of the Medical Sciences Department, Min-istry of Public Health. In addition, contracts have beensigned with some qualified laboratories of local univer-sities to assist in solving the problems of drug quality.
Pharmaceutical Regulations in ThailandContinued from page 7.
The surveillance tasks involve the following activities:• inspection of GMP compliance at manufacturing sites• monitoring of manufacturing process changes to
ensure no adverse effects on the safety or efficacy ofthe medicines
• monitoring of the use of marketed drugs for unex-pected health risks, taking action if risks are de-tected by informing the public, investigating thecause, and removing the drugs from the market
• lot release system is carried out for biological prod-ucts to ensure the consistency of the products
• receiving and handling of complaints• safety monitoring program for new drugs• re-evaluation of pharmaceutical products
Re-Evaluation of Pharmaceutical ProductsEven though drugs have been strictly examined fortheir quality, efficacy, and safety before being ap-proved for marketing, chronological consumption datain a large population, new findings, and pharmaceuti-cal progress may later reveal very serious side effectsthat were not previously seen. A balance betweenefficacy/benefit and potential risks or serious adversereactions is frequently questioned, especially those incombination. The Drug Committee in 1991 appointeda subcommittee to evaluate the registered products.Some criteria have been set and the evaluation processhas been ongoing.
Strategic Directions and ChallengesWhile continuing with efforts to ensure the availabil-ity of safe and effective medicines, the Thai FDA alsotakes active roles relentlessly in many activities. Amongthese are, for example, efforts in pharmaceutical har-monization and initiatives to enhance capacity of thedomestic pharmaceutical industry.
Toward ASEAN PharmaceuticalHarmonization
Thailand along with other ASEAN member countriesmoves toward harmonization of pharmaceutical regu-lations in order to facilitate trade by minimizing tech-nical barriers posted by regulators without compro-mising drug quality, efficacy, and safety.
To achieve the goal, ASEAN Consultative Com-mittee for Standards and Quality-Pharmaceuti-
cal Product Working Group (ACCSQ-PPWG)had agreed to first develop ASEAN CommonTechnical Requirement (ACTR), ASEAN Com-
mon Technical Dossier (ACTD), and Technical Guide-lines, followed by training and relevant capacitystrengthening. For implementing, the ASEAN hadagreed to start from a trial period, then full implemen-tation at the agreed specific timeframe. Along withthese implementations, the training, as well as ques-tion and answer forum also will be provided to ensureappropriateness, applicability, feasibility, andsustainability of the ASEAN agreement’s implemen-
tation. In addition, the ASEAN also will develop someMutual Recognition Agreements (MRAs) in particularissues, e.g., GMP’s Inspection Report, laboratory test-ing report, and will finally be endorsed.
Prior to and along with the trial period for imple-menting the ASEAN pharmaceutical harmonizedguidelines and requirements, the Thai FDA has ar-ranged seminars and meetings to enhance under-standing and know-how to implement the ASEANpharmaceutical harmonized requirement includingmethods and procedures to all relevant stakeholders,both in public and private sectors. Brainstorming andworkshops about the implementation of harmoniza-tion are planned. All comments and suggestions fromall stakeholders are welcome to ensure that the imple-mentation of harmonization requirements will posethe minimum obstacle and impact to all.
Enhanced Capacity of DomesticPharmaceutical IndustryIn the face of rising drug expenditures in Thailand, theMinistry of Public Health has realized the necessity todevelop initiatives to promote accessibility, availabil-ity, and affordability of medicines for Thai people.Currently, the Thai FDA has conducted one importantprogram titled, Promotion of Domestic PharmaceuticalIndustry. Primarily, the program is designed to assurethe public of high quality, safe, and effective genericdrug products manufactured by the local drug indus-try. The ultimate goal of the program is to improve thepotential, capacity, and competitiveness of the localindustry. To achieve the goal, the Thai FDA has devel-oped strategic plans for the program as follows:• Provide and improve infrastructures and facilities
that are necessary for the manufacturing of genericdrug products.
• Create collaborative partnership with related orga-nizations (both public and private sectors) to facili-tate generic drug product production.
• Develop mechanisms to assure quality, safety, andefficacy of generic drug products.
• Promote widespread use of locally manufacturedgeneric drug products among prescribers, dispens-ers, and consumers to substitute imported drugproducts.
• Promote investments on domestic generic drug in-dustry and increase its competitiveness for theinternational market.
Currently, one important activity undertaken by theThai FDA is to upgrade existing bio-equivalence cen-ters to meet the international standards so that we canassure that locally produced generic products are ofsame quality as that of the innovative drug products.Such an activity is an integral part of our endeavor toincrease the competitiveness of our local pharmaceu-tical industry.
2. Current status and practice in pharmaceutical regu-lations and technical requirements in Thailand.Available at: http://wwwapp1.fda.moph.go.th/drug/eng/files/status.pdf
Questions relating to the article may be directed to:Drug Control Division, Food and Drug Administra-tion, Ministry of Public Health, Web site: http://www.fda.moph.go.th.
Pharmaceutical Regulations in ThailandContinued from page 9.
Chiang Mai University – Faculty ofPharmacyChiang Mai 50200 ThailandTel: 66-539443423Fax: 66-53222741E-mail: [email protected]: suthep.pharmacy.cmu.ac.th/
Food and Drug Administration (Thailand)E-mail: [email protected]: www.fda.moph.go.th/
Hauchiew Chalermprakiet University –Faculty of Pharmaceutical Sciences18/18 BangNa-Trad Road, KM 18Bangpli, SamutPrakarn 10540ThailandTel: 66-23126300, ext.1171Fax: 66-23126237Email: [email protected]: pharmacy.hcu.ac.th/
Mahidol University - Faculty of Pharmacy447 Sri Ayudthaya Road,Ratchathevi, Bangkok, ThailandTel: 66-644867790E-mail: [email protected]: www.mahidol.ac.th/mahidol/py/h-pharm.html
Naresuan University - Faculty ofPharmaceutical SciencesURL: www.pha.nu.ac.th/
Prince of Songkla University – Faculty ofPharmaceutical SciencesURL: www.pharmacy.psu.ac.th/
Rangsit University – School of PharmacyPaholyothin RoadMuang-Ake Patumthani 12000ThailandTel: 66-29972222, ext.1422Fax: 66-29972222, ext.1403E-mail: [email protected]: www.rsu.ac.th/pharmacy/pharmweb/
Silpakorn University – Faculty ofPharmacyURL: www.pharm.su.ac.th/
Thailand Pharmaceutical Research &Manufacturers Association (TPMA) –PreMARoom #408/51, 12th FloorPhaholyothin Place Bldg.Phalolyothin RoadSamsennai, PhayathaiBangkok 10400 ThailandTel: 66-26190729 to -32Fax: 66-26190728E-mail: [email protected]: www.prema.or.th
The Federation of ThaiIndustries (FTI)4th floor Zone C Queen SirikitNational Convention Center60 New Rachadapisek RoadKlongtoeyBangkok 10110 ThailandTel: 66-23451000Fax: 66-2345129699E-mail: [email protected]
Thai Medical Device SuppliersAssociation11th Floor, Dr. Gerhard Link Bldg.88 Krungthepkreetha RoadHuamark, BangkapiBangkok 10240 ThailandTel: 66-23794296, 66-23794279Fax: 66-23794297E-mail: [email protected]: www.thaimed.co.th
Department of Medical SciencesMinistry of Public HealthTiwanon RoadAmphur MuangNonthaburi 11000 ThailandTel: 66-25890022, 66-29510000URL: www.dmsc.moph.go.th/defaulte.htm
Pharmaceutical Associations and Organizationsin Thailand
ments with a variety of partners inan effort to boost exports and main-tain high growth, and in 2004 be-gan negotiations on a Free TradeAgreement with the US. Thailand’sindustrial production is orientatedtoward exports with Japan as theleading country for Thai imports,totalling 23.6% of Thailand’s totalexport (Source: CIA fact book,2005).
A Hub Layout Concept for Oral SolidDosage (OSD) Facilitiesby M.P. Brocklebank, J. Lam, and P. Mehta
This articledescribes a newconfiguration forthe layout of anOSD facility thatcloselyintegrates all ofthe facilitycontents plusoptional R&Dandadministrationareas. Itpresents anumber ofoperating andGMPadvantages, aswell as potentialcapital andoperational costsavings.
Introduction
Oral Solid Dosage (OSD) facilities pro-ducing tablets and capsules use well-defined unit operations, regardless ofdifferences in production volumes or
usage of such facilities for single or multi-product manufacture. The contents of such afacility typically include warehousing with re-
ceipt and dispatch areas, clean process rooms(which are conveniently called ‘white’ areas)containing process equipment, clean supportprocess areas for items such as washing, move-ments, and staging, QA/QC laboratory, andblack normal factory finish technical space forprocess ancillary equipment and ventilation,and basic utility supply equipment if it is a
stand alone facility. To gain com-petitive advantage, companies aimto minimize capital and operatingcosts for such facilities, which pri-marily means reducing the size ofsuch a facility, as well as improvingthe operational and internal logis-tics of the facility without compro-mising current good manufacturingand engineering practices.
Process and FacilityOutputs and Size
Dispensing with sieving, blending,granulation, compression to formtablets, capsuling, tablet coating,blister packing or bottle filling,cartoning, and packaging are someof the discrete batch unit operationsin an OSD facility plant. The num-bers, capacity, cycle times of theequipment, and daily operating timewill determine the output of the fa-cility. Different companies may re-quire different facilities whose out-puts can range from 100 million totwo to three billion tablets a year.Increasing equipment capacity byfactors of 5-10 does not greatly affectfacility size (area), though its capac-ity may increase by a factor of 5-10.Increasing plant utilization from aone day shift for five days per weekoperation to a 24 hour, seven days
Figure 1 (a, b, and c).Typical production suitearrangements.
operation will increase outputs by a further three to fourtimes without any significant increase in production areasizing. However, increasing equipment numbers and typesto achieve more capability and flexibility will greatly affectfacility size, and efficiencies in space utilization. Neverthe-less, since the facilities are expensive, any unnecessaryincrease in OSD plant size will have a significant costimpact.
Materials HandlingA key requirement of an OSD plant is solids materialsmovement between each batch unit operation. Movements ofsolids between unit operations can be achieved by gravityflow from ‘directly’ coupled equipment (advantageous inlarge output single product stream plants) or more frequentlyby collecting the batch in an Intermediate Bulk Container(IBC) and moving this to the next stage by emptying thecontents of the IBC by gravity to the equipment.
One or two level plants are common for smaller plants withsmaller batch sizes (e.g., typically up 500L), while the threelevel plant is often used for larger batch sizes (e.g., above1000L).
Process Arrangement -The Cleanroom Suite
For OSD plants, ISO Class 6 (equivalent to Class 100,000)particulate environments are typically used for process equip-ment rooms and adjoining areas where ‘open’ product oritems which could be in contact with the product (includingpeople) are present. Consequently, the process area is typi-cally composed of a ‘clean’ room suite with ‘clean’ corridorsand ‘clean’ process rooms for production equipment. Techni-cal areas are needed for ancillary supporting equipment tothe production equipment, e.g., blowers, heaters, vacuumsystems, being located either ‘behind’ the wall of the produc-tion equipment at the same level and/or in a technical spaceabove the production suite. Technical areas adjacent to pro-duction rooms allow ‘through the wall’ installation of some
process equipment, e.g., coaters. Such desirable facility fea-tures also can facilitate the process equipment installation.
In addition to the production rooms, the cleanroom suitealso will require space for materials staging, wash rooms,store rooms, and sometimes a small operator batch log roomand an in-process control laboratory.
Figure 1a shows a typical arrangement of a plant with asingle corridor and adjacent process rooms, allowing a tech-nical space on either side of them. In Figure 1b, a larger multi-corridor cleanroom suite facility is schematically indicated,and in Figure 1c, a cross section is provided of a single, levelplant with a technical level above it.
Operator access to the cleanroom suite will be via achange room and materials enter or leave via one or moreairlocks.
Filling and PackagingWhile a number of OSD plants may just make the capsules ortablets in bulk to be shipped to other filling and packagingfacilities around the world, most OSD plants incorporatethese operations within them. For multi product plants, it isusual for the filling equipment to be in a cleanroom with the‘line’ then extending into the lower GMP category packagingand cartoning room. Thus, the filling rooms are typicallyaccessed off the process cleanroom suite with the associatedpackaging area next to them accessed directly or indirectlyfrom the warehouse.
Mechanical VentilationOSD plants require large ventilation systems where 15 or soair changes per hour are required for the cleanrooms, andlow humidity may be needed for specific product require-ments. The ventilation systems typically consist of AirHandling Units (AHUs) and their associated control damp-ers and ducting systems, plus chemical dehumidifiers ifrequired. There may be six plus such systems in a facilitysupplying clean areas, support areas, warehouse, labora-tory, and support offices depending on the product(s) andplant type.
AHUs and their ducts are primarily located and distrib-uted in technical areas, which are located near the cleanareas. The conventional approach is to provide a top floor ofthe facility dedicated to AHU system as indicated in Figure1c. This area also may include water chillers and otherservices if the plant is remote from central facility services.Quite often the process and support functions (i.e., the overallfacility) floor area requirement is greater than the spacerequired by the AHU systems, and this can result in under-utilization of the floor created at this level.
Good Manufacturing Practice (GMP)GMP guidelines cover all aspects of manufacturing includingvalidating process methods and analytical control, equip-ment usage, facility layout, environments, storage, documen-tation, labelling, and the required training of personnelemployed. Regardless of the arrangement of the facility, anumber of key principles should be applied, including:Figure 2. Typical conventional facility layout arrangements.
• avoiding mix ups• provide suitable environments• take measures to avoid contamination• provide suitable materials flow around the facility• provide adequate space for operations taking place• design to allow for cleaning• adequate labeling (at point of operation)
Some layouts are better than others, to meet these GMPguidelines, but in practice the following are preferred:
• segregate raw materials and final products• segregate different production suites involving different
classes of products• closed operations where possible• segregate physical barriers or other proven means ‘open’
operations with different products• ensure an orderly flow direction• provide distinct staging areas if required between process
steps• provide cleanable production suites and equipment• provide suitable environments for controlled areas where
products and their active material are stored and pro-cessed
Typical Overall Layout for a ‘Conventional’Design Facility
As stated previously, there is no one single arrangementadopted by facility designers to meet the following issuesaffecting the layout:
• relative location of warehouse to the production and pack-aging suites
• cleanroom suite arrangement• usage of different levels and IBC movements
Figure 3. U SPAH concept.
Figure 4. Ground Floor (Level 1) U SPAH facility layout.
• adjacency of technical areas to production rooms• QC laboratory location• separate raw materials and finished product routes• minimization of expensive ‘clean’ areas• avoidance of ‘white’ areas adjacent to external walls• dispensary location
Other layout considerations, which could be taken into ac-count in facility design, but are often lacking, include visitorviewing access, central supervisor area, and external visibil-ity of process room operation.
Two generic schematic arrangements for OSD plants areshown in Figure 2 indicating potential locations of the keycomponents. In these schemes, the technical area is generallyon the upper floor of the facility.
The Proposed Hub Facility ArrangementOverall ConceptA proposed hub arrangement for an OSD plant has beendeveloped,1 which is believed to offer advantages over con-ventional layouts. It can be applied for plants with outputs of0.2 to two billion tablets a year or more, in principle, where aone or two process level approach is adopted. Its first applica-tion is considered for a ‘standalone,’ greenfield two processlevels plant manufacturing a number of similar class prod-ucts on a campaign basis. The scope and requirement for thefacility includes warehousing, manufacturing, support ar-eas, technical space, and utilities generation, and companyadministration offices. In addition, its scope includes a GMPpilot plant for process R&D plus small-scale manufacture fortrial material.
In developing this layout arrangement, a number of keyattributes and features were sought, namely:
• adopt an overall unidirectional materials flow through theplant starting with raw materials in and final product outwith two (relatively) small warehouses
• maximize adjacency of materials storage with productionsuites
• adopt a central ‘spine’ in the building in both the supportareas and process area around which materials flowed andthe process functional rooms are located
• maximize technical space adjacency to production rooms
• provide an IBC handling and discharge level above theprocess and filling rooms
• integrate the R&D suite into the facility in an optimal way
• close adjacency of the QC/QA laboratory to all operations
• avoidance of a separate upper floor technical area aboveall of the facility footprint
• minimize under-utilized technical plant space and cleancorridors
• provide a visitor viewing gallery through the plant whichmaximizes visibility of the process areas without enteringthem
• analysis of final product only with no inter-stage QA holdpoints, thus minimizing staging area requirements
• optimize facility space need for minimal cost
In order to achieve these aims, the facility concept wasdeveloped by adopting an overall U flow of materials andprocess operations around a central spine at the two buildinglevels, and designated as the U Satellite Process AssuranceHub (U SPAH) layout - Figure 3. There are three geographi-cal zones along the building within the overall ‘U’ flowpattern where materials handling and production operationsare ‘wrapped’ around the central spine which provides accessfor people, the QC laboratory at Level 1 (ground), and peoplecirculation at Level 2. The spine forms the hub in the produc-tion suite.
The proposed hub layout is shown in more detail inFigures 4 and 5 for each of its two levels for the production
Approximate Facilities Comparison(1)
Parameter U SPAH 1 U SPAH 2 Project A Project B Project C Project D
'White' production 20 21 28 17 15 14
Production staging (w+g) 7 7 12 5 5 4
Production support (w+g) 19 14 14 15 15 11
Technical black areas 29 31 39 45 51 47
Warehousing/Dispatch 25 27 7 18 14 24
Total facility area – approx m2 3200 7000 18000 10000 7400 8000
Notes: (1) Numbers are percentage of total facility area required by this part of facilityw 'white' area, i.e., clean area in facility used for production support, e.g., washing areas, stagingg 'grey' area, i.e., areas of facility used for production operations/support, but lower level of cleanliness
Table A. U SPAH vs. conventional facility features.
Figure 5. Upper floor (Level 2) U SPAH facility layout.
scope of the facility which includes dispensing and IBCfilling, two granulation rooms, three tablet rooms, two coaterrooms, one capsule room and three filling lines plus supportareas.
Material Handling and Support ServicesThis consists of warehousing and QC at Level 1 with offices,support areas, and technical space at Level 2. Two separatewarehouses are provided for raw materials and final productsrespectively, with raw materials entering at one side andgoing into the production area and final packaged productcoming out the other side of the ‘U’ into the final productwarehouse and dispatch area. The larger raw materialswarehouse extends to roof height, while the smaller finalproduct warehouse is limited to Level 1.
The QC/QA laboratory is located in the central spine atLevel 1 between these two areas to give it adjacency tosampling, production, and packaging.
Level 2 provides the main space for office staff and peoplecirculation to the production suite by incorporating in the‘spine’ corridor a clean change (gowning) area for productionand R&D staff.
R&D Pilot PlantAlthough the GMP ‘mini’ production suite pilot plant isembedded in the facility, it is located in a discrete separatezone within the building. Given that this part of the facility
is separate, but adjacent with the production facility, it isbelieved that such arrangement has decreased cost andincreased efficiency and control in technology transfer, andsuch implications may lead to corporate competitive advan-tages.
Adjacent to the cleanroom R&D suite is the stabilitychamber room and R&D staff offices with visibility into thesuites and a technical space on the outside wall to facilitateinstallation of supporting items to the processing equipmentin the suite.
Production AreaThe key concept developed is to ‘wrap’ the process roomsaround the central hub in the ‘spine,’ and ‘wrap’ the technicalspace at Level 1 around the clean process rooms. Hence, rawmaterials directly enter the production area via the adjacentpre-dispensary/dispensary on one side of the building withfilling and packaging on the other side of the building suchthat final product then directly enters the final productwarehouse.
This concept provides at Level 1, the minimum cleancorridor and a compact production room arrangement. Pro-cess support rooms have been allocated to the ‘central hub’including a wash area, small process control lab room, super-visor office, and storeroom.
The operational concept is that IBCs are filled in thedispensary and then taken via the lift in the production area
Table B. Comparative area requirements - U SPAH vs. conventional facilities.
SPAH vs. Conventional Facility Features
SPAH Feature Benefit Conventional
Control hub overlooking production suites For monitoring production processes for safety, Usually not availablecompliance, quality
Tech corridor wrap around production suites along Non-intrusive maintenance of thru-the-wall process Usually not wrapped aroundperimeter of plant equipment
Viewing gallery Non-intrusive visitor viewing Usually not available
Uni-directional U-shaped layout of production suites More efficient and ergonomic operations Usually not in a compact U-shaped flowaccording to process flow
QC labs in close proximity to production suites Speeds up QA time Usually not in close proximity
Separate raw materials and finished goods loading areas Eliminate mix-ups Usually shared areas
One cost effective security Hub (concentration) at the Better and cost effective monitoring of people and Material and people flow entrances and exit areentrance monitoring both material and people flow material flow not in close proximity to facilitate one security hub
R&D pilot lab well embedded and integrated with the Facilitate cost, compliance and performance of R&D pilot lab unavailable or not well embeddedproduction facility in single building technology transfer from process lab to production and integrated with the production building
Figure 6. Visitor viewing gallery concept.
for blending at Level 2 and charging to the various equipmentitems below. Different IBCs are then filled with processmaterial after each unit operation in the various processrooms at Level 1. Finally, tablet IBCs are filled with tabletsand then fed from Level 2 to the filling machines below. Theusage of the second floor IBC discharge minimizes the plantfootprint, which was a key requirement.
Within this ‘U’ concept, the number and size of all roomscan be adjusted to suit the amount and size of the equipment.The surrounding technical space has in it support equipmentfor the production equipment such as coater air handlingunits, together with pipework, cabling, and ducts. The fillingrooms are accessed off the clean corridor extension from thehub area.
Level 2 consists of the IBC handling/discharge stationarea around the ‘black’ central hub. The IBC handling areaconsists of lift access, closed IBC discharge stations to thefloor below, IBC blender room and IBC washroom.
Since all operations are nominally closed, and since theproducts are nominated to be of the same activity class, thenit is not considered necessary to segregate each IBC dischargestation in its own room, but a partial height barrier anddifferent time discharge procedures could be used.
Central HubWe believe the central hub area offers a number of advan-tages. At Level 1, it provides a central location for commonfunctions. At Level 2, it provides a central area adjacent toand above the cleanrooms in which to locate their AHU plantin a space efficient way, and it allows circulation around theclean area for personnel in factory ‘black’ clothing. In thiscentral hub arrangement, a viewing gallery as indicated inthe partial building cross section can be provided in thefacility - Figure 6. In this arrangement, visitors and companystaff can have un-precedented direct visual access into mostof the production rooms below and across into the IBChandling area via transparent material ceiling/walls. A con-trol and information room also can be located at Level 2.
This is an advantageous arrangement to companies whodeem accessible visitor viewing into production operations agood feature to support company image and sales.
Utility PlantThe green field site facility necessitates the need for basicutility generation equipment and this equipment has beenlocated at Levels 1 and 2 at the end of the building strategi-cally adjacent to the technical space for cost minimization.These include chillers, air compressors, hot water system,process water, purified water, and electrical MCC room.
Believed Advantages of U SPAH LayoutEfficient Space Utilization and Cost SavingsA key aim of the project was to develop a compliant lowest cost‘lean’ plant design incorporating all of the components manymanufacturers could need in their facility and operation.Cost reductions can be considered both in lower overallfacility floor area for the same production output, and thelower area of the high cost clean areas for production.
A preliminary comparison of this facility against otherdesigns has been carried out using a number of criteria. Inorder to attempt to provide a numerical comparison withother production facilities, we have normalized the new
facility by taking out area allocated to administration andR&D area since these are specific to this facility. Compara-tive values are given in Table A, where we have included twoSPAH sizes, one for 150-600MM tablets a year for the produc-tion scope given and another estimated for two billion tabletsa year containing two dispensaries, two blenders, two granu-lators, four compression stages, two coaters, one capsuler,and four filling lines.
It can be seen that the area ratio profile generally followsthe same shape for all plants. However, it can be seen that theSPAH layout differs to the benchmark and the other plants inthat it appears to have a distinctly lower black technicalspace percentage.
As stated earlier, due to the variations in equipmentnumbers and plant weekly operation times, normalizingfacility area to output is difficult to judge, but from the TableA data, at least a 20% reduction in area appears achievableoverall, which could provide significant savings when fullyserviced facilities typically cost $3000-5000/m2 to construct.
We believe the lower white circulation area can be attrib-uted to the ‘loop’ white corridor in the process area and thelower black technical area percentage can be attributed to theabsence of a space inefficient upper floor technical area forthe ACMV plant since this has been located centrally in the‘hub.’ These two items result in reductions in facility floorarea and consequently provide capital and operating costssavings compared to conventional designs.
GMP and Technology Transfer ConsiderationsWhile all qualified OSD plants in production meet GMPrequirements, the degree of attainment of GMP objectivesabove “basic” levels can vary from plant to plant. It isproposed that the new layout concept has the followingintrinsic GMP advantages over conventional plant designs:
• segregated raw material and final product warehouses
• clear and separate raw materials and final product flowpaths through the plant
• good access control of process personnel to productionareas
• availability of the technical space behind every productionroom, allowing easy use of ‘through the wall’ technology tominimize congestion in the rooms
• convenient location for a centralized information room tofacilitate the implementation of the FDA initiative
• through the wall technology is available for each produc-tion room, allowing less congested process rooms easier toclean for multi product facilities
Another wider GMP consideration is the benefit of the inte-gration of the R&D pilot/development plant with the produc-tion plant. Even though segregation of such activities iscommon, integrating these two operations on one site for
many companies will allow much easier technology transferfrom a regulatory and speed/cost of transfer perspective.
Operational ConsiderationsWe propose there are operational considerations and poten-tial benefits with the SPAH concepts as indicated in Table B.The production suite ‘U’shape around the hub has a numberof advantages. The transparent wall and ceiling at Level 2allows production management outside the cleanroom areasto observe operations and provides visitor access to see thecleanroom operations with no disturbances to operations inthem and the costs incurred by this. This ease and scope ofdirect visual access to the unit operations can facilitatesupervisors or managers to identify and monitor the stageand situation of the unit operations. Such visual accessibilitybenefits can easily be taken advantage of for the entire lifecycle of the plant operation, including equipment hook-up,qualification, validation, assurance of good practice of clean-ing and manufacturing operations etc.
Also the corridor around the process rooms allows easymaterials movement between each room, and with the relativenear location of the IBC lift to all the suites, it allows easylogistical access to the charge points and operations at Level 2.
The central hub area at Level 1 provides a convenientstrategic location for cleaning operations as well as supervi-sor office and in process IPC lab.
Each production room has a rear wall to the technicalspace, and can be easily isolated from ongoing productionoperations for equipment change or new equipment installa-tion and hook up via the technical space.
Project Implementation - Standard DesignMany different dosage form facility designs have been devel-oped to date in terms of content and layout configuration,both horizontally and vertically. Each ‘new’ design costsmoney and time to develop which can significantly affectproject implementation. We believe that this compact ar-rangement could offer a lower cost standard design solutionfor many companies either as a small and strategic newproduct launch facility or as a production facility with theconsequent capital and project cost and time savings whenutilized.
ConclusionMany different designs and sizes of OSD facilities are utilizedby most pharmaceutical companies to manufacture dosageforms. In this article, we have proposed a compact layoutdesign concept. We have made a preliminary comparison ofthe area and facility space usage with a number of otherlayout facilities. Through this comparison, it is believed thatthe hub design offers a smaller and lower cost facility inaddition to some key GMP and operational advantages thatthe SPAH presents. Such advantages and feature benefits –with or without R&D operational support – could initiate theSPAH concept to become an adopted design standard forcompanies embarking on future production facilities or for anew drug launch facility.
About the AuthorsDr. Michael P. Brocklebank has been in-volved in the pharmaceutical industry formore than 30 years both working initiallywith a major manufacturer and then for mostof this time with engineering design andconstruction companies. Over this period, hehas developed a wide knowledge of the devel-oping trends and design principle of a num-
ber of plant types for API, biological science, and dosageprocess. Brocklebank is currently the Manager of the Phar-maceutical Division of Foster Wheeler's Singapore Office. Heis responsible for developing pharmaceutical business inSingapore and Southeast Asia, and is currently serving as theVice President of the ISPE Singapore Affiliate.
Foster Wheeler, 32 Maxwell Road, #02-03, The Whitehouse,Singapore 069 115.
Joseph Lam graduated from Ohio StateUniversity with BS in pharmacy in 1994 andhe has eight years of experience in the phar-maceutical industry. He is currently the Man-aging Director of Beacons PharmaceuticalsPte. Ltd. which is Singapore’s largest genericand contract manufacturing company. He isthe inventor of SPAH System Technology,
and holds other patents. He is also a member of ISPE.Beacons Pharmaceuticals Pte. Ltd., 53 Quality Road,
Singapore 618 814.
Dr. Pranav H. Mehta is a Chemical Engi-neer, PhD (Tech) with more than 10 years ofexperience in the Pharmaceutical manufac-turing industry designing, executing, andcommissioning, API, and biotech. In the lasttwo years, he has been with a consultingorganization and involved in designing anOSD and API plant. Mehta is currently the
Senior Technology Engineer of the Pharmaceutical Divisionof Foster Wheeler's Singapore Office.
Foster Wheeler, 32 Maxwell Road, #02-03, The Whitehouse,Singapore 069 115.
