 outsourcing survey is to gauge overall business trends in the contract manufacturing and research businesses. This year's edition documents an industry that is in good health but facing a new set of dynamics. 

This year's survey results indicate that the market remains strong, but signs suggest that demand may be leveling off. Buyers of contract services at bio/pharmaceutical companies report that their spending will be higher in 2006 than it was in 2005, but not as robustly as it was in 2005 (Figure 1). Only 16% of respondents expect their spending on contract services to decline, but just 38% expect spending to increase by 10% or more. This figure is down from nearly 50% in last year's survey. The spending slowdown reflects, in part, a reduction in the growth of spending overall (Figure 2): 43% of respondents report that spending on contract services is growing at the same rate as overall spending in their area of responsibility. Nonetheless, 39% report that their outsourcing spending is growing more slowly than their total spending.  

 Contract service providers view the market very much like their bio/pharmaceutical clients (Figure 3). Some 51% expect their revenues to grow by 10% or more in 2006—still a strong showing, but down from 62% in the 2005 survey. More vendors expect to grow by less than 10% this year (39% in 2006 

 27% in 2005), but only 10% expect revenues to decline.  

Contractor optimism over new contract signings also is down somewhat this year. Only 48% of contractors expect new contract signings to increase by 10% or more this year, which is down markedly from 62% in last year's survey. Slightly more contractors expect a net decline in new contract signings this year 

Somewhat surprisingly, contractors don't seem to be alarmed by the slower pace of new contract signings, which should be a bellwether for revenue in the coming months and years. Fifty-five percent of contractors expect business opportunities in 2007 to be "somewhat better" or "much better" than they have been in 2006 (Figure 4). Nonetheless, bio/pharmaceutical respondents indicate that contractors shouldn't be too sure: 48% expect contract spending to grow by less than 10% next year, and only 10% expect spending to grow by 20%, which is almost half as many as those who predicted a 20% spending increase last year (Figure 5).  

Contractors aren't resting on their laurels. Bio/pharma buyers report that contractors are aggressively seeking new business. That desire for new business is not necessarily at the expense of price, however. Only a third of respondents indicated that vendors have been willing to cut prices to get their business. Just more than half said that vendors have been asking for new business, but have not been willing to cut price to get it.  

One check on price cutting—and a brake on growth—may be that contractor capacity is growing tighter. The percentage of contractors reporting that they are facing capacity constraints is up to 43% in our 2006 survey 

 35% last year. Finding staff has been the greatest challenge for service providers battling to grow faster, followed by their ability to find capital for investment. The availability of new equipment is less of a problem in 2006 than it was in 2005. 

In our 2006 survey, we continue to track the evolution of procurement practices, which we believe are undergoing fundamental shifts that will impact the contract services industry significantly. Price pressures on the pharmaceutical industry have forced drug companies to take greater control of their costs, and procurement groups are on the front lines of those efforts.  

A major component of this trend is the shift of decision-making power and influence from scientific and technical professionals to sourcing professionals. Service providers are certainly experiencing the growing power of procurement operations: the number of contractor respondents who report that sourcing and procurement groups "control the decision" nearly doubled in 2006, to 22% from 12% in 2005. The portion of respondents that expect sourcing and procurement groups eventually to control decision making grew to 30% from 26% (Figure 6).  

 Procurement and sourcing groups are driving the consolidation of the supplier base that most major pharmaceutical companies have made a critical component of their cost-reduction strategy. Shrinking the supplier base enables companies to reduce vendor management and overhead costs, negotiate better pricing based on the higher volume of purchases, and improve coordination and communication mechanisms.  

Our 2006 survey indicates that vendor consolidation efforts are well underway (Figure 7). Among bio/pharmaceutical company respondents, 29% indicated that they already have begun reducing the number of suppliers they work with, and another 15% expect to begin the process within the next year or so. Nearly half of those client-side respondents report that they consider using new vendors only when they need specialized capabilities or services not currently available to them, and 13% said that they are using only current or recent vendors.  

Contractors are beginning to feel the effect of supply-base consolidation efforts. Among contractor respondents, 20% say that they are gaining business, while 16% report that they are losing business as result of supplier consolidation programs. Nearly half of respondents do not expect consolidation to have much effect on their revenues, which we think is a naïve position for them to take. 

Service providers willing to be proactive and keep knocking on doors should do well, however, because plenty of opportunities still remain. Almost a quarter of respondents report that they are looking to expand their supplier base and are continually identifying and testing new vendors. 

We continue to track bio/pharmaceutical company interest in sourcing from Asia, especially contractors in India and China. Our survey results show a slowly-but-steadily growing interest in sourcing from Asian supplies, but the pace is not yet alarming to Western service providers. 

In this year's survey, 20% of bio/pharma company respondents report that they are actively sourcing now from suppliers in India and China 

 16% in 2005. Slightly more companies aren't sourcing from Asia yet, but are preparing to do so. Thirty-one percent indicated they are either searching for vendors or developing a strategy 

Nearly a quarter of bio/pharma respondents expect 5–10% of their external spending to be with Asian suppliers in two years, a big jump from the 15% reported in the 2005 survey. The share of respondents expecting to spend more than 10% of their budgets in India and China in 2 years was flat at 18% (Figure 8).  

Western contract service providers are not panicking over Asian competitors yet, but most know that they will feel an effect sooner rather than later. In this year's survey, 27% of contractors say they expect to feel a "considerable" impact from Indian and Chinese competitors within two years. That is up from just 15% of respondents in last year's survey and is consistent with the bio/pharmaceutical company findings cited above. Twenty-nine percent of contractors expect to feel a "considerable" impact in five years, again up sharply from 2005. 

 outsourcing survey has documented a slowdown in the growth of demand for contract services, but the revenue and spending projections certainly don't portend an overall decline in the market. The year 2005 represented a significant rebound from the early part of this decade, when a thinning pipeline, funding constraints, and Big Pharma restructuring combined to slow the pace of drug development. The industry has moved well beyond those constraining factors, and new drug development is booming. In addition, Big Pharma is increasingly inclined toward outsourcing as companies seek to reduce costs and free up cash for investments in new products. 

Nevertheless, this year's survey predicts substantial changes in the outsourcing environment, especially for contract services providers. We have confirmed the trends toward consolidated supplier bases, procurement-driven processes, and the opportunities to reduce costs 

 offshore sourcing. My conversations with industry participants have convinced me that nonclinical development and manufacturing service providers have not fully grasped the implications of these trends and have failed to react adequately to their implications.  

CRO and CMO executives need only look at the restructuring of the pharmaceutical chemical industry in recent years to see what awaits them if they don't respond to the new realities of their industry. 

Articles

Context

The principal objective of the annual PharmSource–Pharmaceutical Technology outsourcing survey is to gauge overall business trends in the contract manufacturing and research businesses. This year's edition documents an industry that is in good health but facing a new set of dynamics.

Slowing Growth, Accelerating Changes Mark Contract Services Industry

The chemical synthesis industry is a dynamic, highly segmented, and growing component of the overall outsourcing business. It has a distinctive service platform and customer base compared with chemical suppliers, thus requiring a clear understanding of key success factors, expectations from pharmaceutical company customers, and changing global economic environments. This article describes the current extent and scope of the chemical synthesis business within the framework of increasing regulatory compliance demands and a growing number of pharmaceutical firms that are requesting contract services for their chemical operations. 

Custom synthesis is a business operation conducted by a third-party vendor to build molecules on the behalf of a specific customer that typically has no legal affiliation with the service provider. Custom synthesis is one of the multiple segments within the outsourcing space. Although it is defined as covering only molecule-building activities, custom synthesis represents in of itself a very complex field made by a multitude of segments and subsegments (see Figure 1), which reflect the situations and variants that can be observed, including:  

Figure 1: Custom synthesis comprises several subsegments.									

the type of input supplied—namely, either a service (

the development of an optimized process or a technology package ready to be transferred) or a product (

 early- or late-phase development, launch, maturity, or sunset), which has implications for the quantities supplied as well as customer expectations;

 agrochemical, personal care, pharmaceutical, and electronic chemicals) such that regulatory and vendor qualifications hurdles vary greatly depending on the application.

 Both the customer and the vendor slate in custom synthesis are very broad, reflecting the extreme fragmentation of this arena and the number of situations encountered. Similarly, contractual conditions associated with custom synthesis range widely depending on the scope of the vendor–customer agreement. As an example, the agreement can be either restricted to the conversion on a cost-plus basis of a raw material supplied by the customer or it can cover the vendor's development of the technology processes required to effectively produce the end molecule expected by the customer. 

It is therefore no surprise that, given this extreme fragmentation, the precise boundaries and size of the custom synthesis space are ill defined. The industry sales range is estimated from more than $50 billion to $6–7 billion when the scope is limited to the life-science outlet. 

Some traits clearly differentiate custom synthesis from other segments of the chemicals industry. These include: 					

the one-to-one relationship existing between the vendor and its customer. The vendor supplies the product or the service on an exclusive basis to the end user that typically is either the originator or owns the full rights on the molecule;

the service-type dynamics of the business. The vendor engages in supplying the molecule at the request of its customer. This means that the vendor is entirely dependent on its customer and has either no or limited control on the ultimate fate of the business should the product be withdrawn from the market or should the customer elect to change its supply source;

the limited transparency of the business. Customers typically insist on maintaining confidential dealings with their vendors, probably as a tactic to increase their leverage.

Although it is often associated with fine chemicals and the life science industry, custom synthesis covers all segments of the chemical industry. Custom-synthesis agreements pertain to both large-volume commodity chemicals, supplied in hundreds or thousands of tons as well as to molecules produced at the microgram scale. The end-user base for custom synthesis also is extremely broad, comprising both integrated end users that have their own chemical production facilities and that elect to outsource part of their requirements as well as virtual companies that have no option but to rely on a third-party supplier for their molecular requirements. Examples include start-ups such as Gilead Sciences (Foster City, CA) or NiCox (Sophia Antipolis, France) as well as established pharmaceutical companies such as Wyeth (Madison, NJ) or Solvay  (Brussels, Belgium). 

The vendor slate is equally extremely broad. Several thousand companies supply custom-made chemicals, including both large, diversified chemical companies such as Dow Chemical (Midland, MI) or Lanxess (Leverkusen, Germany)—which have developed a base in custom synthesis as a strategy to leverage their technology base to load assets—as well as players for which custom synthesis is their main business, including, for example, Albany Molecular Research (Albany, NY), Proviron (Oostende, Belgium), and Siegfried (Zofingen, Switzerland). Interesting to note, the number of players in custom synthesis has increased over the past five years. As an example, China and India have become increasingly important forces in custom chemicals, particularly in some product groups such as intermediates. Their entry has contributed to exacerbating competitive intensity levels,  thereby depressing profit margins in the industry. 

The birth date of the outsourcing industry is difficult to pinpoint because outsourcing is viewed as a logical and gradual development. Rather than insisting on full backward integration, end users elect to consider other options for securing access to required inputs (

 materials and services), including relying on external vendors for items traditionally produced in-house. 

 The drivers behind the decision to outsource are multiple, including: 					

a focus on core competencies, often referred to as those operations and processes that allow a company to create value effectively and provide scope for a competitive edge. Manufacturing is rarely considered as one of these operations, contrary to product design, research and development, and marketing;

the new metrics adopted by management following increasing scrutiny by investors and financial analysts pushing for a more effective use of capital. Options to avoid in-house investments in fixed assets such as production plants are increasingly embraced (

 outsourcing to a third party). Interestingly, this drive toward reducing the asset base has prompted some companies to divest part of, if not their entire, manufacturing network;

the strategy of companies to avoid building up their own fixed cost base and instead electing to outsource some operations. Outsourcing is a convenient tool to shift costs to an external entity, thereby enhancing financial flexibility.

the desire to tap the best available resources when and if required, thus avoiding building an in-house  resource base that would be difficult to maintain fully loaded and alert;

advances in information technology and communication systems combined with more-sophisticated management tools. These systems enhance the transparency of internal operations, including those pertaining to activity costs, thereby enabling companies to compare these with alternative options such as outsourcing;

the emergence of viable alternatives to in-house operations. Several vendors focusing their activities around outsourcing services make these their core business (

 Lonza [Basel, Switzerland] in fine chemicals molecule building).

Several new entrants are seeking extensive outsourcing of most of their operations, thereby challenging the dominance of traditional fully vertically integrated contract service providers. This action has forced these companies to reassess their mode of operations, including sourcing. Examples include most emerging biotechnology companies that are focusing on discovery and development. 

 It is difficult to pinpoint the origins of custom synthesis to a specific event. The first indications and reports on custom-synthesis activities date to the 1960s and 1970s. At that time, several end users, including some large pharmaceutical companies, saw the demand for their products surge, outstripping their own synthesis capacity. Parallel to this, the paradigm calling for backward integration among several end-user outlets for chemicals such as pharmaceuticals started to become increasingly challenged. This led several chemical users to rely on third-party vendors for not only "standard" catalogue items but also for exclusive products made under custom-synthesis supply contracts. 

One of the first documented examples of custom synthesis appeared in the late 1970s and early 1980s with the supply of cimetidine intermediates to SmithKline & French (Philadelphia, PA). At that time, the sales volumes for cimetidine (the first anti-ulcer drug treatment with truly curative effects, thus reducing the need for surgery) exploded, outstripping the most wildly optimistic assumptions. Facing the prospect of losing sales volumes resulting from short product availability given chemical synthesis bottlenecks, Smith Kline & French contacted various fine chemicals vendors including Fine Organics (Seal Sands, UK), Heico Chemicals (Delaware Water Gap, PA), and Lonza, asking them to supply on an exclusive basis selected cimetidine precursors such as the imidazole or the cysteamine intermediate. The overall concept and business model applied was simple. The originator provides the process to the vendor and the vendor applies the process in its own facilities. The vendor took advantage of the infrastructure on hand, and the customer avoided investing in its own capacity while securing immediate access to capacity if and when required—a win-win scheme. 

 The custom synthesis arena has evolved and grown greatly. Landmarks in this development path include the following: 

 The pioneering efforts of some companies such as Lonza helped to develop the concept of custom synthesis by positioning themselves as credible alternatives to in-house chemical operations ranging from development to full-scale production. 

 The role played by start-ups in various end-user segments such as pharmaceuticals is growing. For example, although the share of start-ups in total pharmaceutical sales is still modest—not exceeding a few percentage points of the global $500+ billion pharmaceutical turnover—an increasing number of new molecular entities are stemming from this set of players (see Figure 2). Because these companies have no own synthesis capacity, they have no choice but to rely on third-party suppliers for their chemical requirements, thereby providing a new impetus to custom- synthesis demand.  

Figure 2: Share of new molecular entities (NMEs) by type of originator.			

 Globalization and increasing transparency of chemical markets have combined with mounting financial pressures faced by chemical companies and chemicals users. These companies are expected to compress their fixed cost base while also reducing outlays in capital expenditures (capex). The industry is gradually starting to pay more attention to concepts such as economic value-added or the ability to ultimately earn acceptable returns on capital invested. End users are electing to source their chemical requirements from third-party vendors rather than continuing to produce in-house. Examples include personal care companies such as Procter & Gamble (Cincinnati, OH) relying on sources such as China for some of their fatty amines requirements as well as DuPont (Wilmington, DE) outsourcing part of its intermediate requirements for various agrochemicals to the Jayhawk fine chemicals business of Degussa (Jayhawk, KS) or the now defunct Hickson & Welch (Castleford, UK). The rationale behind such outsourcing includes access to key technologies, capex reduction, cost structure optimization, and time-to-market imperatives. 

 Compared with the scenario in the early days, the custom-synthesis industry has become increasingly segmented. For example, in addition to full-scale production of synthetic molecules (organic or inorganic), the industry now also covers biological products as well as related development services. Examples include: 					

mycophenolic acid, an immunosuppresant used in organ grafts originally developed by Syntex (now Roche) and obtained through microbial fermentation. Mycophenolic acid traditionally has been produced on a custom basis by Antibioticos (Settimo Torinese, Italy), a company having a long tradition in fermentation;

etanercept— the fusion protein making the active ingredient "Enbrel," one of the most innovative treatments for degenerative arthritis. Obtained by cell culture, it has been largely sourced through custom-production agreements between the originator Immunex (Seattle, WA) and Boehringer Ingelheim (Ingelheim, Germany). Amgen (Thousand Oaks, CA) and Wyeth have been marketing the product and have  invested in their own capacity over the years. Similarly, UCB (Brussels, Belgium) has entered in a custom production agreement with Lonza for the product "Cimzia" stemming from the former Celltech R&D pipeline.

Although custom synthesis traditionally has been limited to the supply of full-scale volumes, there are increasing activities involving products and services associated with molecules still in early or late development (see sidebar "Outsourcing services for molecule development"). Another important factor in the overall evolution of custom synthesis is the emergence of vendors that provide a full range of outsourcing services, sometimes going beyond molecule building. Examples include vendors in India such as Matrix Laboratories (Hyderabad, India) and Suven Life Sciences (Hyderabad, India), which aim to become full-fledged pharmaceutical custom research and manufacturing (CRAM) partners by providing services in a wide spectrum of molecule-building activities—from microgram up to multiton scale—as well as dosage-form manufacturing. The business rationale is to offer a viable alternative to in-house operations, with the predicament always being the same: Keep in-house only the activities that genuinely contribute to creating value.  

Outsourcing services for molecule development			

Synthesis by far is not the sole activity outsourced by chemical companies and chemical users. Other activities include distribution and logistics, information technology, and human resource management.  

During the past two to three decades, outsourcing in general and custom synthesis in particular has undergone rapid growth. Some industry analysts have envisioned a growing polarization between, at one extreme, virtual companies that outsource their entire chemicals operations and, at the other end, service providers that do not own products but rather supply chemicals to other clusters of companies. This scenario can be legitimately questioned. 

Such a model predicated in the 1990s called for a radical departure from the full backward integration that was typical of most chemical companies or large end users such as pharmaceuticals. Rather, it postulated that chemical operations are best left to specialized vendors for whom chemical production is their core business, thereby enabling them to achieve better economics through a combination of experience and enhanced capital efficiency. Within this framework, the paradigm was that the cost of capital for this set of vendors would be lower than for the end users—investor expectations vested in developers and marketers were higher than those of traditional manufacturing companies. On the basis of these postulates, there have been predicaments from the financial community relayed to senior management that several chemical companies and chemical users would be better off entirely farming out their molecule-building activities and divesting or spinning off their chemical assets rather than relying on specialized third-party vendors.  

In doing so, companies would embrace a "virtual" model that would delegate to selected vendors major parts of, if not the entire, chemical assets and operations. Barring a handful of exceptions (observed mainly in Japan associated with changes in the Pharmaceutical Affairs Law that provides for the decoupling of the marketing license from manufacturing permits, which has prompted the spin-off of the chemical production arm of various pharma companies such as Fujisawa or Yamanouchi), hardly any player has moved so far as to fully embrace the virtual model. Rather, most chemical companies and chemical users now have adopted the concept of strategic sourcing where all options to secure access to the required chemical input are systematically evaluated on a "which is the best" basis. Within this framework, neither in-house production nor outsourcing custom production are viewed as ends 

. The overall objective is to pursue the most attractive avenue possible to access the required input, be it at the development stage or at full scale. To this end, various considerations are taken into account when formulating the sourcing decision (see sidebar "Decision factors in outsourcing").  

The supply structure for custom synthesis or, in general, custom chemicals is extremely broad and heterogeneous. In total, probably several thousand companies provide custom synthesis–related services. These companies include both small family-owned vendors such as Pressure Chemical (Pittsburgh, PA) or ChemShop (Weert, Netherlands) and divisions of diversified chemical groups such as Dow Pharmaceutical Services and Lonza's exclusive synthesis business. Depending on the supplier, custom chemicals can be either the sole business or part of a broader business portfolio. 

 The emergence of companies specializing in supplying custom chemical-related inputs should be viewed within the broader context of the reshaping of the chemical industry (see Figure 3). This reshaping has been prompted by:  

Figure 3: Four models of outsourcing providers, including chemical service providers.									

chronic inability of most chemical companies to recover their cost of capital over a full economic cycle;

evolving expectations from customers or stakeholders;

 As a result, the chemical industry has undergone major changes and consolidation leading to the emergence of focused models.  

For example, chemical utilities, often referred to as "commodity chemical companies," are focusing on large-volume products obtained directly out of basic feedstocks. Their value proposition revolves around low-cost positions, reflecting economies of scale and superior access to key feedstocks. Examples of such players include Methanex (Vancouver, Canada), SQM (Santiago, Chile), and Sabic (Riyadh, Saudi Arabia). 

As another example, product and technology specialists are centering their activities around selected product trees and technology know-how such as fluorination, nitration, or oxo condensations. Their product range largely revolves around forward commodities and value-added intermediates. Levers for value creation are represented by distinctive technology capabilities and focus. Examples of such players include the polymer intermediates business of Lonza, the building blocks group of Degussa (Düsseldorf, Germany), and the diphenols activities of Rhodia (Paris, France).  

Examples of genuine strategic partnership agreements between end users and vendors are relatively rare in custom synthesis.

Custom Synthesis and Custom Chemicals

The PharmTech Annual Event, held June 12–15 in Somerset, New Jersey, provided attendees with expert advice and useful strategies ranging from clarifying the roles and responsibilities of a sponsor and contractor to establishing effective due diligence and quality agreements, handling regulatory issues, and managing risk. Attendees were given opportunities to take part in small-group interactive discussions, and each session provided relevant "toolkits," including take-home templates for quality agreements, quality contracts, and process models.  

A successful outsourcing project begins with understanding the relevant contract definitions, regulations, and responsibilities of the sponsor and contract service provider. Applying an organization's strengths to achieve a "win–win" contract begins with clarifying a company's needs "the sooner the better" according to James Lodigiani, senior vice-president of global sales at Cardinal Health (Somerset, NJ, 

). Defining the contract situation also requires developing an appropriate course and the timing for this course. Templates, for example, can help speed up the process and clarify contract terms. As Lodigiani observed, a legal review is "critical for both sides" to maintain a successful contracting partnership. 

Defined as "the proper and sufficient earnest effort to accomplish what is undertaken," due diligence produces the necessary information for making a well-informed choice of a contract business partner, according to Roby Blasini, a former corporate consultant. Three components must be considered when assembling the optimal due diligence team: the

component, which includes determining the scope and nature of the project and developing the confidentiality agreement; the 

 component, which involves gathering the expertise of various scientific personnel such as biologists, formulation scientists, and process engineers; and the 

 component, which includes selecting an experienced quality professional, preferably with auditing experience. To select the appropriate contractor, quality must not only be part of the process, but "quality must be a component from the start in order to maximize its effectiveness in the process," said Blasini. 

These three components (business, technical, and quality–compliance) form the basis of the evaluation process as well as the analysis of results. The business evaluation, noted Blasini, includes evaluating the geographical location, the length of time the service provider has been in business, the company's economic health (

 its public annual reports), and the scope of the project. When evaluating the business results, the business team members will analyze the results of the assessment with an emphasis on cost. Technical evaluation and results analysis should take into account the provider's capabilities and facilities. Finally, the evaluation of quality and compliance should include a review of the service provider's regulatory history (

 inspections, consent decrees), though it should not be an in-depth quality-system audit. Instead, it should be "a high-level review of the facilities and overall quality systems of the potential contractor," said Blasini. A risk assessment is essential in evaluating quality and compliance results. On the basis of the recommendations by these three components of the due diligence team, sponsor companies can make a selection of the preferred contractor. 

Blasini followed with a presentation about the process for developing quality agreements. Also known as "technical agreements" in the European Union (EU), a 

 is "an arrangement between two or more parties with respect to the quality responsibilities of the participants, with the business objective being a service or product supply," he said. Whereas in the European Union quality agreements are required and regulated by EU directive and the European good manufacturing practices (GMP) guide, they are not required by the US Food and Drug Administration, though they are "an expectation."  

After determining the scope of the agreement, which may include product or material suppliers or service providers, companies can compose a quality agreement using one of three formats: a tabular format, a legal-style format, or a combination of both. Moreover, the quality agreement can serve as a "stand-alone agreement" such as in situations in which there is no business contract, or it can serve as a part of the supply and service agreement. As Blasini pointed out, the difference is that the latter contains quality content only, whereas as a stand-alone agreement, the document contains a table of contents, quality content, a confidentiality section, a chain of custody, a survival clause, a termination agreement, a procedure for amendment, and other agreements and legal sections.  

Agreements with manufacturers, for example, can apply to the following categories: equipment qualification, sampling, cleaning and process validation, batch release, storage and transport of product, certificate of analysis, certificate of compliance, complaints, and annual product reviews. 

Outsourcing is an investment, and maximizing the return on that investment is the goal of every contractual project. Karen Ginsbury, president of PCI Consulting Israel Ltd. (Petach-Tikva, Israel), outlined the key quality factors involved in an outsourcing project and provided recommendations on how to maximize value. For example, good communication, especially in the initial contact with an outsource service provider, is essential. "To have quality in the output, you must have face-to-face meetings before kick-off," she said. 

Ginsbury outlined the regulatory responsibilities of contract givers and contract acceptors as provided in EU GMP, Chapter 7. Companies must be "exceedingly precise" when defining the scope of the outsourced project. "You need to discuss the scope with relevant functions within your company and put out a document agreed upon and signed by these parties," said Ginsbury. For example, if the project is to involve process development in addition to production, companies must determine whether they or the contractor will own the developed process. "If the latter, you may be paying royalties forever, so be aware of this up-front and define it in the contract." 

Sponsor companies must take into account several quality issues when the contract is drawn up. For example, "manufacturing, analytical, and distribution records as well as reference samples should be kept by or be available to the contract giver." In addition, if there are complaints, it must be clear who is responsible for investigating the complaints, and the relevant records pertaining to quality "must be accessible and specified in the defect or recall procedures of the contract giver," said Ginsbury. 

During the course of the project, several questions also will have to be addressed. "It is critical to define who is responsible for purchasing raw materials: contract giver or acceptor. Who performs sampling and analysis, and will analysis be further outsourced? What are your company's requirements for access to the outsourcing facility during production, and will you be allowed to send a representative? What if a deviation or malfunction occurs and must be investigated? All of these aspects should be addressed in the technical quality agreement," said Ginsbury. "At the end of the day, outsourcing is a partnership. If the contract giver and contract acceptor can work with this concept of mutual benefit, both will give their best to ensure that the other party receives what they expected, wanted, and needed." 

Every outsourcing investment entails an element of risk, and with FDA's current emphasis on risk management, learning strategies to control and maintain that risk was of special interest to attendees. The final presentation was on defining and managing risk, given by Charles F. Carney, consultant for Seraphim Life Sciences Consulting LLC, who provides a detailed description of this topic as a separate article in this issue. 

Quality must not only be part of the process, it must be a component from the start.

Outsourcing Toolkits from the PharmTech Annual Event

This article proposes a block diagram process model that can be used to ensure that the information and input from each part of the organizational elements are collected and summarized according to standard categories of work tasks in GMP-regulated organizations. It briefly addresses some of the means to achieve quantitative and qualitative risk evaluations, including the use and limitations of the risk table for summarizing risk factors and management options. It emphasizes the importance of quantitative assessments where possible and the scientifically valid and defensible collation and analyses of the data.  

In 2002, the US Food and Drug Administration pointed out the need for a new paradigm in drug development. As part of its description of a new order for current good manufacturing practices (CGMPs) regulations, the agency introduced its desire to begin using a risk-based approach for drug development, manufacturing, and control. The final report for this initiative was issued in 2004 (1).  

"Risk" is now a synonym for "opportunity" in the pharmaceutical industry. Risk definition, assessment, mitigation, and elimination for all manufacturing and control processes should become the basis for thinking about product development and life-cycle management for every product. When a company can show—with supportive technical reasoning and meaningful, carefully and correctly evaluated data—that it completely understands the manufacturing process for its product and can mitigate or eliminate altogether the risk factors for failure through materials and process controls, then making future changes in manufacturing procedures becomes much easier to perform, document, and report. True assurance of consistency and value for each successive product lot will ensure greater patient confidence, disease treatment with minimum excursions, and overall cost reductions in terms of surveillance, documentation preparation and review, and industrial waste generation (

failed batches, recalls, and reprocessing). 

This understanding must extend to all of the materials, facilities, personnel practices, and procedures that contribute to the manufacture of the product. This knowledge applies not only to work done in-house but also to work performed by a contract manufacturing organization (CMO). For products already on the market, it may be difficult and cost prohibitive to establish retrospectively the scientific understanding for the materials, equipment, and procedures used to establish a complete risk assessment and risk-management action plan (RiskMAP) (2). Nonetheless, early development is the ideal time to start accumulating knowledge and understanding about any new product. 

The concept of risk assessment and management applies not only to technical matters such as material quality assessment and lot-to-lot consistency, analytical control capability, machine performance and control, influence of facilities, and manufacturing procedures. Risk assessment also applies to interactions between organizations such as those for the procurement of starting materials and those for the procurement of manufacturing and control services for the production of the product. 

A systematic and objective process for performing the risk assessment and preparing or executing the RiskMAP must be invoked by each organization. This article proposes a block diagram process model that can be used to ensure that the information and input from each part of the organizational elements are collected and summarized according to standard categories of work tasks in GMP-regulated organizations. The author also briefly addresses some of the means to achieve quantitative and qualitative risk evaluations, including the use and limitations of the risk table for summarizing risk factors and management options. It emphasizes the importance of quantitative assessments where possible and scientifically valid and defensible data collation and analyses.  

 is a small word with a large connotation. It is used in many contexts in the pharmaceutical industry, but primarily in a nonquantified format in each case. To achieve the highest usage level of risk-based development and manufacturing as intended by FDA, industry must develop quantifiable risk algorithms for each function and process within the function. 

 vary and give little guidance for quantification. For example, some dictionary definitions include: 					

Only the latter definition provides an idea that risk can be calculated according to a probability algorithm. Nonetheless, all the definitions connote a negative outcome whether or not it is quantifiable, and that further implies an element of action or process that might result in a negative result. 

				An Internet search yielded the following definitions for "risk":									

potential future harm that may arise from some present action;

an expected after-the-fact level of regret (4).

These definitions also imply the negative consequence of a process or action. 

The banking industry in 2004 recognized and defined risk associated with organizational interactions in the following definition:  

is the loss resulting from inadequate or failed internal processes, people and systems, or from external events (5).			

FDA recently provided guidance with respect to the need for an action plan to identify, evaluate, and manage such consequences of risk in the following definition:  

Risk Management Action Plan [FDA RiskMAP] 

is the strategic safety program to meet specific goals and objectives to minimize risks for a product while preserving its benefits (2).			

Thus, there is an expectation that a pharmaceutical company will proactively and systematically identify risks that might negate some deliverable quality attribute of a product and have a program in place to prevent or minimize these identified risks. Although negative impact on product quality will be linked most directly to the technical aspects of production and control, one must also consider the effect of negative organizational performance leading to decreased product quality as a potential true root cause.  

Facilities, production machines, and analytical equipment require a human interface in their use to produce and control the medicinal product. Personnel and procedural systems, whether in the sponsor organization or in any of the supportive supplier organizations, affect the final quality of that medicinal product. Therefore, the RiskMAP must take such organizational elements into account, at least implicitly, as well as the technical elements. These are paraphrased from the FDA guidance as follows and apply throughout the full life cycle of the product. 

Goals target the achievement of outcomes to mitigate known safety risks.

Objectives are the pragmatic, specific, and measurable steps taken to achieve a goal.

The program requires evidence-based risk identification, assessment, and characterization throughout the product's life cycle.

Medicinal product development and registration are a continuum of tasks. This process was recently described in a pictorial that shows the pathway leading from product concept through design through development and execution (6, see Figure 1). This pictorial clearly shows the concept of quality by design for any medicinal product and the need for proactive management of all suppliers, whether they supply materials or services. It shows that each of the functional areas must be evaluated with respect to "evidence-based risk identification, assessment and characterization" (2).  

Figure 1: Process pathway from product concept and definition to market supply. This pictorial has been modified to indicate that risk is a factor in each of the segments of the development pathway. Source: ref. 6. 			

Three schematic elements, "manage supplier quality," "build the product to customer needs," and "control process and product conformance," are the most pertinent elements for any discussion about sourcing product manufacturing and analytical control testing at a CMO. 

Humans have the ability to conceptualize fairly complex systems and tools. Such ideas are termed 

and apply to physical objects, to regulation of interactions between people either in the allowance or constraint of actions, and to processes for achieving intellectual objectives (

solving quadratic equations) or building useful items.  

The identification and assessment of risk and development of a risk management plan is a process. Such evaluation determines the degrees of freedom that one has in the performance of that particular process. It is fairly clear that the risks are greater in number and potentially more complex when a sponsor uses a CMO whether for manufacturing operations, control testing, warehousing and distribution, or all of these. The organizational complexity and inherent risk of negative results are great within the sponsor company and can be even more complex and risky when the sponsor chooses to use another company to supply materials or services. 

Our mental model of CGMPs has evolved over time. A pictorial view of one mental model of the process for product development and marketing is shown in Figure 2.  

Figure 2: A mental model of the evolution of the product development process over time. White boxes indicate a concept just before the introduction of good manufacturing practices (GMPs). White and green boxes indicate a concept at the introduction of GMPs. White, green, and blue boxes indicate an addition of qualification and verification to developmental aspects of current good manufacturing practices (CGMPs). All boxes indicate contemporary CGMPs, including the RiskMAP concept.			

 GMP regulations were first introduced by FDA in 1963 and strengthened during the next decade with a major revision in 1978 that increased requirements for review and rationalization of processes and data and associated validation. GMPs were originally developed for commercial products but also were implicitly applicable to the production of clinical trial supplies for products that were not yet approved for the market place. The application of CGMPs to clinical trial materials was explicitly expressed by FDA in its guidance issued in 1991 (7). When FDA decided in the early 1990s that something such as validation was needed for product development, industry responded that this was not a realistic option and devised the concepts of qualification and verification (8). The concept of risk-based development and life-cycle management was added in 2002 in the FDA statement on CGMPs for the 21st century (1). 

The quantitative assessment of potential negative outcomes associated with a risk factor, after the risk factor has been identified, is the best approach. This method is not always possible or applicable, however. When quantification is possible, the appropriate statistic algorithm also must be used for analytical representation. Not all outcomes can be represented as a Gaussian distribution. Outcomes may be singular in nature or else the distribution of outcomes may be skewed. Some negative outcome distributions may not be describable in a simple mathematical form. A detailed understanding of the root causes of the negative outcomes associated with any particular risk factor must be achieved. Then, an accurate analytical representation of these negative outcomes and the potential for experiencing such negative outcomes for any product must be developed and used. 

The physical chemistry of each material used in the manufacture of a product must be understood and controlled. The effect of the various machine-running parameters on the manufacture of the product must be understood in relation to the intended product. And, the interrelation of the material characteristics and specifications with the operating conditions, which also includes the facility environment, operating conditions and controls, must be well understood and managed to manufacture a product to meet its quality expectations. The quality expectations are defined by a multidimensional response surface of quality-attribute specifications for each of the materials and equipment or facilities factors. The limits of the quality specifications define the desired response surface of all parameters that must be met for the quality of the product to "meet specifications." This response surface lies within a multidimensional space and must be established during product development. 

 is the current terminology for this response surface. 

 is the current terminology for achieving the repeatable manufacture of a product for which quality parameters will always be achieved and lie, in the aggregate, in that design space. The definition and control of design space is a large topic that will be addressed elsewhere by a consortium of universities and FDA (9). Nonetheless, the assessment and management of risk to minimize or eliminate the negative impacts will be an important aspect for the full definition of any design space. Therefore, a methodology and process for performing this assessment and describing the management plan must be developed in parallel.  

 Although there is every reason to believe that we can achieve an understanding of the risk factors and root causes for failures associated with physical, chemical, and apparatus or facilities parameters, there is not such confidence for understanding and describing failures associated with organizational interactions in a quantitative and predictable way. But, industry must make an attempt to categorize and identify organizational risks related to human interactions as well as the technological risks associated with material and machine interactions. For example, the development of a quality agreement as part of the contract development between a sponsor and CMO seems risk free until one begins to discuss the definitions and understanding of quality terms by the personnel in the two organizations.  

The risk of failure can be exacerbated by issues such as a lack of completeness in the determination of true root cause of an error or insufficient investigation suitable for resolution of out-of-specification results during stability evaluation. Failure to recognize any result as suspect may lead to product failures and recalls. Communication and training are very important in any organization and the concepts become critical for interactions between two organizations. Tracking of successes and failures in product performance also should include an evaluation of the successes and failures in information exchange and communication for understanding within and between the organizations. Just as there are specifications, acceptance criteria, and critical control criteria for the material parameters and machine or facility operating conditions, so should the specifications, acceptance criteria, and critical control criteria for organizational interactions be identified and managed.  

Quantitative estimates for negative outcomes can be developed from existing in-house data or from benchmark evaluations. For example, if moisture content for a granulation is a risk factor for a product, then the control of moisture content of all of the ingredients, including active pharmaceutical ingredients (APIs), excipients, and added granulation solvents, may contribute to successful product manufacture. If the moisture content of an excipient is a known contribution to failure, then finding a supplier that can deliver the material with the appropriate moisture content every time is crucial. The risk of failure by that supplier can be estimated from the number of times the sponsor (in-house data) or other sponsors (benchmark data) received material from the supplier that did not meet the moisture content criterion. A similar approach can be used for establishing a quantitative representation of risk for particle-size distribution, particle shape, related substances, and other parameters. Similar concepts can be defined for API and packaging components as well. 

Formulation scientists often evaluate machine-running parameters for their formulations but often do not put an assessment of risk into this evaluation. The critical control criteria for finished goods such as content uniformity or dissolution for solid oral products can be affected by machine-running parameters, as well as by material characteristics. Compression running speed, compression dwell time, and segregation patterns for the granulation resulting from bulk granulation feed reservoir configuration and machine vibration are well recognized as factors. Currently, industry does not have good methods for quantifying the potential for negative results for these known risk factors in any given formulation. Developing such methods should be the goal for future product development. 

Once identified, risks often must be ranked to know which are most critical and must be taken care of first. This ranking is often done in a 

-quantitative way with the establishment of the risk priority number for each risk: 

 is how serious is the potential failure mode for the customer (numeric value between 1 for least serious and 5 for very serious);

 is the likelihood that a cause will happen and will result in failure mode during the intended life and use of the product (numeric value between 1 for lowest likelihood and 5 for highest likelihood);

is the likelihood that current controls (design and process) will not detect the cause of the failure mode or of the failure mode itself (numeric value between 1 for good detection capability and 5 for poor detection capability).

This tool was originally defined in the failure mode and effects analysis (FMEA) program developed by the National Aeronautics and Space Administration for evaluating rocket and guidedmissile development programs and is well described in quality management literature and in Internet postings (10). Its use may be the most quantitative possible in some situations. Nonetheless, assumptions should be carefully and clearly challenged. For example, the potential failure of the O-ring on shuttle Columbia was recognized but considered a very low value risk. Similarly, the observation of the break-off of the approximate 1.3-lb piece of oxygen tank insulation foam on the shuttle Challenger also was discounted as a very low risk factor. It seems difficult to estimate a failure distribution curve when considering either of these two events, especially when knowing that a very serious outcome such as multiple deaths may happen but doesn't always occur. The original estimates of shuttle loss were in the range of 0.0001–1%, in the absence of data and based on the potential threat that something catastrophic would happen while in orbit. Nonetheless, the risk of loss can be estimated today at 1.8% based on those two lost flights out of the total 113 flights to that date of the Challenger flight. And, both losses occurred on launch or re-entry, not while in orbit.  

Block diagram process model for risk assessment for interactions with a CMO 

Process models come in many forms ranging from simple line drawings to more complex geometric displays. In all cases, the process model is intended to provide a pictorial representation of the interactions, requirements, inputs, and outputs for the tasks associated with the process. Such models are bounded by a beginning and an end, which means that the model allows the freedom to describe it in a very high-level or a very detailed view. It allows for breaking a large process into multiple parts to provide the required detail for each of these parts. Because all pharmaceutical firms require multiple processes to reach overall objectives for the organization, the process model allows for defining each subprocess as a contributing input to the "main" process, thus allowing the various functional groups to show all of the detail they need for any particular process to which they contribute.  

Risk assessment and management applies to technical matters as well as to interactions between organizations.

Defining and Managing Risk for Projects with a Contract Manufacturing Organization

At the end of June, in Shanghai, I had the chance to talk to teams from a dozen or so Chinese manufacturers of pharmaceuticals and bulk intermediates. Every one of them told me that their goal was to make their company, and their country, the world's preferred pharmaceutical manufacturing partner. Many of these executives are returnees, bringing back some Western know-how and a Western viewpoint.  

In an interview last month, I asked FDA Deputy Commissioner Janet Woodcock, MD, what one single message she'd like to send drug manufacturers in this country. "The pharmaceutical industry," she said, "is going to have to start looking at manufacturing as a competitive advantage, not as a way just to get product out the door." 

Woodcock was talking about the United States' long tradition of technical innovation, the need to return to that tradition in the manufacture of drugs, and FDA's progress towards new regulatory standards to make that possible. 

 In China, where even skilled workers may earn a tenth of what they'd make in the West, the Shanghai executives were talking about meeting established CGMP quality standards at half the price of other world producers.  

So, which competitive advantage do you pursue? The technological edge or the cost edge? Technology costs money and effort, and there's no guarantee that, at the end of the day, the new systems will produce the benefit in productivity or quality that you were counting on when you started. The cost edge can pay off immediately, but following that path leads to an immediate commoditization of the contract service, as competitors find their own low-cost vendors and the combined buyers relentlessly squeeze supplier profits. If it's possible to have irrational exuberance in capital markets, it's also possible to have irresponsible economy in supplier relationships. 

described how automotive parts suppliers are struggling in a deeply commoditized market (1). Eleven companies—including Delphi Corp., Dana Corp., Tower Automotive Inc. and Collins & Aikman Corp.—have defaulted since 2002, according to the Journal. 

Delphi, you may recall, was spun out of General Motors in 1999. In an evolution that might be familiar to the US pharmaceutical industry, the newly independent parts-maker was immediately subjected to relentless price pressure that threatened to crush it. And Delphi is not alone. 

"If you squeeze the suppliers a little bit more, there may be even fewer of them left," the Journal quotes an industry analyst. "The suppliers don't produce bad parts, but they are in trouble because they have been squeezed . . . it's a sheer formula for disaster." 

 Over the past year, I've made a point of visiting contract manufacturing facilities in the United States, too. They are very good at what they do. They are also preternaturally attuned to their clients' moods. And what their clients want, for the most part, is no-frills production at the lowest possible price . . . or, if possible, below the lowest possible price. The clients are not generally interested in financing the research and equipment that would be necessary to create the kind of flexible, innovative manufacturing systems emerging from the new US regulatory formulas.  

Yet, in theory, the contract manufacturers, with multiple campaigns on multiple products and their pressured profit margins, have much to gain from adopting new technology. The question is, will the market allow them the breathing space? 

 1. I. Mcdonald and S. Ng, "Heard on the Street: Auto-Parts Makers May Give Clearer Picture of Industry Health—Or an Underlying Lack Thereof," 

From The Editor

Industry must look at manufacturing as a competitive advantage, not just a way to get product out the door.

Pharmaceutical Technology-08-01-2006

Slowing Growth, Accelerating Changes Mark Contract Services Industry

Custom Synthesis and Custom Chemicals

Outsourcing Toolkits from the PharmTech Annual Event

Defining and Managing Risk for Projects with a Contract Manufacturing Organization

Outsourcing and Competitive Advantages: Lessons from the Road