In-house vs. contract manufacturing
05/01/2007
Deciding whether to produce a product in-house or outsource it to a vendor is a complex decision that every small pharmaceutical manufacturer must face
By Sarah Fister Gale
As a pharmaceutical start-up matures, it must face an unavoidable decision: whether to develop, test, and produce its new product in-house or to outsource it. For many, this is a complex question with many-faceted answers. Some choose to outsource active ingredient manufacturing but produce finished product on-site. Some perform early-phase development in-house but partner with a larger vendor for clinical trials and commercial manufacturing. And still others outsource it all, keeping knowledge capital on staff and relying on vendors to do the rest.
This is a critical decision for any start-up that can and will shape the future of the company, and there are many factors to consider, says Bob Dream, the New York-based director of manufacturing and life sciences for CH2M Hill, a global leader in full-service engineering, construction, and operations. “The issue is bigger than simply ‘should I build a facility or outsource to a vendor,’” he says. “For some projects, you can really wrestle with the decision.”
Before any company can make this decision, there are many factors to consider. The leadership team should begin by defining short- and long-term goals for the company and the product, including timelines and deadlines for product completion and forecasted growth expectations. They should also evaluate just how difficult or proprietary the product formula is to develop or reproduce.
“Once you know your objectives and goals, you can figure out the most cost-effective decision,” Dream says.
Having these criteria defined will lead most companies in the most logical direction; however, it’s not the end of the process. For example, the cost to build vs. the cost to outsource may look good early on, but the long-term costs of paying vendors could tip the financial balance in the other direction. There are also issues around flexibility, control of the process, and trustworthiness of partners, all of which can impact the ROI of the in-house or outsource decision.
Regardless of the direction chosen, either path has its share of issues and concerns to consider.
Keeping it in the family
Building a facility in-house for development and/or commercial manufacturing has definite benefit, says Dream. It gives the company total control over quality and quantity of the product, enabling it to increase or decrease volume, change scope, or adjust timelines in response to market demands; and it ensures the product is the top priority for the manufacturing facility, not just a small percentage of a much larger client roster. Building and running a production facility, however, is not an easy endeavor. There are major up-front costs, audits, regulatory issues, and a significant learning curve to overcome in order to get the operation up and running-all while remaining focused on developing a critical first product.
Fortunately, says Robert Ogara, director of architecture for manufacturing and life sciences at CH2M Hill, there are many levels of facilities to meet a variety of size and regulation requirements.
For example, if the company plans only to do development in-house but outsource manufacturing, it may have more flexibility in terms of size and requirements. “In early-phase development, regulations are a little softer,” he says.
Incubator systems for small-batch products or minor ingredient development may involve primarily disposable sterile products and can utilize portable softwall cleanrooms, which are less expensive to own and operate than a fixed facility and can be up and running much more quickly. However, they will not offer a rigorous enough environment when production ramps up to the next phase.
“Large-scale production areas need a more complex space, which may include glove boxes or a segregated cleanroom,” Ogara says. They must also meet stricter requirements from Food and Drug Administration (FDA), OSHA, and Environmental Protection Agency (EPA) regulations. “Regulatory compliance issues drive design decisions.”
If a company expects to do more than a few small batches of early phase product per year, it needs to look at what’s required for GMP compliance and factor in foreign regulations if the product will be sold globally. If potent compounds are used, OSHA guidelines are critical and will require stringent designs for waste-stream handling, surface-contaminant monitoring, and airflow and air quality designs, which is where much of the up-front investment is spent.
“The HVAC system is one of the most significant costs of a cleanroom space,” Dream points out. “A lot of times the first estimate we do is based on the air classification requirements. It’s a significant factor in building a cleanroom.”
Cross-contamination is another factor to consider, as the production space will require segregation for ingredients, equipment, and people, particularly if it is a multiproduct production space.
But these are all just broad topics to address with a developer who can help determine what a company’s production needs are based on the type of drug it is producing and in what quantities, Ogara says. Different drugs and ingredients require different cleanroom classifications, which dictate the size and number of allowable airborne particles, circulation or airflow requirements, ingredient handling and storage, surface materials and coatings, sanitation options, and gowning areas.
A company will also want to configure the space for optimal flow by, for example, placing media buffer suites above purification areas and creating production lines to minimize handling. “However, the most elegant solution isn’t always the most cost-effective,” Dream says. “With a start-up, you want to do the minimum in order to meet requirements.” That may mean creating a single-level suite with some double handling or back and forth movement, which doesn’t cause segregation issues but is more awkward. “It may not be the ideal process environment, but for small operations on a tight budget it can be more cost-effective.”
Future upgrades may also be factored into the design plan, although Dream says 90 percent of the time companies are wrong about their future expansion plans. “It’s better to build to accommodate future growth rather than build in plans for future growth,” he says. He encourages clients to create skeletal spaces in open areas centered around lockers, gowning areas, ingredient storage, and main circulation and electrical conduits to create a logical path for future growth.
 A scientist conducts research at a pharmaceuticals facility. Photo courtesy of CH2M Hill. |
Dream and Ogara help their clients through all of these evaluations, often beginning with the ideal state-of-the-art space and working backward to find the compromise between need and budget.
Dream suggests the best way to achieve your goals without spending a lot of extra money and time is to work with a designer at the feasibility stage to generate an analysis of your needs-as well as what would be nice to have-in order to see the implications of scope and schedule variances. “If you spend two months up front figuring out exactly what you need, you can make that time back-and then some-by controlling scope creep during the execution phase.”
How to choose a vendor
Despite the benefits of owning the production process, for many small companies an in-house facility is unrealistic or out of the realm of their business strategy. For those companies, giving up production control and committing to annual contracts for defined quantities are small sacrifices for the comfort of having a third-party research or manufacturing expert take over this part of the process.
Choosing to use a contract manufacturing organization (CMO) or a contract research organization (CRO) for any phase of a drug’s development, however, should involve just as much due diligence and investigation as building an in-house facility, advises Sam Tetlow, managing partner of Clearview Ltd., an investment firm for life sciences in Research Triangle Park, NC.
He boils the decision down to two key factors: the level of difficulty of the manufacturing process and the impact on the new chemical entity revenue getting scaled up and to market. “In particular, there are a lot of activities around the difficulty level to consider,” he says.
For example, if a product requires a sophisticated production method, you may have fewer vendors to choose from and the vendor will likely require more oversight. Also, you will pay a premium for technical expertise.
 Research is conducted in this lab in the mid-central U.S. Photo courtesy of CH2M Hill. |
Tetlow has defined a detailed approach to choosing an outsource partner for any phase in a pharmaceutical production process, including early-phase development, clinical trials, active ingredient manufacturing, or commercial manufacturing. The process begins with defining performance criteria, making sure they are quantifiable and measurable.
Some issues to consider during this step include whether the protocol is defined and validated; whether validation triggers the next phase of the project; and whether there is a periodic review of the plan as it relates to established milestones.
Tetlow also suggests defining a communication plan up front, with expectations for frequency and format, that states how often the vendor team will communicate with key members of the client team; who will be in charge of communication; whether these meetings will be face-to-face, over the phone, or via e-mail; and what the vendor’s process will be if its facility falls out of compliance or if an adverse event occurs-especially in Phase III trials. “There has to be regular interaction at the project manager level on both sides for the relationship to work,” Tetlow says.
The specification report defining a company’s expectations for the vendor should include sufficient levels of detail to describe its goals and the steps to achieve them, as well as provisions for one-to-one comparisons from multiple respondents, advises Tetlow. The spec should further identify any non-standard evaluation parameters and internal evaluation criteria, with values assigned to each criterion for later calculation.
“Using a matrix, you can then quantitatively evaluate vendors’ responses, placing criteria for success on the X axis, and respondents on the Y axis,” says Tetlow. “This will result in a short list of vendors.”
Using a structured quantitative process for evaluating vendors gives small companies a much better chance at success with their third-party partnerships. “You have to think like an enterprise that needs operational excellence,” Tetlow says. “If you don’t know what’s out there, you won’t get the best deal.”
That can translate into time and money lost when dealing with vendors that can’t handle production requirements, that need hand-holding through the development phase, or worse, get shut down or bought out in the middle of your product’s production cycle.
“By going through a rigorous up-front evaluation of partner options and giving them well defined criteria, you create a structure for the relationship that they have to use to get the work done,” he says. “If you don’t go through this process, you won’t be sure what you’re getting, the criteria won’t be defined, and the inclination will be to over-manage the process.”
Making the most of gut instinct
Although most small companies know that a formal review process, with requests for proposals and facility audits, is the best way to choose a vendor, many will admit they don’t have the time, manpower, or resources to go through that process. For those companies, a few good vendor recommendations and gut instinct are the tools they rely on to choose vendors, says Steve Pondell, director of manufacturing for Encysive Pharmaceuticals, a Houston, TX-based pharmaceutical company that discovers, develops, and commercializes small-molecule drugs for cardiovascular and inflammatory diseases.
“Working with a CMO when you’re a small organization is very different than when you’re one of the pharmaceutical giants,” he says. “You may only have one person who’s your expert, who owns the whole process.”
He should know. After spending 22 years at Abbot Laboratories, a well-known drug discovery and development company, he joined Encysive, where every element of the process is outsourced and often there are only one or two people managing whole departments.
Not only does that lay a heavy burden on the scientists, who have little access to peer review, but it also means there are fewer people on the team to help select and manage the vendor relationships. Those constraints play a big part in how Encysive, and many other small companies, choose their partners.
Pondell admits that because Encysive is a smaller operation, the team didn’t conduct a formal review and bidding process when it chose a CMO for its new product development phase. Instead, the team relied on word-of-mouth recommendations from industry peers, reviewing a few CMOs that had either been recommended by colleagues or had been used by team members at previous jobs. “At best, you could say it was an informal process,” he says.
Despite the lack of formality, Pondell says the relationship has been successful, in part because he had several pre-existing criteria for CMO selection that he naturally adhered to in his decision-making process.
“There are several key issues that small companies need to consider when choosing a vendor,” he says. “The first is geographical.”
While the lure of cheap labor in foreign markets may seem attractive, it can be a counterproductive choice for small companies with limited human resources, Pondell points out. “The time difference alone can cut a productive work day in half if the vendor is in Europe; while the short time overlap in China, Singapore, or India can make daily communication very difficult.”
The time it takes to physically get to these countries can also add days to the management process and put key team members out of touch during critical times. “You can’t just pop into a facility in Asia,” he says. “While larger companies may have a dedicated person to manage an offshore vendor relationship, small companies can’t afford to send key team members out of the country for days at a time. It’s just not worth it.”
Instead, Pondell always tries to stay local for his CMOs. His most recent vendor relationship is with a facility located an hour away from Encysive’s offices. “I can get there any time I need to; we work in the same time zone; we speak the same language; and we work under the same business culture.”
Pondell also looks for development groups that have excellent customer service skills and will pay attention to his needs, however small they are. “For product development, customer service is critical,” he says. “CMOs are never overstaffed and you don’t want your needs to get lost in the shuffle.”
Small fish, small pond
Pondell also prefers to work with smaller vendors with limited client rosters, assuming they have the expertise he needs. He recognizes the allure of larger CMOs, but says a small company needs to balance its desire to work with the most experienced or well known vendor with that vendor’s size and schedule to be sure its products are going to be a priority to the vendor.
Whether a company chooses a larger vendor with dozens of clients, or a smaller firm with one or two clients, Pondell believes it should take a hands-on management approach to the relationship to ensure its priorities are met. “If you take a hands-off approach to having your process developed, it could come back to you three months later and be a disaster,” he says.
At Encysive, that means an active pharmaceutical ingredients (API) scientist works directly with the production team at the development house while it develops Encysive’s product methods and validation processes to ensure the plant staff stays on task and focused on Encysive’s needs.
Pondell and others on the executive team also communicate frequently with the vendor team. They meet monthly to review status updates, evaluate milestones, and address problems that have arisen. During active times, weekly meetings take place to keep tabs on progress and issues as they arise. Pondell also knows he can call the vendor and get answers any time he needs them, which he feels contributes largely to his satisfaction with the relationship.
“A successful partnership with a CMO is largely about how they approach you,” he says.
“A lot of companies are internally focused and if you don’t go through their system, they won’t be responsive. Our CMO, on the other hand, is very customer-focused. I know I can call my project manager at any time and he will be my advocate in the plant. That’s what keeps you going during rocky times.”
Customer service isn’t the only factor to consider, however. Depending on the company’s current goals and the phase of the project, other issues such as technical expertise, experience, price, and quantity all play major roles in the decision-making process. For example, Encysive works with a particular supplier known for its technical expertise. “We’re blind to their prices because they’re so good,” Pondell says. “If we have a problem, we know they can fix it.”
That’s especially important for a small company developing its premier product, but only during early phases, Pondell says. When it comes time to transition to commercial manufacturing, his criteria change dramatically. “Once you are into manufacturing, and you have your process defined, you need someone who can deal with the day-to-day issues of manufacturing at a cost-effective price.”
At that point, a company should consider volume price, ability to deliver, quality processes, and reliability. He says it can be hard to convince scientists to give up technical expertise in exchange for manufacturing professionals, but, as the “commercialization person” on the team, he sees it as his responsibility to show them the value of making the switch.
“It’s a mistake to use development people to commercialize because it’s a totally different business,” he says. And even if a development firm is moving into commercialization, he prefers to use a firm with more experience. “I’m not going to trust my ‘baby’ to a company just learning how to do commercial manufacturing.”
Whether you produce a product in-house or outsource it, experts agree that the keys to success are making informed decisions about partnerships, choosing vendors who will support your needs, and staying in constant communication about progress or break-downs in the system.
“Small companies have a lot more at stake,” Pondell says, “and a lot more to lose if something goes wrong,”
Mistakes to avoid with a CMO
When choosing a vendor for any part of a new drug production process, it's easy to make mistakes. Experts offer this advice on avoiding some of the most common pitfalls.
- Choosing a CMO that is too busy. Small companies with small projects are a low priority for overly busy CMOs. “The biggest and best CMO may not be the best choice for you if it means getting lost in the shuffle,” says Sam Tetlow, managing partner of Clearview Ltd. “That’s the biggest risk for a small start-up.”
- Putting intellectual property at risk. There is always a danger in giving your intellectual property to a third party. To avoid leaks, define control protocols for the formula, including requirements for a unique project manager, erasing all white boards after use, and creating a circle around the people who have access to the information. Also think carefully about using a CMO that has clients making competing or similar products.
- Choosing a vendor that is too small or inexperienced. If the product is complex or difficult to make, you need to balance finding a small CMO that can pay attention to your needs with getting the best technical expertise from your vendor. Small CMOs also have a higher risk of getting shut down due to compliance issues, or getting bought by larger firms, which can throw off your production cycle. “This is a low-risk-by-high-severity issue,” Tetlow says, adding that companies can mitigate these risks by demanding proof of GMP compliance in the product specs and auditing standard operating procedures prior to agreeing to the contract.
- Letting price cloud your judgment. In the development phase, when a product and method are being defined, talent and experience can be more important than price, says Steve Pondell, director of manufacturing for Encysive Pharmaceuticals. “Small companies need access to technical expertise and premium service during development and that means premium prices,” he says. Having the equipment, talent, and experience to deal with problems during development can determine whether you get a product to market on time-or at all. However, that changes during the commercial manufacturing phase, when a good price is one of the most important factors. “It’s okay to leave a great development house for a good manufacturer,” he says. “It’s a totally different process.”