May 1, 2008 (Vol. 28, No. 9)
Angelo DePalma Ph.D. Writer GEN
Current Trends Favor Smaller, Less Capital-Intensive Manufacturing Operations
Among the uncertainties in biotechnology, constructing a new facility is perhaps the riskiest venture of all. Plants must be constructed in anticipation of, rather than in response to, a product’s success. As a consequence, developers of biological products are hedging their bets by adopting strategies in which design, building, and qualification overlap. Because a plant’s ultimate purpose may not be known until after the ribbon-cutting, owners are also designing in more flexibility than they might have a decade ago.
These days, construction and renovation are as likely to be driven by market and technology forces as by general capacity needs. “With biotech, large capital projects are particularly risky as their planning must begin long before a product’s anticipated approval,” says Gary Nagamori, managing director at architecture and engineering firm CUH2A. “The probability that a company will use a product for its intended function is small, so big facilities represent high risk.”
Trends that are changing biotech facility design and construction include outsourcing, the growing use of disposable process equipment, improvements in protein titers and yield, and emerging patient-specific or personalized treatments.
Designing around disposable equipment confers a level of versatility that was unavailable for facilities designed 10 or 15 years ago. Personalized medicine and high titers, in particular, have rendered moot the capacity-crunch concerns of the early 2000s. Smaller batches, in turn, provide greater flexibility and responsiveness to other market forces.
“In this scenario architects and engineers are tasked with designing out risk while satisfying the demand for collaboration and interoperability,” Nagamori says. “Risk, therefore, becomes a design driver.”
One countertrend is the ongoing shortage of skilled labor, which means companies are always training new workers. So paradoxically, as processes become more efficient and adaptable, companies must factor in 30–40% more room for technicians and scientists in training.
More Complex than Commercial Space
While commercial/educational facilities devote approximately 20% of resources to mechanical, electrical, plumbing, air handling, and heating and cooling utilities, biotech facilities must factor these critical components at 50–60% of total construction costs. “Commercial systems may just have hot and cold water,” observes John Planz, project executive at Suffolk Construction. “A research lab or manufacturing facility has those, plus systems for purified water, distilled water, and perhaps even reverse osmosis.”
A diverse construction firm, about 20% of Suffolk’s work is in biotech. Its most recent project was building a research facility for the Torrey Pines Institute for Molecular Studies in Port St. Lucie, Florida. Suffolk also works for universities and municipal school systems.
Air handling is also an order of magnitude more complex in biological labs and production suites, particularly when containment becomes an issue for dangerous or sensitive biological materials. Office building space changes its air three times an hour, but at Torrey Pines the exchange rate is as high as six times per hour.
The facility includes a vivarium for test animals, with even higher air exchange requirements. At Torrey Pines the vivarium has the highest cost concentration, ranging up to $1,000 per sq.ft. compared with general lab space at close to $400 and a school for perhaps $200. “You can see why these facilities are so challenging,” says Planz.
Because of their complexity, large laboratory projects are generally undertaken in design-build mode, where the builder and design teams are either the same or managed under one umbrella. For the Torrey Pines project Suffolk hired the entire team, including the architect and subcontractors. “Coordination between the designer, contractor, and subcontractors must be seamless. Everyone has to know where their space is,” Planz says.
For areas with high biosafety requirements, the design-build process involves creating full-scale mockups of the laboratory space and a complete walk-through of every operation undertaken therein. This can add 25% to timelines, but Suffolk makes up for this in part by beginning construction as the documents for various areas of the facility become finalized. This is only possible, says Planz, through design-build.
Managing and Scheduling Big Projects
MannKind is completing a facility in Danbury, CT, where it hopes to manufacture its Phase III inhaled insulin product, Technosphere® Insulin.
The Danbury plant is a brownfield project consisting of 250,000 sq.ft. of renovated and newly constructed space. Construction has progressed rapidly, with time from groundbreaking to qualification projected at 18 months. The success has been due to good planning and project definition, which has limited the number of changes related to throughput, unit operations, or product flow paths, says Steve Hall, senior director of project management. “As a result we have been able to do our jobs without being second-guessed.”
For plant design, MannKind hired CRB Consulting Engineers, which interviewed scientists and engineers regarding workflows. From that survey, CRB generated the construction documents.
Commissioning and qualification normally occurs after construction is completed. MannKind compressed these activities by overlapping them somewhat with construction using Primavera 6.0 (P6™), a scheduling program from Primavera Systems. As the project moved forward, MannKind populated Primavera 6.0 with more than 7,000 “activities” coded to allow report generation specific to commissioning, qualification, or construction, and scheduling, with the goal of minimizing total elapsed time.
“Streamlining is important for meeting our timelines but complicated as well,” notes Hall. “So many details must be accounted for to ensure that a system is safe and ready to be turned over to the next group.”
The MannKind plant contains about 100 systems. Primavera allows project managers to pick any one of them to determine where it stands in the construction sequence, how it interrelates with other systems, and what activities need to be done to move it into the testing phase.
For example, the commissioning team can examine the schedule to see if the tank farm system will be ready for them on a specific date. The scheduling entry for that item also incorporates related activities that the commissioning team needs to do, such as writing protocols, walk-downs, and system testing. The Primavera schedule “knows” when the system will be ready for turnover and when the antecedent activities must be completed.
The Primavera package could have included design functions, but by the time the software was adopted the design team had already implemented its own schedule in a different software package. MannKind, however, convinced the construction team to combine its schedule of 1,500 activities into the Primavera master schedule. Once merged, any change affecting one activity is automatically reflected on other jobs, through what Hall describes as a ripple effect.
Primavera manages programs around costs, time, resources, contracts, documents, and risk. As MannKind demonstrated, the program works best in top-down mode, when implemented from an early stage and managed by the building owner. Contractors and subcontractors may, of course, employ Primavera (or similar products) to manage their activities, but the benefits are lessened compared to when the entire project is managed under Primavera.
“By the time a subcontractor comes on board the project will already be under way,” notes Richard Sappé, industry manager for Primavera. “That’s part of the legacy of our industry. We have owners who hold final control through contracts and checklists, and so many players held together by contractual obligations who need to work together and resolve issues that come up.”
Built on Flexibility
Bayer Healthcare began construction of a new clinical manufacturing facility at its Berkeley, California campus two years ago and received a Certificate of Occupancy in June 2006. Building 66, as the plant is known, consists of four environmentally controlled zones for inoculum preparation, fermentation/isolation, and two purification suites.
The facility will manufacture clinical-grade proteins, generate master working cell banks, and support process development, optimization, and validation.
Each zone has dedicated air handlers and waste drains as well as multiple systems to allow equipment to be cleaned in place. Segregation enhances flexibility by enabling one zone to produce GMP material while the other operates in development mode.
Bayer built flexibility into the 37,000 sq.ft. facility to accommodate either pilot-scale development or clinical manufacturing, through such essentials as process cooling, process gasses, and clean utilities, and by designing in easier removal and installation of equipment. Building 66 is designed to allow for a mirror-image expansion to the east, as needed.
Bayer Berkeley occupies 45 acres, two-thirds of which are in use. The remainder will serve for future expansion. The company’s history in the Bay Area extends back more than 100 years, beginning as Cutter Labs. In 1974, Bayer acquired Cutter and decided to focus the Berkeley operation on biotechnology. Bayer has invested heavily in the site, bringing in new equipment and capabilities while constantly upgrading. “GMP facilities have a limited lifecycle,” notes Klaus Weisenberger, Ph.D., vp of engineering for capital projects at Bayer.
In its Berkeley expansion, Bayer tried to create a campus atmosphere. The building, which cost $50 million and will take three years to build and qualify, creates an operational bridge between Bayer’s research and clinical initiatives by serving as a scale-up and clinical material manufacturing facility.
“We strongly desired to exploit organizational synergies between different groups,” says Dr. Weisenberger. Part of that goal is a seamless product lifecycle, from bench to pilot to manufacturing, with facile knowledge transfer between groups.
Also recently added to the Berkeley campus is a new, $100 million sterile fill and finish facility. The 45,000 sq.ft. plant houses the final stages of the manufacture of Kogenate®, Bayer’s recombinant Factor VIII hemophilia A product. This facility is in the early stages of licensure.
Bayer has developed a close relationship with the city of Berkeley over the 100 years it has operated there. Bayer is the second-largest employer in the city, after the University of California-Berkeley. Facility projects are planned after carefully considering community desires with respect to building characteristics and general layout.
Flexibility also plays into the design and construction of facilities with lower capacity needs. When contract manufacturer Eden Biodesign won a U.K. government contract to design and operate the U.K. National Biomanufacturing Center, it envisioned a facility that could handle 95% of processes and operations that Eden’s clients would likely request for the foreseeable future. Eden enjoyed significant latitude in design, but by virtue of operating under a government grant, not in cost.
Now completed, the 43,000 sq.ft. plant is small by cGMP standards but packed with capability. Its three manufacturing suites, three process development areas, and six analytical/QC laboratories handle workflows for cell culture, microbial fermentation, stem cells, and virus-based processes. The center’s largest process is a 200 L fermentation.
“We took ten candidate processes that we’d operated or worked on and used those to create a paper-based plan for the facility at each stage of the design,” says project manager Derek Ellison, Ph.D.
Disposable processing was essential for achieving these broad capabilities. “We could never have achieved the degree of flexibility and segregation required of this facility without single-use equipment,” Dr. Ellison reports. “The cost of segregation alone would have been immense.”
Although the center focuses on early clinical-phase material and process development, it was designed so its GMP suites could be upgraded for larger manufacturing batches. Dr. Ellison points out that with rising protein titers, the advent of highly potent biologics, and the emergence of personalized medicine, the line between development batches or pilot plants and full-scale manufacture is blurring, a fact that favors smaller, less capital-intensive facilities.