In 2007, the biopharmaceutical market represented ~$71 billion: 10% of the entire pharmaceutical market. Therapeutic proteins and monoclonal antibodies (MAbs) account for 98% of all biotherapeutics in development, the rest being blood proteins and enzymes — all the products of recombinant DNA technology. Before the recession hit full on, growth of this market was estimated by some at ~15%. (Now it's hard to predict at all.) Making biotech drugs consumes huge amounts of time and money, but they satisfy unmet medical needs, many that cannot be addressed any other way. Now the biotherapeutics industry is being transformed by lean manufacturing, operational excellence, and the FDA's quality-by-design (QbD) mandate. The emphasis is on cross-functional design and development teams, new approaches to automated processes and technologies, and even ways to integrate upstream and downstream processes. It's all indicative of a truly maturing industry.

Drug Discovery, Innovations, and Product Development

Small/virtual biotechnology companies need to design drug/biologic programs that can not only minimize development time, but also convince investors of their value. Achieving regulatory milestones is critical to a product's success. James Weston (senior vice president of Talaris Advisors LLC) organized a session for the 2010 BIO International Convention's achieving regulatory approval track called “Development Strategies for Novel Drugs and Biologics: Doing It Right the First Time,” based on the work his consultancy does.

“Talaris Advisors helps small/virtual companies get results earlier and faster. We cover the entire development process — manufacturing, preclinical or nonclinical, clinical, and regulatory — but focus on the front end, typically the preclinical early development stage, and designing processes to take clients through major milestones. We look at ways to accelerate getting there and help them make efficient use of their capital, and we look at regulatory strategies and all agencies worldwide. We look at preclinical strategies: what's needed, when, and why. And we also take a look at manufacturing strategies, at the most efficient process designs. For small companies, it's key to know what type of process is needed and when. We also incorporate the QbD paradigm into the overall process. Investors want to see the product get developed, not necessarily the company. And we provide the expertise to get that product where it needs to go in a very short time, so an investor doesn't have to invest in building an infrastructure.”

Weston will be the session chair, and his panel tentatively includes Jean-Yves LeCotonnec (CEO of Trisco Integrated Services in Switzerland); Michael Webb (managing director of Exponential Pharma Ventures); Susan Flint (senior vice president of drug development for Talaris); and Mark Hurt (Talaris' chief medical officer). “This will be a working interactive session,” he says. “We'll talk about practical examples in obtaining worldwide regulatory approval of a biologic development program, then how we can implement those programs cost-effectively using a company's executive team.”

Bioinformatics: What is the current status of computing for biotechnology? What lessons have been learned in applying information technologies to bioindustry? What will be the impact of petascale computing? What does the future hold now that computing power will not be a limiting factor, and how can it transform the nature of industry–academic collaborations? Diana Dummitt (associate director of advancement at the University of Illinois College of Medicine at Urbana–Champagne) organized a session for the convention's exciting science track called “Petascale Computing, Computational Science, and the Biotechnology Industry,” which aims to answer many of those questions. Her panel of experts will briefly review the status and potential of computational science; describe the capabilities of Blue Waters (the world's most powerful supercomputer for open scientific research — capable of 1,000-trillion operations/second — 500× faster than today's most common supercomputers, and expected to go online in 2011); and discuss the potential for supercomputing to address challenges the industry is facing. Among those experts will be Thom Dunning, director of the Institute for Advanced Computing Applications and Technologies and the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign.

Dunning told us, “Computing power will continue to increase over the next decade in accordance with Moore's Law. But there's been a major change in how this increased computing power is obtained. From the 1970s to 2004, increasing numbers of transistors on each chip was accompanied by frequency increases. It is estimated that 80% of increased computing power came from that increase in frequency, with the number of transistors providing the remaining 20%. From 1993 to 2004, chip frequencies rose from 66 MHz to nearly 4 GHz. In the early 2000s, leakage current raised its ugly head, and it's preventing further increases in frequency (because of overheating). So chip vendors began to place multiple computer cores on each chip. Quad-core chips are now available, and eight-core chips are being introduced. Increased computing power is now tied to increasing parallelism.”

Computers play three fundamental roles in biology, Dunning said, “First, they enable the modeling of biological systems, from biomolecules to organs to populations. Increased computing power will lead to more accurate computational models of these systems and processes, allowing computational biologists to provide insights into their behavior that would be difficult to obtain otherwise.”

Second, Dunning said, computers enable biologists to manage, analyze, and correlate increasing quantities of data. “This started with the Human Genome Project, which would not have realized its goals without the computing power to assemble and analyze gene sequences.” Thanks to investments in that project, sequencing technology rapidly advanced. “Data from those new technologies offered tremendous opportunities for understanding the relationships between species, the origin of human diseases, and so on.” But all those data and their associated analysis require a lot of computing power. “Some bioinformaticians estimate that petaflops of computing combined with petabytes of data will be required.”

Another application of high-performance computing is in personalized medicine. Its foundation is the correlation of genomic data with the effectiveness of therapeutics and other treatments. “This will require the analysis of massive quantities of data,” said Dunning. “Again, petabytes of data will be needed and teraflops to petaflops of computing power required for the analysis.” Increasing computing power will also continue to affect drug and process development laboratories, Dunning said, providing more functionality and greater automation than in the instruments of today.

Dummitt told BPI contributing editor Lorna McLeod, “There's often a disconnect between the type of research traditionally conducted in academic laboratories and the biotech industry's interest in getting things to market quickly. In our conversations with panelist Andrea Hunt (vice president of regenerative medicine at Baxter Healthcare), it became apparent that there is significant controversy about the role computational science might play in the biotech industry. It also seems to me that some attempts might not have been well grounded in research to engage pharmaceutical companies in this science. They didn't quite fit the problems pharma had.”

According to Hunt, “The power of computational science in biotech and pharma development sounds like a winner. How can you go wrong with so much more data being processed at faster and faster rates? Yet while we continue to have faster processing capabilities, the number of drugs approved per year in the United States is declining, and the time and costs to get them approved has continued to increase. The challenge is in targeting our data analysis to be more useful in screening target compounds or ultimately improving time to development. Perhaps we need to engage in conversations that include earlier and more diverse discussions between pharma/biotech professionals from multiple disciplines and computational scientists to take advantage of these new processing capabilities.”

EX#CITING SCIENCE SESSIONS