July 1, 2010

A Drug Is Not a Jet Plane, or Is It?

In a previous Pharma of the Future? blog entry, we described the Factory of the Future program created by the US Air Force in the 1970s. The success of the Factory of the Future program grew from these innovations:

  • A systematic approach to problem definition;
  • Development of an information technology (IT) infrastructure to make information accessible to all stakeholders;
  • Use of modeling and simulation to formalize the relationships between design parameters and performance; and
  • Development of efficient and sustainable solutions to manufacturing challenges.

Each change alone was not sufficient to overhaul the system.

Even though biological systems are more complex than aeronautical systems, similarities abound between the pharmaceutical industry of today and the aerospace industry of the 1970s: low productivity; late-stage failures; withdrawal of new products after commercialization; and high cost of implementing new processes.

Every sector of the drug industry has tried to increase productivity or reduce costs. Many of these initiatives have merit, ranging from new technologies and better IT to outsourcing and off-shoring, but they have created “islands of innovation” that are analogous to the “islands of automation” that were a problem in the aerospace industry. To solve the pharmaceutical crisis, innovations must be integrated into a radically new enterprise for drug discovery, development, and commercialization.

Science-based businesses, such as the pharmaceutical and biotechnology industries, have unique challenges in managing and mitigating risk. The heterogeneous nature of their knowledge base requires knowledge integration across wide areas of expertise. In addition, the rapid pace of scientific progress requires effective strategies for cumulative learning. Use of modeling and simulation was a key improvement in the aerospace industry, and model-based drug development can be an equally important innovation for the drug industry.

Pharmacometric modeling combines knowledge of disease state, biomarkers, and findings from preclinical and clinical studies with knowledge of placebo responses and drop out rates to gain insights into the determinants of efficacy and safety outcomes. These disease-drug models, particularly those that are detailed regarding disease mechanisms and drug effects, support knowledge integration, risk assessment, and informed decision-making. Continuous reassessment of models as new information becomes available can guide design of studies needed to close knowledge gaps.

To realize the benefits of model-based drug development, pharma companies must address current limitations. Timely results are delayed or derailed by a lack of infrastructure. Knowledge management systems and attendant data definitions are inadequate. Lack of sufficient computing capacity, particularly for complex models, adds greatly to the time needed to develop models. Scientists who have skills in both the technical aspects of modeling and the strategic implications of the results are difficult to find. Few university programs are available to train these scientists.

While time, money, and continued experience could address some of these limitations, changes in management approaches to program governance and decision making are needed. The innovations of model-based drug development must be implemented with full attention paid to its effects on the upstream and downstream components of the larger enterprise, including manufacturing, marketing, preclinical and clinical research, and regulatory affairs. This integration is critical to achieving a new model-based drug development paradigm, and this integration is one of our key motivations for the Pharma of the Future? program at Cognigen.

Be sure to come back for the next Pharma of the Future? blog entry: Not Too Complicated for Words. And don’t forget to peruse the previous post: Disambiguation.


You have chosen to email the article entitled:

A Drug Is Not a Jet Plane, or Is It?

Your Name:     * Your Email:    *

Friends Name: * Friends Email: *


Please enter the word you see in the image below:



* = Required Field

Commenting is not available for this article.