MIT'S PROGRAM ON THE PHARMACEUTICAL INDUSTRY


Annual Report of Research and Educational Programs
1996

2. New and Ongoing Research

The pharmaceutical industry speaks of drugs as being "in the pipeline" and-since our founding- POPI has employed the analogy of the pharmaceutical pipeline as a means of understanding the activities of firms in the industry. This pipeline model has evolved into a useful tool for organizing our multidisciplinary research efforts.

The activities of pharmaceutical firms "in the pipeline" comprise three primary steps, or stages. The first stage include drugs discovery and development. This is followed by the manufacturing stage. Finally, in its marketplace stage, the pipeline includes promotion, pricing, market strategy, outcomes research, and pharmacoeconomics.

Fig. 1 pipeline

It would be an oversimplification to consider these steps in the pipeline to occur strictly in sequence. There is feedback that takes place across the indicated steps and interactions that occur among them. Also, there are important issues and themes addressed in POPI research that transcend the specific steps shown here, but are nonetheless important throughout the pipeline. One example is regulation, which continues to be important in discovery, development, manufacturing, and marketing.

The projects described in this annual report illustrate the broad scope of our examinations. Each has an advisory committee-with individuals from industry, government, and academia-to provide guidance and feedback on the research.

NEW RESEARCH

This report presents four new POPI research projects. Three address issues related to accelerating drug discovery and development. One also incorporates concerns central to manufacturing productivity. Another project explores a new way of approaching the process innovation component of manufacturing.

"Technology Transfer" and the Integration of Technological Innovation in the Pharmaceutical Industry

Faculty
Thomas Allen (management)
Arnold Barnett (statistics)
Iain Cockburn (economics)
Charles Cooney (chemical engineering)
Stan Finkelstein (medicine)
Rebecca Henderson (management)
Ralph Katz (management)
Robert Rubin (medicine)
Anthony Sinskey (biology)
Scott Stern (management)

This new project seeks a better understanding of innovation in pharmaceuticals as it occurs in academic, industrial, and government research and development settings. Several main questions are explored: How are ideas brought to fruition? How do ideas become products? How does technological innovation developed in one place become a part of the work in another? How might the efficiency of this process be improved?

Innovation may be considered to flow through four distinct stages. The first is the generation of ideas in basic, clinical, and applied research. These ideas may be prompted, for example, by the more general goal of understanding life processes or the more specific goal of understanding the causes and cures of a given disease more fully. The second stage is communication of these ideas. Researchers, wherever they work, must communicate ideas among themselves to integrate the essential pieces of knowledge which eventually will be developed into a specific technology. At the third stage, development of clinical applications, researchers and technologists respond to societal needs with practical applications of ideas for the diagnosis and treatment of disease. In private industry, the economics of the marketplace have a significant influence on the directions of these research applications. Fourth, the packaged technology moves out of the research community to become a product-a market good that is the object of production on a large scale.

In the field of medical technology, these stages of innovation often occur within a unique environment that raises many complex issues regarding the need for collaboration across organizational and disciplinary lines and the ethical implications of working with human subjects.

Objectives. This work builds on a foundation set in prior POPI research projects. The starting point is that products with proven success in the marketplace must have depended to a greater or lesser degree on contributions made in various industrial, academic, and governmental settings. To gain a better understanding of the nature of drug development and opportunities for improvement, one must determine how the contributions of different organizations and institutions are integrated and identify the critical success factors. In this regard, it is particularly important to learn how successful relations were developed among the participating organizations-and to understand what may have caused failures.

The principal objective of the research is to determine the relationship between today's highly "successful" pharmaceuticals and the three broad contexts in which their component innovations arose: innovations by researchers working exclusively in private, government, or university settings; innovations resulting from informal interactions across these different constituencies; and innovations resulting from formal arrangements among these constituencies.

Working through a number of specific examples, the team is building a model of the innovation process. From this, points of leverage will be identified that could lead to improved performance of the contributing R&D organizations, and thus enhance the entire process.

Approach. The research examines the 20 or so drugs that account for some 80 to 90 percent of sales in the U.S. pharmaceutical market. The team's first step is to identify the advantages of these drugs over their predecessors, if any. Next, the researchers trace back through the history of these drugs to identify their enabling events and where the important innovations were developed. Funding sources-public, private, or a combination of both-are identified. The team explores how information about the innovation was communicated, and records the chronology of that communication. From this information, a model of the innovation process begins to emerge.

The next step is to look at products and determine in what organization products using the innovation were first made, the extent to which innovation in industry depended on earlier innovations in government or university laboratories, and the lag time between innovations. The team explores the elements of public-private interaction that make a difference in the development of a product, as well as the factors that make some firms to be better able to take advantage of public-sector research than others. Issues related to the management of intellectual property rights are also addressed.

In addition to these issues, the team will explore ways to improve technology transfer between the public and private sectors, including specific improvements that might be made in the licensing process.

For comparison, the team is also choosing a sample of failures, that is, cases where firms thought a project would lead to success but ultimately canceled the project. Several questions are then posed: Why was the project canceled? What are the elements of failure? Was the project missing some key element of understanding?

Through in-depth interviews with scientists and others involved in projects at present, the team is identifying the paths through which to trace the innovation process back, as well as the component parts of that process-documenting the process through further interviews and additional literature searches where necessary.

The research comprises two phases. The initial pilot phase, already underway, guides the design of a more extensive research program. The second phase, which is scheduled to begin in the summer of 1997, will encompass more extensive research that addresses systematically the technological innovation process in pharmaceuticals and biotechnology and seeks to uncover objective causes of failure.

Drug Development in a New Environment

Faculty
Rebecca Henderson (management)
Scott Stern (management)
Iain Cockburn (economics)
Ernst Berndt (economics)
Stan Finkelstein (medicine)

The pharmaceutical industry finds itself in a period of heightened uncertainty. Recent and prospective changes in the method of financing healthcare, rapid industry consolidation, increasing requirements for clinical trials and the testing of new drugs, and other regulatory measures by both U.S. and foreign governments present the industry with several major challenges. For example, the expansion of health maintenance organizations (HMOs) and other managed care arrangements, and the increasing importance of pharmacy benefit managers (PBMs), are changing the way healthcare is financed and delivered.

There have also been significant advances in medical science and technology, both in understanding biological mechanisms and in the more traditional approaches to the screening of drugs as possible agents for activity against biochemical targets. One result is that there are more biologic pharmaceutical products-as opposed to synthetic organic products-being made than ever before, and major breakthroughs in healthcare through gene therapy, cellular transplants, and other advanced technologies are on the horizon.

These broad changes have had a substantial impact on the way research and development is organized within and among firms in the pharmaceutical industry. Anecdotal evidence suggests that the combined effect of these changes may increase competition and reduce prices in the industry, and this may decrease both the incentives for pharmaceutical research and the number of drugs that ultimately come to market.

Objectives. Several specific questions are posed regarding the behavior of firms in the new environment, and how drug development is changing.

Approach. The team is conducting in-depth, focused interviews with a wide variety of industry participants. This is the primary source of data. These data will be supplemented with a wide variety of secondary data sources, including IMS sales figures and data on patents, scientific papers, and regulatory approvals, and with a "snapshot" of internal firm data describing the research portfolio for a single year (probably 1996). The team will also build on earlier work, which focused on a sample of 10 well-established U.S. and European pharmaceutical firms, by gathering data from a broader set of firms and seeking to understand how smaller, more "biotechnology"-oriented firms are responding to the new environment.

Anticipated outputs. Several different kinds of outputs are anticipated as the research proceeds. First, the team will summarize results in brief reports and presentations to firms that are participating in the study. Second, the researchers will prepare refereed academic papers that explore the organizational and competitive implications of the data. Third, the team will draft papers summarizing the implications of the results for a more public policy-oriented audience.

Effectiveness and Efficiency of the Drug Development Process

Faculty
Thomas Allen (management)
Arnold Barnett (statistics)
Charles Cooney (chemical engineering)
Stan Finkelstein (medicine)
Robert Rubin (medicine)
Scott Stern (management)

The costs associated with the drug development process have skyrocketed. Current research by Stewart Myers of MIT (under the auspices of POPI) finds the cost of developing a successful new drug to be in the vicinity of $450 million, considering actual expenditures on discovery, preclinical and clinical testing, and the cost of capital. Only five years ago, a major study placed the price at a much lower $231 million.

With pressures in many markets to keep prices low, drug firms need to be more efficient in their spending on drug development. Similarly, firms need to be more efficient with respect to the time to market. The size of the target market for virtually any drug in development is at least in the range of hundreds of millions of dollars per year. Consequently, delays in bringing a drug to market can easily cost a million dollars a day.

Objectives. This research aims to examine whether information can be obtained earlier in the drug development process that would have a substantially favorable effect on efficiency (i.e., time and cost). For example, drug firms specify their clinical testing strategy early in the process, based largely on what they know or expect will be demanded by regulators and the marketplace. It may be possible that new technologies-including, but not limited to, medical imaging used in concert with Phases I and II testing-could reduce the requirement for more extensive Phase III trials. Further, firms must plan early on to manufacture enough drugs for testing, and "production" decisions made at that time can become "frozen" into specifications for later full-scale manufacture (particularly its timeline and cost). For this reason, these decisions should be based on as complete information as possible. Finally, there is the increasing use of contract research organizations (CROs), to which drug firms outsource parts or even all of their clinical development and regulatory management functions. This decision has major cost and scheduling implications, and should be examined very carefully.

Approach. Working closely with selected pharmaceutical firms with both U.S. and foreign drug development operations, the research team will explore these issues in great depth, collaborating with individuals who play major roles in drug development in these firms. The inquiry includes a selection of drugs from different therapeutic areas, small molecules, and large molecules. Through structured interviews, the team is gathering historical information on drug development projects that have been successful (i.e., that have resulted in a drug entering the market) or have failed.

Within chosen therapeutic areas, the team will develop an instrument to compare similar drug developments first in terms of their time and cost to develop. The next step will be to explore back through the process to determine the points at which key pharmacokinetic and pharmacodynamic information was available in each development. The timing of this information will be related to the overall schedule and cost of the development. The researchers will also examine the crucial "make or buy" decisions with respect to contract research. Within firms and within therapeutic areas, comparisons will be made to determine the degree to which outsourcing certain elements of the development can reduce development time and cost.

Capturing Competitive Advantage through Pharmaceutical Manufacturing

Faculty
Stephen Byrn (pharmaceutical engineering)
Charles Cooney (chemical engineering)
G.K. Raju (chem. eng., management)

If looked at as process innovation, manufacturing becomes important over a much broader period of the drug development process than that related only to production. Early on, process innovation affects the production of the relatively small quantity of the product used for clinical testing. When a pharmaceutical firm submits an Investigational New Drug (IND) Application to the Food and Drug Administration, it must specify how it will manufacture the drug to obtain the samples to be used in the investigation, including ranges of dissolved oxygen levels, temperature levels, and pH levels. This technical information submitted early in the development process can have a significant impact on cost, time, and quality-and is difficult to change later.

Firms that are locked in to manufacturing a certain way find that they are unable to compete when process innovation begins again, that is, at the end of a product's patent, when a generic manufacturer demonstrates the ability to produce the product more efficiently.

Process innovation throughout the lifecycle of the product, though, could have positive cost-saving consequences for pharmaceutical firms.

Objectives. The research team is exploring ways, through process innovation, to move drugs more quickly through clinical trials, capture what is learned more effectively, and continue to learn and improve on manufacturing during the period of patent protection. The ultimate objective is to reduce the number of experiments required to determine the optimal manufacturing process.

Approach. The team looks at a set of fermentation experiments conducted to move a product to market, for example, measuring dissolved oxygen, oxygen uptake rate, and respiration quotient (carbon dioxide versus oxygen). These three indicators tie into physiological states. Using an automated system to collect data, and designing subsequent experiments automatically through machine learning, the team removes the human biases from choosing which experiments and to conduct how much change theree should be from one experimental run to the next.

For example, say the experiments measure temperature and pressure. Scientists in the laboratory today are likely to start at certain levels based on past experience. A different approach might be to conduct fewer experiments-all over the spectrum, without this human bias-and then narrow down the region for further experimentation. A measure of performance- which could be some surrogate for cost, quality, or time-is taken from one experiment to the next.

This approach aims at realizing a significant decrease in the overall number of experiments. If the same information can be obtained much faster, the time to market can be reduced substantially.

Using evidence gathered at 15 companies through POPI's project to benchmark best practices in pharmaceutical manufacturing, the team looks at existing and potential opportunities for cost savings.

Preliminary findings. Company data and the simulated university laboratory environment suggest significant reductions in the number of experiments-up to 50 percent-are possible. The implications for the cost of goods in the manufacturing process, and for time to market, are considerable. Overall cost savings, in turn, are potentially quite large.

One company reduced its cost of goods by about 20 percent. Assuming that number over the lifecycle of the product, the cost savings would total as much $300. If three to six months can be saved in time to market, and assuming an industry-standard $1 million per day in sales for a major drug-it could translate into as much as $200 million in additional savings.

Future work. The team has arranged that all projects initiated under the new Consortium for the Advancement of Manufacturing in Pharmaceuticals (CAMP) will incorporate these concepts. Further, CAMP projects will include benchmarking against the industry's best practices and an analysis of the cost implications of the targeted process innovations.


Please send comments and suggestions to popi-www@mit.edu

[Back to POPI Home Page]