CPSA Digest 2002

Proceedings -Tuesday, October 8, 2002

TuOA1: Plenary

The Economics of Pharmaceutical Innovation: Trends in Costs, Risks and Returns

Background:
The Tufts Center for the Study of Drug Development (CSDD) has been tracking drug development costs for 25 years. The newest research on development costs reveals a price tag of $802 million per drug. Rising costs throughout the development process have pushed costs to this level. The average drug takes 12 years to develop and must overcome a very selective process. Tufts research reports that of every 5,000 potential new drugs tested in animals, only five are promising enough to be tested in humans; only one of those five is eventually approved for marketing.

This presentation by Dr. Joseph DiMasi focused on four aspects of the Economics of Pharmaceutical Innovation:

• Risks in new drug development
• R&D costs per approved new drug
• Returns to new drug development
• Impact on cost from improvements in the development process

Analysis

A. Risks in New Drug Development
The total Dollars spent on pharmaceutical research & development are increasing at a greater rate than the number of new chemical entities (NCE) introduced each year. Shown in Figure†1 are data collected from 1963 through 2001 to illustrate this point.

As each drug progresses into the next phase of development, a "Phase Transition Probability" is associated with that move. For example, 71% of the drugs examined in Phase I make the transition into Phase II. Likewise, it has been determined that 44% of the drugs examined in Phase II make the transition into Phase III, and 69% of the drugs in Phase III reach NDA (New Drug Application) Approval.

Why do drugs fail to make it into clinical development by the time of IND (Investigational New Drug) filing? There are four major categories for this failure: economics, efficacy, safety and "other". A comparison of data from two time periods, 1981-1986 and 1987-1992, has been made. Although data are missing from the last ten years, the trend from 1981 to 1992 has been a slight increase (+4%) in failures due to economics and efficacy, a slight decrease in failures due to safety (-2%) and a larger decrease due to "other" (-6%). Thus, the recent impact of lead optimization and high throughput screening for efficacy is not shown by these data (source: DiMasi, Clin. Pharmacol. Ther. 69(5) (2001) 297-307.

B. R&D Costs per Approved New Drug
Investing in new drug development presents a risk that may never pay off, or that pays off only at a later date. For example, suppose that an investor is asked to spend $400 million on an R&D project but the returns on that investment will not be seen for 10†years. Most investors would demand a quicker return on their investment. The costs of R&D have grown substantially over the years, even when adjusted for inflation. The trends for fully allocated capitalized cost per approved drug are shown in Figure 2, comparing 1970s approvals with those from the 1980s and 1990s. Note that the current cost of drug development is estimated at $802 million per drug.

Note that only 1 in 5 drugs that enter the clinic make it through the regulatory approval process and reach the marketplace. The costs for clinical research are rising at a much more rapid rate than those for discovery and preclinical research. These costs are a result of many factors, including a greater number of clinical studies, rising numbers
of required volunteers to complete these studies, the greater complexity of clinical trial designs (more procedures per patient), more testing against comparative drugs and more treatments associated with chronic and degenerative diseases. Compared with 10 years ago, patient numbers have increased from 3,567 (1990-1992) to 5,507 (1994-1995) and to 5,621 (1998-2001).

C. Returns to New Drug Development
The market lifetime of a new drug can be as long as 20 years, but a peak is seen by year 10 in terms of a drugís revenue. As patent protection expires, revenues generated from a new drug decrease at an accelerated rate. It is commonly seen that only a few drugs are "blockbusters" and sustain a company and the industry. For example, the top 10% of sales performers account for 50% of value; also, only 30% of marketed drugs
earn back their R&D costs (defined as $500 million in this assessment).

D. Impact on Cost from Improvements in the Development Process
Productivity assessments have been made by company (anonymously), in terms of both technical and commercial productivity (source: DiMasi, Drug Information Journal 34(4) (2000) 1169-94). About 6 of 15 companies perform at or above the industry average. Cost reductions to drug development can potentially result from higher clinical success rates. Also, reduced development time by about 3 years is predicted to be able to reduce costs by about 30%. The value of this cost and time reduction would be a drug on the market much sooner, working to return the cost of investment. Shown in Figure 3 are estimates of the required reductions in time, clinical success rate and out of pocket preclinical costs that must be made in order to reduce the total costs of development by $100 million or $200 million.

Conclusions
The drug development process has been, and continues to be, a process that is risky, lengthy and costly. The costs for performing R&D have increased but so have the returns to those projects that are successful in the marketplace. Pharmaceutical innovation is a highly uncertain process, even for highly diversified and large firms. There are payoffs for improvements to this drug development process, in terms of reduced costs from greater success rates.

References and Links
Tufts Center for the Study of Drug Development
http://csdd.tufts.edu
The Tufts Center for the Study of Drug Development is an independent,
academic, nonprofit research group affiliated with Tufts University.
Founded in 1976, Tufts CSDD is internationally recognized for its
scholarly analyses and thoughtful commentary on pharmaceutical issues.
The mission of Tufts CSDD is to provide strategic information for drug
developers, regulators, and policy markers on improving the quality and
efficiency of pharmaceutical development, research and utilization. Read
the CSDD cumulative bibliography from 1976-2002 at
http://csdd.tufts.edu/InfoServices/Centerbib.pdf (requires Adobe Acrobat)

J. A. DiMasi, "Risks in new development: approval success rates for
investigational drugs." Clin. Pharmacol. Ther. 69(5) (2001) 297-307.

J. A. DiMasi, "New drug innovation and pharmaceutical industry
structure: trends in the output of pharmaceutical firms." Drug
Information Journal 34(4) (2000) 1169-94.

Return to Proceedings »