Ralph Nader and James Love
Testimony before the Special Committee on the Aging
of the United States Senate
February 24, 1993
Table of Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Summary of Comments. . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Federally Funded Health Care R&D . . . . . . . . . . . . . . . . . . . . 5
Investments in Clinical Trials 5
1991 Priority New Drug Approvals . . . . . . . . . . . . . . . . . . . . 6
Federal Funding of New Cancer Drugs. . . . . . . . . . . . . . . . . . . 8
Financial Disclosure . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Drug Pricing Models. . . . . . . . . . . . . . . . . . . . . . . . . . . 9
The Taxol Fair Pricing Negotiations. . . . . . . . . . . . . . . . . . 10
Appendix A
Role of Federal Funding for Ten 1991 Priority FDA Drugs. . . . . . . . 29
Definition of Priority Drugs 29
1991 Priority Drugs 29
Aredia 30
Ceredase 31
Fludara 33
Foscavir 35
Ganite 36
Nipent 37
Supprelin 38
Survanta 40
Ticlid 41
Videx 42
Sources 43
Appendix B
Notes on Federal Policies to Transfer Ownership of Federally Funded R&D 45
Appendix C
Notes on the Cost of New Pharmaceutical Inventions . . . . . . . . . . 49
Comparing Federal Investments to Private Investments 50
Tables
NIH and PMA Expenditures on Clinical Trials. . . . . . . . . . . . . . 13
Wholesale Prices of 1991 FDA NME Priority Drugs. . . . . . . . . . . . 14
NCI Role in the Development of 37 Cancer Drugs . . . . . . . . . . . . 15
NCI's Comparison Drugs for Purposes of Pricing Taxol . . . . . . . . . 17
Average Investment Lags before Drug Approval . . . . . . . . . . . . . 18
Figures
[GOPHER SITE NOTE: FIGURES ARE NOT INCLUDED IN THIS FILE, EMAIL
MIKE@TAP.ORG TO RECEIVE PAPER COPY]
1. Sources of Funding for U.S. Health Care R&D . . . . . . . . . . . 19
2. 1991 FDA Priority NME Drug Approvals. . . . . . . . . . . . . . . 20
3. National Cancer Institute Research Role for 37 Cancer Drugs . . . 21
4. Companies that Market 34 Cancer Drugs Developed with Funding from
National Cancer Institution . . . . . . . . . . . . . . . . . . . 22
5. NCI Role in the Development of 11 Cancer Drugs Marketed by Bristol-Myers
Squibb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
6. Opportunity Cost of One Dollar Invested in R&D. . . . . . . . . . 24
7. Opportunity Cost of One Dollar Invested in R&D. . . . . . . . . . 25
8 Cost Multiplier Adjusted for Risk . . . . . . . . . . . . . . . . 26
9. Cost Multiplier to Adjust for Risk and Opportunity Cost of
Capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
10. Estimated Shares of Drug R&D Costs. . . . . . . . . . . . . . . . 28
Introduction
Mr. Chairman and members of the Committee, thank you for the opportunity to testify today. Your leadership on the issue of drug pricing is richly appreciated by the millions of American consumers who pay for drugs as consumers, as purchasers of health insurance, and as taxpayers.
Over the past two years the Center for Study of Responsive Law has undertaken a review of federal policies and practices regarding the transfer of federally funded medical Research and Development (R&D) to the private sector. Our initial research involved a number of detailed case studies of particular drugs, which identified numerous problems with federal policies and laws regarding the commercialization of federally funded drug R&D. More recently, we have expanded our research efforts to address the following questions:
1. What is the extent of the federal government's role in the funding of new drug development?2. Do federal laws and policies regarding the allocation of property rights from federally funded drug R&D protect the public interest?
3. What changes in federal laws and policies are needed to ensure that drugs developed with federal funds are priced fairly?
4. What new analytical and management approaches are needed to control prices of drugs developed with federal funds?
Summary of Comments
Our comments today will address the following points.
One firm, Bristol-Myers Squibb, has benefitted the most from the NCI new drug program. Of the 34 cancer drugs developed with federal funding, 11 are marketed by Bristol-Myers Squibb, including the recent blockbuster drug Taxol.
In comparing the relative contributions of the government and the private sector in the development of new drugs, it is important to recognize ways that industry spokesman manipulate the data. For example, studies of the industry's costs of developing new drugs typically adjust nominal expenditures for inflation, risk and the opportunity cost of capital. In contrast, the government's costs of drug development are frequently presented in nominal terms, without any adjustments for inflation, risk or the opportunity cost of capital. As a result, many observers have a grossly distorted view of the economic value of the government's drug research investments. For example, some studies report industry Phase I investments at 11 times the initial nominal cash outlays, while government agencies often report drug development costs that only reflect nominal cash outlays to contractors, and ignore the government's costs of intramural research.
While there is ample evidence that private sector prices for drugs are excessive, there is very little data available to the government to determine fair prices. It is easier to determine what is unfair than
what is fair. In order to determine fair prices, the government needs better economic data on the pharmaceutical industry, including the costs and risks of development, manufacturing and marketing drugs, and it also needs to develop better methodologies for determining fair prices.
The recent attempt by NCI to determine a fair price for Taxol illustrates the primitive nature of NIH efforts in this regard. Taxol was discovered, manufactured and tested in humans by the National Cancer Institute. NCI gave Bristol-Myers Squibb (BMS) a Cooperative Research and Development Agreement (CRADA) that assigned to the firm the exclusive rights to commercialize all NCI past and future Taxol research. BMS, the NCI's favorite partner in drug development, paid nothing for the CRADA and will pay the government no royalties on its Taxol sales. However, BMS did agree to a "fair pricing" clause in the Taxol CRADA.
BMS's only contribution to the NDA approval for Taxol was to supply NCI with approximately 17 kilos of Taxol, and to process the paperwork for the NDA. When the NDA was approved in December 1992, BMS announced a price of $4.87 per milligram. The cost of a completed Taxol treatment will exceed $10,000 for some patients.
The Taxol price was the product of bizarre negotiations between NCI and BMS. NCI claimed that BMS simply refused to disclose any information on its development, research or marketing costs. NCI then gave BMS a list of the monthly wholesale prices for 15 arbitrarily chosen drugs and told BMS to price Taxol below the median for the group.
To consider the adequacy of the NIH's "fair pricing" methodology, consider the following facts. NCI was able to produce Taxol in small research quantities at $.60 per milligram prior to its 1989 agreement with BMS, using the same third party contractor as is used by BMS. According to the BMS contractor's recent filings with the Securities and Exchange Commission (SEC), it is under contract to produce approximately 400 kilograms of Taxol for BMS by August 1994, for which the firm expects to be paid $100 million. The wholesale value of 400 kilograms of Taxol, on the other hand, is $1.948 billion.
Since BMS is able to manufacture Taxol for about $.25 per milligram -- about 5 percent of the current wholesale price, the company's cost of providing 17 kilograms of Taxol to NCI for research purposes (which BMS owns the rights to) was less than $5 million.
In the face of criticism of the Taxol price from Representative Ron Wyden and the Taxpayer Assets Project, BMS claimed that it had made "huge" investments in the development of Taxol, which were in excessive $114 million. However, while the company refused to provide any accounting of where these "huge" expenditures had gone, the $114 million
figure clearly was based upon BMS long term contracts to supply Taxol for manufacturing purpose, and not for the research costs of the drug, which were largely borne by taxpayers. Moreover, NCI's claim that it was denied access to the BMS data reflects its contempt for its responsibilities to determine fair consumer prices for government funded drugs. The fact that BMS refused to provide financial data to NCI was a regrettable but hardly unsurmountable obstacle. While NCI is not run by rocket scientists, a child with a fourth grade education and a pencil and paper could have easily estimated BMS's development and manufacturing costs from a review of publicly available SEC documents.
Federally Funded Health Care R&D
Health care R&D is the second largest element of federally funded R&D -- only defense R&D ranks higher. The Bush Administration's fiscal year 1993 budget included more than $12 billion for health care R&D. About 70 percent of all federal health care R&D expenditures are funded through the National Institutes of Health. The remaining expenditures are funded through other agencies in the Department of Health and Human Services (HHS), as well as in the Departments of Defense, Energy, Veterans Affairs, the National Science Foundation (NSF), the National Aeronautics and Space Administration (NASA), and other federal agencies.
The federal share of all national health R&D expenditures is estimated to be about 42 percent, compared to 47 percent for private industry. The remaining 10 percent is funded by non-profit foundations and other sources. (see figure 1).
The federal government's role in the development of new drugs spans a wide range of activities, encompassing nearly all aspects of drug development, such as the discovery of new therapeutic agents, clinical testing of drugs in humans, and the development and refinement of manufacturing techniques. The notable exception concerns the final step of drug development, which is the request for an FDA New Drug Application (NDA), which is required before the drug can be commercially marketed.
Investments in Clinical Trials
Among the more interesting figures are the federal expenditures on human use clinical trials, a relatively advanced area for drug research. From fiscal year 1989 to 1993 NIH expenditures on clinical trials grew from $495.5 million to $869 million, an increase of 75 percent. By comparison, the members of the Pharmaceutical Manufactures Association (PMA) reported spending $1,555 on clinical trials in 1989 (the most recent year for which data are available). (see Table 2).
Other federal agencies, ranging from ADAMHA to the Department of Defense also finance clinical trials of pharmaceutical products.
1991 Priority New Drug Approvals
The Taxpayer Assets Project is currently examining the role of the federal government in funding the development of therapeutically important new drugs. This study will eventually cover all FDA priority new drug approvals from 1989 to 1992. Based upon our preliminary results from 1991, we conclude that the federal government's contributions to the development of new drugs is more important than is commonly believed, and is particularly important for drugs that represent the best advances in therapeutic value, or which treat the most serious illnesses.
In 1991 the Food and Drug Administration (FDA) gave marketing approval to 327 new and generic drugs and biologic products. Of this total, the majority were for variations on existing drugs or generic forms of existing compounds. Thirty of the approvals were for New Molecular Entities (NMEs), which are defined by the FDA as drugs distinctly different in structure from those already on the market.
The FDA's priority review system determines the order in which pharmaceutical products will be examined for marketing approval. The first system of classification concerns the efficacy of the drug, since, for example, there is no need to rush an antibiotic through the FDA approval system if it has the same therapeutic effect as penicillin. Until 1992 drugs were classified by the FDA as either A, B or C -- according to the drug's therapeutic gain over other drugs already on the market.
Class A - Drug offering significant therapeutic gainClass B - Drug offering moderate therapeutic gain
Class C - Drug offering little or no therapeutic gain
The FDA also gave two additional priority classifications for drugs used to treat particular ailments.
Class E - Drugs for the treatment of a severely debilitating or fatal diseaseClass AA - Highest priority classification for AIDS Drugs
The efficacy ratings may be combined with the E or AA ratings. Videx, for example, is a Class A,AA,E drug, indicating that it is an important therapeutic gain over other drugs on the market that is used in the treatment of a severely debilitating or fatal disease associated with AIDS.
Our study of federal funding of new drug development focused on the 1991 FDA NME approvals that received FDA Class A, AA or E ratings. These included five Class A drugs, nine Class E drugs, and two Class AA drugs. The combined number of priority drugs was ten. Brief background information about the drugs is given in Appendix A. including are the trade and generic names, the indication for which the drug is used, the FDA efficacy and priority classification, a discussion of the federal government's role, if any, in the drug's development, and the price of the drug (which is normalized as the wholesale cost of a completed course of treatment or one year of treatment, whichever is less). When possible, a single price for an average treatment was used. In four cases it was necessary to estimate ranges of prices based upon different treatment regimes. These data are summarized in Table 2. (see also figure 2).
Although we expected the government's role to be more pronounced among the priority new drug approvals, we were surprised at the strength of the results.
There were a number of other surprises as well. Among the FDA NME priority drugs approved in 1991, those that were developed with federal funding were priced considerably higher than those developed without federal funding. Drugs developed without federal funding were priced at $321 to $2,376, while drugs developed with federal funding were priced at $368 to $546,000. Among the seven priority drugs developed with federal funding, four were priced at $9,350 or more -- nearly four times as high as the most expensive drug developed without federal funding -- and only one was priced less than $1,000.
As expected, the federal government's Orphan Drug Act is an important factor. Six of the priority drugs, including four of the five Class A drugs benefitted from the exclusive marketing protections of the Orphan Drug Act. Moreover, four of the five highest priced drugs received Orphan Drug marketing protection.
Federal Funding of New Cancer Drugs
The Taxpayer Assets Project is also able to report on the federal government's role in the development of the 37 cancer drugs which have been developed since the beginning of the National Cancer Institute's new drug program in 1955. This analysis is based upon information provided by the National Cancer Institute.
In the April 1989 issue of the International Journal of Radiation Oncology, Biology and Physics, Bruce Chabner and Dale Shoemaker from NCI's Division of Cancer Treatment examined 37 cancer drugs. Based upon the publication dates of research, they concluded that 16 of the drugs were discovered prior to the beginning of the NCI's new drug program, while 21 of the drugs were developed afterwards. Chabner and Shoemaker identified the institution that discovered the antitumor agent, as well as NCI's contribution to the drug's preclinical and clinical research. Dr. Shoemaker and Dr. Saul Shepartz from NCI have recently provided unpublished updated information on 16 cancer drugs which were approved for marketing after the 1989 article was written. The results are reported in Table 3.
Of the 37 cancer drugs developed since 1955, the federal government was directly or significantly involved in the preclinical development of 18, and played some role the preclinical research for 10 others. In only 9 cases was NCI not involved at all in the preclinical research. When the drugs reached the stage for clinical research, NCI's role was even more pronounced. NCI played an important role in the funding of clinical research for 34 of the 37 drugs, or 92 percent of the entire group. (see figure 3).
Bristol-Myers Squibb benefitted greatly from the NCI's new drug program. Among the 34 drugs which received NCI funding, 11 are marketed by Bristol- Myers Squibb. No other drug company sells more than two cancer drugs which were developed with NCI funds. (see figure 4). Moreover, NCI played a decisive role in the preclinical research for seven of the 11 Bristol-Myers Squibb drugs, and providing significant role in the clinical research on all 11 drugs. (see figure 5).
Financial Disclosure
In order to control drug prices in an intelligent manner, the government needs better economic data on the costs and risks of development, manufacturing and marketing drugs.
The financial disclosure should identify the revenues the firm receives for the sale of the drug and the costs the firm incurs in developing, manufacturing and marketing the drug. Moreover, it is important that the firm provide historical data which shows when research and development expenses were incurred, relative to such important benchmarks, such as the beginning of Phase I, II, or III trails, the award of Orphan Drug or Patent protection, or
other events which will help evaluate the risks the firm undertook in the development of the drug.
The historical information will be important to determine how much of the industry's expenditures on the development of a drug occur at the riskiest phases. Investment before clinical trials is a higher risk than investment after clinical trials. Investments in Phase I trials are more risky than investments in Phase II trails. Investments after an FDA NDA approval are quite different than investments before an NDA approval. Research before a patent is more risky than research after a patent. If consumers or policy makers want to sort out the industry's claims about the risks involved in the development of drugs, it needs detailed information about the amounts and timing of investments, juxtaposed by the relevant milestones.
Congress should also require this disclosure to be made as public as possible.
Under most most NIH CRADAs the government promises to hold the company's financial data confidential. PHS often refuses to disclose even the terms of royalty agreements, let alone company financial data.
Financial disclosures should be as widely available as possible. The information is needed for policy research, and it is also needed to evaluate the performance of government officials who work for the taxpayers.
The drug industry vastly overstates its proprietary interests in financial data. In many cases the data are only secret from the public. Firms that purchase industry trade journals or which subscribe to expensive industry surveys such as those provided by IMS on drug revenues, can easily obtain good economic data on competitors. Companies themselves regularly publish detailed information about drug revenues in their own annual reports and SEC filings. If investors can receive this information, why not provide it to a broader public in a systematic and useful format. Moreover, in many cases the firms are protected against competition by patents or other government monopolies, such as the Orphan Drug or the exclusive agreements that were used for Taxol.
Secrecy about the economics of pharmaceutical drugs is necessary to protect the industry's political interests, rather than their business interests.
Drug Pricing Models
The development of new drugs is a complicated process that involves risks and judgments. Prices that are set ex post should reflect these considerations, and should not create inefficient ex ante incentives. Policy makers need to be sensitive to the stages where price competition is possible, such as when a Phase III drug is licensed to the private sector, as well as cases where price competition is impossible, such as when a private firm already holds a patent of an agent that is the subject of government research.
We are currently working with Professor David Genesove, a member of the economics faculty at the Massachusetts Institute of Technology, and other experts on the development of models that can be used by the federal government to:
a) determine if an exclusive agreement is in the public interest;
b) increase price competition in negotiations on exclusive agreements; and
c) determine reasonable prices for drugs developed with mixed funding.
Preliminary findings from this research are expected in April.
The Taxol Fair Pricing Negotiations
The recent attempt by NCI to determine a fair price for Taxol illustrates the primitive nature of NIH efforts to control prices for government funded drugs.
Taxol was discovered, manufactured and tested in humans by the National Cancer Institute. NCI gave Bristol-Myers Squibb (BMS) a Cooperative Research and Development Agreement (CRADA) that assigned to the firm the exclusive rights to commercialize all NCI past and future Taxol research. BMS, NCI's favorite partner in drug development, paid nothing for the CRADA and will pay the government no royalties on its Taxol sales. However, BMS did agree to a "fair pricing" clause in the Taxol CRADA.
BMS's only contribution to the NDA approval for Taxol was to supply NCI with approximately 17 kilos of Taxol, and to process the paperwork for the NDA. When the NDA was approved in December 1992, BMS announced a price of $4.87 per milligram. The cost of a completed Taxol treatment will exceed $10,000 for some patients.
The Taxol price was the product of bizarre negotiations between NCI and BMS. NCI claimed that under the Taxol CRADA, BMS was not required to disclose any information on its development, research or marketing costs. Rather than focus on BMS's actual costs, NCI officials told the firm that it expected the drug to be priced in the range of other cancer drugs. NCI submitted to BMS a list of 15 drugs and their estimated "monthly wholesale cost." (see Table 4). BMS was asked to set the price so that one month of Taxol treatment would cost no more than the median price for the group. NCI was, in essence, telling BMS that it could price Taxol, a government funded drug invention, the same as other cancer drugs, regardless of where the funding came from, and regardless of how "fair" those prices were.
The median cost of the drugs on the list was $1,776. The highly arbitrary nature of the "list" is evident. There is one non-cancer drug on the list, human growth hormone (hGH) -- which is estimated to cost $2,520 per month. If a lower priced drug had been used instead, such as Ganite, Aredia or Ticlid, the median price would have been the $1,030 for carboplatin -- priced 42 percent lower than the $1,776 median obtained when hGH is included. Moreover, why did NCI choose the "monthly cost" when other comparisons, such as the cost of a completed treatment, are available, and why didn't NCI take into consideration such relevant factors as the size of the client population, or the types of intellectual property rights? No one from NCI wants to answer such questions.
Ironically, the lowest priced drug on the list, Johnson & Johnson's levamisole, is widely considered to be excessively priced, even at $100 per month, since the same drug has long been used to deworm sheep for about $.06 per pill -- a small fraction of the $6 per pill that Johnson & Johnson now charge human patients.
How did NCI's "list" compare to the actual prices set by BMS? A patient who receive eight vials per treatment cycle (a common dose), will pay $1686.92 a month for Taxol, a few dollars less than the $1,776 median cost from NCI's list.
To consider the adequacy of the NIH's "fair pricing" methodology, consider the following facts. NCI was able to produce Taxol in small research quantities at $.60 per milligram prior to its 1989 agreement with BMS, using Hauser Chemical Research, the same third party contractor that is currently used by BMS. According to a Hauser stock prospectus dated September 25, 1992, the firm was under contract to produce approximately 400 kilograms of Taxol for BMS by August 1994, for which the firm expected to be paid approximately $100 million. The wholesale value of 400 kilograms of Taxol, on the other hand, is $1.948 billion.
Since BMS is able to manufacture Taxol for about $.25 per milligram -- less than 6 percent of the current wholesale price -- the company's cost of providing 17 kilograms of Taxol to NCI for research purposes (which BMS owns the rights to) was probably less than $5 million.
In the face of criticism of the Taxol price from Representative Ron Wyden and the Taxpayer Assets Project, BMS claimed that it had made "huge" investments in the development of Taxol -- in excess of $114 million. However, while the company refused to provide any accounting of where these "huge" expenditures had gone, the $114 million figure clearly was based upon BMS long term contracts with Hauser, Weyerhaeuser and others to supply Taxol for manufacturing purpose, and not for the research costs of the drug, which were largely borne by taxpayers. The fact that BMS refused to provide financial
data to NCI was a regrettable but hardly unsurmountable obstacle. Since NCI officials were intimately knowledgeable about the cost of obtaining Taxol from third party sources, including BMS's own contractor, NCI could easily have estimated BMS's costs of supplying Taxol for the NCI sponsored clinical trials and compassionate use program, which were BMS's only real contributions to the Taxol NDA.
At some point Members of Congress need to examine the NCI/BMS relationship to better understand why this company has received so much from our government.
Year NIH PMA
1988 487.6 1,333.4
1989 495.5 1,555.0
1990 609.7
1991 745.2
1992e 843.8
1993e 868.8
Source: NIH Budget Formulation Office and PMA's 1989-
1991 Annual Survey Report.
Orphan
Drug Efficacy Priority Status Price
Drugs Developed With Government Funding
Ceredase A A,E yes 57,960-546,000
Fludara A A,E yes 9,350
Foscavir B AA,E no 21,214
Ganite B E yes 368
Nipent A A,E yes 33,600
Supprelin A A yes 5,136-7,126
Videx A A,AA,E no 1,745
Drugs Developed Without Government Funding
Aredia B E no 312-468
Survanta C E yes 594-2,376
Ticlid B E no 803
Notes: Unit drug prices were based upon wholesale price per unit,
as reported in the 1992 Drug Topics Red Book (Medical Economics
Data: Montvale, N.J.: 1992) or from industry sources. Treatment
regimes obtained from industry sources and the 1992 Physicians
Desk Reference, Edition 46, (Medical Economics Data: Montvale,
N.J. 1992). In all cases we have reported the costs of a
completed course of treatment or a year of treatment, whichever
was less. FDA efficacy and priority ratings are given the "Drugs
Approved in 1991," FDA Talk Paper, January 15, 1992.
DRUG DISCOVERY DATE MARKETING NCI NCI
INSTITUTION OF COMPANY PRECLINICAL CLINICAL
NDA ROLE ROLE
______________________________________________________________________________
Vincristine Lilly 1963 Lilly none major
Pipobroman NCI 1966 not marketed discovery major
Hydroxyurea NCI 1967 Squibb discovery major
Cytarbine Upjohn 1969 Upjohn significant major
Procarbazine Roche 1969 Hoffman-LaRoche none major
Mithramycin NCI 1970 Pfizer discovery major
o,p - DDD NCI 1970 Bristol discovery major
Bleomycin IMC 1973 Bristol little major
Doxorubjcin Farmitalia 1974 Adria little major
Dacarbazine NCI 1975 Dohme discovery major
Lomustine NCI 1976 Bristol discovery major
Carmustine NCI 1977 Bristol discovery major
Cisplatin Michigan 1978 Bristol significant major
State
L-Asparaginase Cornell 1978 Merck significant major
Daunorubicin Farmitalia,1979 Ives little major
Rhone Poulenc
Streptozotocin NCI 1982 Upjohn discovery major
VP-16 Sandoz 1983 Bristol little major
Leuoprolide Abbott 1985 Abbott little major
alpha-interferon England 1986 Roche little major
alpha-interferon England 1986 Schering little major
Mitoxatrone NCI 1987 Lederle discovery major
Ifosfamide Asta-Werke 1988 Bristol-Myers little major
Squibb
Carboplatin NCI 1989 Bristol-Myers significant major
Squibb
Flutamide Schering- 1989 Schering-Plough none none
Plough
Zoladex ICI 1989 ICI none none
BCG 1990 Connaught Organon none major
Teknika
GM-CSF Immunex 1990 Immunex Hoechst none major
Roussel
HMM American 1990 US Bioscience significant major
Cyanamid
Idarubicin Farmitalia 1990 Adria Labs. none none
Levamisole Janssen 1990 Janssen little major
Fludarabine Southern 1991 Berlex discovery major
Phosphate Research Institute
(NCI Contractor)
G-CSF Amgen 1991 Amgen none major
Pentostatin Parke-Davis 1991 Parke-Davis significant major
(NCI Contractor)
VM-26 Sandoz 1992 Bristol-Myers little major
Squibb
IL-2 Cetus 1992 Chiron none major
Taxol NCI 1992 Bristol-Myers discovery major
Squibb
Sources: Bruce A. Chabner and Dale Shoemaker, "Drug Development for Cancer:
Implications for Chemical Modifiers," International Journal of Radiation
Oncology, Biology and Physics, April 1989, Volume 16, Number 4, pp. 907-909.
Dr. Dale Shoemaker and Dr. Saul Schepartz, Division of Cancer Treatment,
National Cancer Institute. Taxpayer Assets Project.
monthly
wholesale
drug price
gm-csf 4,095
c-csf 3,010
deoxycoformycin 2,880
hGH 2,520
mitoxantrone 2,010
ifosfamids 1,960
idarubicin 1,929
bcg 1,776
carboplatin 1,030
fludarabine 720
cisplatin 680
hexamethylmelamine 612
zoladex 322
flutamide 216
levamisole 100
Median price 1,776
Mean price 1,591
years from mean
phase start phase
Phase to approval length
Preclinical 11.8 3.6
Phase I 8.2 1.3
Phase II 6.9 2.0
Phase III 5.0 3.0
NDA Review 2.5 2.5
Source: J.A. DiMasi et al., "Costs of innovation in
pharmaceutical industry," Journal of Health Economics 10 (1991).
[HTML Note: Pages 19- 28 are graphs, email love@tap.org for a paper copy]
Definition of Priority Drugs
In 1991 the Food and Drug Administration (FDA) gave marketing approval to 327 new and generic drugs and biologic products. Of this total, the majority were for variations on existing drugs or generic forms of existing compounds. Thirty of the approvals were for New Molecular Entities (NMEs), which are defined by the FDA as drugs distinctly different in structure from those already on the market.
The FDA's priority review system determines the order in which pharmaceutical products will be examined for marketing approval. The first system of classification concerns the efficacy of the drug, since, for example, there is no need to rush an antibiotic through the FDA approval system if it has the same therapeutic effect as penicillin. Until 1992 drugs were classified either A, B or C according to the drug's therapeutic gain over other drugs already on the market.
The FDA also gave two additional priority classifications for drugs used to treat particular ailments.
The efficacy ratings are combined with the E or AA rating. Videx, for example, was a Class A,AA,E drug, indicating that it is an important therapeutic gain over other drugs on the market that is used in the treatment of a severely debilitating or fatal disease associated with AIDS.
1991 Priority Drugs
Our population of 1991 priority drugs includes all ten FDA NME approvals that received FDA Class A, AA or E ratings. Of these, five were Class A drugs, nine were Class E drugs, and two were Class AA drugs.
Trade Name: Aredia Generic Name: pamidronate disodium Marketing company: Ciba-Geigy Date of FDA marketing approval: October 31, 1991Indication Aredia is used to treat cancer-related hypercalcemia. This disease is a serious problem in the clinical management of patients with cancer. Almost one-half of women with metastatic breast cancer eventually develop hypercalcemia. FDA Classification Efficacy: B Priority: E Orphan Drug status: no Federal Involvement discovery of agent: no pre-clinical research no clinical research: no No federal research or funding was used in development of Aredia. Intellectual Property Rights Aredia was discovered by Ciba-Geigy researchers in England and subsequently patented in the U.S. Drug Economics Mild hypercalcemia is treated with a single 60mg infusion of Aredia costing $312. A severe case of hypercalcemia costs $468 using a single 90mg infusion.
Trade Name: Ceredase
Generic Name: alglucerase
Marketing Company: Genzyme Inc.
Date of FDA marketing approval: April 5, 1991
Indications
Ceredase is used in the treatment of Gaucher's disease, an inherited
metabolic disorder effecting the production of a critical bodily enzyme.
Gaucher's disease is severely debilitating disease that causes
hematologic disorders, enlargement of the liver and spleen, bone erosion
and pain. A significant number of Gaucher's disease victims in the U.S.
are Ashkenazic Jews.
FDA Classification
Efficacy: A Priority: A,E
Orphan Drug status: yes
Federal Involvement in Development
discovery of agent: yes
pre-clinical research: yes
clinical research: yes
A 1992 study, conducted by the congressional Office of Technology
Assessment (OTA), described three very important breakthroughs in the
treatment of Gaucher's disease. First, in the mid-1960's Dr. Roscoe
Brady of the NIH isolated the glucocereborosidase enzyme that Gaucher
sufferers lack. Second, during the 1970's NIH researchers discovered a
method for harvesting the enzyme from the human placenta. Third,
government scientists discovered a chemical modification that greatly
improved the effectiveness of the enzyme. This modified form became
Ceredase. In addition to this developmental research, the National
Institutes of Health conducted numerous clinical trials during the
1980's.
Intellectual Property Rights
Because extensive research about alglucerase has been published, the
drug is non-patentable. Genzyme has seven years of marketing
exclusivity under the Orphan Drug Act.
Drug Economics
According to the Congressional Office of Technology Assessment (OTA),
the treatment cost to patients, most of whom must take the drug
indefinitely, ranges from $57,960 to $546,000 per year.
According to OTA, the cost of producing, manufacturing, and marketing
costs for Ceredase was $1.90 per unit in 1992. The unit cost of
production is expected to fall with higher scale production. The
wholesale price of Ceredase is $3.50 per unit, a markup of 84 percent
over the 1992 production costs. OTA estimates that Genzyme spent $29.4
million on R&D for Ceredase. A significant portion of these
expenditures were for the acquisition of alglucerase manufacturing
facilities.
OTA estimates that 2,100 to 11,000 persons are candidates for treatment
of Type 1 Gaucher Disease.
Trade Name: Fludara
Generic Name: fludarabine phosphate
Marketing Company: Berlex Laboratories
Date of FDA marketing approval: April 1991
Indications
Fludara is used in the treatment of chronic lymphocytic leukemia, an
often fatal disease.
FDA Classification
Efficacy: A Priority: A,E
Orphan Drug Status: yes
Federal Involvement in Development
discovery of agent: yes
pre-clinical research: yes
clinical research: yes
The Division of Cancer Treatment (DCT) at the National Cancer Institute
(NCI) conducted extensive research on converting ara-A, an acute
leukemia drug limited by its rapid degradation in the body into a more
insoluble and useful drug. In 1982 two Department of Health and Human
Service workers discovered a way to convert ara-A into a more insoluble
compound, fludarabine phosphate.
After concluding preclinical toxicology and pharmokinetic studies, the
DCT proceeded with clinical trials on Fludarabine in 1983. A number of
clinical trials have been conducted or sponsored by DCT, NCI; 8 are
currently active. There have been nine Phase I and three Phase I-II
trials in adults and an additional 4 Phase I-II trials have been
conducted in children. Fludara was approved on the basis of two single
arm studies conducted at the Southwest Oncology Group, a branch of NCI's
cooperative research group, and M.D. Anderson working under federal
grants.
Intellectual Property Status
The Department of Health and Human Services received a patent for the
drug in 1982. In 1984 the National Technical Information Service (NTIS)
licensed the patent to Dupont Chemical Company and Triton Biosciences
who later sold their rights to Berlex Laboratories the current marketer
of Fludara. As a result of Orphan Drug status, Berlex enjoys exclusive
marketing rights through 1998.
Drug Economics
Fludara is administered for five consecutive days. This cycle is then
repeated every 28 days, or 13 times a year. The daily dose is 25mg per
m2 of body surface. A person with a body surface of 1.7m2 would use
five vials per treatment, or 325 vials per year. The cost of one vial
is $143.85. Some patients will receive the drug for the rest of their
lives. The cost of a year of treatment for a person of average size is
$9,350.
Trade Name: Foscavir
Generic Name: foscarnet sodium
Marketing Company: Astra Pharmaceuticals
Date of FDA marketing approval: September 27, 1991
Indication
Foscavir is used in the treatment of the cytomegalovirus, an
opportunistic infection associated with HIV. It is the most prevalent
eye infection and leading cause of vision loss among AIDS patients.
FDA Classification
Efficacy: B Priority: AA,E
Orphan Drug status: no
Federal Involvement
discovery of agent: no
pre-clinical research: no
clinical research: yes
Of the five clinical trials evaluated for the NDA filed in June 1990,
two were sponored by NIH, at a cost of $3.05 million. NIH research on
Foscavair continued after Astra received FDA marketing approval. To
date the U.S. government has spent an estimated $21.8 million on
Foscavir clinical trials.
Intellectual Property Rights
Foscavir's patent is held by Astra Pharmaceuticals.
Drug Economics
Foscavir is administered in two stages. The induction therapy is
60mg/kg of body weight 3 times per day for 21 days. This is followed by
a 90mg/kg maintenance dose that is given indefinitely. A 110 pound
person would receive 1,737 grams of Foscavir over the period of one
year. The wholesale price of Foscavir is $73.28 for 6 grams. The cost
of 21 days of induction therapy followed by 344 days of maintenance
therapy is $21,214.
Trade Name: Ganite
Generic Name: gallium nitrate
Marketing Company: Fujisawa Friar
Date of FDA marketing approval: January 17, 1991
Indications
Ganite is used in the treatment of cancer-related hypercalcemia. This
disease is a serious problem in the clinical management of patients with
cancer. According to some researchers, almost one-half of women with
metastatic breast cancer eventually develop hypercalcemia.
FDA Classification
Efficacy: B Priority: E
Orphan Drug status: yes
Federal Involvement in Development
Discovery of agent: yes
pre-clinical research: yes
clinical research: yes
Researchers at Memorial Sloan-Kettering Institute for Cancer Research in
New York City discovered gallium nitrate's effectiveness in the
treatment of hypercalcemia. They were working under a general research
grant given to Sloan-Kettering by the National Cancer Institute. The
research was conducted with gallium nitrate provided by the Division of
Cancer Treatment of NCI. The clinical trials were held at Sloan-
Kettering using NCI and FDA grants.
Intellectual Property Status
As a result of the research performed at Sloan-Kettering with NCI
funding, Sloan-Kettering applied for and received a patent on the use of
gallium nitrate to treat calcium homeostasis disorders in July of 1985.
This and Orphan Drug status ensure Fujisawa Fria exclusive marketing
rights.
Drug Economics
Ganite is administered at a dose of 200mg/m2 of body surface for five
days. Four 500mg vials, priced at $92 would be necessary to treat a
person with a body surface of 1.7 m2, for a total cost of $368.
Trade Name: Nipent
Generic Name: pentostatin
Marketing Company: Parke-Davis
Date of FDA marketing approval: October 15, 1991
Indication
Nipent is used in the treatment of Hairy Cell Leukemia. This chronic
illness affects approximately 2500 patients in the U.S. About 500-600
new patients are diagnosed yearly. Hairy Cell Leukemia has a median
survival period of approximately four years from the time of diagnosis.
FDA Classification
Efficacy: A Priority: A,E
Orphan Drug status: yes
Federal Involvement in Development
discovery of agent: Parke-Davis with funding from NCI
pre-clinical research: yes
clinical research: yes
Even though Parke-Davis discovered pentostatin the National Cancer
Institute was the most significant organization involved in the
development. According to Dr. Saul Schepartz of NCI, the federal
government played a "significant" role in the pre-clinical research of
pentostatin. The approval of Nipent was based on Phase II data from 133
patients during five clinical trials, with most of the clinical work
conducted at M.D. Anderson Hospital and by Cancer Leukemia Group B with
NCI funding.
Intellectual Property Status
Parke-Davis applied for and received a patent for pentostatin.
Drug Economics
Nipent is administered at a dose of 4mg/m2 of body surface once a week
for 28 weeks. The cost of one vial is $1,200. The cost of 28 weeks of
treatment is $33,600.
Trade Name: Supprelin
Generic Name: histrelin acetate
Marketing Company: Roberts Pharmaceuticals
Date of FDA marketing approval: December 24, 1991
Indication
Supprelin, a synthetic agonist of the gondatropin releasing hormone
(GnRH) is used in the treatment of central or unexplained precocious
puberty, a condition in which young children develop secondary sexual
characteristics of adolescents.
FDA Classification
Efficacy: A Priority: A
Orphan Drug status: yes
Federal Involvement
discovery of agent: yes
pre-clinical research: yes
clinical research: yes
Jean Rivier of the Salk Institute, a non-profit research organization,
discovered histrelin acetate while conducting government funded research
on the basic mechanisms of GnRH's. The Salk Instute conducted pre-
clinical toxicology and animal studies as well as producing the first
batches of histrelin acetate for the earliest clinical research. Dr.
William Krawley, of Massachussets General Hospital, in the midst of
conducting ten years of NIH funded research on GnRH's, used histrelin
acetate in the treatment of precocious puberty. This research was the
largest group of data used in the NDA for histrelin acetate.
Intellectual Property Rights
Patent #4244946, issued in January 13, 1981 expires January 13, 1998.
Orphan Drug protection expires on December 24, 1998.
The Salk Institute licensed their patent for histrelin acetate to six
pharmaceutical companies. All but Johnson and Johnson (J&J) gave up
their license. Ortho pharmaceuticals, a subsidiary of J&J, sold its
license to Robert's Pharmaceutical, the current marketer of Supprelin.
Drug Economics
Histrelin acetate is purchased in a 30 day kit that includes syringes
and medicine. The two most commonly used solutions are 500 mg/ml and
1000 mg/ml costing $428 and $594, respectively. A years supply of the
smaller dose costs $5,136, the larger dose costs $7,126. Since this drug
retards the onset of puberty, a four year old child with precocious
puberty might have to take the drug until the normal onset of puberty,
11, 12, or 13 years of age.
Trade Name: Survanta
Generic Name: Beractant
Marketing Company: Ross Laboratories
Date of FDA marketing approval: July 1, 1991
Indication
Survanta is a drug used to treat critically ill premature infants with
Respiratory Distress Syndrome (RDS). RDS afflicts approximately 50,000
of the 250,000 premature infants born in the U.S. each year and kills
approximately 5,000 of those infants annually.
FDA Classification
Efficacy: C Priority: E
Orphan Drug status: yes
Federal Involvement
discovery of agent: no
pre-clinical research: no
clinical research: no
No federal research or money was used in the development of Survanta.
Intellectual Property Rights
The Tokyo Tanabe Co. discovered Survanta in Japan and subsequently
applied for and received a patent in the U.S. Abbott Laboratories, the
parent company of Ross Laboratories, licenses this drug from Tokyo
Tanabe. Also, Orphan Drug status guarantees exclusive marketing rights
to Ross Laboratories.
Drug Economics
Infants can receive between one and four vials of Survanta at a cost of
$594 to $2,376.
Trade Name: Ticlid
Brand Name: ticlopidine hydrochloric acid
Marketing Company: Syntex Laboratories
Date of FDA Marketing Approval:
Indication
Ticlid is used for the management of thrombotic strokes, a common and
often fatal affliction.
FDA Classification
Efficacy: B Priority: E
Orphan Drug status: no
Federal Involvement
discovery of agent: no
pre-clinical research: no
clinical research: no
No federal research or money was used in the development of Ticlid.
Intellectual Property Status
Ticlid was discovered by Sanofi, a French pharmaceutical company.
Syntex licensed Ticlid from this company and re-evaluated previous
studies and conducted its own pre-clinical and clinical trials.
Drug Economics
One 250mg tablet is taken twice a day indefinitely. The wholesale cost
of each tablet is $1.10. The annual cost of Ticlid therapy is $803.
Trade Name: Videx
Generic Name: ddI
Marketing Company: Bristol-Myers Squibb
Date of FDA Marketing Approval: October 9, 1991
Indications
Videx is one of three antiviral drugs used to combat the AIDS virus.
FDA Classification
Efficacy: A Priority: A,AA,E
Orphan Drug status: approved but withdrawn by Bristol-Myers Squibb
Federal Involvement in Development
discovery of agent: yes
pre-clinical research: yes
clinical research: yes
The National Cancer Institute identified the activity of ddI against HIV
in the laboratory, performed preclinical development, and initiated the
first clinical trials in humans. NCI estimated that its costs for
developing ddI were $6.8 million through the end of 1990.
Intellectual Property Rights
In 1989 the Department of Health and Human Service received a patent for
ddI. On February 1, 1988, NCI signed an exclusive license with BMS
giving the company exclusive rights to market the drug for ten years
following the first commercial sale.
Drug Economics
BMS estimates the average wholesale cost for a patient receiving ddI
will be $1,745 per year.
Sources
At present the federal government does not publish information that routinely identifies the federal role in the development of new drugs. In order to establish what role, if any, the government played in the development of the drug, it was necessary to survey a wide range of sources. Often the pharmaceutical companies themselves down played or refused to acknowledge the government's role in the development of the drug. For example, a spokesperson for Roberts Pharmaceutical refussed to acknowledge any federal role in the development of Supprelin, while in fact, the drug was discovered at the Salk Institute and clinical testing was carried out by the Salk Institute and Massachusetts General Hospital with funding from NIH.
It is typically necessary to contact government agencies or the non-profit institutions that were directly involved in the research to identify the nature of government funded research in the development of the drug. The National Institutes of Health and its member agencies were helpful when asked specific questions, although only the Division of Cancer Treatment of the National Cancer Institute had routinely tracked its role in the development of new drugs.
In most cases it is necessary to research the history of the each drug's development through medical journals. Much of this research was assisted by the National Library of Medicine's MEDLINE database. It was also helpful to examine drug patents. The Department of Health and Human Services publishes Approved Drug Products, which lists all products which have been approved for marketing under Sections 505 and 507 of the Federal Food, Drug, and Cosmetic Act, including patent numbers and expiration dates. Patents were obtained from the Patent and Trademark Office (PTO). The PTO maintains an online system which can search the full text of all U.S. patents, but it does not provide public access to the system except in its reading rooms -- at a rate of $40 per hour plus $.25 per patent page. This service is also available from commerical data vendors, at very high prices. For example, the PTO estimates that a typical searcher would pay $340 per hour to search patents on LEXPAT, a service based upon the PTO's publicly funded database, which is marketed by Mead Data Central.
The Research Documentation Section, Information Systems Branch, Division of Research Grants at the National Institutes of Health maintains the CRISP database, which contains data on the research programs supported by the U.S.
Public Health Service from 1972 to the present. This database includes research grants, contracts, and cooperative agreements funded by NIH and ADAMHA, selected research grants funded by the Centers for Disease Control, FDA, the Health Resources and Services Administration, and the Agency for Health Care Policy, as well as selected information on the intramural programs of NIH, ADAMHA and FDA. The CRISP database is available online from commercial vendors such as Dialog, at prices too high for us. However, NIH does respond to mailed requests for searches. While the CRISP database is a helpful source of information that should be more readily available to researchers, the data are often only suggestive. For example, for many government grants or contracts, numerous drugs are under investigation, and it is not possible to obtain specific allocation of funding for individual drugs.
Drug prices were based upon wholesale price per unit, as reported in the 1992 Drug Topics Red Book, or from industry sources, and treatment regimes obtained from industry sources and the 1992 Physicians Desk Reference. In all cases we have reported the costs of a completed course of treatment or a year of treatment, whichever was less.
Information on the revenues from drug sales and the company's costs of development, manufacturing and marketing are incomplete. Revenues from individual drugs are collected by IMS, a Philadelphia research firm, and are readily available to the drug companies themselves, but the data are licensed with substantial restrictions against redissemination and are too expensive for most non industry researchers. In several cases, company SEC filings and annual reports contain public disclosure of information on individual drug revenues and company R&D or manufacturing expenditures. While federal agencies have a wealth of information about hospital and doctor services, no federal agency has responsibility for collecting or disseminating information about the economics of the pharmaceutical industry.
Policies concerning the allocation of intellectual property rights from federally funded research have changed markedly since World War II. There once was a broad consensus that federally funded research should be liberally shared with the public, either by allowing research to enter the public domain, or by licensing property rights, such as patents, on a non-exclusive basis. In recent years this has given way to a more proprietary approach.
Many of the important changes in federal policy occurred in the 1980's. These include the enactment of the Stevenson-Wydler Technology Innovation Act and the Bayh-Dole University and Small Business Patent Procedures Act (both passed in 1980) the 1984 mendments to the Bayh-Dole Act, and the Federal Technology Transfer Act of 1986. President Reagan also issued a number of Federal Executive Orders and memorandums which broadened the rights of government contractors and grant recipients to obtain titles to federally funded research products and grant exclusive licenses for commercial development.
These measures accomplished the following:
i. In those cases where federal research funds were managed by non- government personnel, contractors or grant recipients were given title to a wide variety of government funded research rights, with few restrictions on how these rights could be exploited.ii. In those cases where the federal government retained title to the research products (such as research carried out by government employees or by contractors or grant recipients who were not qualified to take title to the research) federal agencies were told to place an increasing emphasis on exclusive marketing agreements with firms to commercialize the research products.
Patent, copyright, trade secret status and other measures are used to control access and ownership of government funded research. Private corporations routinely obtain exclusive marketing privileges on everything from government funded software and technical databases to inventions of new drugs.
Rationales for this more proprietary approach to federal research are varied, but two stand out, in terms of the frequency to which they are employed and the degree to which they are believed:
i. the need to provide incentives for the commercial development of a product, andii. the need to prevent foreign interests from benefiting from U.S. funded research and development.
While these rationales are contradictory (the need to prevent foreign companies from developing U.S. technology suggests that sufficient incentives already exist for commercial development), they are often used simultaneously as a catch-all to justify policies.
These policy guidelines, which apply to all areas of federally funded research, have received wide bipartisan support in Congress, due in no small part to aggressive lobbying and generous political contributions by companies in defense, energy, pharmaceutical, and other fields, which have benefitted from these transfers. Critics worry that universities are becoming too concerned with potential commercial rewards from research, and are abandoning important types of basic research or restricting the free flow of information among scientists. There are also concerns that the large emphasis on exclusive marketing agreements will harm consumers, or lead to anticompetitive practices, such as blocking the development of related technologies, where companies which would use the government research product as an input for a
different product. Others have questioned the policies on the grounds that taxpayers should receive greater benefits from its investments. For example, in a minority report on the 1980 Bayh-Dole Act, which gave small businesses and non-profit institutions title to inventions developed under federal grants or contracts, Congressman Jack Brooks said:
Nowhere are these issues as important as in the medical field, where the public investment is enormous, and the product will not be a better microwave oven, but better health or even life itself. On the one hand it is important that there are proper incentives to develop new medicines and other health care technologies. On the other hand, consumers of medical technologies are uniquely vulnerable to monopoly pricing.The major problem I have with H.R. 6933 is that it violates a basic provision of the unwritten contract between the citizens of this country and their government; namely, that what the government acquires through the expenditure of its citizen's taxes, the government owns. Assigning automatic patent rights and exclusive licenses to companies or organizations for inventions developed at government expense is a pure giveaway of rights that properly belong to the people.
. . .The federal government has the equivalent of a fiduciary responsibility to the taxpayers of the country. Property acquired with public funds should belong to the public. Deviations from that fundamental principle should be allowed only where a compelling justification can be shown and where the voice of the public can be heard in protest. This legislation stands that principle on its head by automatically conveying title or the exclusive right to use public property to private entities and placing the burden on the federal government to demonstrate that a retrieval of those rights is in the public interest.
The Federal Orphan Drug Act is also an important mechanism for allowing private firms to obtain monopolies on government funded inventions. Since we have previously addressed this issue at length to other congressional committees, we will reference our earlier statements, and add only a few comments. Today the Orphan Drug Act is primarily viewed as a mechanism to obtain seven years exclusive marketing rights (a feature that was considered unimportant when the legislation passed a decade ago). Thus, the Act is most important when company's do not hold patents for the drugs. A common reason that a drug cannot be patented is because researchers have already reported results in academic journals. This is more likely the result of research funded by the government than research funded by the industry, since the drug companies are known for exercising tight controls on a researcher's right to publish before applications for patents have been processed.
Because much government funded research ends up in the public domain (although less now than before), the Orphan Drug Act has become an important mechanism to obtain patent like protection on substances that cannot be patented, because of government funded research which has been published in academic journals.
There are many things that the federal government can do administratively to correct the woeful inadequacies of agency policies on the pricing of government funded pharmaceutical inventions. The federal government can develop better approaches to negotiating or setting prices on government funded medical inventions. It can collect better information on the economics of drug development and profits, and it can place more emphasis on competition, by requiring agencies to make findings that exclusive agreements are necessary, or by restructuring the negotiations for exclusive contracts so that companies negotiate on prices before partners are chosen for exclusive arrangements (getting more mileage from the ex ante competition that exists in these awards).
But there are also many problems that cannot be solved without addressing the reforms in the Bayh-Dole University and Small Business Patent Procedures Act or the Orphan Drug Act -- two deeply flawed laws that have very powerful constituencies. So long as the federal government grants titles to federally funded research conducted by universities and other contract and grant recipients, the Congress will not be able to control prices for many government funded drugs. For example, the 1990 NIH Health Care R&D budget included $4.18 billion in work performed by universities and another $1.8 billion performed by other non-profit institutions. Until the Bayh-Dole Act is changed, the federal government will be severely restricted in its ability to control prices for many important taxpayer-financed drugs inventions.
Consider, for example, the recent agreement between the Scripps Research Institute and Sandoz, a Swiss pharmaceutical firm. Sandoz has agreed to pay Scripps $300 million for the exclusive right to commercialize new technologies developed by Scripps over a ten year period beginning in 1997. Since Scripps receives about $100 million per year in NIH funding, the agreement works like a reverse case of Industrial Policy. The U.S. taxpayers will be funding research that Sandoz will own the rights to, and will be free to sell back to them at unregulated prices. The Sandoz agreement, moreover, is pretty common stuff, distinguished only the amount of dollars involved. Many of the major research institutions now have full time staff whose job is to dream up similar industry partnerships, without regard to the national interest, or the effect of such agreements on consumers or taxpayers.
Most recent studies of the costs of developing new drugs have focused on the "green field" cost of drug development. That is, the studies look at the industry's cost of drug development, when the company finances all development, including investments in pre-clinical research. Because there is little data in the public domain on the economics of drug development, the best known research has been carried out by economists who work closely with drug companies, relying upon their consulting relationships to gain access to confidential data. The study which has received the widest exposure on Capital Hill is a 1991 paper by Joseph DiMasi, Ronald Hansen, Henry Grabowski and Louis Lasagna, researchers from Tufts University,the University of Rochester and Duke University. We will refer to this as the DHGL study.
DHGL's analysis has its strengths and weaknesses. The strength of the analysis is the availability of project level information from 99 "new chemical entities" (NCE), provided by 12 U.S. owned pharmaceutical companies. The weakness of the study is the lack of data on pre-clinical investments, which were estimated from aggregate data using assumptions that may or may not be accurate. The bottom line of the DHGL study is that the cost of developing a NCE is estimated to be $231 million. This figure, which is widely quoted, includes both cash outlays, adjusted for inflation, and imputed costs. The imputed costs include an adjustment for the "dry hole" risk of investments, and also the company's opportunity cost of capital, assumed to be a real rate of 9 percent. For example, if a company invests a dollar at a point where the chance of success is only 25 percent, it is counted as four dollars, and then increased again to reflect the opportunity cost of money. In their report, the average inflation adjusted cash outlay of $2.134 million for Phase I trials is adjusted for both risk and opportunity costs, and counts as $17.8 million.
We are currently undertaking an analysis of the DHGL study, as well as other estimates of the costs of development for new drugs, and we are not prepared at this time to provide the Committee with our findings. But there are several factors which the Committee should keep in mind.
1. First of all, the DHGL estimate of $231 million was far higher than previous estimates. For example, a 1987 study authored by Wiggins for the PMA, estimated the cost to be $125 million, and a 1979 study by Hansen estimated the cost to be $54 million (in 1976 dollars). Earlier studies, many of which were flawed on methodological grounds, came up with far lower figures.The fact that the DHGL study was considerably higher than previous estimates isn't itself a criticism of their finding. It is conceivable, for example, the costs of developing new drugs is itself increasing -- as evidenced by the rapid escalation in the prices of new drugs -- as well as the new incentives to invest that those high prices create. If so, Congress should redouble its efforts to control drug prices, particularly when the company does not bear the full costs and risks of development.
2. Next, the Committee may be interested to know that more than half of the $231 million estimated by DHGL was assigned to the opportunity cost of capital. That is, DHGL's estimate that the cash outlays, adjusted for inflation and risk, are $114 million, and the rest represents the profits that the company needs to satisfy investors. Thus, if a company were to receive $231 million in profits from drug sales in its first year, it would have "broken even" on the drug, including the necessary shareholder returns.3. Third, the Committee will probably be surprised to learn that DHGL assumes that two- thirds of all costs are related to the pre-clinical stages of development. This figure is really the softest part of their analysis, but it does suggest that the cost of developing drugs that have already entered the clinical testing phase is quite low. As a related issue, by the time a drug enters Phase III trials, DHGL conclude that more than 85 percent of the cost of the drug has already been paid for.
In comparing the relative contributions of the government and the private sector in the development of new drugs, it is important to recognize ways that industry spokesman manipulate the data. For example, as described above, studies of the industry's costs of developing new drugs typically adjust nominal expenditures for inflation, risk and the opportunity cost of capital. In contrast, the government's costs of drug development are frequently presented in nominal terms, without any adjustments for inflation, risk or the opportunity cost of capital. As a result, many observers have a grossly distorted view of the economic value of the government's drug research investments. For example, some studies report industry Phase I investments at 11 times the initial nominal cash outlays, while government agencies often report drug development costs that only reflect nominal cash outlays to contractors, and ignore the government's costs of intramural research.
To illustrate the types of adjustments that are made, we will assume that inflation is steady at 4 percent, the industry's opportunity real cost of capital is 9 per cent, and the average investment lags behind drug approval are the same as are used by DHGL, as reported in Table 5.
Figures 6 and 7 illustrate the inflation and return adjustments to a dollar of R&D investment, based upon different phases of drug development. Thus, for example, to evaluate the economic value of government investments in Phase I research, it would be appropriate to multiply the nominal outlays by 2.53 to reflect inflation and the cost of capital.
In Figure 8, further adjustments are made for the risks of development, based upon the DHGL estimates of the probabilities of success at in Phase I, II and III. In Figure 9 is the combined effect of all cost adjustments, including those for inflation, investment return, and investment risk. Thus, for example, in evaluating the economic value of the a government investment in Phase I, it would be appropriate to multiple the nominal cash outlays by 11.
Mathematically, the adjusted cost of drug development investment in Phase j can be expressed in terms of the nominal outlays, the Phase risk given by the probability of success Pj, the rate of inflation i, opportunity cost of capital r, and the years to drug approval, given by N.
(nominal outlay) n
Adjusted Cost = __________________ (1 + r + i )
j
P
j