Reproduced, with permission, from:
Farnsworth, N. R. 1988. Screening plants for new medicines. Chapter 9 in Biodiversity, ed. E.O. Wilson. Washington, D.C.: National Academy Press.
SCREENING PLANTS FOR NEW MEDICINES
NORMAN R. FARNSWORTH
Research Professor of Pharmacognosy, Program for Collaborative Research in the Pharmaceutical Sciences, University of Illinois at Chicago, Chicago, Illinois
The U.S. pharmaceutical industry spent a record $4.1 billion on research and development in 1985, an increase of 11.6% from 1984 (Anonymous, 1986). In the same year, the American consumer purchased in excess of $8 billion in community pharmacies for prescriptions whose active constituents are still extracted from higher plants (Farnsworth and Soejarto, 1985). For the past 25 years, 25% of all prescriptions dispensed from community pharmacies in the United States contained active principles that are still extracted from higher plants, and this percentage has not varied more than 1.0% during that period (Farnsworth and Morris, 1976). Despite these data, not a single pharmaceutical firm in the United States currently has an active research program designed to discover new drugs from higher plants.
THE GLOBAL IMPORTANCE OF PLANT-DERIVED DRUGS
Approximately 119 pure chemical substances extracted from higher plants are used in medicine throughout the world (Farnsworth et al., 1985) (see Table 9-1[a], [b], [c], [d]). At least 46 of these drugs have never been used in the United States. For the most part, the discovery of the drugs stems from knowledge that their extracts are used to treat one or more diseases in humans. The more interesting of the extracts are then subjected to pharmacological and chemical tests to determine the nature of the active components. Therefore, it should be of interest to ascertain just how important plant drugs are throughout the world when used in the form of crude extracts. The World Health Organization estimates that 80% of the people in developing countries of the world rely on traditional medicine 1 for their primary health care needs, and about 85% of traditional medicine involves the use of plant extracts. This means that about 3.5 to 4 billion people in the world rely on plants as sources of drugs (Farnsworth et al., 1985). Specific data in support of these estimates are difficult to find, but the few examples that are available are quite revealing.
THE IMPORTANCE OF HERBAL DRUGS
In Hong Kong
In the small British colony Hong Kong (1981 population, 5,664,000), there were at least 346 independent herbalists and 1,477 herbal shops in 1981 (Kong, 1982); that same year, there were 3,362 registered physicians and 375 registered pharmacies. Chinese herbalist unions in Hong Kong claim to have a membership of about 5,000 (Kong, 1982). It is claimed that Hong Kong is the largest herbal market in the world, importing in excess of $190 million (US) per year (Kong, 1982). About 70% of these herbal products are used locally, and 30% are reexported. They fall into three roughly equal categories: ginseng products, crude plant drugs other than ginseng, and over-the-counter drugs and medicated wines (Kong, 1982). By comparison, about $80 million worth of Western-style medicines were imported into Hong Kong during the same period. Kong (1982) calculated that the average Hong Kong resident spends about $25 (US) per year for Chinese medicines.
The system of traditional medicine in Japan, known as Kampo, is an adaptation of Chinese traditional medicine. Kampo formulations are essentially multicomponent mixtures of natural products, primarily plant extracts. In 1976 more than 69 kinds of Kampo formulae were introduced into the National Insurance Scheme in Japan, and this number has doubled since that time. The total expenditure for all types of pharmaceutical products in Japan was approximately $8.3 billion (US) in 1976, whereas only about $12.5 million (US) was spent on Kampo medicines. Thus in that year, Kampo medicines in the Japanese health care system amounted to only about 0.15% of total pharmaceutical expenditures. In 1983, total pharmaceutical expenditures in Japan were valued at about $14.6 billion (US) and those for Kampo medicines increased to about $150 million (US). Hence, in 7 years, expenditures for Kampo medicines in the Japanese health care system increased to about 1% of total pharmaceutical expenditures (Terasawa, 1986).
In a survey of 4,000 Japanese clinicians conducted in 1983, 42.7% of the respondents reported that they used Kampo medicines in their daily practices. As with most systems of traditional medicine, the applications of Kampo are most successful in the treatment of chronic diseases, most of which are difficult to treat successfully with Western type medicine. Conditions for which traditional medicine is most frequently used include chronic hepatitis, climacteric disorders, common cold, bronchial asthma, high blood pressure, constipation, autonomic insufficiencies, allergic rhinitis, diabetes mellitus, gastritis, headache, and bowel dysfunction (Terasawa, 1986).
In the People's Republic of China
The People's Republic of China includes one-fourth of the world's population. In 1974 I was privileged to visit that country as a member of the Herbal Pharmacology Delegation--the third of nine scientific exchange delegations set up by former President Nixon when he first visited that country. Since then, I have returned to the PR(' in 1980 and again in 1985. It is obvious that the system of Chinese traditional medicine, in which the use of plant extracts to treat disease is extremely important, remains today as an important element in providing adequate primary health care for this populous country. Some of the value of Chinese medicine is most likely its use as a placebo, but I for one am convinced that the vast majority of plants used in this system have constituents that produce real therapeutic effects.
THE SEARCH FOR NEW PLANT DRUGS
There is a great deal of interest in and support for the search for new and useful drugs from higher plants in countries such as the People's Republic of China, Japan, India, and the Federal Republic of Germany. Virtually every country of the world is active in this search to a limited degree. However, in light of its size and resources, the United States must be regarded as an underdeveloped country with regard to productivity and programs designed to study higher plants as sources of new drugs, both in terms of industrial and university-sponsored research.
Estimates of the number of higher plants that have been described on the face of the Earth vary greatly--from about 250,000 to 150,000. How many of these have been studied as a source of new drugs? This is an impossible question to answer for the following reason. The National Cancer Institute in the United States has tested 35,000 species of higher plants for anticancer activity. Many of these have shown reproducible anticancer effects, and the active principles have been extracted from most of these and their structures determined. However, none of these new drugs have yet been found to be safe and effective enough to be used routinely in humans. The question then arises, could any of these 35,000 species of plants contain drugs effective for other disease states, such as arthritis, high blood pressure, acquired immune deficiency syndrome (AIDS), or heart trouble? Of course they could, hut they must be subjected to other appropriate tests to determine these effects. In reality, there are only a handful of plants that have been exhaustively studied for their potential value as a source of drugs, i.e., tested for several effects instead of just only one. Thus, it is safe to presume that the entire flora of the world has not been systemically studied to determine if its constituent species contain potentially useful drugs. This is a sad commentary when one considers that interest in plants as a source of drugs started at the beginning of the nineteenth century and that technology and science have grown dramatically since that time.
As shown in Table 9-1[a], [b], [c], [d], the 119 plant-derived drugs in use throughout the world today are obtained from less than 90 species of plants (Farnsworth et al., 1985). How many more can be reasonably predicted to occur in the more than 250,000 species of plants on Earth?
Use of the NAPRALERT Data Base
It is possible to present certain types of data showing the relative interest in studying natural products as a source of drugs by means of the NAPRALERT data base that we maintain at the University of Illinois at Chicago (Farnsworth et al. 1981, 1983; Loub et al., 1985). This specialized computer data base of information on natural products was derived from a systematic search of the world literature. Data that can be retrieved from the system include folkloric medicinal claims for plants, the chemical constituents contained in plants (and other living organisms), the pharmacological effects of naturally occurring substances, or the pharmacological effects of crude extracts prepared from plants. More than 80,000 articles have been entered into the data base since 1975, and about 6,000 new articles are added each year. The system contains folkloric, chemical, or pharmacological information on about 25,000 species of higher plants alone.
Pharmacological Interest in Natural Products
To give some idea as to the interest (or lack thereof) in studying the pharmacological effects of natural products, we can cite the following data from NAPRALERT. In 1985, approximately 3,500 new chemical structures from natural sources were reported. Of these, 2,618 were obtained from higher plants, 512 from lower plants (lichens, filamentous fungi, and bacteria), and 372 from other sources (marine organisms, protozoa, arthropods, and chordates) (Table 9-2). A significant 56.6% of the new chemicals obtained from lower plants (primarily antibiotics produced in industrial laboratories) were reported to have been tested for biological effects. About 23.9% of those obtained from marine sources, protozoa, arthropods, and chordates were studied for biological effects, but only 9.5% of the new structures obtained from higher plants were tested for pharmacological effects. The probable reasons for the low, 9.5% figure are that a majority of these discoveries were reported from university laboratories where the interest is mainly on chemistry, where there is less interdisciplinary research (i.e., botanists, chemists, and biologists working in collaboration), and where routine testing services for pharmacological activity are not readily available.
Why is there so little interest and activity in plant-derived drug development in the United States? An attempt will be made to answer this question, but first it is important to describe briefly some of the more fruitful approaches to drug discovery from higher plants.
Approaches to Drug Discovery from Plants
There are many approaches to the search for new biologically active principles in higher plants (Farnsworth and Loub, 1983). One can simply look for new chemical constituents and hope to find a biologist who is willing to test each substance with whatever pharmacological test is available. This is not considered to be a very valid approach. A second approach is simply to collect every readily available plant, prepare extracts, and test each extract for one or more types of pharmacological activity. This random collection, broad screening method is a reasonable approach that eventually should produce useful drugs, but it is contingent on the availability of adequate funding and appropriate predictable bioassay systems. The last major useful drugs to have reached the marketplace based on this approach are the so-called vinca alkaloids, vincristine sulfate (leurocristine) and vinblastine sulfate (vincaleukoblastine). Vincristine is the drug of choice for the treatment of childhood leukemia; vinblastine is a secondary drug for the treatment of Hodgkin's disease and other neoplasms.
Vincristine was discovered by Gordon H. Svoboda at the Lilly Research Laboratories. In January 1958, Svoboda submitted an extract of the Madagascan periwinkle plant [Catharanthus roseus (L.) G. Don] to a pharmacological screening program at Lilly (Farnsworth, 1982). This was the fortieth plant that he selected for inclusion in the program. Vincristine was marketed in the United States in 1963, less than 5 years after a crude extract of C. roseus was observed to have antitumor activity. In 1985, total domestic and international sales of vincristine (as Oncovin$) and vinblastine (as Velban$) were approximately $100 million, 88% of which was profit for the company (G. H. Svoboda, personal communication, 1986).
This discovery of new drugs from higher plants is one of the few that has evolved from a random-selection broad pharmacological screening program. For example, in the very expensive research and development effort undertaken by the National Cancer Institute described above, not one useful drug has emerged.
Recently we analyzed information on the l l9 known useful plant-derived drugs to determine how many were discovered because of medicinal folkloric information on the plants from which they were isolated. In other words, what correlation, if any, exists between the current medical use of the 119 drugs and the alleged medical uses of the plants from which they were derived? As shown in Table 9-1[a], [b], [c], [d], 74% of the 119 chemical compounds used as drugs have the same or related use as the plants from which they were derived. This does not mean that 74% of all medical claims for plants are valid, but it surely points out that there is a significance to medicinal folklore that was not previously documented.
Thus, in my opinion, future programs of drug development from higher plants should include a careful evaluation of historical as well as current claims of the effectiveness of plants as drugs from alien cultures. Such information is rapidly disappearing as our own culture and ideas permeate the less developed countries of the world where there remains a heavy dependence on plants as sources of drugs.
LACK OF INTEREST IN NEW DRUG DISCOVERY PROGRAMS FROM PLANTS
Why is there such a reluctance to initiate new programs involving plants as sources of drugs in the United States, where we have the most sophisticated pharmaceutical industry in the world and where expenditures for drug development are staggering? In my conversations with staff from U. S. pharmaceutical companies, the following reasons seem to be consistent:
What really seems to be the problem is that most pharmaceutical firms, as well as decision-making offices in government agencies, lack personnel who have a full understanding and appreciation of the potential payoff in this area of research. For example, new programs in drug development are usually initiated by the presentation of a proposal by a research staff member before a group of peers and research administrators. Following is one possible scenario: Dr. E. Z. Greenleaf prepares his arguments for a new drug development program at the ABC Pharmaceutical Corporation in which he proposes to study plants as a source of new drugs. His approach to the program is to examine written medicinal folklore to obtain information on plants allegedly used by primitive peoples for certain specified diseases. He might even be brave enough to suggest that the ABC Pharmaceutical Corporation hire one or two physicians to travel to Africa, Borneo, New Caledonia, or other exotic areas to live with the people for a year or so. During this period, Drs. U. Canduit and 1. M. Reliant would observe the witch doctors treating patients and then would make their own diagnoses of each patient and conduct follow-up observations on outcome. When improvement is noted, they would record which plants had been used to treat the patients. These plants would then be collected and sent to the Research Laboratory of the ABC Pharmaceutical Corporation located in Heartbreak, Colorado, for scientific studies. Total cost of such a 5-year program would be less than the cost of a new jet fighter.
- To recover the costs of developing such drugs, solid patent protection must be secured. It is generally believed that natural products cannot be patented with the same degree of assurance as can synthetic compounds. This of course cannot be a valid deterrent, since patent protection for vincristine and vinblastine was sufficiently secure that the Eli Lilly Company had exclusive marketing rights to these substances for the full term of patent protection.
- Most promising plants seem to be indigenous to developing countries, many of which do not have stable governments and thus cannot provide assurance that there will be a continued supply of the raw material needed to produce the useful drugs. This of course may be true in a strict sense; however, as history shows, it is rare when a useful plant grows only in one isolated developing country. In the course of developing a full program involving plants as a source of raw material, it would be normal logic to immediately seek sources of the useful plant from a large number of geographic areas. Cultivation programs should also be initiated. In the early stages of development of vincristine and vinblastine, the plant source C. roseus was collected from many different countries of the world and was also cultivated in eastern European countries and in the United States.
- There is reputed to be biological variation from lot to lot of plant drugs, but scientific documentation for this statement is difficult to find. This does not appear to be a problem affecting any of the plant sources required for production of the 119 drugs listed in Table 9-1[a], [b], [c], [d].
The second scientist from the ABC Pharmaceutical Corporation to make a new program presentation is Dr. Adam N. Molecule. He uses a long sequence of chemical equations to illustrate his theory that he can synthesize a series of chemical analogs based on computer analysis of structure-activity relationships in which his theoretical compounds will react favorably with specific receptor sites. He illustrates his plan with a full color videotape presentation of the computerized sequence of events that he hopes will take place at the molecular level. There is nothing left to the imagination. Molecule's computer produces a flowchart projecting the full costs of each stage of the synthesis at 2-month intervals. Everything is predictable, based on a percentage of projected sales should the end product prove to be a useful drug, and ensuring at least a 75% profit margin.
At the end of the two presentations, management must decide on whether to follow the folkloric line of Dr. E. Z. Greenleaf or the molecular biology-computer graphic-theoretical approach of Dr. Adam N. Molecule. Since Dr. Greenleaf is probably the only person in the room with a background and appreciation for his approach and most of the scientists in attendance are well trained and highly skilled synthetic chemists, biochemists, and molecular biologists, it is not difficult to predict which program will he approved and implemented.
Higher plants have been described as chemical factories that are capable of synthesizing unlimited numbers of highly complex and unusual chemical substances whose structures could escape the imagination of synthetic chemists forever. Considering that many of these unique gene sources may be lost forever through extinction and that plants have a great potential for producing new drugs of great benefit to mankind, some action should be taken to reverse the current apathy in the United States with respect to this potential.
1 Traditional medicine is a term loosely used to describe ancient and culture-bound health practices that existed before the application of science to health matters in official, modern, scientific medicine or allopathy.
Anonymous 1986. Pharmaceutical R&D Spending by US Industry Hits $4.1 Billion, Setting Record as do Sales. P. 5 in Chem. Mark. Rep. February 3, 1986.
Farnsworth, N. R. 1982. Rational approaches applicable to the search for and discovery of new drugs from plants. Pp. 27-59 in Memorias del 1er Symposium Latinoamericano y del Caribe de Farmacos Naturales, La Habana, Cuba, 21 al 28 de Junion, 1980. Academia de Ciencias de Cuba y Comisin Nacional de Cuba ante la UNESCO, UNESCO Regional Office, Montevideo, Uruguay.
Farnsworth, N. R., and W. D. Loub. 1983. Information gathering and data hases that are pertinent to the development of plant-derived drugs. Pp. 178-195 in Plants: The Potentials for Extracting Protein, Medicines, and Other Useful Chemicals. Workshop Proceedings. OTA-BP-F-23. U.S. Congress, Office of Technology Assessment, Washington, D.C.
Farnsworth, N. R., and R. W. Morris. 1976. Higher plants--the sleeping giant of drug development. Am. J. Pharm. Educ. 148(Mar.-Apr.):46-52.
Farnsworth, N. R., and J. M. Pezzuto. 1983. Rational approaches to the development of plant-derived drugs. Pp. 35-63 in Proceedings of thc Second National Symposium on the Pharmacology and Chemistry of Natural Products, Joao Pessoa, Brazil, November 3-5, 1983. Paraiba University, Joao Pessoa, Brazil.
Farnsworth, N. R., and D. D. Soejarto. 1985. Potential consequences of plant extinction in the United States on the current and future availability of prescription drugs. Econ. Bot. 39(3):231-240.
Farnsworth, N. R., W. D. Loub, D.D. Soejarto, G. A. Cordell, M. L. Quinn, and K. Mulholland. 1981. Computer services for research on plants for fertility regulation. Korean J. Pharmacogn. 12:98-110.
Famsworth, N. R., O. Akerele, A. S. Bingel, D. D. Soejarto, and Z.-G. Guo. 1985. Medicinal plants in therapy. Bull. WHO 63:965-981.
Kong, Y.-C. 1982. The control of Chinese medicines--a scientific overview. Yearb. Pharm. Soc. Hong Kong 1982:47-51.
Loub, W. D., N. R. Farnsworth, D. D. Soejarto, and M. L. Quinn. 1985. NAPRALERT: Computer handling of natural product research data. J. Chem. Inf. Comput. Sci. 25:99-103.
Terasawa, K. 1986. The present situation of education and research work on Traditional Chinese Medicine in Japan. Presentation at the International Symposium on Integration of Traditional and Modern Medicine, Taichung, Republic of China, May 22, 1986.