Methods of drug development

Austel and Kutter (1980) suggest that there are three methods for obtaining lead compounds or chemical structures for drug development:

by reference to compounds and structures that are known to demonstrate activity relative to the disease target, by random screening of compounds and structures - the 'empirical' approach, and by utilizing knowledge of the biomolecular processes that play an important role in disease - 'rational' drug design.

Two major sources of leads that are based on known activity are the use of naturally-occurring compounds that regulate biological functions within organisms and the improvement of existing medicinal preparations. Whereas Austel and Kutter (1980) consider these to be efficient routes to drug discovery, their advantage relative to other techniques depends on the extent to which the existing applications of these compounds (naturally or by intervention) and preparations are already optimized.

The exploration of plants with traditional medicinal uses falls under this initial method of developing new chemical leads. As traditional uses often rely on rough formulations of crude extracts, they leave considerable room for optimization with the tools employed in modern pharmaceutical R&D. By isolating the active compound, and conducting research on formulation and toxicology, herbal preparations may eventually be transformed into valuable pharmaceuticals. Modification of natural compounds may lead to semi-synthetic analogues of plant-based compounds that offer further improvements in drug efficacy or safety.

All medicinal preparations, from a historical vantage point, used to be derived from plants. With the advent of the search for 'magic bullets' in the late nineteenth and early twentieth century, the isolation and optimization of active chemical principles from plants became the dominant mode of drug discovery. Over the course of the twentieth century exploration of medicinal plants has no doubt resulted in the discovery of many important pharmaceuticals. N. Farnsworth (unpublished data) suggests that approximately three-quarters of the 121 useful prescription drugs developed from plants were discovered following their earlier use in indigenous medicine, that is with the aid of ethnobotanical knowledge about the plants' potential uses.

The 'empirical' approach to drug development offers a different approach to the exploration of the chemical diversity inherent in plants and other natural products. Tests for efficacy against a particular disease target are devised which will enable the researcher to empirically evaluate the potential efficacy of a range of available compounds. A knowledge of the underlying biomolecular processes that cause a disease is not necessarily required to make a discovery with this 'trial and error' approach. In contrast, 'rational' drug design originates from scientific investigation of the underlying basis of a given disease. The presumption of this approach is that along with an understanding of the mechanisms involved in disease comes the ability to actually 'design' a compound that will alter the course of disease.

In contrasting the potential effectiveness of the rational or empirical approach with the pursuit of an ethnobotanical approach it is useful to consider how the stocks of important inputs relevant to each approach to the R&D process, such as knowledge, technology and biodiversity, are likely to change over time. Looking at the stock of ethnobotanical knowledge there are a number of observations that can be made. First, the stock of ethnobotanical information appears to be very much a non-renewable resource, that is, the rate of 'recruitment', the generation of new knowledge, is likely to be very low relative to the rate at which it has been 'consumed' by eager pharmacognists over the past century. The low rate of regeneration of knowledge is exacerbated by the intrusion of 'modern' medicine and modern lifestyles into indigenous cultures. This diverts demand away from traditional products towards non-traditional remedies, but more importantly it reduces the number of potential 'healers' in the next generation, thus further retarding the rate of development of new ethnobotanical knowledge. Finally, development pressure and changing lifestyles lead to the complete loss of existing information as healers die without transferring their knowledge to a new generation. This 'mortality' rate of the stock of ethnobotanical information is particularly acute in indigenous societies in the developing world (Plotkin, 1988).

In comparison, both 'rational' and 'empirical' approaches tend to exhibit 'renewable' characteristics. Both approaches are capable of rapidly, at least in comparison with ethnobotanical information, generating new active principles that target novel diseases (e.g. AIDS), disease agents that become resistant to existing treatments (e.g. malaria), or diseases for which treatments have never been well optimized.

In the case of the random screening of natural products, Aylward and Barbier (1992) argue that the sheer scale of the resource, in the order of 10 to 100 million species, and the continuing evolution of new screens and new disease targets implies that biodiversity will never be fully explored for its pharmaceutical potential. This argument clearly applies to the plant kingdom which is estimated to comprise some 250-500000 species. Although, Farnsworth and Morris (1976) agree that it is practically impossible to actually determine when a particular species has been fully investigated for its pharmaceutical properties, they make a 'guesstimate' that just 5000 plant species have been exhaustively explored. Also, note that the 'empirical' approach opens up the possibility of investigating the full range of biodiversity and not merely the plant kingdom.

In 'rational' drug design synthetic compounds are derived based on an understanding of the mechanisms of disease action. This approach relies on a stock of knowledge in the fields of medicine, chemistry molecular biology, etc. The acquisition of knowledge and technology (embodied knowledge) in these fields has grown rapidly over the past century; some might even argue exponentially in recent years. It should come as no surprise then that the pharmaceutical (and biotechnology) industry relies heavily on 'rational' design.

As a consequence, the rational and empirical approaches to drug development do not face the non-renewability constraint as does the ethnobotanical approach. It would, nevertheless, be premature to argue that there are not important unexplored ethnobotanical leads available to science. As the pharmaceutical industry experiences a renewed wave of interest in exploring natural products, ethnobotanical investigation will no doubt play a role. Eli Lilly's recent equity investment of $4 million in Shaman Pharmaceuticals, a California-based company that exclusively screens ethnobotanical leads, indicates that the industry cannot afford to ignore such sources of leads as the existing drug pipelines dry up.

It must be recognized, nonetheless, that over the past century a good number of traditional uses of medicinal plants have been investigated. If, as seems likely, the most promising ethnobotanical sources are explored first, then as more and more ethnobotanical leads are investigated not only will the absolute number of potential pharmaceutical leads be reduced, but with each additional discovery the chances of a further discovery will diminish. Thus, although an efficient, initial strategy in the development of pharmaceuticals would be to explore and optimize existing medicinal uses of plants the relative attractiveness of such a strategy would be expected to decline over time.

What about the 'empirical' and 'rational' approaches? Which is best? The empirical and rational approaches clearly stem from different preconceptions about the best way to go about developing new drugs but the comparison may often be overstated by advocates of either paradigm. The portrayal of these two approaches as mutually exclusive, by competing factions within the research community, often spills into the popular debate. Witness the following titles of articles from reputable periodicals heralding the triumphal ascent of rational drug design: 'The Reign of Trial and Error Draws to a Close' and 'The Drug Industry Moves from Discovery to Design' (Waldrop, 1990; The Economist, 1984).

Absolute fidelity to one or another of these research paradigms may be of great importance within the research community. To consumers in the health care marketplace, however, it is of little importance whether drug development occurs through trial and error or rational design. What matters to the patient and prescribing physician is access to safe, efficacious and low cost medicines. Thus, the casual observer may be perplexed by the degree of scientific arrogance embodied in statements such as:

Gone are the days when a fortune could be made by patiently sifting a lorry load of soil. Pharmaceutical research now has to be rational, and that means science-based.

(Lancet, 1981 as quoted in Gross, 1983)

The impressions left by such statements may obscure the fact that the rational and empirical approaches to drug design are ideal types. A parallel exists in the ideologies expoused by economists: free-markets versus state planning. In reality, only mixed economies exist. Scientific 'values' or biases may lead individuals or institutions to argue that one or the other approach is better in theory but reality may dictate the retention and application of both approaches.

Developing screens against which compounds can be tested is likely to involve some knowledge regarding the cause of the disease; likewise, the evaluation of screening results may lend itself to improving knowledge about the disease which in turn enables improved screening methods. From the other end of the spectrum, the 'rational' design of molecules still requires testing, whether in vitro or in vivo, before undertaking additional testing in the clinic. The 'French' abortion pill RU486 is a synthetic molecule that was developed with a particular type of activity in mind; however, its name hides the fact that RU486 was no less than the 35 486th such molecule screened for the desired activity. 'Rational' does not mean perfect. As a result, there is likely to be an element of trial and error even in the most 'rational' approach to drug design.

A truly reasonable as opposed to 'rational' approach to drug development might recognize that scientific ideologies aside, what really matters is developing safe and cost-effective pharmaceuticals utilizing a judicious application of each approach. Cost-effectiveness will depend on many factors that change over time, including available technological, biological and human resources and it would therefore seem premature to bury the empirical approach purely on the basis of scientific prejudice that it is not 'scientific'. Similarly, whether a successful compound is derived in the chemist's laboratory or in nature's laboratory should not be significant in and of itself. In the next section a number of the factors impinging on the exploration of plants and other natural products are explored with this view in mind.

Plants and other natural products as chemical leads

Lead compounds from nature that are of interest to the pharmaceutical industry are drawn from the secondary metabolites produced by living organisms. Primary metabolites are the principal chemical constituents, such as amino acids, that are common to all living organisms. Secondary metabolites are more complex compounds that are generally common only to a particular family, genus or even species (Balandrin et al., 1985). As secondary metabolites of different types are present in all organisms, the full range of biodiversity has the potential for yielding new compounds of medicinal interest. A discussion of the use of natural products in the development of new pharmaceuticals is often largely associated with the use of plants, particularly those found in tropical rainforests. This is in part a result of the popular appeal of the argument for conservation that is based on preserving potential cures for AIDS, cancer and other life-threatening and debilitating diseases; however, beneath the gloss of multimedia efforts such as the movie Medicine Man there does lie a substantial argument for associating the derivation of pharmaceuticals with plants.

In the nineteenth century, before the advent of the 'pharmaceutical' industry, all medicinal preparations were derived directly from nature, mostly from plants. Since then, however, the pursuit of plant-based pharmaceutical applications has been of a cyclical nature (Findeisen and Laird, 1991). The development of synthetic chemistry led early drug researchers to abandon plants in the search for 'magic bullets'. More recently, methods for 'rational' drug design have led researchers back to the laboratory once more.

The rise and fall of research into the medicinal properties of plants in the mid- to late twentieth century is symbolized by the experience of an American drug firm, Eli Lilly. In the 1960s, Eli Lilly and Co. developed the anti-cancer drugs vinblastine and vincristine from the rosy periwinkle, Catharanthus roseus. These 'miracle' drugs for leukemia were produced as a result of a program screening ethnobotanical leads. Shortly thereafter, Lilly terminated its plant screening program. In late 1992, Lilly evidently had a change of heart regarding the potential of phytochemistry by striking a deal with Shaman Pharmaceuticals. Shaman conducts extensive investigations of ethnobotanical leads before bringing promising plants to its laboratories in California for screening, identification and isolation of active compounds. In return for a $4 million equity investment, Eli Lilly will assist Shaman in developing potential anti-fungal agents over the next 4 years.

As a result of the lack of interest by major players in the industry during the 1970s and 1980s, major new drugs developed from plant sources in the past 20-30 years are limited in number. The only major plant (and marine organism) screening program during this period took place at the instigation of the publicly funded US National Cancer Institute. The NCI program and the development of taxol and camptothecin, two novel agents discovered by the NCI program, are chronicled in the next section. In the past few years new technological developments, including advances in high throughput screening techniques, have rekindled the interest of pharmaceutical companies in the exploration of plants for novel chemical compounds. In addition, popular concern over the richest source of plant chemical diversity, the tropical rainforests, may have added to the impetus to explore the potential of plants. Plants, however, are not alone in providing a potential cornucopia of active chemical agents. Despite its neglect in the popular press microbial diversity has long been an important and well-known source of lead compounds for the pharmaceutical industry. Marine diversity has also been targeted by drug researchers although its potential remains largely unexplored. Finally, a large reservoir of species diversity, the insects, is almost completely unexploited.

Microbial diversity is a rich source of natural products chemistry, a source which has seen a consistent level of exploitation by industry's R&D departments since World War II. Penicillin, a host of subsequent antibiotics and many other products have been produced from microbial sources. Mevacor, a breakthrough cholesterol-lowering drug with sales of over $100 million in 1991, is just one of four products recently developed by the microbial screening program at Merck (Merck & Co., 1992). While the extent of microbial diversity is largely unknown, microbial diversity may equal or exceed that of all other diversity: currently expected to be in the order of from 10 to 100 million species. Recent improvements in screening technologies complement the 'chemical inventiveness' of microorganisms in generating many new leads for drug development (Nisbet, 1992).

The collection of microbial samples is accomplished by sampling different microenvironments such as soils, detritus, etc. A simple scoop of material may yield thousands of different species. Once collected there is little reason to return to the original site as microorganisms can normally be cultured through fermentation processes. Freedom from the difficulties of resupply is probably an important factor in explaining the sustained level of interest by the pharmaceutical industry in fermentation products relative to other natural products. The need to return for additional collections of plants or marine organisms as research continues and as full-scale production becomes necessary can be a costly, risky and frustrating experience for industry. Making use of microbial diversity avoids these complications and keeps the research and development process in-house.

The full diversity of marine organisms is largely unknown. Current species estimates imply that only 20% of species are of marine origin; however, this may reflect a bias towards terrestial research in systematics. Grassle (personal communication in Ray, 1988) suggests that deep sea fauna may rival tropical forests in species diversity. Rinehart (1992) reports that marine macroorganisms and microorganisms produce a 'dizzying array' of secondary metabolites. The hunt for drugs at sea is, however, constrained by the relatively large cost of obtaining specimens, the lack of techniques for ex situ reproduction of marine invertebrates and the difficulty of locating the source of activity. Many of the compounds isolated from macrospecies actually are produced by microorganisms.

The NCI incorporated marine organisms into its early program, screening approximately 16000 extracts made from 3000 different species during the 1972-80 period. The collection of marine samples for screening resumed in 1986. In addition, Mallinckrodt and Laird (1992) cite a number of pharmaceutical companies that have initiated marine screening programs, including Bristol Myers Squibb, Merck & Co., Rhone Poulenc Rorer and SmithKline Beecham.

Arthropod or insect diversity is a new and unexplored area in natural products chemistry. Eisner (1990) suggests that arthropods may hold considerable amounts of material of interest to medicinal chemists. As a source of species diversity, arthropods far outstrip their terrestrial plant and animal counterparts. Wilson (1988) reports that arthropods make up just over one-half of the 1.4 million species described to date. In its recent arrangement with Costa Rica's National Biodiversity Institute, Merck & Co. has agreed to pay $1 million for the right to investigate the chemical properties of not only plant and microbial species, but those of insects as well.

In summary, it is important to understand that plants are just one of the potential ways in which drug screening programs can capitalize on the chemical diversity inherent in ecological and biological diversity.

Not only are synthetic compounds and natural products potential substitutes, and hence competitors for R&D investment, but within the category of natural products the potential exists for substitution and competition. In the past few decades the exploration of plants has taken a back seat to that of microbial diversity. The relative merits of each major group (in terms of cost and availability of initial supply, cost and reliability of resupply, and industry perceptions about the chances of success and the degree of appropriability of returns) will determine the direction and magnitude of industry's R&D investment over time.

These strategic concerns may be less important in the case of publicly funded research. The US NCI was the last large natural products program to shut down and the first to reopen its program in the latest up and down cycle of interest in natural products research. We, therefore, turn to a consideration of their past successes and failures with plants and the status of their current efforts in this area.

Was this article helpful?

0 0
100 Health Tips

100 Health Tips

Breakfast is the most vital meal. It should not be missed in order to refuel your body from functional metabolic changes during long hours of sleep. It is best to include carbohydrates, fats and proteins for an ideal nutrition such as combinations of fresh fruits, bread toast and breakfast cereals with milk. Learn even more tips like these within this health tips guide.

Get My Free Ebook


  • Quinzio
    How can i develop drug from traditional medicinal plants?
    2 years ago
  • Caradoc
    How to formulate a drug from plants?
    2 years ago

Post a comment