Magazine
Sign up for your free subscription and keep up-to-date.
Upcoming Events
Newsletters
Stay updated on the latest news and technologies with Bioscienceworld's newsletters.
Five to choose from.
Selection of the Best Candidate for Development
By Elizabeth B. Vadas, PhD
The Holy Grail of every drug-discovery group is to find the perfect molecule that can fulfill all requirements of specificity, potency, safety and efficacy, resulting in a major breakthrough in targeting a specific disease. While achieving this goal is indeed the major motivation that brings medicinal chemists, biologists, biochemists and other discovery scientists to the laboratory day after day, it must be recognized that finding the “perfect” molecule may be an elusive goal.
Beyond having the above mentioned characteristics, the molecule must fulfill a number of other requirements before it can be declared a development candidate. These requirements include the appropriate biopharmaceutical and pharmacokinetic properties, stability in the solid state and in solution — alone and in the presence of excipients — and the potential for a commercially viable synthetic route, just to name a few.
Upon examination of the different requirements corresponding to the needs of each of the scientific disciplines that play a role in developing a molecule into a therapeutic entity, it becomes clear that the best development candidate is unlikely to be the ideal molecule as envisaged by any of the disciplines. Often the best development candidate represents a best compromise. Using a team approach, discovery and development scientists must jointly evaluate the risks inherent in the molecule, decide what imperfections of the molecule may be tolerated and design a strategy to minimize risks during the development process.
In this context it is interesting to note why compounds fail in preclinical and clinical development. Approximately 30% of molecules fail due to lack of efficacy. Eleven per cent fail due to findings in animal toxicology and 10% of molecules are discontinued because of adverse experiences in humans. Commercial reasons are cited in 5% of the cases. Remarkably, approximately 39% of molecules fail due to “poor drug-like” properties.1 The major drug-like properties are classified as biopharmaceutical, stability and manufacturing related properties. Table 1 lists some of the major concerns to which attention must be paid during the compound selection activities in order to avoid selecting a molecule with poor drug-like properties.
Table 1. Areas of major concern regarding molecule selection.
The primary clinical concerns are safety, tolerability and efficacy. Also important considerations for the clinical scientists are potency/dose, pharmacokinetics, metabolism, drug interactions and potential side-effects. The toxicologist is concerned with the appropriate selection of the toxicology species, generally based on similarity of metabolism to man as assessed by liver microsomal studies. The dose range to be covered in order to provide adequate exposure and safety margins is of major concern, as is the means by which the doses will be administered. Bioavailability, PK and PD characteristics must be considered. The toxicologist’s goal is to demonstrate a clear “no-effect” level, as part of a well-defined toxicity profile that is relevant to man.
The bulk drug concerns focus on the synthetic route, its complexity, the yield, potential scale-up issues, impurities, consistency and commercial viability. Chemical stability and solubility are of major concern, along with the physicochemical properties of the molecule such as crystallinity and polymorphism. The dosage form concerns include route of administration, dose range and dose strengths, crystallinity, solubility, chemical and physical stability, biopharmaceutical properties, the analytical methods that will need to be developed and the processes by which the formulations will be made.
Figure 1 illustrates the complex relationships among some of the properties of a new chemical entity one must consider during the candidate selection process.
The above concerns are but a sampling of the requirements a potential development candidate must fulfill. The concerns often overlap among different disciplines but they may be the cause for conflict also. It is imperative that clear expectations are set from the very beginning, and a consensus is established regarding the following questions:
How long will it take to discover the “perfect” compound?
What imperfections can be tolerated?
What is the cost of accommodating the imperfections?
As an illustration of the potential approaches to take when selecting a development candidate, one can use the example of the biopharmaceutical properties of a hypothetical compound slated for oral drug delivery. Let us assume that the molecule has very low aqueous solubility and good permeability; that is, it is a biopharmaceutical class 2 molecule.2 If the expected dose range falls within reasonable limits — for example, approximately 10 to 300 mg per unit dose — the low aqueous solubility may be accommodated by a number of different formulation approaches or by crystal engineering. Possible approaches include particle size reduction, incorporation of wetting agents in the dosage form, use of solubilizing agents or self-emulsifying systems, complexation or crystal modifications. None of these approaches is trivial — each has its own liabilities, or may result in licensing in third-party technology at an added cost. Should the molecule be rejected as a viable development candidate and the search continued for a better substance? It is very likely that employing one of these approaches will have a positive outcome and will result in a faster development process than the continued search for a molecule with significantly better aqueous solubility while retaining all other desirable properties.
Let us assume that the molecule has reasonable aqueous solubility but poor permeability; in other words, it is a class 3 compound in the biopharmaceutical classification system. This “imperfection” represents a more severe challenge to the toxicologist, the formulation scientists and the clinicians.
Nevertheless, it may be a difficulty that could be overcome by a limited number of formulation approaches such as the use of lipid-derived excipients. The options here are much more limited and their utility will depend on a number of other factors, such as the potency and physicochemical properties of the molecule and the desired dose strength(s). Nonetheless, before abandoning such a molecule, it is important to assess the totality and complexity of its properties by an interdisciplinary team composed of both discovery and development scientists.
An interesting case is the hypothetical molecule having both poor aqueous solubility and poor membrane permeability; that is, a biopharmaceutical class 4 compound. Should this molecule be rejected outright? The answer is likely to be yes, but in rare instances even such a highly imperfect molecule may be considered.
Selection of a development candidate does not take place in a vacuum. The compromises researchers make and the imperfections and associated risks they may be prepared to tolerate often will depend on the therapeutic area and on the maturity of the research program within that therapeutic area. The discovery effort may be aimed at a new therapeutic area of either a yet-to-be validated target or an unproven mechanism of action in a disease area where treatment options are non-existent or very limited. In this case, even an imperfect molecule may represent a major medical breakthrough and choosing it may be justified. In contrast, one may be working in an area where the therapeutic target is validated and a number of treatment options are available. In this instance, incremental improvements in therapy are sought and a second- or third-generation molecule must have few, if any, imperfections if it is to be selected as a development candidate and if it is to be therapeutically and commercially viable.
References:
1) Kennedy, Tony. “Managing the drug discovery/development interface.” Drug Discovery Today 2.10 (1997): 436-444.
2) Loebenberg, Raimer and Gordon Amidon. “Modern bioavailability, bioequivalence and biopharmaceutics classification system. New scientific approaches to international regulatory standards.” European J. of Pharmaceutics and Biopharmaceutics 50 (2000): 3-12.
Elizabeth Vadas, PhD has more than 25 years of drug-development expertise. She was executive director of Pharmaceutical Research and Development at Merck Frosst Canada Ltd. (Kirkland, QC) with responsibilities for product development for all new chemical entities discovered in the Montreal, Que. research centre. In mid-2002, Vadas took early retirement to establish her own consulting company, InSciTech Inc. (Dorval, QC). Currently, she works with several drug-discovery companies, both in the U.S. and Canada, providing development expertise. Vadas obtained her PhD in physical chemistry at McGill University (Montreal, QC). She is a fellow of the American Association of Pharmaceutical Scientists (Arlington, VA) and past president of the Canadian Society for Pharmaceutical Sciences (Edmonton, AB). She is the recipient of a number of scientific and management awards.