Metabolism, Pharmacokinetics and Toxicity of Functional Groups

Impact of Chemical Building Blocks on ADMET

By Dennis A. Smith

The Royal Society of Chemistry

Copyright © 2010 Royal Society of Chemistry
All rights reserved.
ISBN: 978-1-84973-016-7

Contents

Chapter 1 Drugs and their Structural Motifs Alexander A. Alex and R. Ian Storer, 1,
Chapter 2 ADMET for the Medicinal Chemist K. Beaumont, S. M. Cole, K. Gibson and J. R. Gosset, 61,
Chapter 3 Carboxylic Acids and their Bioisosteres Amit S. Kalgutkar and J. Scott Daniels, 99,
Chapter 5 Sulfonamide as an Essential Functional Group in Drug Design Amit S. Kalgutkar Rhys Jones and Aarti Sawant, 210,
Chapter 6 Influence of Aromatic Rings on ADME Properties of Drugs Deepak Dalvie, Sajiv Nair, Ping Kang and Cho-Ming Loi, 275,
Chapter 7 Influence of Heteroaromatic Rings on ADME Properties of Drugs Deepak Dalvie, Ping Kang, Cho-Ming Loi, Lance Goulet and Sajiv Nair, 328,
Chapter 8 Peptidomimetics and Peptides as Drugs: Motifs Incorporated to Enhance Drug Characteristics Tracey Boyden, Mark Niosi and Alfin Vaz, 370,
Chapter 9 Pharmacokinetics and Metabolism of Compounds that Mimic Enzyme Transition States Iain Gardner, Chris Barber, Martin Howard, Aarti Sawant and Kenny Watson, 390,
Chapter 10 Alcohols and Phenols: Absorption, Distribution, Metabolism and Excretion Zhuang Miao and R. Scott Obach, 460,
Chapter 11 Future Targets and Chemistry and ADME Needs Dennis A. Smith and David S. Millan, 486,
Subject Index, 512,

CHAPTER 1

Drugs and their Structural Motifs

ALEXANDER A. ALEX AND R. IAN STORER

Pfizer Global Research and Development, Ramsgate Road, Sandwich, Kent, UK, CT13 9NJ

1.1 Introduction

The major focus of the research-based pharmaceutical industry is the discovery of safe, efficacious, new chemical entities (NCEs) for therapeutic targets. The pharmaceutical industry can look back at a history of successful innovations, indicated by the fact that there are currently just over 1400 unique drugs on the market.

The success of the industry can be measured in, for example, the increase in life expectancy in men and women over the last four decades. For instance, a child born in the United States in 2005 can expect to live nearly 78 years (77.9 years). The increase in life expectancy represents a continuation of a long-running trend. Life expectancy has increased from 75.8 years in 1995 and from 69.6 years in 1955. (www.cdc.gov/nchs/pressroom/07newsreleases/ lifeexpectancy.htm). Although there are multiple factors which potentially contribute to the increase in life expectancy, like for example diet and life style, the development and availability of new drugs appear to have made a substantial contribution.

Equally impressive, the impact of the industry can also be highlighted by the increase in five-year-survival rates for cancer when diagnosed 1975–1977 compared to when diagnosed in 2000 (www.phrma.org/files/PhRMA% -202009%20Profile%20FINAL.pdf). Between 1975 and 1979, the five-year survival rate for cancer was just 50%; by 2000, survival had risen to 67%. Survival is increasing dramatically for many forms of cancer. The rate of five-year survival went up 21% for breast cancer, 42% for prostate cancer, 28% for colon and rectum cancer, and 25% for lung and bronchial cancer.

Drug discovery is a complex multivariate process, but the basic requirements for orally administered NCEs include novelty and patentability, intrinsic potency, oral bioavailability, no toxicological effects in humans, and a significant advantage over existing accepted therapies (if applicable). A schematic representation of the drug discovery process in the United States is shown in Figure 1.1.

Although it is possible to predict, with varying accuracy, what a NCE will do when orally administered to humans, the full potential of a NCE is not known until it has been tested in clinical trials. Therefore, any investment made will not yield any return until the NCE is on the market, which could be in the region of ten years after patenting, and for the majority of compounds there will be no return at all to offset the enormous costs of drug discovery and development. Therefore, drug discovery is a high risk business with massive, long-term up-front investments aiming at discovering the few blockbusters that are on the market at any one time. In addition, the pharmaceutical industry is one of the most research-intensive industries; in the United States, an average of 16% of sales is spent on R&D, second only to the aerospace industry (www.nsf.gov/ statistics). The global pharmaceutical market is worth $553.4 billion in the top ten markets alone (Table 1.1).

The top ten marketed drugs and their revenue between June 2007 and June 2008 are shown in Table 1.2; they account for a total of $67.4 billion, which is only 12.2% of total sales in the top ten markets.

Among the top ten drugs, Pfizer’s Lipitor is by far the biggest seller, $5.5 billion ahead of a cohort of three drugs, Plavix, Nexium and Serentide with sales of $8.3, $7.7 and $7.5 billion, respectively. The top ten therapies are shown in Table 1.3 and account for 36.5% of global sales. The annual sales figures indicate that oncologics are by far the biggest revenue stream for the pharmaceutical industry, followed by lipid regulators. Interestingly, Pfizer’s Lipitor alone accounts for almost half of lipid regulator sales.

Historically, big pharmaceutical companies delivered. The secret of their success was simple: pharmaceutical companies brought a huge number of innovative products to the market that genuinely helped sick people, and so were readily prescribed, which generated solid sales. Even during times of economic hardship, drugs continued to be an essential purchase. During this flourishing period from the mid-1980s to the beginning of this decade, major drug companies routinely generated double-digit growth in sales year after year.

However, the pharmaceutical industry’s investment in R&D has also risen steeply over the last 20 years, with R&D spending of $47.9 billion in 2007 compared with $26 billion in 2000 and $8.4 billion in 1990, and an average cost of $1.3 billion for bringing a new drug to market — an increase of 65% since 2000 (www.phrma.org/files/PhRMA%202009%20Profile%20FINAL.pdf). Despite this increased investment in research and development, the number of new molecular entities (NMEs) has not increased in line with rising investment; in fact it declined between 1996 to 2008 from 54 to 21 (Figure 1.2).

Figure 1.2 could suggest that there has been a significant decline in innovation rates in the pharmaceutical industry over the last decade. The reasons for this decline have been reviewed extensively and several causes have been indicated as contributors to the R&D decline. Among these are for example submaximal optimisation of resources and the inability to control costs as well as negative impact of mergers and acquisitions, which have been grouped together as factors internal to R&D. Alongside these, external reasons for the decline include evolving healthcare, regulatory burden, lack of regulatory harmonisation as well as changes in tolerance for risk. Looking back over recent decades, total approvals by the US Food and Drug Administration (FDA) reached a record high of 381 entities in the decade between 1995 and 2004 compared with the two previous decades (241 in the decade 1985–1994 and 190 in the decade 1975–1984). Thus, it would appear that a myopic focus on near-term performance has given rise to a perception that bears very little relationship to the actual innovation rates of the pharmaceutical industry in the last decade.

However, the issue of high attrition rates in drug discovery and development still remains, without which the innovation rates would be even higher and, potentially, would keep better track with the enormous increases in R&D investment. Only about 11 % of compounds entering clinical development ever reach the market, being withdrawn for reasons such as efficacy (25%), toxicology (24%), clinical safety (12%), drug metabolism and pharmacokinetics (DMPK, 8%), formulation (1%) and portfolio-related and other reasons (30%). Therefore, out of the 70% of failures caused by specific effects, the majority of 61% can be attributed to lack of efficacy, toxicology and clinical safety, whereas DMPK (physicochemical properties, or drug likeness, of the drug candidate itself) accounts for only 8% of attrition. However, the actual proportion may be higher since some reported attrition, which was attributed to lack of efficacy, might be due at least in part to poor DMPK. A similar proportion of 7% was discussed as having inappropriate absorption, distribution, metabolism and excretion (ADME) properties among NCEs between 1964 and 1985. In addition, apparently only about 30% of marketed prescription drugs produce revenues that match or exceed average R&D costs.

The apparent decrease in productivity in the entire pharmaceutical industry has put enormous financial pressures on individual companies and their share price — one of the measures of confidence of investors in future profitability. Although the underlying reasons for this decline in productivity are complex, many factors have been suggested, such as for example increasing clinical development costs, FDA approval standards and political pressures on drug pricing.

One of the key reasons for the decline in productivity is without doubt the high rate of attrition at all stages of the drug discovery process from failures in the early pre-clinical stages to the very expensive late stage failures in the clinic or even post-launch. Although exact figures on attrition in drug discovery are difficult to derive due to the sparseness of publicly available data, it is clear that success rates of discovery projects over the last decade, perhaps in part due to the very high attrition rates, have not been able to match expectations in terms of productivity targets.

Therefore, attempts to reduce attrition early in the drug discovery process have been a major focus over the last decade. During that time, the application of guidelines linked to the concept of drug-likeness (in particular absorption) such as the ‘rule of five’ (see Section 2.1.1 for details) has gained wide acceptance as an approach to reducing attrition in drugs. However, despite this acceptance, an analysis of recent trends revealed that the physical properties of molecules that are currently being synthesised in leading drug discovery companies differ significantly from those of recently discovered oral drugs and compounds in clinical development. This was particularly notable for lipophilicity, where the consequences of a significant increase include a greater likelihood of lack of selectivity and attrition in drug development. Physicochemical properties of molecules are completely under the control of medicinal chemists and can be easily calculated for very large numbers, in some cases for hundreds of thousands of designed structures prior to synthesis.

Close monitoring of physical properties during a drug discovery programme and compound series selection based on orthogonal attrition risks as indicated by compound properties and chemical scaffold may provide the medicinal chemist with opportunities to significantly reduce attrition rates, which are currently estimated at 93–96%.

In this chapter, we focus on the relationship of molecular properties and functional groups of compounds on their interactions with biological targets, which can potentially impact on their pharmacological profile and their potential attrition risks.

1.2 Launched Drugs

The relationship between chemistry, biology and medicine has been a remarkably productive one over the past century since Paul Ehrlich pioneered the idea of systematically searching for drugs. By screening just over 600 synthetic compounds, Ehrlich discovered arsphenamine (Salvarsan) in 1909 which, at the time, greatly improved the treatment of syphilis. Since then, there have been a large number of very significant breakthroughs, for example penicillin (1941), cortisone (1949), benzodiazepines (1960), beta blockers (pronethalol, 1967), anti-histamines (cimetidine, 1977), ACE inhibitors (cap-topril, 1981), insulin (1982), statins (lovastatin, 1987), HIV (zidovudine, AZT, 1987), COX-2 inhibitors (celecoxib, 1999) and kinase inhibitors (imatinib, 2001). Between 1983 and 2007, 907 different NCEs were approved as drugs.

In their elegant analysis of drug targets, Overington et al. found the number of unique launched drugs to be 1357, of which 1204 were considered to be ‘small-molecule’ drugs. Of those, 803 can be administered orally. The analysis included data up to the end of 2005; the number of small molecule drugs has since increased by 21 in 2006 and 19 in 2007, resulting in a total number of launched small molecule drugs of 1244. Of the 1204 drugs used in the 2005 analysis, 1065 were assigned protein molecule targets believed to be responsible for the efficacy of the drug. The data is summarised in Table 1.4.

In the following we discuss the target space and chemical space of drugs separately, but it should be pointed out that these are not separate ‘spaces’ but are interlinked through common, complementary properties. These are, for example, the steric complementarity of a small molecule with a binding site — not only in terms of shape but also in terms of electrostatic interactions and physicochemical properties. This principle of complementarity of chemical and biological space has been discussed extensively elsewhere and will not be expanded upon as part of this chapter.

1.2.1 Target Space of Launched Drugs

The first analysis of the draft sequence of the human genome resulted in an estimate of ~31 000 protein-coding genes; the current estimate has dropped to 22 287 genes. It is generally estimated that 3000 of these are druggable. The relationships of drugs and their targets has been studied extensively. In this context, the term ‘chemogenomics’, described as ‘the discovery and description of all possible drugs for all possible drug targets’, has been coined.

Chemogenomics has been identified as a new approach that can guide drug discovery based on integration of all information within a protein family, for example sequence, structure–activity relationship (SAR) data and protein structure. This allows very efficient cross-SAR analysis and exploration between targets that share small molecule inhibitors, leading to identification of new lead structures. Chemogenomic approaches to drug discovery effectively explore the observation that similar receptors bind similar ligands and have shifted traditional receptor-specific studies towards a more cross-receptor view of pharmaceutical research. Chemogenomic approaches have been exemplified recently for cardiovascular diseases as well as for kinases.

The druggable genome has been initially quantified as 483 small molecule drug targets, with a later figure suggesting that number to be between 600 and 1500. However, it has also been shown that out of these potential drug targets, only a total of 324 drug targets account for all classes of approved therapeutic drugs. This number is reduced further to only 186 for targets of approved oral small molecule drugs. The gene family distribution of current drugs is shown in Figure 1.3.

The concept of druggability, which has been used widely in recent years, postulates that since the binding sites on biological molecules are complementary with their ligands in terms of volume, topology and physico-chemical properties, then only certain binding sites on putative drug targets will be compatible with high-affinity binding to compounds with drug-like properties. The extension of this concept to the whole genome analysis led to the identification of the druggable genome. This is the expressed proteome predicted to be amenable to modulation by compounds with drug-like properties. However, it needs to be noted that the meaning of the term druggability has broadened beyond its generally accepted definition to signify very different aspects along the discovery, development and clinical pipeline.

A very useful categorisation of druggability has been published by Sugiyama which differentiates between the druggable genome and druggable proteins, and the druggability of compounds in terms of their molecular properties. The usefulness of the concept of druggability from a medicinal chemistry standpoint has been summarised by pointing out that the rule of five (Ro5) and its extensions have generated awareness about the importance of pharmacokinetic parameters for drug discovery and development. In addition, the concept of druggability has led to the realisation that there may be whole families of proteins for which it is either extremely challenging or impossible to design small molecules with acceptable oral bioavailability.

Another concept related to druggability is ligand efficiency, which generates a quantitative relationship between drugs and their biological targets, and is defined as the binding energy per non-hydrogen atom in a particular molecule. This concept can be very useful for lead selection by normalising binding energy for molecular weight, but also for differentiation between gene families that have a high or low probability of binding Ro5-compliant small molecules based on an analysis of experimentally determined ligand-binding energies for a particular target. These concepts of maximal affinity and ligand efficiency have been developed further into a computational approach to predict druggability.

In the past ten years or so, expectations in the pharmaceutical industry have been raised as many companies have invested significantly in high-throughput technologies that would make use of information derived from the sequencing of the human genome. Therefore, it would seem that companies are now well-placed to take advantage of the discovery of new targets that have appeared in the post-genomic era. However, there appears to be a reduced likelihood of delivering a preclinical drug development candidate against a new target, which could lead to a temptation to concentrate on more established targets to reduce risk in current development portfolios. More recent in silico approaches such as high-throughput electronic biology may help in identifying, for example, previously unknown complex relationships between targets as well as compounds and targets in biological pathways on a large scale in order to support many parallel work streams in a drug discovery portfolio.

(Continues…)Excerpted from Metabolism, Pharmacokinetics and Toxicity of Functional Groups by Dennis A. Smith. Copyright © 2010 Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.