Table of Contents  

Abou-Gharbia: Discovery of first-in-class therapeutics – venlafaxine (Effexor®), desvenlafaxine (Pristiq®), temsirolimus (Toresil®), ILS-920 and gemtuzumab ozogamicin (Mylotarg®)


Drug discovery and development is a challenging and complex process involving the dedicated multidisciplinary efforts of many research and development functions. Breakthroughs in innovation and process refinements have dominated drug discovery during the last decade, which have been aimed at increasing efficiencies and, thus, reducing cycle time. Despite these technological advances, the number and diversity of new chemical entities (NCEs) approved for human use has not kept pace between the 1980s and 2000s. Business models vary between large pharmaceutical corporations, smaller biotechnology companies, government research groups and academic drug discovery laboratories, but the general drug discovery process that they all follow is essentially the same model (Figure 1).1,2


The drug discovery process.


The drug discovery process

The first step is to identify a biological target whose pharmacological manipulation is expected to impact beneficially on a disease state. The target must be relevant to the disease and ‘druggable’. Approximately 1500 drugs have been approved for use in humans3 (1204 small-molecule drugs and 166 biological agents). These drugs work through their actions on 324 unique biological targets. A seminal paper by Hopkins and Groom in 20024 introduced the concept of the ‘druggable genome’ and suggested that 600–1500 of the approximately 30 000 human genes could be relevant, druggable targets. These results did not take into account the possibility of multiple targets from a single gene through complex formation, splice variation, post-translational modification and the existence of multiple receptor states for receptors, e.g. ion channels.5 Thus, although the number of predicted druggable genes is significantly smaller than the total genome, the chances are good that novel, viable drug targets still await discovery and exploitation. Many tools and techniques have been applied to identify potential drug targets.6 Bioinformatics, database mining, genetic association and phenotypic screening have all been applied successfully, and precedence from known drugs and probe molecules can also assist in new drug target identification. Nearly 25% of known drugs have no known primary target or, at best, lack a well-defined mechanism of action.7 Drug repositioning has become a popular subject for the drug discovery sector and is one of the missions of the recently formed National Center for Advancing Translational Sciences (Bethesda, MD, USA).8 Examples of successfully repurposed drugs include gabapentin, ropinirole, buprenorphine (Temgesic®, Reckitt Benckiser, Slough, UK), thalidomide (Thalidomide Celgene®, Celgene Ltd, Uxbridge, UK), minoxidil (Rogaine®, McNeil Products Ltd, Maidenhead, UK) and aspirin.

Once a potential new drug target is identified it must be validated and shown to be relevant to the disease state.9,10 It is of utmost importance to gain a high degree of confidence that the target is a good one before launching into a costly drug discovery campaign. There are obvious genetic considerations in target validation, such as whether or not the target is expressed in tissues that are involved in the disease and at an appropriate age for the patient. But a target must also meet other criteria to be considered ‘validated’. It is important to show that manipulation of the target (either enhancement or inhibition of its action) produces a biological change that will impact positively on the disease with a minimum of deleterious side-effects. This can be done in a number of ways. The target should be characterized at as many levels as possible, including biochemical, cellular, isolated tissue and in vivo. Transgenic animal models, in which the target has been either ‘knocked out’ or ‘knocked in’, are widely used to associate the target with a disease-related phenotype. One recent development in the field of transgenics is the use of cost- and time-efficient zebra fish in place of murine transgenics for some diseases. When small-molecule or biological tool compounds are not available to study the pharmacology of the target, antibodies, antisense technology and RNA interference (siRNAs) can often be used to inhibit the activity or expression of the target. Questions of specificity and uniqueness must be addressed. Are there alternative pathways that the disease can employ if a drug is given for a candidate target? Will manipulating the target primarily affect the disease, or will there be deleterious side-effects that may limit dose or use of drugs affecting the target? Another issue that goes hand-in-hand with reliability is ‘druggability’. Not all potential drug targets will suitably interact with small molecules or biological agents. When the three-dimensional structure of a target is known but that of potential binding sites is not, computational methods can be employed to suggest likely locations.11,12 Virtual screening of focused small-molecule- or fragment-based libraries can provide additional confidence in a target before a costly high-throughput screen is attempted. Detailed thought processes for validating potential drug targets have been proposed in the literature. Wisely, the value of human insight of experienced structural biologists and medicinal chemists is still apparent in many of the druggability rubrics.13

A plethora of papers and reviews discuss the hit identification, hit-to-lead and lead optimization activities associated with drug discovery. Some have represented that process as crafting a key to fit into a three-dimensional lock (see Figure 1). An important aspect that enhances the chances of success is the identification of druggable scaffolds,14 whether those scaffolds come from virtual screening, fragment-based design, high-throughput screening or rational design based on known drugs or probes. To that end, Lapinski’s ‘rule of 5’ and its variations still play an important role in triaging and prioritizing potential chemical scaffolds for initiation of structure–activity relationship (SAR) campaigns.15 High-throughput chemistry and in vitro biology have made it possible for drug discovery teams to gather and analyse a tremendous amount of SAR information in their quest for potent, selective, efficacious drug candidates. The establishment of high-throughput in vitro physicochemical absorption, distribution, metabolism and excretion (ADME) screens, and in silico ADME, has allowed medicinal chemists to prioritize chemical scaffolds and optimize drug-like properties simultaneously with pharmacological activity, thereby identifying structure–property relationships in addition to SARs.16,17 This ‘multidimensional optimization’ strategy has led to a decrease in the number of compounds terminated from clinical trials for unsatisfactory pharmacokinetics.18 Optimization of the early leads into developmental drug candidates is the cornerstone of drug discovery and development. Several medicinal chemistry approaches have been utilized successfully to optimize initial leads identified via screening of natural products, compound libraries and rational- and/or structure-based drug design.19,20

Discovery of innovative therapeutics – selected case studies

Our drug discovery efforts over past two decades led to the discovery of eight marketed drugs and several drugs under clinical evaluation. Many of these drugs are ‘first’ in class. They include venlafaxine (Effexor®, Pfizer, Surrey, UK), the first serotonin–noradrenaline reuptake inhibitor (SNRI), as well as the second-generation SNRI desvenlafaxine (Pristiq®, Pfizer, Surrey, UK), and three cancer drugs – gemtuzumab ozogamicin (Mylotarg®, Pfizer, Surrey, UK), the first conjugate antibody anticancer therapy; temsirolimus (Toresil®, Pfizer, Surrey, UK), the first mammalian target of rapamycin (mTOR) inhibitor; and the tyrosine kinase inhibitor bosutinib (Bosulif®, Pfizer, Surrey, UK). The identification of tigecycline (Tygacil®, Pfizer, Surrey, UK) by this team ushered in a new era for the treatment of resistant bacterial infection, the short-acting anti-insomnia drug zaleplon (Sonata®, Meda Pharmaceuticals Ltd, Takeley, UK) provided much needed relief to patients and the non-steroidal oestrogen receptor modulator bazedoxifene (Conbriza®, Pfizer, Surrey, UK) opened up new options for the treatment of postmenopausal osteoporosis (Figure 2).


Innovative drugs – six are first-in-class therapeutics.


This report will highlight some of the approaches that were successfully utilized in optimizing early leads into drug candidates. Ligand-based design utilizing biogenic amines serotonin and noradrenaline led to the discovery of the first-in-class SNRI antidepressants venlafaxine and desvenlafaxine. Natural products-based drug design utilizing macrocycle rapamycin led to the discovery of two novel immunophilins – temsirolimus, an anticancer drug for treatment of renal cell carcinoma; and ILS-920, a neuroprotectant and neuroregenerative drug for treatment of stroke – as well as the discovery of gemtuzumab ozogamicin, the first Food and Drugs Administration (FDA) approved humanized antibody–drug conjugate.

Discovery of first-in-class serotonin–noradrenaline reuptake inhibitor antidepressants

Alterations in biogenic amine neurotransmitter levels in the brain have been implicated in the pathophysiology and pharmacotherapy of a number of neurological and psychiatric disorders.21 Throughout the years the biogenic amines have provided medicinal chemists with a solid starting point for their design of innovative therapeutics (Figure 3).


Biogenic amines as a starting point for drug design. Reproduced with permission from Abou-Gharbia MA, Childers WE. The Discovery of Effexor® and Pristiq®. In: Fischer J, Ganellin CR (eds.) Analogue-Based Drug Discovery II. Wiley-VCH: Weinheim; 2010, pp. 507–24.31 Copyright Wiley-VCH Verlag GmbH & Co. KGaA.


Major depressive disorder (MDD), commonly referred to as ‘depression’, presents as a spectrum of emotional and physical symptoms.22,23 This prompted a flurry of research activities aimed at discovering small-molecule antidepressants that would modulate 5-hydroxytryptamine (5-HT) receptor and noradrenaline deficiencies. One approach to that end is to increase synaptic neurotransmitter levels by inhibiting reuptake of biogenic amines through their presynaptic transporters (Figure 4).24,25 Indeed, the observation that imipramine and other tricyclic antidepressants block noradrenaline reuptake spurred scientists to more fully explore the potential of reuptake inhibitors.26,27 Early work focused on designing selective reuptake inhibitors of 5-HT and noradrenaline. Efforts targeting serotonin-selective reuptake inhibitors (SSRIs) led to the development of many currently prescribed antidepressant drugs and tool compounds such as fluoxetine (Prozac®, Eli Lilly and Co. Ltd, Liverpool, UK), paroxetine (Paxil, GlaxoSmithKline, Brentford, UK) and sertraline (Zolft, Pfizer, Surrey, UK), while research aimed at identifying selective noradrenaline inhibitors (NRIs) produced agents such as nisoxetine, maprotiline (Deprilept®, Lundbeck, Copenhagen, Denmark) and reboxetine (Edronax®, Pfizer, Surrey, UK).


Schematic diagram of 5-HT and noradrenaline neurotransmission and reuptake. Reproduced with permission from Abou-Gharbia MA, Childers WE. The Discovery of Effexor® and Pristiq®. In: Fischer J, Ganellin CR (eds.) Analogue-Based Drug Discovery II. Wiley-VCH: Weinheim; 2010, pp. 507–24.31 Copyright Wiley-VCH Verlag GmbH & Co. KGaA.


These agents were perceived to possess a more favourable side-effect profile than monoamine oxidase inhibitors (MAOIs) and tricyclic antidepressants, which often demonstrated cardiovascular and anticholinergic adverse drug effects.28 However, like MAOIs and tricyclics, SSRIs and NRIs demonstrated a time lag between initiation of treatment and the onset of antidepressant action.29,30 However, while more recent clinical data suggest that some agents may induce improvement in a few measures of depression earlier than others (especially in severely depressed patients),31 the goal of a definitive and robust rapid onset of action remains elusive. This prompted us to focus our efforts on the discovery of more effective and tolerable antidepressants.

Discovery of venlafaxine

Multitarget ligand-based designs have been successfully utilized in the search for effective antidepressant drug candidates. While the majority of antidepressant research in the 1980s focused on design of novel SSRIs, we focused our efforts on the design of multi-target SNRIs.32 We capitalized on the existing structural similarities between one of our earlier opiate analgesic leads, ciramadol, and antidepressants that were under clinical Investigation.33,34 Ciramadol was further optimized by reducing its chiral centres from three to one and incorporating an alkylamine pharmacophore in a three-step synthesis to create venlafaxine (Effexor, Figure 5).


Structural evolution of venlafaxine. Reproduced with permission from Abou-Gharbia MA, Childers WE. The Discovery of Effexor® and Pristiq®. In: Fischer J, Ganellin CR (eds.) Analogue-Based Drug Discovery II. Wiley-VCH: Weinheim; 2010, pp. 507–24.31 Copyright Wiley-VCH Verlag GmbH & Co. KGaA.


Venlafaxine represented the first-in-class of SNRIs approved for treatment of MDD. It demonstrated efficacy and tolerability in a broad range of patients, including those who were difficult to treat. Since its launch in 1997 as ‘Effexor XR’, it has become the mainline therapy for MDD and has been shown to have higher remission rates than SSRIs.35

Discovery of desvenlafaxine

The remarkable clinical efficacy seen with venlafaxine has benefited over 20 million patients, and studies have shown that patients resistant to treatment with other antidepressants often respond to venlafaxine.36 The success of venlafaxine encouraged the discovery of other non-tricyclic SNRIs, including duloxetine (Cymbalta®, Eli Lilly and Co. Ltd, Liverpool, UK), milnacipran (Savella®, Forrest Laboratories, New York City, NY, USA) and, most recently, the active enantiomer of milnacipran, levomilnacipran (Fetzima®, Forrest Laboratories, New York City, NY, USA). Venlafaxine is metabolized in man by cytochrome P450 2D6 to give the major active O-desmethyl metabolite desvenlafaxine (Figure 6). Desvenlafaxine is also a mixed SNRI,37 although the compound is somewhat more selective for the serotonin transporter than venlafaxine (desvenlafaxine is 10 times more potent at inhibiting serotonin uptake than noradrenaline, while venlafaxine is three times more selective for serotonin uptake). Owing to genetic polymorphisms in the 2D6 subtype, different people metabolize via 2D6 at different rates, resulting in phenotypic classifications such as poor metabolizers (PMs) and extensive metabolizers (EMs). Clinical data revealed that significantly higher plasma levels of desvenlafaxine (in terms of both Cmax and area under the curve) were seen in EMs than in PMs following administration of venlafaxine (see Figure 6b).37


(a) Structures of venlafaxine and desvenlafaxine. (b) Pharmacokinetic data for administration of venlafaxine. (c) Pharmacokinetic data for administration of desvenlafaxine (data taken from Preskorn et al. 200937). AUC, area under the curve. Reproduced with permission from Abou-Gharbia MA, Childers WE. The Discovery of Effexor® and Pristiq®. In: Fischer J, Ganellin CR (eds.) Analogue-Based Drug Discovery II. Wiley-VCH: Weinheim; 2010, pp. 507–24.31 Copyright Wiley-VCH Verlag GmbH & Co. KGaA.


As a result, the dose of venlafaxine for patients initiating treatment has to be titrated until steady-state plasma levels of both the drug and active metabolite are reached. To overcome this issue, Wyeth36 developed the active metabolite desvenlafaxine as a follow-on to venlafaxine. From the data shown in Figure 6c, it can be seen that direct administration of desvenlafaxine results in similar plasma levels in both EMs and PMs. Additionally, the lack of a role for 2D6 in the metabolism of desvenlafaxine translates to a reduced risk for 2D6-associated drug–drug interactions, including those likely to occur with other antidepressant drugs that inhibit CYP 2D6, such as paroxetine. This is pertinent because patients suffering from depression are often co-treated with multiple antidepressant drugs.

We embarked on a parallel discovery and clinical programme to advance O-desmethy venlafaxine for development to be made available to patients. This led to marketing the succinate salt of desvenlafaxine as Pristiq in 2008. Like venlafaxine, desvenlafaxine demonstrated efficacy not only in treating MDD but also in alleviating pain of depressed patients, which distinguishes this SNRI class of drugs from traditional SSRI drugs. Desvenlafaxine is now well on its way to being the antidepressant of choice and achieving the blockbuster status once enjoyed by its predecessor, venlafaxine.

Discovery of first-in-class immunophilin drugs

Natural products and their derivatives have historically been invaluable as a source of innovative therapeutics. Natural products-based drug discovery reached its peak during the 1970s and 1980s, and of the small-molecule NCEs introduced to the market between 1981 and 2002, roughly half (49%) were natural products, semi-synthetic natural product derivatives or synthetic compounds based on natural product pharmacology.38,39

Advances in technology have enabled natural product-based drug discovery to operate on a more rational basis, and this has spurred a renewed interest in natural products research, which is now feasible using precise synthetic methods and high-resolution analytical tools.40,41 Genetic engineering of biosynthetic pathways of biologically active natural product scaffolds can provide new starting points for optimization of these privileged structures. Natural product-based drug discovery was successfully applied to rapamycin, a novel immunosuppressant natural product with a unique mechanism of action.42 It binds to the effector protein FKBP and forms a complex that binds to another effector protein, mTOR (Figure 7).43 Rapamycin was marketed in 1999 as Rapamune® (Pfizer, Surrey, UK) for the treatment of transplantation rejection.


Binding of rapamycin to FKBP and mTOR.


Discovery of temsirolimus

We embarked on a non-high-throughput screening approach looking at our rapamycin equity in order to expand its therapeutic utility. The X-ray co-crystal structure of rapamycin bound to the FKBP12/mTOR complex revealed that the region of the molecule surrounding the C-42 hydroxyl was relatively solvent exposed and could be altered without affecting the tertiary complex.44 Thus, a series of aqueous solubility-promoting substituents were appended to the C-42 hydroxyl to give compounds that retained immunosuppressant and antiproliferative activity. By applying a semi-synthetic strategy to the C-42 secondary alcohol functionality, a broad array of rapamycin analogues were synthesized45 (> 700) and tested for in vitro potency and in vivo efficacy. Among these, 31 compounds were selected for later-stage in vivo efficacy studies. This led to the identification of three advanced leads – 42-carbamate, ester and hindered ester – out of which CCI-779 was nominated as a candidate for clinical trials. CCI-779 was synthesized from rapamycin in two steps, as shown in Figure 8. It has a hydrophilic diol ester tethered to C-42 bringing in additional solubility, added H-bonds to FKBP and crystallinity. The hindered nature of the C-42 distal ester slowed down the hydrolysis and thereby eliminated the β-elimination problem. CCI-779 possessed good chemical and pharmaceutical properties and was very active in murine transplantation and oncology models (in vitro potency and in vivo efficacy). Compound CCI-779 was found to inhibit mTOR signalling via CCI/FKBP/mTOR complex formation in vitro, resulting in the arrest of treated tumour cells in the G1 phase. In vivo testing with compound CCI-779 showed sustained inhibition of human tumour growth (U87 glioblastoma) in nude mice dosed daily for 5 days every 2 weeks. In vivo testing showed that T-cell response to dinitrofluorobenzene recovered 1 day after five daily doses of CCI-779, indicating that it had no lingering immunosuppressive effects. CCI-779, known as temsirolimus, was approved by the FDA (USA, May 2007) for the treatment of advanced renal cell carcinoma. Temsirolimus is the first approved immunophilin cancer therapy that specifically targets mTOR, a key protein in cells that regulates cell proliferation, cell growth and cell survival. Temsirolimus works along two pathways in the mTOR cascade to provide both antiproliferative and antiangiogenic activity. Renal cell carcinoma accounts for 2–4% of all cancers and the 5-year survival rate for advanced renal cell carcinoma is approximately 5–10% and the disease is estimated to be the cause of approximately 13 000 deaths annually in the USA alone. The discovery of temsirolimus added to the armament of drug therapies available to patients suffering from renal cell carcinoma in whom treatment has historically had limited success. Our innovative approach and the success of temsirolimus led other companies to develop and market similar water-soluble anticancer rapamycin analogues such as everolimus (Afinitor®, Novartis, Grimsby, UK) and ridaforolimus (Deforolimus®, Merck-Ariad, Whitehouse Station, NJ, USA).


Functionalization of C-42 alcohol led to the discovery of temsirolimus (Toresil).


Discovery of ILS-920

It is well established that immunophilins, such as FK-506, demonstrate neuroprotective activity in animal ischaemia models. A key event in this process is thought to be binding to FKBP-12, which can trigger a cascade of events, although the exact mechanism is not fully known. One hindering factor of using these compounds as neuroprotectants is their immunosuppressant activity.4547 However, it is possible to achieve neuroprotection using immunophilins without immunosuppression. Efforts aimed at exploring the role of these non-immunosuppressive immunophilin ligands in neuroprotection revealed that, while binding to FKBP-12 was still a key step, formation of a tertiary complex with calcineurin may not be responsible for mediating neuroprotection and neuroregeneration.

The journey from rapamycin to ILS-920 is a good example of how a class of molecules can have multiple therapeutic applications. The discovery of ILS-920 (Figure 9) began with a few key observations. Both rapamycin and the structurally related FK506 (see Figure 9) were reported to show neuroprotective effects in vitro. However, while that in vitro activity translated into efficacy in a rodent model of ischaemic stroke for FK506, rapamycin was not effective in the same study. It was generally accepted that the immunosuppressive activity shown by FK506 precluded it from development for stroke. Other immunophilins that bound to FKBP12, such as GPI-1046 (see Figure 9), also demonstrated some neuroprotective activity in vitro but were not immunosuppressant, leading some to hypothesize that the neuroprotective activity seen with FK506 and GPI-1046 may be the result of binding to other targets. A breakthrough for Wyeth48 came with the isolation of a macrocyclic derivative structurally related to rapamycin, 3-normyridamycin (see Figure 9). 3-Normyridamycin showed little binding to FKBP12 and no immunosuppressant activity, but was potently neuroprotective and neuroregenerative in a cellular model of Parkinson’s-like neurodegeneration. These results encouraged Wyeth to embark on a reassessment of its immunophilin analogue library using a pair of phenotypic assays that assessed neuroprotective and neuroregenerative potential. Hits were cross-screened for immunosuppressant activity. This exercise identified the non-immunosuppressive rapamycin analogue WAY-124466 (see Figure 9), a derivative in which the polyene moiety, thought to interact with mTOR, had been modified using Diels–Alder chemistry.


Structures of immunosuppressive and non-immunosuppressive immunophilins.


Our efforts in designing non-immunosuppressant neuroprotectant immunophilins focused on the pharmacophore that binds to the mTOR effector protein. Structural manipulation targeting the C1–C4 triene of rapamycin utilizing Diels–Alder chemistry with a variety of dienophiles led to the synthesis of novel non-immunosuppressant neuroprotectant rapamycin derivatives. One of these, the semi-synthetic adduct ILS-920, lacked immunosuppressant properties and demonstrated good brain penetration. It was synthesized in a three-step synthesis by reacting rapamycin with nitrosobenzene followed by hydrogenation of the six-membered cyclic adduct (Figure 10).49

Figure 10

Synthesis of ILS-920.


ILS-920 promoted survival of E16 primary rat cortical neurons in culture and stimulated neurite outgrowth in that system (Figure 11). The compound demonstrated neuroprotection in a rat transient middle cerebral artery occlusion focal ischaemia model in a dose-dependent manner (Figure 12). In that model, animals treated with ILS-920 showed reduced neurological deficits compared with placebo-treated animals, as measured by sensorimotor performance. In a permanent focal ischaemia model, that improvement in sensorimotor performance lasted 3 months longer with ILS-920 than with the placebo. Taken together, these data suggest that ILS-920 not only functions as a neuroprotectant, but also promotes functional recovery. These exciting results will be featured in a forthcoming article.


Neuroprotective/neuroregenerative activity of ILS-920 on E16 primary rat neurons in culture. (a) Effect of ILS-920 on neuronal survival. (b) ILS-920 promotes neurite outgrowth.


ILS-920 reduces infarct volume in a rat transient middle cerebral artery occlusion model.


Graziani et al.49 have provided evidence that that the neuroprotective and neuroregenerative activities of ILS-920 may involve binding to the effector protein FKBP-52. In addition, inhibition of L-type calcium channel currents may contribute to the compound’s neuroprotective activity.49,50 Thus, the initial discovery of a natural product from a soil sample from Easter Island ultimately led not only to additional drugs with different indications, but to the discovery of new signalling pathways that remain an area of intense research and new discoveries even today.

Gemtuzumab ozogamicin – the first antibody-targeted anticancer cytotoxic agent

Our research scientists embarked on innovative research aimed at exploiting calicheamicin cytotoxicity for cancer therapy. This led to designing the acid-sensitive hydrazone–disulphide linker group that made development of the first anticancer antibody–drug conjugate, gemtuzumab ozogamicin, possible (Figure 13). This innovative multifunctional linker allowed for conjugation of the potent cytotoxin calicheamicin to external lysine residues of an antibody raised to the cancer cell surface protein CD33. The linker was designed to hydrolyse within the acidic environment of cancer cells, selectively delivering the cytotoxic ‘warhead’ to where it would be most efficacious with minimal toxicity to normal cells.42


Gemtuzumab ozogamicin – a small-molecule antibody conjugate. Calicheamicin is linked to an antiCD33 antibody by the AcBut Linker.


Gemtuzumab ozogamicin demonstrated impressive selectivity for CD33-associated cancer cells, was assessed in vitro and in vivo and was advanced to clinical evaluation, which led to it receiving FDA priority approval in 2000.42 Many leukaemia patients received the drug over a decade. Gemtuzumab ozogamicin was voluntarily withdrawn from the market in 2010 following FDA recommendation as a clinical trial raised concern over its safety. As it is under accelerated approval, the company is required to conduct additional clinical trials after approval to confirm the drug’s benefit. However, as testimony to its benefit and potential, Pfizer and the FDA continue to work together to identify dosing regimens that will make the drug available to needy patients once again. Moreover, the pioneering breakthroughs realized with gemtuzumab ozogamicin and the cancer-selective linker strategy developed by our research team have been mimicked in the design of other antibody–drug conjugates such as brentuximab vedotin (Adcetris®, Takeda UK Ltd, High Wycombe, UK) and trastuzumab emtansine (Kadcyla®, Roche Products Ltd, Welwyn Garden City, UK).

Concluding remarks

During my 30-year journey working in drug discovery in industrial and academic institutions, I have experienced vast changes in drug discovery approaches in which new NCE drug candidates were discovered. The trend with the most impact was shifting medicinal chemistry emphasis from only optimizing biology of synthesized molecules into optimizing drug-like properties and the ‘absorption, distribution, metabolism, excretion and toxicity’ profile led to a reduction in the attrition rate of compounds in development. Adopting this approach resulted in the discovery of novel SNRI antidepressant drugs that have been of benefit to over 20 million patients worldwide and many other innovative drugs that have been referred to in the introduction.

The non-traditional approaches to natural products-based drug discovery are now being adopted, such as the use of fractionated libraries that result in broader assay compatibility and fewer interferences, and lead to faster identification and isolation of active leads. Future natural products drug discovery leads will not only capitalize on semi-synthetic approaches but will focus on a ‘customized natural products approach’ wherein new lead molecules, once thought to be inaccessible by semi-synthesis, are generated through biosynthesis and genetic engineering. We successfully adopted both strategies, which resulted in the discovery of the anticancer drug temsirolimus, the neuroprotectant drug candidate ILS-920 and several new natural products derivatives currently in development. Specifically, our work supports the hypothesis that modification of rapamycin at the mTOR-binding region can provide non-immunosuppressive compounds with potent neuroprotective activity and significant efficacy in an animal model of ischaemic stroke. The role of natural products as privileged scaffolds for semi-synthesis has been underexplored in recent years, but remains a useful complement to modern approaches to medicinal chemistry. Combined with cell-based screening, the preparation of biologically active rapamycin analogues has yielded the clinical candidate ILS-920, a neuroprotectant for stroke therapy. This is a great, unmet medical need as only 5% of stroke patients ever receive tissue plasminogen treatment, and still there are no available marketed neuroprotective agents for ischaemic stroke.

Current and future approaches include non-traditional approaches to the discovery of drug candidates. This will include bridging the gap between small-molecule and protein drug candidates by applying multi-platforms (small molecules, vaccines and proteins) in an integrated strategy to tackle a given disease target. As more innovative new drugs acting at novel targets advance through development, it becomes increasingly important to establish a correlation between preclinical efficacy in animal models and efficacy in humans. Translational medicine is rapidly emerging in many biomedical research institutions as a tool to bridge this gap between preclinical and clinical studies, often referred to as ‘from bench to patient bedside’. Medicinal chemists are playing a pivotal role in the discovery of biomarkers, from radiolabelled synthesis efforts in imaging studies to the development of sophisticated instrumental analysis methods and biochemical assays in metabolomic investigations, which are used to measure biomarker levels in biological fluids.

Thus, while drug discovery hurdles remain many and high, the future holds great promise for this industry. We will continue discovering major breakthrough therapies that will address many unmet medical needs, such as disease-halting drugs for the treatment of Alzheimer’s disease, drugs to eradicate AIDS and drugs to attack resistant cancers, treat strokes, multiple sclerosis and other debilitating diseases.


I am indebted to many researchers with whom I was privileged and blessed to have worked with during my career journey at Wyeth, and to my current researchers at the Moulder Center for Drug Discovery Research at Temple University (Philadelphia, PA, USA). Their creativity, dedication and hard work were instrumental in the discovery of several drug discovery breakthroughs that led to many first-in-class drug candidates. Sincere thanks to my colleague Wayne Childers, not only for providing constructive feedback and suggestions to the content of this review, but also for his dedicated tireless efforts in medicinal chemistry during the past 25 years.



Mariani G, Bruselli L, Kuwert T, et al. A review on the clinical uses of SPECT/CT. Eur J Nucl Med Mol Imag 2010; 37:1959–85.


Drug Discovery and Development. Understanding the R&D Process. URL: (accessed 8 May 2013).


Overington J, Al-Lazikani B, Hopkins A. How many drug targets are there? Nature Rev Drug Disc 2006; 5:993–6.


Hopkiins A, Groom C. The druggable genome. Nature Rev Drug Discov 2002; 1:727–30.


Kubiyni H. Drug research myths: hype and reality. Nature Rev Drug Disc 2003; 2:665–8.


Hughes J, Rees S, Kalindjian S, Philpott K. Principals of early drug discovery. Pharmacology 2011; 162:1239–49.


Gregori-Puigjane E, Setola V, Hert J, et al. Identifying mechanism-of-action targets for drugs and probes. Proc Natl Acad Sci USA 2012; 109:11178–83.


Boguski M, Mandl K, Sukhatme V. Drug discovery, repurposing with a difference. Science 2009; 324:1394–5.


Smith C. Drug target validation: hitting the target. Nature 2003; 422:341–7.


Chen X-P, Du G-H. Target validation: a door to drug discovery. Drug Discov Ther 2007; 1:23–9.


Halgren T. Identifying and characterizing binding sites and assessing druggability. J Chem Infect Model 2009; 49:377–89.


Nisium B, Sha F, Gohlke H. Structure-based computational analysis of protein binding sites for function and druggability prediction. J Biotech 2012; 159:123–34.


Campbell S, Gaulton A, Marshall J, et al. Visualizing the drug target landscape. Drug Discov Today 2010; 15:3–15.


Keller T, Pichota A, Yin Z. A practical view of ‘druggability’. Curr Opin Chem Biol 2006; 10:357–61.


Lapinski C. Lead- and drug-like compounds: the rule-of-five revolution. Drug Discov Today 2004; 1:337–41.


Kerns E, Di L. Drug-Like Properties: Concepts, Structure Design and Methods. Burlington: Academic Press; 2008.


Gleeson M, Hersey A, Hannongbua S. In-silico ADME models: a general assessment of their utility in drug discovery applications. Curr Topics Med Chem 2011; 11:358–81.


Kola I, Landis J. Can the pharmaceutical industry reduce attrition rates? Nature Rev Drug Discov 2004; 3:711–15.


Parker C, Zhang J-H. High-throughput screening for small-molecule drug discovery. In: Gad SC (ed.) Development of Therapeutic Agents Handbook. New Jersey: John Wiley & Sons Inc.; 2012, pp. 147–79.


Davis D, Evans-Illidge RA, Quinn EA, Ronald J. Guiding principles for natural product drug discovery camp. Future Med Chem 2012: 4:1067–84.


Stahl SM. Essential Psychopharmacology: Neuro Scientific Basis and Practical Applications, 2nd edn. Cambridge, UK: Cambridge University Press; 2000.


Andrade L. Caraveo-Anduaga JJ, Berglund P, et al. The epidemiology of major depressive disorder episodes: results from the international consortium of psychiatric epidemiology (ICPE) surveys. Int J Methods Psychiatr Res 2003; 12:3–21.


Schildkraut JJ. The catecholamine hypothesis of affective disorders: a review of supporting evidence. Am J Psychiatry 1965; 122:509–22.


Richelson E. Biological basis for depression and therapeutic relevance. J Clin Psychiatry 1991; 52:4–10.


Ressler KJ, Nemeroff CB. Role of serotonergic and adrenergic systems in the pathophysiology of depression and anxiety disorders. Depres Anxiety 2000; 12(Suppl. 1):S1–12.<2::AID-DA2>3.0.CO;2-4


Mann JJ. The medical management of depression. N Engl J Med 2005; 353:1819–34.


Khawam EA, Laurencic G, Malone DA. Side effects of antidepressants: an overview. Curr Drug Ther 2006; 73:351–61.


The World Health Organization. The World Health Report 2001: Mental Health: New Understanding, New Hope. Geneva: World Health Organization; 2001.


Machado-Vieira R, Salvadore G, Luckinbaugh DA, Manji HK, Zarate CA. Rapid onset of antidepressant action: a new paradigm in the research and treatment of major depressive disorder. J Clin Psychiatry 2008; 69:946–58.


Gourion D. Antidepressants and their onset of action: a major clinical, methodological and pronostical issue. Encephale 2008; 34:73–81.


Abou-Gharbia MA, Childers WE. The Discovery of Effexor® and Pristiq® (Chapter 21). In: Fischer J, Ganellin CR (eds.) Analogue-Based Drug Discovery II. Weinheim: Wiley-VCH; 2010, pp. 507–24.


Yardley JP, Husbands GEM, Stack G, et al. 2-Phenyl-2-(1-hydroxycycloalkyl) ethylamine derivatives: synthesis and antidepressant activity. J Med Chem 1990; 33:2899–905.


Lecrubier Y. Clinical utility of venlafaxine in comparison with other antidepressants. Int Clin Psychopharmacol 1995; 10(Suppl. 2):29–35.


Nelson JC. Managing treatment-resistant major depression. J Clin Psychiatry 2003; 64(Suppl. 1):5–12.


Thase ME, Entsuah R, Rudolph R. Remission rates during treatment with venlavaxine or selective serotonin reuptake inhibitors. Br J Psychol 2001; 178:234–41.


Deecher DC, Beyer CE, Johnston G, et al. Desvenlavaxine succinate: a new serotonin and norepinephrine reuptake inhibitor. J Pharmacol Exp Ther 2006; 318:657–65.


Preskorn S, Patroneva A, Silman H, et al. Comparison of the pharmacokinetics of venlafaxine extended release and desvenlafaxine in extensive and poor cytochrome P450 2D6 metabolizers. J Clin Psychopharmacol 2009; 29:39–43.


Newman DJ, Cragg GM. Natural products as sources of new drugs over the last 25 years. J Nat Prod 2007; 70:461–77.


Koehn FE, Carter GT. The evolving role of natural products in drug discovery. Nat Rev Drug Discov 2005; 4:206–20.


Kanafani ZA, Corey GR. Daptomycin: a rapidly bactericidal lipopeptide for the treatment of gram-positive infections. Expert Rev Anti-Infect Ther 2007; 5:177–84.


Rini B, Kar S, Kirkpatrick P. Temsirolimus. Nature Rev Drug Discov 2007; 6:599–600.


Abou-Gharbia M. Optimization of natural product leads into drug candidates. In: Sener B (ed.) Biodiversity. London: Kluwer Academic; 2002, pp. 60–70.


Choi J, Chen J, Schreiber SL, Clardy J. Structure of the FKBP12- rapamycin complex interacting with the binding domain of human FRAP. Science 1996; 273:239–42.


Skotnicki JS, Leone CL, Smith AL, et al. Design, synthesis and biological evaluation C-42 hydroxyesters of rapamycin. Clin Cancer 2001; 7:S3749.


Steiner JP, Connolly MA, Valentine HL, et al. Neurotrophic actions of nonimmunosuppressive analogs of immunosuppressive drugs FK-506, rapamycin and cyclosporine. Nat Med 1997; 3:421–8.


Gold BG, Densmore V, Shou W, Matzuk MM, Gordon HSJ. Immunophilin FK506-binding protein FKBP-52 (not FK506-binding protein 12) mediates the neurotrophic action of FK506. J Pharmacol Exp Ther 1999; 289:1202–10.


Lyons WE, George EB, Dawson TM, Steiner JP, Snyder SH. Immunosuppressant FK-506 promotes neurite outgrowth in cultures of PC12 cells and sensory ganglia. Proc Natl Acad Sci USA 1994; 91:3191–5.


Summers MY, Leighton M, Liu D, Pong K, Graziani EI. 3-Normeridamycin: a potent non-immunosuppresive immunophilin ligand is neuroprotective in dopaminergic neurons. J Antibiot (Tokyo) 2006; 59:184–9.


Pong K, Graziani E, Liang S, et al. ILS-920, a novel, nonimmunosuppressive rapamycin analog, stimulates neuronal survival and outgrowth and promotes cellular and functional recovery following cerebral ischemia. 22th Annual Meeting of the Society for Neuroscience, 3–7 November 2007, San Diego, CA, Poster 598.


Ruan B, Pong K, Jow F, et al. Binding of rapamycin analogs to calcium channels and FKBP-52 contributes to their neuroprotective activities. Proc Natl Acad Sci USA 2008; 105:33–8.

Add comment 

Home  Editorial Board  Search  Current Issue  Archive Issues  Announcements  Aims & Scope  About the Journal  How to Submit  Contact Us
Find out how to become a part of the HMJ  |   CLICK HERE >>
© Copyright 2012 - 2013 HMJ - HAMDAN Medical Journal. All Rights Reserved         Website Developed By Cedar Solutions INDIA