Discovery and Development of Novel Drugs

Abstract

Drug discovery and development process is nowadays conducted in relatively standardised sequence of phases, starting with Discovery and being followed by Preclinical, Clinical and Non-Clinical Development. Discovery phase is divided in Hit Finding, Lead generation, Lead Optimisation and Candidate Identification Phase. Main drivers of the whole process are regulatory requirements and the aim to eliminate the unnecessary spending by early elimination of unlikely drug candidates. Marine products, once purified, isolated and produced in required quantities, follow the same route as any other synthetic drug.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Human life expectantly has increased significantly during the course of the last century, at least partially due to the discovery of large number of drugs and therefore, the 20th century can be rightfully called true pharmaceutical century. Although number of diseases that were incurable hundred years ago can be easily cured today, modern medicine is still facing significant changes in its aim to ensure good quality of life and extended life-span. New and better drugs are needed to address diseases of the ageing population, such as degenerative diseases and cancer, as well as to overcome continuous threat of novel infections caused by micro-organisms resistant to current therapies.

Throughout the human history, nature was the obvious place to search for medicines. Based on a “trial and error” approach, traditional medicine has collected a body of knowledge that was passed from generation to generation. In the last century, supported by the modern science, people have continued that quest in even more methodical and controlled manner. This resulted in 1453 drugs that have obtained FDA approval till the end of 2013 (Kinch et al. 2014).

First modern drugs, such as hormones or antibiotics, have provided a basis for the next generations of semisynthetic and synthetic drugs. Initially, compounds were synthetized and tested individually to be followed by chemical libraries produced by combinatorial chemistry (Seneci et al. 2014) and, most recently, DNA-encoded libraries (Mullard 2016). Initial generations of modern drugs are not very selective and numerous serendipitous observations related to their add-on activates inspired 1988 Nobel Laureate in Physiology and Medicine, Sir James Black, to say: “The most fruitful basis for the discovery of a new drug is to start with an old drug”. Furthermore, novel informed insights and technological platforms have provided basis for repositioning of some relatively old drugs, as well as for optimising current drug discovery and development paradigm.

Beforehand, compounds were tested for their activity on animals and sometimes humans, whereas compound screening in vitro systems was subsequently introduced once scientific and technological progress made it possible. Extensive exploration of pathophysiology of human diseases led into several decades of focused TARGET BASED DRUG DISCOVERY. In vitro-centric target based approach resulted with lower than expected number of new drugs due to the very high attrition rate of drug development process. Relatively recently, hope of better predictability has been raised by introduction of PHENOTYPIC ASSAYS that are not focused on specific targets but reconstruct the elements of human disease as accurately as possible.

Evolution of drug discovery and development in modern times has been catalysed by progress and/or limitations in the several areas: our understanding of disease ethiology, knowledge about potential druggable targets, advancement of technologies related to drug pharmaceutical R&D, as well as diversity and number of scaffolds that can be utilised for derivatisation of new drugs. Whereas majority of these areas have been advancing significantly, availability of new natural scaffolds/compounds has been mostly limited by availability of natural sources. Although humans have been attempting to understand and utilise marine resources for therapeutic purposes since ancient times, modern drugs are predominantly based on terrestrial natural origins. Nowadays, pharmaceutical R&D is increasingly turning towards the sea as a source of new therapeutics with the aim to expand innovation potential and maximise utilisation of great advancements in science and technologies. Marine environment represents almost unlimited source of biodiversity and novel bioactive natural products, with structural and chemical features generally not found in terrestrial natural products. To date, a rather limited number of drugs with marine origin have been put to the market or are currently in late phases of development (Table 1). However, relatively rich early-phase pipeline suggests that we are going to see many more of them in the near future (Mayer et al. 2016; http://marinepharmacology.midwestern.edu/clinPipeline.htm).

Table 1 Marine-derived marketed pharmaceuticals and late stage pipeline (6)

Although every disease and every scientific approach have their particularities, drug discovery and development process itself is nowadays conducted in relatively standardised sequence of discovery and development phases. They are driven by regulatory requirements and the aim to avoid the unnecessary cost by early elimination of unlikely drug candidates. Marine products, once purified, isolated and produced in required quantities, follow the same route as any other synthetic drug (Martins et al. 2014).

2 Project Planning Phase

Basis for every successful drug discovery project is clear understanding of the problem that is intended to be solved and what would successful outcome look like. This type of clarity is usually achieved by precise description of the medical and market needs, target product profile (Table 2) and understanding of the competitive environment. These elements should provide exact answers to the below listed questions.

Table 2 Example of target product profile (TPP)

• Medical need

  • Which medical problem is intended to be solve?

  • How relevant would the intervention be for the human health?

  • Are there already treatment options available?

• Market need

  • Are there other drugs which already address the problem?

  • Would anybody be ready to pay for the new drug with the expected superiority in comparison with exiting treatment options?

• Competition/Comparators

  • Are there other drugs in the market or in the drug discovery and development pipeline that address that particular need? Which feature(s) of those drugs should be improved?

• Target product profile (desired versus. minimal)

  • What should new drug be like and which criteria should it fulfil? (Table 2)

3 Where to Start?

One should always start with the disease and disease related knowledge. The better understanding of medical problem, the better will be a selection of relevant drug targets and phenotypic testing systems. Nowadays, a wealth of patient and disease relevant information is present in a public domain collected by clinical and basic researchers, such as human genome project data, genome wide association studies, numerous population studies, etc. Many diseased and healthy human tissues have been compared so far by using number of different technologies, either directly upon removal from donor’s body or after cultivation in ex vivo culture systems.

If drug discovery programme is focused around a particular target, the process starts with TARGET VALIDATION. The goal of these efforts is to confirm that particular target is relevant for the pathophysiological process of interest and that its modulation would not present unnecessary health hazard. This is typically done, in addition to extensive literature search, via genetic interventions (e.g. gene knock in/out, siRNA, CRISPER, etc.) and/or using tool compounds within particular biological systems (Bunnage et al. 2013). Both approaches have their limitations. Genetic intervention takes whole protein out of equation and not only its particular function that drug substance actually targets, possibly resulting with an overestimation of potential drug activity, no matter positive or negative. On the other hand, usefulness of tool compounds in target validation process depends highly on their selectivity and potency. Ultimate target validation can be done only in clinical disease setting when selective drug, with appropriate PK/PD properties, is administered to carefully selected patient population.

Even if drug discovery process starts with PHENOTYPIC APPROACH, there is always a clear notion to seek for drug-target(s) interaction that can explain activity in such phenotypic system. Current paradigm of the modern drug discovery stipulates understanding of drug target(s) and affected pathways in order to decrease likelihood of negative surprises later in the drug development process as well as to enable a rational drug design. In contrast to a single target approach, phenotypic-based approach can result with identification of several targets that are hit by the some compound and combination of which is optimal for achievement of the desired effect.

Once biological strategy is defined and decision has been made weather to use the target-based assays or phenotypic assays, one needs to select which molecules to test in such tests.

4 Drug Discovery

Typical discovery process (Fig. 1) is stepwise approach which starts with hit finding phase. HITS are molecules that show activity in primary screening/test system in dose response manner and demonstrate certain structure-activity relationship (SAR).

Fig. 1

Drug discovery process. HTS High Throughput Screening, ADME Absorption Distribution Metabolism Elimination, GLP Good Laboratory Practice, PK Pharmacokinetics, PK/PD Pharmacokinetic vs. Pharmacodynamics, PPD Pharmaceutical Product Development

Next phase is frequently called Lead generation phase. During this phase hit molecules are optimised for various properties such as potency, selectivity, drug-like physical and chemical properties, etc. A LEAD is a molecule that shows activity and selectivity in vitro and first evidence of in vivo activity. During the Lead optimisation phase lead is further optimised until it reaches desired pharmacodynamic (PD) and pharmacokinetic (PK) properties, both in vitro and in vivo, and becomes a CANDIDATE molecule. Toxicity of that molecule is evaluated in non-GLP toxicological tests and PK/PD relationship is explored in animal efficacy models. In addition to biological profiling, assessment of chemical and pharmaceutical developability is performed in order to estimate the likelihood that final drug product containing such active principle can eventually reach the market. If the molecule criteria, it becomes a PRECLINICAL CANDIDATE. Such molecule is made in larger quantities in order to Preclinical development program, performed in line with Good Laboratory Practice (GLP), which include GLP toxicology, ADME (Absorption, Distribution, Metabolism and Elimination) and PK. A molecule that successfully completes preclinical development phase is ready to enter clinical phase of the drug development and therefore is called a CLINICAL CANDIDATE.

The clinical candidate is then progressed through comprehensive clinical development programme, and, in if it meets required criteria, it is submitted to regulatory bodies for approval.

4.1 Chemical Tactics—Where to Get First Hits?

Several past decades have been characterised by high throughput screening of large chemical libraries composed of hundreds of thousands of random compounds or relatively smaller focused libraries, likely to hit particular type of biological targets, such as kinases, GPCRs, etc. Large compound libraries are mainly made by combinatorial chemistry and their significant portion might not have drug-like physical–chemical properties. On the other hand, when drug target is well explored, rational drug design can lead chemists in the process of making specific compounds with higher likelihood to interact with the target and its active site. Relatively recently, screening of fragment-based libraries has proven to be good strategy in revealing small fragments that interact with a target active site. In the next step, those hit fragments are combined in larger, more potent and selective molecules. Finally, number of chemical structures linked to DNA, combined in DNA-encoded libraries, are currently being used for screening in various biological systems, DNA being unique coding tag and chaperon (Mullard 2016).

In addition to testing pure compounds, some researchers test extracts obtained from various natural sources, including marine organisms (Martins et al. 2014). Such extracts most likely contain molecules that have never been synthetized by researchers and therefore bare potential to hit the targets that have not been conquered so far and reveal novel chemical space. Furthermore, such extracts present mixture of molecules which can have additive or synergistic effects in particular biological assays. Complexity is further added when testing natural extracts in phenotypic assay and observed activity could be result of multiple molecules and multiple targets. Next step, isolation of active structure(s), is very labour intensive process which is frequently unsuccessful or results with structures which cannot be made by chemical synthesis. Sometimes, decent quantities of active molecules can be produced by using biotechnological approaches. Although there is a whole universe of microbes out there, that produce secondary metabolites, we do not have optimal knowledge and/or technology to cultivate them. Please find more details in an excellent review by Martins et al. (2014).

4.2 Biological Tactics—Screening Cascade

Drug discovery and development attrition rate is extremely high with only one out of 100,000 synthetized molecules reaching the market in approximately 15 years. Sequences of biological tests that are meant to select the best molecules, most likely to be successful at the end, are called screening or testing cascades (9). If properly designed, testing cascade should reflect target product profile and eliminate inadequate molecules as early as possible, reducing unnecessary spending in the following, always more expensive, stages. Complexity and predictive potential of the testing cascades varies among various therapeutic areas and various diseases. For example, potent antibiotics can be spotted already in in vitro systems, whereas efficacy of drugs addressing various psychiatric diseases can only truly be tested within clinical trials and therefore, later therapeutic area is known to be associated with significantly high number of failures during Clinical development.

Testing starts in in vitro systems which address all three biological aspects of every drug: PD, PK and toxicology. Cascade starts with more simple and high throughput screening (HTS) assays and gradually moves to more complex systems (Jelic et al. 2013). Typically, first answer one looks for is which compound interacts with the target and that answer can be very frequently obtained in biochemical cell free assay (Fig. 2). Furthermore, it is import to recognise if there is functional consequence of that interaction, as well as, how potent and how selective is the compound.

Fig. 2

Target-based testing cascade

Cellular system adds complexity and only the compounds that are able to reach their biological target and exhibit activity in such living system are selected for further progression. Observed activity is a combined result of several properties of the molecule, next to its target interaction, such as, ability to cross the cell membrane or accumulate in particular subcellular compartment, if necessary for the action. Furthermore, cells are used to evaluate cytotoxicity potential.

Next level of complexity is addressed in whole blood assays, which, in addition to mix cell population, obviously contain various extracellular components that compound will be exposed to during clinical trials. For many years, in vitro testing has been done on immortalised cell lines, which provides excellent platform for particular target or pathway related testing, but seriously lack similarity with the disease conditions. Lately, significant efforts have been put into setting up disease-relevant phenotypic assays on primary cells and/or human tissues, with the aim of early evidence of disease-relevant effects.

In parallel with in vitro efficacy and toxicity testing, compounds are screened for their plasma protein binding potential, CYP inhibition, solubility, permeability, microsomal and hepatocyte stability in order to identify potential PK liabilities very early in the process.

After successful completion of in vitro testing, an evidence of activity in complex living organism is required in order to manage the risk for further progression of the compound and to increase probability of positive clinical outcome. For proper in vivo evaluation of the drug candidate molecules, it is essential to fully understand the limitations of the animal efficacy models and recognise that animal models reflect particular pathophysiological elements/pathways and do not represent equivalent of human disease.

In vivo compound testing on laboratory animals provides information on effective dose and route of administration and generates toxicology, safety pharmacology and PK data. That information allows a ranking of potential drug candidates in “head to head” comparison and generates proof of concept for regulatory authorities, as a preparation for “first in human” studies. Establishment of PK/PD relationship and PK/PD modelling provides basis for the clinical dose prediction.

5 Drug Development

Drug development (Fig. 3) is a complex set of activities with a goal to transform a promising drug candidate into new drug. In comparison to drug discovery phase, it is highly regulated and closely monitored by regulatory authorities. Since drug development represents substantial investment, with no guarantee of success, it is crucial that risks are identified and eliminated at the earliest possible stage, or understood and managed while caring forward, in order to build confidence for subsequent investments. In parallel, the progress of the project is continuously assessed via a number of milestones at which obtained results are confronted with the target product profile, potential competition and the market developments.

Fig. 3

Drug development process. PoC Proof of Concept, PK Pharmacokinetics, PK/PD Pharmacokinetic vs. Pharmacodynamics

Drug development can be divided into preclinical development, Clinical development and Non-Clinical development.

6 Preclinical Development

The main goal of Preclinical development is to determine ultimate safety profile of experimental drug candidate with purpose of, as much as possible, limiting risks of its use in humans during clinical trials. This critical phase includes a number of activities required to progress a new drug candidate through toxicology, pharmacology and pharmacokinetic testing. All these activities must be performed in a regulatory compliant manner to ensure that quality of data and the safety of human subjects involved in future clinical trials.

During preclinical studies (Table 3), it is especially important to:

Table 3 Preclinical Development

• Define initial safe dose and dose escalation schemes for application in humans

• Identify target organs that are at risk for potential toxicity

• Explore if observed toxicity is reversible

• Identify safety parameters for clinical monitoring

The other important aspects of Preclinical development are related to the formulation development, manufacturing and procurement of clinical samples of experimental drug.

7 Clinical Development

Clinical programme is a set of studies designed to investigate the benefits and risks of a specific drug in human subjects and patients. While preclinical testing provides basic evidences about drug’s safety and efficacy, it cannot replace studies focused on drug interaction with human body. Therefore, the results of clinical trials represent the crucial set of data to support registration and eventual use of future drug in clinical practice.

Human testing of experimental drugs, before drug is approved for consumer sale, is typically conducted in three phases. Each phase is regarded as separate trial, and after completion of each of them, investigators are required to submit results and obtain approval from the regulatory body to progress into next phase.

Phase I of clinical trials is a first point at which an experimental therapy is administered to people and represents initial testing in a small number (usually 100 or less) of healthy volunteers. These trials are primarily focused on assessing if drug is safe for use in human subjects. In addition to monitoring of side effects, the scope also includes evaluation of how the drug is absorbed, metabolised and excreted, as well as to determine the safe dose range for use in the next phases.

Studies on healthy volunteers can also provide very valuable information linked to “proof of mechanism (PoM) in humans”, confirming that drug, once administered to humans, at particular dose and dosing schedule, triggers drug mechanism-related cascade of events.

Phase II studies evaluate safety and efficacy in a small number of patients (typically 100–300). The most Phase II studies are designed in the way in which the study group of patients who receives the experimental drug is compared with patients receiving either inactive substance (placebo), or a drug that is considered as the standard of therapy. Many of these studies are “double-blinded”, meaning that neither the patients nor investigators have information on who has received the investigated drug. Besides information about drug behaviour, safety of therapy, side effects and potential risks, the scope of trials is to determine the most effective dosing scheme for the experimental therapy as well as optimal method of delivery.

Phase II clinical trials are usually divided in Phase IIA studies, exploratory (non-pivotal) studies that have clinical efficacy, pharmacodynamics or biological activity as primary endpoint, and Phase IIB studies, that are definite dose ranging studies in patients with efficacy as primary endpoint. These days, results of Phase IIA are considered as “proof-of-concept (PoC) in humans”, and represent a major go-non-go milestone in drug development process.

Phase III studies are to demonstrate safety and efficacy of experimental therapy in a large group of patients. The goal of these pivotal studies is to generate statistically significant evidence of safety and efficacy, as required for approval. The goal is also to establish overall risk–benefit relationship of investigational medicine. Finally, the scope of Phase III studies also includes studies that will serve as a basis for labelling instructions and post marketing commitments.

Phase III studies usually enrol thousands of patients across multiple clinical sites around the world, and represent the largest investment in the whole drug discovery and development process.

8 Non-clinical Development

In parallel to the clinical development programme a number of non-clinical activities are performed with aim to support clinical trials and to generate data for approval. Some of these activities represent continuation of activities that have been initiated during preclinical development, and other follow the project evolution. Non-clinical activities can be divided in four major categories: safety, DMPK, chemical/substance development and pharmaceutical development (Table 4).

Table 4 Non-clinical development activities

Drug development is exceptionally complex; lengthy and costly process in which project planning and project management play the role of utmost importance. Although clinical part of the project progresses mostly in consecutive manner, supporting extensive and complex set of non-clinical activities has to be carefully planned and executed in order to ensure fast progression, informed decision-making at milestone points, and stringent cost control.