What is the drug discovery process?

(6.6min read)

Why do we make medicines?  Medicines are needed to treat, cure, and prevent medical conditions.  If there are no medicines, academic and industrial partners engage in the expensive and lengthy process that is known as drug discovery (Figure 1).  It can take around 12-15 years to develop a new medicine, and involves scientists from many disciplines.  In this article we will highlight the different stages of the pre-clinical drug discovery process along with who is involved at each stage. 

 

Figure 1 Drug discovery process. Discovery science typically takes place prior to start of a drug discovery process with academic institutes identifying the therapeutic target. Area in red indicates pre-clinical drug discovery process and the highlight of this article.

 

Drug discovery begins when there is an unmet medical need.  Typically, biologists identify a mechanism that can either slow and/or cure the disease in question (Discovery Science).  Once a therapeutic target is identified, it is checked to ensure it meets the requirements for drugability, then a drug discovery project can begin. 

The process (Figure 2) begins by identifying a suitable chemical starting point, through the process known as “screening”.  Utilising target and phenotypic assays, screening is the process that involves monitoring the effect of a disease target when exposed to a chemical compound.  Biochemists and/or cellular biologists will screen chemical start points (usually > 10 000 compounds curated into libraries).    The aim of screening these compounds is to find hits (“hit generation”), which are compounds that possess activity against the disease but will require optimisation to improve this activity. 

 

To find a “hit” compound, libraries of chemical starting points are screened against the disease target (or against an infected cell or a cell with a suitable reporter for the target of interest. When using target-based screening, active compounds should be screened in a cell-based model to ensure the desired phenotype has been achieved.
Figure 2 To find a “hit” compound, libraries of chemical starting points are screened against the disease target (or against an infected cell or a cell with a suitable reporter for the target of interest. When using target-based screening, active compounds should be screened in a cell-based model to ensure the desired phenotype has been achieved.

 

Medicinal chemists, who have a background in organic chemistry, will triage the hits to ensure only the best compounds are taken forward for further development.  The chemist will ensure the compounds do not have flags for toxicity (software tools can assist in identifying PAINS, AMES, hERG etc), ability to be synthesised (including derivatives, and in usually, free from intellectual property claims.  Typically, the number of hits taken forward is low compared with the number of compounds screened (excess of 5 chemical series). 

The drug discovery team will undergo a “design, make and test” cycle in the “hit-to-lead” phase, (Figure 3), changing the compounds to meet the desired target candidate profile.  In the test phase, drug metabolism and pharmacokinetic (DMPK) in vitro scientists will begin screening the more promising compounds.  These experiments assist the scientists in determining if their compound will be soluble, absorbed, and stable when taken into the human body.  Only compounds that meet these criteria are suitable for next phase, the pharmacokinetic (PK) studies where the compound will be tested in its first living system (often a mouse or rat). 

 

Figure 3 A typical process to identify a compound suitable to reach in vivo studies. Areas highlighted in green are unique to structure-based programmes where the project has begun from target-based screening.

 

In vivo DMPK scientists work with healthy mammalian models to determine what the body does to the compound.  Will the compound reach the systemic circulation (bioavailability if dosing orally)?  How long is the compound in the circulation for (half-life)?  A PK model MUST be performed prior to the compound being dosed in an efficacy model (mammalian host infected with disease of study). This way if the compound is not efficacious it would not be down to its PK properties but simply it does not work. 

This stage of the process can reveal PK issues and the compound(s) must be passed back to the medicinal chemists for further optimisation.  This moves the project into the lead optimisation phase.  The design, make, test cycle will continue once more but this time there will be greater emphasis on the in vitro DMPK to assist in overcoming the PK issues.  These experiments may differ from the previous round, and often cost significantly more.  Although lead optimisation does require further in vivo experiments, the use of animals is still kept low by utilising a number of in vitro DMPK experiments.. As the compound makes its way into pre-clinical development, the compound will be subject to a battery of toxicological studies to ensure its safety e.g. when dosed for long periods of time, in combination with other medications, or maximum dosage.  Only once a compound has reached this stage can it be considered for clinical trials. 

Approximately 1 in 100 drug discovery projects make it to the market, pushing costs for new medicines to over $1 billion.  Projects often fail due to unforeseen efficacy reasons and/or safety.  But if scientists follow the drug discovery process, it will assist in identifying those projects that are worth pursuing and those that should be halted or abandoned.