Pharmaceutics is the overall process of developing a new chemical entity into an approved therapeutic that is safe and effective to treat or prevent a disease. It is a complex process requiring multiple scientific, medical, legal, commercial and regulatory expertise. On average, it typically takes at least ten years for a new drug to complete the process from initial discovery to the retail marketplace, with clinical trials alone taking six to seven years on average. The Tufts Center for the Study of Drug Development announced it had calculated that it costs pharmaceutical companies $2.6 billion to develop a new drug.[1] This figure has been subject to question, as the raw numbers on which they base the analysis have not been made available. The driving factor of this estimate is the cost of failed drug development attempts, with 80% of new compounds discontinued during their development. The steps involved in bringing a new chemical entity to the market as a drug involve the stages of discovery, chemical optimization and production, preclinical testing, pharmacology and toxicology testing, early clinical trials establishing proof of concept, registration or pivotal clinical trials showing safety and efficacy in the target patient population and finally regulatory approval.

The discovery phase of the pharmaceutics process involves establishing a receptor, enzyme, protein, RNA, DNA or protein that modulates a biochemical process, as a potential contributor to the pathophysiology of a disease process. They often base this on literature reviews of basic science investigating disease mechanisms. For example, the immune regulating receptor PD-1 and its ligand PD-L1 were implicated in reducing the immune system's response to cancer.[2]The next stage in the development of PD-1 therapy was then to develop monoclonal antibodies targeted against an area of those proteins that would assure good binding and inhibition of effect. They established in-vitro and in-vivo models to show antibody binding and reactivation of the immune system towards a cancer model. For traditional small molecules, a similar process occurs but usually with a screening of activity against a target (typically a receptor or enzyme) against a library of compounds. Enzyme inhibitors such as kinase inhibitors against  BCR-ABL1 in chronic myelogenous leukemia (CML) and acute lymphoblastic leukemia with the Philadelphia chromosome are just one example.[3]

The chemical optimization and production phases focus on improving the drug and manufacturing characteristics of the drugs found in the discovery phase. With antibodies, for example, this would include the selection of the proper sub-type of antibody to prolong its half-life in the body and to enhance or suppress complement fixation or antibody-dependent cell cytotoxicity (ADCC).[4]This would also be the step to determine what type of cell and bio-incubator should be used to produce the product. Chemical optimization for small molecules involves changing the chemical structure such that it enhances the therapeutic effect while minimizing the interaction of the drug with CYP450s and transporter protein and HERG or sodium channels involved in cardiac conduction. At this stage is when screening for interactions across a large panel of receptors and transporters takes place, and when patents get submitted. An important issue regards the use of the excipients, often considered as inert, but these have caused problems both in the pediatric and adult population.

Preclinical testing, pharmacology, and toxicology testing involve using models of the disease tested. Often these animal models harbor close approximations of the human disease, such as human tumor implants in immune-incompetent mice (xenograft tumor models) or are a relatively poor approximation of human pathologies, such as the forced swimming test. Standard pharmacology models involving testing for cardiovascular, CNS and renal effects are performed in in-vitro and in-vivo models such as I(Kr) potassium channel antagonism and prolongation of the QT interval is assessed in telemetered dogs. These are often not required for targeted drugs such as antibodies. Toxicology testing involves standard protocols with doses and duration’s that extend beyond the expected human dosing (in animal lifetimes). These are often used to determine the first dose used in human trials and determine any particular toxicity concerns that will require monitoring in the early clinical trials. 

Initiation of early clinical trials after establishing proof of a concept commence after an Investigational New Drug Application (IND) has been reviewed and approved by the FDA. Similar procedures are used in the EMA, Japan, and other countries; although some nations limit the testing of new drugs for the first time. The initial studies in humans can be in actual patients if the toxicity of the drug or its target would not be appropriate to use in volunteers. Increasing doses of drug are used to establish the drugs' safety and determine the drugs' pharmacokinetics (the duration and extent of the drug in the body) as well as measure the drugs' effect on biomarkers (serum or tissue enzymes, histology, or any other measure of drug effect such as glucose in diabetes). This data will be used to select doses to be brought into a proof of concept or phase 2 studies in a select patient population. Clinical trials like the TeGenero disaster, are fortunately the exception, and not the rule. In this study, the first volunteers received a dose that was 500 times smaller than the safe dose as established in animal studies. However, all six human volunteers faced life-threatening conditions, involving multiorgan failure, which required transfer to the intensive care unit. As a result, the implementation of new precautions and toxicity calculations became standard since that 2006 incident.[5]

Registration or pivotal clinical trials demonstrate safety and efficacy in the target patient population. These trials may encompass thousands of patients, or only a small number (as in ultra-rare diseases). The outcomes vary between pathologies and range from overall survival (in oncology indications), to a reduction in the incidence of asthma exacerbations requiring hospitalizations, to effects on a biomarker like HbA1C in patients with diabetes. Patient selection is crucial in order to match the drug effects with the target patient population, but the data from the studies must also be able to be extrapolated to the broader patient population. Endpoint selection is crucial in that it must reflect a meaningful alteration in patient outcomes but also be statistically achievable with the study design. These studies often take several years to complete and analyze.

Once the entire drug development package is complete with reports from discovery, preclinical, chemistry, manufacturing, and clinical a regulatory submission takes place that integrates all of this data in a standard format. The submission can be many terabytes of data. Regulatory approval can take around a year with often an outside review by experts (FDA Advisory Board) and a public hearing. If all goes well the drug receives marketing approval and becomes available for patients.