The development sequence, and where nonclinical work fits

In the United States, the clinical investigation of a previously untested drug is generally divided into three phases, which the regulation notes may overlap.^[S2] Those trials are conducted at investigative sites under the oversight of an institutional review board, whose purpose the regulation states as protecting the rights and welfare of the human subjects involved, and with informed consent obtained under a separate part of the same title.^[S4] None of that begins until the nonclinical work is complete and submitted.

From the start of development to the end takes about 10.5 years on average, made up of roughly 31 months of nonclinical work followed by roughly 95 months of clinical trials, according to the Congressional Budget Office.^[S3] The nonclinical portion is the smaller share of that time, but everything downstream depends on it being accepted.

The sequence

Discovery comes first. Chemical and biological approaches are used to identify compounds that act on the target of interest, and the output of this stage is a candidate, or a small set of them, to take forward.

Nonclinical studies come next, in laboratory animals and in vitro. Their regulatory purpose is specific: the application to begin clinical work must contain adequate information about the pharmacological and toxicological studies, on the basis of which the sponsor has concluded that it is reasonably safe to conduct the proposed clinical investigations.^[S1] That is the test the nonclinical package has to meet. Data from the clinical phases that follow support a new drug application or, for a biological product, a biologics license application.

What the nonclinical package has to establish

The nonclinical package answers a defined question rather than an open one: is there adequate pharmacological and toxicological information to support a conclusion that the proposed clinical investigation is reasonably safe to start.^[S1] In practice that means characterizing what the compound does, at what exposure, in which species, and what happens at doses above the intended clinical range. Studies that support this conclusion are conducted under the good laboratory practice regulation, which applies study by study.^[S6]

What nonclinical data can and cannot do is worth stating plainly. They establish the exposures at which effects and adverse findings occur in the species tested, which is what the starting dose in humans is derived from. They do not establish that a compound will be effective in patients, and they do not guarantee that adverse effects seen in humans will have appeared in animals first. The most common failure in development is a compound that worked in a disease model and did not work in patients, and no amount of nonclinical rigour removes that possibility entirely.

Summary

Development runs from discovery through nonclinical studies to three clinical phases, and takes about a decade in total.^[S3] Each stage answers a different question, and the regulation defines what the nonclinical stage has to deliver.^[S1][S2]

For a sponsor placing nonclinical work, the practical consequence is that the studies have to be designed against the regulatory requirement rather than against a general standard of thoroughness. The question to ask of any proposed study is which part of the application it supports, and whether it will be conducted under the good laboratory practice regulation.^[S1][S6]