Chemistry Reference
In-Depth Information
the early stages of drug development, reducing costs and time, driving researchers
to compounds with better viability to debut as a drug [1].
The development of a new drug starts with around 10 thousand compounds and
10% (1,000) of these compounds have good ADMET and safety properties for
clinical trials. 1% (10) of these 1,000 may be approved in clinical trials, becoming
drug candidates, and around 10% of this, 10 drug candidates, get to the market as
a drug after regulatory agencies approval. These long years of work do not
guarantee their success due to constant withdrawal of drugs from the market
resultant from unacceptable side effects (in some cases, death) that are not
identified in clinical trials [2].
The ADME/Tox properties enables adjustment of the therapeutic profile of a new
compound, especially during its development. The absorption properties can
define if the compound will be used orally, by intramuscular injection or by other
route. Likewise, the distribution affects the efficiency of this compound, since it
must reach the site where it is supposed to act. With an adequate bioavailability,
the metabolism, excretion and toxicity properties of a drug help to determine the
efficiency, drug administration intervals, interaction with other substances and
side effects. All these properties can shape or even stop the development of a new
compound as a drug [3].
In silico
predictions have become a very attractive approach in order to model and
evaluate ADME/Tox properties when compared to traditional experimental
methods due to its lower cost and faster data prospection.
In silico
data can be
evaluated even before the compound's synthesis or biological assays, avoiding the
ethical barriers of biological trials with animals, saving time and money in the
evaluations of compounds that would not reach the market [4].
The major disadvantage of current
in silico
methods is the dependence on a solid
data set obtained from experimental procedures. Such data is hard to obtain due to
the wide complexity and the diversity of biological systems, nullifying substantial
data obtained with animal models or
in vitro
assays (equivalent to humans) [1, 3,
5, 6].
