Chemistry Reference
In-Depth Information
2.4 Phase III Activities
2.4.1 Evaluation of Fragments
The key to any successful drug discovery effort, whether lead like or fragment based, is
the evaluation of the screen hits and the development of a path forward. There are two
main components of this evaluation: (1) compositional SAR and (2) ligand efficiency.
Compositional SAR is the traditional SAR that most chemists are familiar with: how do
changes in the hit molecules correlate with changes in their potency? FBDD has several
advantages over LLDD in developing compositional SAR. First, the number of compounds
typically screened is in the range of a few hundred to a few thousand, which is smaller
than most typical lead-like libraries.
[
58, 90 92
]
This make evaluation simpler than for lead-
like libraries. One of the important aspects that is often overlooked in SAR evaluation
is the SAR of what was not a hit (inverse SAR). With relatively small libraries, both
compositional SAR and inverse SAR can be evaluated fairly quickly. The first point for
evaluating SAR is to ensure that compounds (both hits and non-hits) pass quality stand-
ards. There are many anecdotal accounts of SAR being developed when the original hit(s)
had poor quality (wrong compound, impurity, etc.). It is also at this time that the hypo-
theses generated during compound assessment are evaluated. This is important because
of the highly iterative nature of FBDD. If the output of the screen does not meet pre-
conceived notions it does not mean that it is untrue. Serendipity is the best friend of
drug discovery.
Fragment libraries are typically smaller because lower complexity compounds
(fragments) have a higher probability of matching a target protein-binding site, result-
ing in higher hit rates.
[
5
]
Even with much smaller libraries, the higher hit rate means that
data for a fewer number of hits have to be scrutinized. Lower complexity compounds have
their own inherent problems in developing SAR, i.e. how does one quantify the complexity
of non-complex molecules?As discussed in Chapters 8 and 9, there are abundant and robust
computational tools available to help in this process.
So, what can be expected from general screening of fragment-like molecules? It has
recently been shown that the observed hit rates for fragment screens were 10-1000 times
higher than conventional biochemical screens.
[
58
]
This bodes well for 'blind' screening of
orphan, undruggable or unvalidated targets. Thus, even from a small library, there are usu-
ally sufficient data to generate compositional SAR or hypothesize SAR, that can be rapidly
evaluated by further virtual and/or real-world screening. It is at this point that orthogonal
screens are typically initiated. 'Nested' libraries have their greatest impact during itera-
tion and in generating further SAR. Compositional SAR can thus be rapidly generated and
evaluated, but again, this can only be carried so far without intense medicinal chemistry
efforts. Culling of fragments for elaboration becomes an important step in the process.
Ligand efficiency.
Ligand efficiency is the other main component for the evaluation of
FBDD efforts. When multiple lines of SAR exist to be followed up, it makes the most
sense to prioritize highest the most ligand-efficient compounds.
[
93
]
In order to evaluate or
rank fragments or any compound, ametric called 'ligand efficiency'has been proposed.
[
93, 94
]
There exists a plethora of ligand efficiency evaluators,
[
94
]
but they all work on the same basic
principle, the logarithm of the activity divided by molecular size. Molecular size can be
molecular weight or heavy atom (i.e. non-hydrogen) count (HAC). We prefer HAC because
