A large number of
benzodiazepine derivatives have been synthesised and their
structure-activity relationships explored in detail. This chart contains binding data for benzodiazepines and related drugs investigated by
Roche up to the late 1990s (though in some cases the compounds were originally synthesised by other companies such as
Takeda or
Upjohn). Other benzodiazepines are also listed for comparison purposes, but it does not however include binding data for; • Benzodiazepines developed in the former Soviet Union (e.g.
phenazepam,
gidazepam etc.) • Benzodiazepines predominantly used only in Japan (e.g.
nimetazepam,
flutoprazepam etc.) • 4,5-cyclised benzodiazepines (e.g.
ketazolam,
cloxazolam etc.), and other compounds not researched by Roche • Benzodiazepines developed more recently (e.g.
remimazolam,
QH-ii-066,
Ro48-6791 etc.) • "Designer" benzodiazepines for which
in vitro binding data are unavailable (e.g.
flubromazolam,
pyrazolam etc.) While binding or activity data are available for most of these compounds also, the assay conditions vary between sources, meaning that in many cases the values are not suitable for a direct comparison. Many older sources used animal measures of activity (i.e. sedation or anticonvulsant activity) but did not measure
in vitro binding to benzodiazepine receptors. See for instance Table 2 vs Table 11 in the
Chem Rev paper, Table 2 lists
in vitro pIC50 values matching those below, while Table 11 has pEC50 values derived from
in vivo assays in mice, which show the same activity trends but cannot be compared directly, and includes data for compounds such as
diclazepam and
flubromazepam which are not available in the main data set. Also note; • IC50 / pIC50 values represent binding affinity only and do not reflect efficacy or pharmacokinetics, and some compounds listed are GABAA antagonists rather than agonists (e.g.
flumazenil). • Low IC50 or high pIC50 values indicate tighter binding (pIC50 of 8.0 = IC50 of 10nM, pIC50 of 9.0 = IC50 of 1nM, etc.) • These are non subtype selective IC50 values averaged across all GABAA receptor
subtypes, so subtype selective compounds with strong binding at one subtype but weak at others will appear unusually weak due to averaging of binding values (see e.g.
CL-218,872) • † indicates a predicted value from
in silico modelling. • Finally, note that the benzodiazepine core is a
privileged scaffold, which has been used to derive drugs with diverse activity that is not limited to the GABAA modulatory action of the classical benzodiazepines, such as
devazepide and
tifluadom, however these have not been included in the list below. 2,3-benzodiazepines such as
tofisopam are also not listed, as these act primarily as
AMPA receptor modulators, and are inactive at GABAA receptors. == Table ==