A detailed Structure-Kinetic Relationship study
Here is a series of papers that I think are a very nice, detailed example of how you can use kinetics in your design. The work was published by Almirall researchers on their CRTh2 antagonist program a few years ago (BMCL2014_5111; BMCL2014_5118; BMCL2014_5123; BMCL2014_5127; EuropJMedChem2016) (As a side note, the sheer amount of synthetic work -especially on heterocycles- summarized in those papers is already noteworthy)
The authors really set out from the get-go to characterize the residence time of their compounds and this consitutes one of the most exhausitve body of published residence time data.
I guess the real take home message for me was that your SKR transfers from one series to the next (see below with the addition of a difluoromethyl group)
And you can also have some synergy (in this case it is all the more striking as the Kd does not really change while the residence time increases by a log unit)
As mentioned, there is a lot of data to look at and probably better examples then the ones I chose... definitely worth your time.
Of note, researchers from Novartis worked on related series and commented on the Almirall results (ACSMedChemLett2017) suggesting that the impressive dissociation half-lives were measured at room temp and could be less impressive when measured at body temp. With this in mind, it was nice to see sustained PD effect in vivo (PharmRes2016) with the Almirall compound LAS191859.
There is still a big debate over the relevance of optimizing residence time (DrugDiscovTodayTech2015; DrugDiscovToday2017 ;Peter Kenny's blog) and I must admit I have not had the opportunity to use this sort of data in a real live project to make up my mind on the subject but I think the Almirall papers are a very nice case study that should be tought in Medchem classes.
#drug #drugdiscovery #medicinal #medicinalchemistry
Retired, so I am no longer able to provide services on behalf of Walter Ward Consultancy & Training
7 年When different compounds are characterized in the same assay, often IC50 ∝ Kd = koff/kon For same binding site, compounds in same SAR cluster usually differ in koff, with similar kon values, so that a lower koff gives proportionally lower Kd (eg streptavidin, HIV protease, protein methyltransferase DOT1L, see Copeland RA, 2016, Nat Rev Drug Disc 15, 87). In my experience a > 10x change in kon suggests a possible break point in SAR (eg using a different binding pocket on the same target, or binding to a different conformation of the same pocket). For example, compounds which use the selectivity pocket (SP) of p38α bind over 30-fold more slowly than those which bind only in the purine site. Filling the SP gives the DFG-out conformation & prevents activation by (PoA) , linked with better transfer of potency into cells, because ATP competes more weakly in PoA rather than inhibition of activated p38 (Sullivan JE et al 2005, Biochemistry 44, 16475).
Scientific Leader | Drug Hunter | Financial Educator
7 年Some beautiful examples are given in this overview! Sure, as suggested in the commentary that accompanied the original article, the T1/2 might not be so impressive at body temperature. However, shifting T1/2 by 10X at room temp must provide a boost in similar comparisons at body temp.
Freelancer and Independent Interdisciplinary Science Consultant
7 年We need to season our insatiatiable desire for "good" news and shallow positivity with an proper understanding of failure.
You also need too be thinking about the consequences of reducing on-rates if your focus is drug discovery. You might find this article to be informative: https://doi.org/10.1016/j.drudis.2017.07.016
Principal Scientist at Astrazeneca
7 年Great post thxs!