Imtesty,

Thanks for posting that article. It provides a great example of two different types of BET inhibitor selectivity: Bromodomain (BD)-selectivity and BET protein selectivity. BET proteins (i.e. BRD2, BRD3, BRD4 and BRDT) each contain two BDs (BD1 and BD2). Figure 6A of the Cell paper shows the binding affinities of different BET inhibitors for BD1 and BD2 of two different BET proteins: BRD2 and BRD4. They compare two novel BRD4-selective BET inhibitors, ZL420 and ZL454, against the pan-BET inhibitor JQ1 and the BD2-selective BET inhibitor apabetalone (aka RVX-208). 

One type of selectivity is BD-selectivity. Apabetalone is a BD-2 selective BET inhibitor that binds BD2 with greater affinity than BD1. Within the same BET protein molecule, apabetalone will bind BD2 more strongly than BD1. However, apabetalone does not seem to discriminate amongsts BRD2-BD2, BRD3-BD2, BRD4-BD2 or BRDT-BD2. The Picaud et al 2003 PNAS paper details this lack of BET protein selectivity. Resverlogix and/or Zenith have also mentioned in previous presentations that they have BD1-selective inhibitors in their library too. Also recall that Resverlogix/Zenith have another published BD2-selective compound (RVX-297) that is effective for acute inflammation and auto-immune disorders. To my knowledge, only one other company has a BD-selective BET inhibitor in clinical trials. Abbvie has ABBV-744, a VERY BD-2 selective BET inhibitor, in clinical trials for onocology. However, not all BD2-selective BET inhibitors are created equal. Different inhibitors can have different structures, bind to different regions of the BD acetyl-lysine binding pocket, elicit different conformational shifts in the BET protein upon binding, have different affinities for BD2 vs. BD1 and have different pharmacokinetic/tissue distribution properties. In other words, different BD2-selective inhibitors may have significant or very little overlap in the tissues exposed, genes/proteins modulated, and diseases affected. 

The other type of selectivity is BET protein selectivity. BRD2, BRD3, BRD4 and BRDT are all BET proteins that each contain a BD1 and a BD2. ZL420 and ZL454 are novel BRD4-selective inhibitors that bind BRD4 with much greater affinity than BRD2. I'm not sure if they tested for BRD3 or BRDT. However, ZL420 and ZL454 bind both BD1 and BD2 of BRD4 with high affinity; they do not seem to have BD-selectivity. As mentioned above, apabetalone does not seem to have BET protein selectivity, but it does have BD-2 selectivity.

Pan-BET inhibitors are not selective. A pan-BET inhibitor like JQ1 binds with equal affinity for either BD1 or BD2, and bind equally well to different BET proteins. In other words, pan-BET inhibitors like JQ1 as well as Zenith's ZEN-3694 and pretty much every other BET inhibitor in oncology with the exception of ABBV-744 bind and inhibit without selectivity to either BD1 or BD2, and without selectivity for a specific BET protein.