I’m not familiar enough with Poser but the thing about DCC is the ability to move assets from app A to app B with little or no loss to design fidelity so as long as app A plays by the rules of the FBX protocol then it would be a edit solution worth looking into.
So if an app has a usable tool set then the first step is to see if it does have a suitable DCC export ability.
The need for optimization though should be the responsibility of the target application, in this case the Unreal 4 closed edit environment, and with Auto LOD for static mesh in 4.14 it can be assumed that Auto skeletal mesh LOD should be made available in a future release. In general a closed edit environment is well suited for controlling the kind of outputs needed as to asset fidelity and performance so what ever app B can do as far as a procedural process could and should be added to the feature request list and preformed in UE4.
Personal opinion wise though I would not feel comfortable recommending Poser as part of a development pipeline due to the lack of community support or company driven promotion and taking into consideration the power of DS alone to do other things with it, that does not cost a cent as far as start up costs goes, at $175 dollars it seems to me to be a case of buying bottle water.