I think anisotropy may be used for quantitative ligand binding study. If you consider the background mathematics of binding analysis for a simple protein ligand system (see the attachment), you see that we need a measurable parameter that will quantify, for say, a ratio of bound ligand per mole of receptor ([ligand]bound/[Protein]total).
We have not used it for quantification/calculation of binding constants; but in the following paper we have got a trend in anisotropy change which suggests we can use anisotropy change as well for binding study.
It is commonly used for binding studies (see the protein binding database (http://www.bindingdb.org/bind/index.jsp) which allows you to query based on binding technique). It is usually referred to as a fluorescence polarization assay.
Largest issues are need to have a fluorescent ligand, detecting weak interactions, and problems with non-specific binding (the last two are common to most assays)