If you can attach a low-molecular weight visible-wavelength absorbance dye or radioactive label to the protein, you can use that to quantify it in the released sample. I would not recommend using dye fluorescence for this application because the antibody binding could quench the fluorescence (the dye could be a fluorescent one, but you would use the absorbance of the dye, not the fluorescence). The only risk is that attachment of the label could affect antibody binding. Ideally, the label would be attached somewhere distant from the antibody epitope. It would also be good if you could assure that all protein molecules have the same number of label molecules attached, to avoid bias. The best way to achieve this is to label all surface accessible Cys residues with a maleimide- or iodoacetamide-based reagent.

Dear Adam. Nice to hear from you again :)

Is is possible for you to share a few works ?

Secondly, is it possible to recapture the released antigen - antibody conjugate via antibody attachment to a modified surface and then follow the sandwich ELISA protocol to access the concentration of the protein ?

I don't know of any references for the method I suggested. However, you can purchase kits for labeling proteins with dyes from various companies, such as the Molecular Probes division of Thermo-Fisher.

Your idea of using a sandwich ELISA sounds like it could work if you have two antibodies with different, non-interfering epitopes and from different animal sources.