The Proper Orthogonal Decomposition (POD) technique has been widely applied to Computational Fluid Dynamics (CFD) formulation to obtain reduced-order model for unsteady aerodynamic applications. However, it is worth point out that the robustness and accuracy of these reduced-order models are strictly related to the reference parameters from which POD modes have been derived. Thus any variation of these parameters (Mach number, angle of attack...etc) requires a new computation of the ROM which devalues the effectiveness of the POD method in industrial application. The objective of this paper is to analyse different ROM adaptation scheme describing critically the advantages and the drawback of the different approaches and their application in fluid-structure interactions problems. After a general presentation of the application of the POD method to the linearized Euler equations, the mathematical formulation of the interpolation schemes is presented. First a simple test on a 2D wing section is performed in subsonic and transonic behaviour. Finally the performances of the different interpolation strategy are tested on an industrial test case.