PROF (Prediction Refinement with Optical Flow) is a method to improve motion vectors by using gradient information. Since the optical flow method has the assumption of brightness constancy generally, it may not be useful when coding tools considering variable brightness are applied. In this contribution, some additional applicable conditions for PROF are proposed. Since DMVR and BDOF have been applied when bi-prediction with CU weights (BCW) selects default weights and the coefficients and offsets for the explicit weighted prediction (WP) is not existed, the same conditions are introduced to PROF. As an experimental result on CTC, the average of decoding times became 1% faster on LB configuration. The losses of coding efficiency were 0.03% on RA configuration and 0.02% on LB configuration.