The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants poses an ongoing threat to the efficacy of vaccines and therapeutic antibodies. Mutations predominantly affect the receptor-binding domain (RBD) of the spike protein, which mediates viral entry. The RBD is also a major target of monoclonal antibodies that were authorised for use during the pandemic. In this study, an in silico approach was used to investigate the mutational landscape of SARS-CoV-2 RBD variants, including currently circulating Omicron subvariants. A total of 40 single-point mutations were assessed for their potential effect on protein stability and dynamics. Destabilising effects were predicted for mutations such as L455S and F456L, while stabilising effects were predicted for mutations such as R346T. Conformational B-cell epitope predictions were subsequently performed for wild-type (WT) and variant RBDs. Mutations from SARS-CoV-2 variants were located within the predicted epitope residues and the epitope regions were found to correspond to the sites targeted by therapeutic antibodies. Furthermore, homology models of the RBD of SARS-CoV-2 variants were generated and were utilised for protein-antibody docking. The binding characteristics of 10 monoclonal antibodies against WT and 14 SARS-CoV-2 variants were evaluated. Through evaluating the binding affinities, interactions, and energy contributions of RBD residues, mutations that were contributing to viral evasion were identified. The findings from this study provide insight into the structural and molecular mechanisms underlying neutralising antibody evasion. Future antibody development could focus on broadly neutralising antibodies, engineering antibodies with enhanced binding affinity, and targeting spike protein regions beyond the RBD.