Reliability-based optimization addresses the cost-effective integrity design of structures in the presence of inherent uncertain parameters. Processing this class of problem is challenging from the computational burden to determine the failure probability of structures violating the limit-state function. This paper presents the novel reliability-based optimization approach that advantageously couples a comprehensive learning particle swarm optimization (CLPSO) algorithm with a Gaussian process regression (GPR) model. In essence, the proposed method iteratively performs the CLPSO with deterministic parameters based on the most probable point underpinning the limit-state function iterative updated by the reliability evaluation process. The GPR model approximates from the design data given by CLPSO the spectrum of the limit-state function under uncertain parameters, and hence enable the significant reduction of Monte-Carlo simulations for failure probability estimation. What is more is that a so-called expected feasibility function is maximized to systematically refine the GPR model by locating the new sampling points in the region with high-reliability sensitivity leading to the more accurate prediction of failure probability. The reliability-based optimization terminates as when the resulting failure probability reaches some allowable threshold. The CLPSO is primarily adopted in the optimization process for the GPR hyperparameters and the expected feasibility function. Some numerical examples are provided to illustrate the applications and robustness of the proposed schemes in performing the reliability-based structural optimization.