The integrated thermal protection system (ITPS) is a complicated system that addresses both mechanical and thermal considerations. An M-pattern folded core sandwich panel packed with low-density insulation material provides inherently low mass for a potential ITPS panel. Herein, we identify the most influential geometric parameters and establish a viable, computationally efficient optimization procedure. Variables considered for optimization are geometric dimensions of the ITPS, while temperature and deflection are taken as constraints. A one-dimensional (1D) thermal model based on a modified form of the rule of mixtures was established, while a three-dimensional (3D) model was adopted for linear static analyses. Parametric models were generated to facilitate a design of experiment (DOE) study, and approximate models using radial basis functions were obtained to carry out the optimization process. Sensitivity studies were first conducted to investigate the effect of geometric parameters on the ITPS responses. Then optimizations were performed for both thermal and thermal-mechanical constraints. The results show that the simplified 1D thermal model is able to predict temperature through the ITPS thickness satisfactorily. The combined optimization strategy evidently improves the computational efficiency of the design process showing it can be used for initial design of folded core ITPS.