We investigate a new method (QTA-PIMC) for global optimization on complex potential energy surfaces which combines the path integral Monte Carlo method with quantum and thermal annealing. This method is applied to the BLN protein model (Honeycutt, J. D.; Thirumalai, D. Biopolymers 1992, 32, 695). We show that this new approach outperforms simulated (thermal) annealing (SB) and that in fact SA is a subset of our method. This means that we could invoke the effects of both quantum and/or thermal annealing in the global optimization process, hence the name quantum thermal annealing (QTA-PIMC). The simulation results also suggest that QTA-PIMC scales better than SA in terms of system size in the search for the global minimum.