In the present molecular modelling study, recently discovered 36 selective urea derivatives were considered to develop pharmacophore based 3D-QSAR model coupled with quantum mechanics (QM) calculations to uncover the essential structural features of urea molecules for mPGES-1 inhibition. The 3D-QSAR model was selected on the basis of highest values of external predictability parameters i.e. Q2 (0.775) and Pearson-r (0.912). The model also showed the highest values of R2, 0.985; F-value, 306.3 and least SD, 0.147. The selected model was further validated for its external prediction power by calculating κ, κ′, Ro2 and R′o2. The contour maps generated against the selected QSAR model helped to interpret the important molecular sites of urea derivatives where the suitable structural modifications would help in better complementary fit to the active site of mPGES-1, in turn would improve the potency of newly designed molecules.