We present a new idea of testing the validity of the Friedman-LemaƮtre-Robertson-Walker metric , through the multiple measurements of galactic-scale strong gravitational lensing systems with type Ia supernovae in the role of sources . Each individual lensing system will provide a model-independent measurement of the spatial curvature parameter referring only to geometrical optics independently of the matter content of the universe . This will create a valuable opportunity to test the FLRW metric directly . Our results show that with hundreds of strongly lensed SNe Ia observed by LSST , one would produce robust constraints on the spatial curvature with accuracy \Delta \Omega _ { k } = 0.04 comparable to the Planck 2015 results .