Biomechanical　CT　(BCT),　i.e.,　quantitative　computed　tomography-based　finite　element　analysis　(QCT-FEA),　promises　an　improved　technique　over　bone　mineral　density　(BMD)　in　predicting　bone　strength　and　the　risk　of　osteoporotic　vertebral　fractures.　However,　most　of　the　BCT　models　only　consider　a　uniform　compressive　loading　condition　and　they　have　not　been　validated　for　Chinese　subjects.　This　study　examined　the　ability　of　BCT　to　predict　wedge　fracture-related　vertebral　flexion　strength　in　a　cohort　of　Chinese　cadaveric　vertebrae.　Twelve　human　vertebrae　were　scanned　with　dual　energy　X-ray　absorptiometry　(DXA)　and　QCT　to　measure　areal　and　volumetric　BMD,　respectively.　To　produce　wedge　fractures,　the　cadaveric　vertebrae　were　experimentally　loaded　until　failure　under　a　15°　flexion.　Vertebral　flexion　stiffness　and　strength　were　measured　from　the　force-displacement　curve.　Voxel-based　heterogeneous　FE　models　of　the　vertebrae　were　created　and　virtually　tested　in　uniform　compression　and　15°　flexion　to　compute　compressive　and　flexion　strength　(and　stiffness),　respectively.　The　predictions　of　vertebral　flexion　strength　with　BMD　or　BCT　measures　were　evaluated　with　linear　regression　analyses.　Results　showed　weak　correlations　between　experimentally-measured　flexion　strength　vs.　DXA-aBMD　(R2　=　0.26)　or　QCT-vBMD　(R2　=　0.39).　However,　there　were　strong　correlations　between　experimentally-measured　flexion　strength　vs.　BCT-computed　vertebral　strength　under　either　flexion　(R2　=　0.71)　or　compression　(R2　=　0.70)　loading　conditions,　although　flexion　reduced　the　BCT-computed　vertebral　strength　by　9.2%.　These　results　suggest　that,　regardless　of　whether　a　uniform　compression　or　a　flexion　loading　is　simulated,　BCT　can　predict　in　vitro　vertebral　flexion　strength　better　than　BMD.　©　2022　IPEM