We have previously reported a series of μ-opioid receptor (MOR) agonist/δ-opioid receptor (DOR) antagonist ligands to serve as potential nonaddictive opioid analgesics. These ligands have been shown to be active in vivo, do not manifest withdrawal syndromes or reward behavior in conditioned-place preference assays in mice, and do not produce dependence. Although these attributes are promising, these analogues exhibit poor metabolic stability in mouse liver microsomes, likely due to the central tetrahydroquinoline scaffold in this series. As such, a structure-activity relationship (SAR) campaign was pursued to improve their metabolic stability. This resulted in a shift from our original bicyclic tetrahydroquinoline core to a monocyclic benzylic-core system. By eliminating one of the rings in this scaffold and exploring the SAR of this new core, two promising analogues were discovered. These analogues (5l and 5m) had potency and efficacy values at MOR better or comparable to morphine, retained their DOR-antagonist properties, and showed a 10-fold improvement in metabolic stability.