Conventional genotoxic anti-cancer drugs target the proliferative advantage of tumor cells over normal cells. This kind of approach lacks the selectivity of treatment to cancer cells, because most of the targeted pathways are essential for the survival of normal cells. As a result, traditional cancer treatments are often limited by undesirable damage to normal cells (side-effects). Ideal anticancer drugs are expected to be highly effective against malignant tumor cells with minimal cytotoxicity toward normal cells. Such selective killing can be achieved by targeting pathways essential for the survival of cancer cells, but not normal cells. As cancer cells are characterized by their resistance to apoptosis, selective apoptosis induction is a promising approach for selective killing of cancer cells. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is a promising tumor-selective anticancer drug. However, the congenital and acquired resistance of some cancer cell types, including malignant melanoma cells, currently impedes effective TRAIL therapy, and an innovative approach that can override TRAIL resistance is urgently required. Apoptosis is characterized by cell shrinkage caused by disruption of the maintenance of the normal physiological concentrations of K+ and Na+ and intracellular ion homeostasis. The disrupted ion homeostasis leads to depolarization and apoptosis. Recent evidence suggests that depolarization is an early and prerequisite event during TRAIL-induced apoptosis. Moreover, diverse natural products and synthetic chemicals capable of depolarizing the cell membrane exhibit tumor-selective killing and TRAIL-sensitizing effects. Here, we discuss the role of depolarization in selective killing of cancer cells in connection with the emerging concept that oxidative stress is a critical mediator of mitochondrial and endoplasmic reticulum dysfunctions and serves as a tumor-selective target in cancer treatment.