The incidence of melanoma is increasing over the years with a still poor prognosis and the lack of a cure able to guarantee an adequate survival of patients. Although the new immuno-based coupled to target therapeutic strategy is encouraging, the appearance of targeted/cross-resistance and/or side effects such as autoimmune disorders could limit its clinical use. Alternative therapeutic strategies are therefore urgently needed to efficiently kill melanoma cells. Ferroptosis induction and execution were evaluated in metastasis-derived wild-type and oncogenic BRAF melanoma cells, and the process responsible for the resistance has been dissected at molecular level. Although efficiently induced in all cells, in an oncogenic BRAF- and ER stress-independent way, most cells were resistant to ferroptosis execution. At molecular level we found that: resistant cells efficiently activate NRF2 which in turn upregulates the early ferroptotic marker CHAC1, in an ER stress-independent manner, and the aldo-keto reductases AKR1C1 ÷ 3 which degrades the 12/15-LOX-generated lipid peroxides thus resulting in ferroptotic cell death resistance. However, inhibiting AKRs activity/expression completely resensitizes resistant melanoma cells to ferroptosis execution. Finally, we found that the ferroptotic susceptibility associated with the differentiation of melanoma cells cannot be applied to metastatic-derived cells, due to the EMT-associated gene expression reprogramming process. However, we identified SCL7A11 as a valuable marker to predict the susceptibility of metastatic melanoma cells to ferroptosis. Our results identify the use of pro-ferroptotic drugs coupled to AKRs inhibitors as a new valuable strategy to efficiently kill human skin melanoma cells.