Distant metastases are detrimental for cancer patients, but the increasingly early detection of tumors offers a chance for metastasis prevention. Importantly, cancers do not metastasize randomly: depending on the type of cancer, metastatic progenitor cells have a predilection for well-defined organs. This has been theorized by Stephen Paget, who proposed the "seed-and-soil hypothesis", according to which metastatic colonization occurs only when the needs of a given metastatic progenitor cell (the seed) match with the resources provided by a given organ (the soil). Here, we propose to explore the seed-and-soil hypothesis in the context of cancer metabolism, thus hypothesizing that metastatic progenitor cells must be capable of detecting the availability of metabolic resources in order to home in a secondary organ. If true, it would imply the existence of metabolic sensors. Using human triple-negative MDA-MB-231 breast cancer cells and two independent brain-seeking variants as models, we report that cyclooxygenase 7b (Cox7b), a structural component of Complex IV of the mitochondrial electron transport chain, belongs to a probably larger family of proteins responsible for breast cancer brain tropism in mice. For metastasis prevention therapy, this proof-of-principle study opens a quest for the identification of therapeutically targetable metabolic sensors that drive cancer organotropism.