Cancer stem cells (CSCs) have been hypothesized to represent the driving force behind tumour progression and metastasis, making them attractive cancer targets. However, conclusive experimental evidence for their functional relevance is still lacking for most malignancies. Here we show that the leucine-rich repeat-containing G-protein-coupled receptor 5 (Lgr5) identifies intestinal CSCs in mouse tumours engineered to recapitulate the clinical progression of human colorectal cancer. We demonstrate that selective Lgr5+ cell ablation restricts primary tumour growth, but does not result in tumour regression. Instead, tumours are maintained by proliferative Lgr5− cells that continuously attempt to replenish the Lgr5+ CSC pool, leading to rapid re-initiation of tumour growth upon treatment cessation. Notably, CSCs are critical for the formation and maintenance of liver metastasis derived from colorectal cancers. Together, our data highlight distinct CSC dependencies for primary versus metastasic tumour growth, and suggest that targeting CSCs may represent a therapeutic opportunity for managing metastatic disease.
Ablation of Lgr5+ cancer stem cells does not result in regression of primary colorectal tumours, but prevents the formation and maintenance of metastasis in the liver.