The majority of breast cancer mortality is due to recurrent metastatic disease, which is likely seeded by a dormant minimal residual disease (MRD) state. Better understanding of the cell state during dormancy and subsequent recurrence is thus necessary to improve patient outcomes. We first identified ferroptosis, a non-apoptotic form of cell death, as preferentially mutated in metastases in a cohort of paired primary and metastatic breast cancers. CRISPR-Cas9 screening of genes that can regulate ferroptosis identified guides that target PNPLA2, GOT1, and GPX4 as being selected for in late dormancy and recurrence. We confirmed that PNPLA2, GOT1, and GPX4 deletion accelerate recurrence and demonstrated that this acceleration can likely be attributed to an increased probability of reentering the cell cycle. Using a model of Her2-downregulation induced breast cancer dormancy, we also discovered that dormant cancer cells are ferroptosis-resistant. PNPLA2, GOT1, and GPX4 deletion were capable of further increasing ferroptosis resistance, even though the genes themselves had previously been thought to be anti-ferroptotic. In addressing what features of dormancy may make it a ferroptotic-resistant state, we discovered that reduced to oxidized glutathione ratios are higher in dormant cells, potentially making them more capable of handling oxidative stress. PNPLA2, GOT1, and GPX4 deletion even further increased this ratio in dormancy, suggesting that glutathione may partially explain the ferroptosis resistance in these cells as well. Analysis of human breast cancer patient data confirms the clinical relevance of our findings: patients with copy number loss of PNPLA2, GOT1, or GPX4 have worse overall and recurrence-free survival. Together, our work implicates three genes in the process of recurrence for the first time and identifies a unique feature of the dormant tumor cell state. The surprising dormant cell resistance to ferroptosis inducers and counterintuitive effects of deleting ferroptosis-suppressing genes may suggest that new therapeutics targeting ferroptosis in cancer will need to carefully consider the possible outcomes of the treatment on dormant cancer cells.