Abstract P3-02-28: A combination approach to target residual tumors via ferroptosis induction

Background: Selection pressure from chemotherapy can result in small populations of resistant cells, even in tumors with apparent clinical response. These residual tumor cells can tolerate cytotoxic treatments, persist in a reversible, semi-dormant, diapause-like state, and ultimately drive tumor recurrence. Pharmacologically targeting these residual tumor cells has the potential to prevent or significantly delay tumor relapse, but it remains a clinical challenge. Methods and Results: We report ferroptosis induction as a targeting strategy to eradicate residual tumor cells following chemotherapy. Through focused chemical library screening using an in vitro residual tumor model, we identified ferroptosis inducers that are effective against the residual tumor cells, but not the bulk proliferating tumor cell population. Ferroptosis induction eradicated residual tumor cells generated by treatments with various chemotherapeutic agents and radiation across multiple breast cancer subtypes. The hypersensitivity to ferroptosis induction in residual tumor cells is driven by elevated levels of ferrous irons, the catalyst of ferroptosis. To evaluate the therapeutic potential of ferroptosis induction in vivo, we established residual tumors in triple-negative breast cancer (TNBC) PDX models following treatment with doxorubicin and cyclophosphamide (AC). In TNBC PDX models, we demonstrated that ferroptosis induction with the single agent RSL3, a glutathione peroxidase 4 (GPX4) inhibitor, effectively eliminated the residual tumor cells and significantly delayed tumor recurrence post-chemotherapy. The time from initiation of treatment to recurrence for AC vs. AC+RSL3 groups was: 65.3+/-3.9 vs. 81.6+/-5.4 days, p=4.1E-06, for PDX-3107, and 81.4+/-4.9 vs. 102.9+/-12.3 days, p=1.8E-06, for PDX-3887. Importantly, simultaneous suppression of the two parallel ferroptosis defense mechanisms mediated by GPX4 and ferroptosis suppressor protein-1 (FSP1), through the combined use of RSL3 and iFSP1, further delays tumor recurrence in PDX models. The time to recurrence in AC vs. AC+RSL3/iFSP1 groups was 49.8+/-8.4 vs. 113.1+/-5.8 days, p=5.0E-09, for PDX-3204. Notably, combined treatment with RSL3/iFSP1 led to complete clearance of the residual tumors in 9 out of 11 mice bearing PDX-3887, with no tumor recurrence observed up to four months of follow-up. Conclusion: Our study demonstrates that ferroptosis induction is an effective strategy for eradicating residual tumors and delaying tumor recurrence in TNBC following cytotoxic chemotherapy.