Abstract Ovarian cancers grow in suspension in the ascites fluid, and contain a small population of ovarian cancer stem cells (OvCSC), which are resistant to therapy. Due to the rarity of OvCSCs, we developed a 3D hanging drop platform, in which as few as one ALDH+ CD133+ cell (isolated from primary malignant ascites) can be stably incorporated into 3D spheroids. Our platform can be utilized to quantify drug sensitivity of chemotherapeutic agents in the context of OvCSCs, distinguish drug responses for the same drugs between several patient samples, and model patient-specific tumor re-emergence, making it uniquely suited for the development of personalized therapeutics. Three patient samples (Pt259, Pt224, Pt152) were evaluated and robust proliferation rates were observed in spheroids, ranging from 5.3 fold to 8.4 fold. By Day 7, ALDH+ CD133+ cells had differentiated within spheroids to form progeny of ALDH- CD133-, ALDH+ CD133-, and CD133+ ALDH- cells while maintaining an ALDH+ CD133+ population. Each patient-derived spheroid demonstrated a different composition of these progeny, which were similar to those observed in the patient samples. OvCSC spheroids had differing responses to drug treatments (cisplatin, ALDH targeting compound 673A, and JAK1/2 inhibitor ruxolitinib). Combination of cisplatin/673A targeted ALDH+ and CD133+ in all patient samples. Pt259 samples were maximally sensitive to cisplatin/673A, while Pt224 and Pt152 were more resistant (20-40% higher viability). Combination dose of cisplatin/ruxolitinib targeted CD133+ populations. By isolating cells that escaped chemotherapy, we created a spheroid model to study tumor re-emergence. ALDH+ populations re-emerged to a lower extent compared to original OvCSC spheroids, while CD133+ populations did not recover at all. Spheroids formed from the most platinum-sensitive cells (Pt259) and the most platinum-resistant cells (Pt152) following cisplatin/673A treatment were also serially passaged over 7 cycles in 7 weeks to characterize CD133+ and ALDH+ populations and evaluate their ability to reform spheroids, effectively modeling tumor re-emergence in vitro. ALDH+ OvCSC progeny reliably repopulated within these spheroids despite initial depletion following treatment. Over six serial passages, ALDH+, CD133+, and ALDH+ CD133+ populations gradually returned to original and even higher than levels seen in original patient samples. Lastly, OvCSC spheroids initiated tumors in immunodeficient mice at 100% success with only 10 spheroids injected. These tumors demonstrated a distinct response to therapy that corresponds with responses seen in spheroids, indicating that our model may be a means to screen tumors for personalized drug selection. Our patient-derived low-cell-number OvCSC spheroid platform can be utilized to study tumor biology, to model tumor re-emergence after primary chemotherapy, and to identify new targeted therapeutics from a personalized medicine standpoint. Citation Format: Shreya Raghavan, Pooja Mehta, Michael Bregenzer, Maria Ward Rashidi, Elyse Fleck, Lijun Tan, Karen McLean, Ronald Buckanovich, Geeta Mehta. Patient-specific evaluation of chemoresistance and tumor recurrence using ovarian cancer stem cell spheroids. [abstract]. In: Proceedings of the AACR Conference: Addressing Critical Questions in Ovarian Cancer Research and Treatment; Oct 1-4, 2017; Pittsburgh, PA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(15_Suppl):Abstract nr A33.