Abstract Introduction: The Halichondrin B analog, eribulin (Halaven), is FDA approved for third line treatment of breast cancer after anthracycline/taxane therapy. Previously, the PPTP identified eribulin as having significant broad-spectrum activity against in vivo xenograft models of sarcoma and acute lymphoblastic leukemia (ALL) at drug exposures similar to those observed in patients (Kolb EA et al., Pediatr Blood Cancer. 60:1325-32, 2013). In this study we investigated both dose response and schedule and compared eribulin with vincristine, an antimicrotubule agent used in curative treatment of childhood soft tissue sarcomas. Methods: Eribulin was evaluated against four sarcoma xenograft models (Rh30R, Rh41 rhabdomyosarcomas and EW-8, CHLA-258 Ewing sarcomas) in CB17 SCID mice using metrics for response previously developed in the PPTP. Eribulin was administered IP at 1, 0.5 and 0.25 mg/kg using a Q4D x 3 schedule or at 1 mg/kg Q4D x 2, with cycles repeated at day 21. Vincristine was administered at 1 mg/kg using a Q7D x 6 IP schedule with a period of observation for all treatment groups up to 12 weeks. Results: Eribulin was highly active against 3 of 4 sarcoma models on the Q4D x 3 schedule. For Rh30R, both the 1.0 and 0.5 mg/kg doses produced CRs, with regrowth beginning at week 9. For EW-8, eribulin was highly effective, and even at the 0.25 mg/kg dose 7 of 10 mice bearing EW-8 xenografts had maintained CRs at week 12. For CHLA-258, 7 of 10 mice bearing CHLA-258 tumors remained in CR at week 12 at the 1.0 and 0.5 mg/kg dose level, while the 0.25 mg/kg dose did not induce CR. In contrast, Rh41 tumors were relatively resistant to eribulin at doses less than 1 mg/kg. To more closely mimic the clinical schedule for eribulin (day 1 and 8, repeated every 21 days), a similar schedule (Q4D x 2 repeated at day 21) was evaluated at the 1 mg/kg dose level. Reduced dosing only marginally decreased the effectiveness of eribulin against Rh30R, EW-8 and CHLA-258 tumor models, with similar frequencies of CR and maintained CR at the end of 12 weeks observation. In general, time to regrowth was decreased by 1 week in each tumor model compared to the 1 mg/kg Q4D x 3 schedule. The Q4D x 2 schedule was inferior to the Q4D x 3 schedule against Rh41 tumors where responses were equivalent to the 0.5 mg/kg dose on the Q4D x 3 schedule. In comparison to eribulin, vincristine treatment showed similar efficacy against Rh30R tumors. Vincristine induced CRs against EW-8, but regrowth was more common for vincristine than eribulin after completion of treatment. For CHLA-258, vincristine failed to induce CRs in most treated animals, and rapid tumor regrowth occurred 1 to 2 weeks after the last dose of vincristine. Vincristine was ineffective against the Rh41 xenograft model. Conclusions: Eribulin demonstrated high level antitumor activity against 3 of 4 sarcoma xenografts over a 4-fold dose range as well as when the frequency of drug administration was reduced to more closely mimic clinical schedules of administration. Eribulin demonstrated equivalent or superior activity to vincristine in all models tested. High-level activity for eribulin was observed using doses and schedules that produce systemic exposures below those that are achievable in humans, supporting a potential therapeutic window for eribulin against selected childhood sarcomas. Future preclinical studies with eribulin will examine the agent in combination with other cytotoxic agents, and clinical trials are currently being planned (Supported by award NO1-CM-42216 from the NCI). Citation Format: Anders E. Kolb, Gorlick Richard, Jianrong Wu, Kurmasheva T. Raushan, Houghton Peter, Smith A. Malcolm. Pediatric Preclinical Testing Program (PPTP) stage 2 evaluation of eribulin, a novel anti-microtubule agent. [abstract]. In: Proceedings of the AACR Special Conference on Pediatric Cancer at the Crossroads: Translating Discovery into Improved Outcomes; Nov 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2013;74(20 Suppl):Abstract nr A81.