Abstract Introduction: In developing any new drug, non-invasive assays of efficacy provide vital insight into activity with respect to dosing and pharmacodynamics of action. We have previously demonstrated the utility of dynamic bioluminescence imaging (dBLI) to evaluate VDA activity [1,2]. The current study, which evaluated the vascular-disrupting activity upon treatment with the promising novel benzosuberene-based agents KGP18 and KGP265 [structurally inspired by the natural product, combretastatin A-4 (CA4)] led to significantly decreased and delayed light emission indicative of vascular disruption. Methods: Human breast cancer cells (MDA-MB-231), prostate cancer cells (PC3), and lung cancer cells (A549) along with mouse breast tumor 4T1 cells were implanted in male SCID/Balb/c mice. The dose of KGP265 ranged from 5 to 15 mg/kg and the well studied VDA combretastatin A-4P (CA4P; 120 mg/kg) was used as positive control. Dynamic BLI was performed at baseline, 4, 24, 48 and 72 hours after IP administration of each VDA. Fresh D-luciferin was administered at each time point. As a control, additional tumor bearing mice were injected with the same volume of saline. Results: Dose escalation indicated that higher doses produced more effective vascular shutdown, and doses of 3 mg/kg KGP18 and 7.5 mg/kg KGP265 appeared optimal, being effective while causing little observable toxicity with the MBA-MD-231 tumor model. The optimal dose of KGP265 was 5 mg/kg for 4T1, and 10 mg/kg for A549 and PC3. Treatment of the 4T1, PC3 and A549 tumors with the above dose of KGP265 greatly reduced signal output in BLI images. By 4 hours, about 90% of signal was reduced and the signal stayed down for the next 48 hours post-injection. In vivo observations were confirmed by post mortem histology. Treatment with KGP265 led to significantly decreased and delayed light emission indicative of vascular disruption in different tumor types. The BLI signal was reduced by 90% at 4 hours compared to baseline. Light emission remained significantly depressed up to 48 hours and 72 hours, whereas CA4P showed substantial recovery by 24 hours and further recovery by 48 hours. Meanwhile light emission was highly consistent following administration of saline as control with reproducible kinetics and intensities. Conclusion: The results indicate that KGP18 and KGP265 cause significant vascular disruption. KGP265 is a phosphorylated prodrug of KGP18; it has enhanced water solubility, and maintains substantial vascular disrupting efficacy. Compared to CA4P (optimal dose 120 mg/kg), the new VDAs are effective at lower doses and achieve similar imaging results, suggesting enhanced efficacy. We are continuing to explore efficacy and therapeutic response through combination studies. Acknowledgment: CA4P was kindly provided by Dr. David Chaplin of OXiGENE. Supported by NIH R01CA140674