Scintigraphic imaging of the hypoxia marker (99m)technetium-labeled 2,2'-(1,4-diaminobutane)bis(2-methyl-3-butanone) dioxime (99mTc-labeled HL-91; prognox): noninvasive detection of tumor response to the antivascular agent 5,6-dimethylxanthenone-4-acetic acid.

5,6-Dimethylxanthenone-4-acetic acid (DMXAA) and combretastatin A4 phosphate (CA-4-P) markedly inhibit tumor blood flow in mice and are both currently in clinical trial. One of the challenges in clinical evaluation of antivascular agents is the monitoring of tumor blood flow inhibition in individual patients. This study investigates, using mouse models, whether a new marker for tissue hypoxia, (99m)technetium-labeled 2,2'-(1,4-diaminobutane)bis(2-methyl-3-butanone) dioxime (99mTc-labeled HL-91; Prognox)] has potential for the scintigraphic monitoring of tumor response to antivascular agents. Determination of radioactivity in dissected tissues 3 h after DMXAA (80 micromol/kg) or CA-4-P (227 micromol/kg) was injected indicated that both drugs inhibited blood flow (86RbCl uptake; 84 and 87%, respectively) and increased 99mTc-labeled HL-91 levels (350 and 300%, respectively) selectively in murine RIF-1 tumors. Planar imaging of 99mTc-labeled HL-91 3 h after DMXAA injection showed a dose-dependent increase in tumor levels above a threshold of 50 micromol/kg; this same threshold was observed for the inhibition of tumor blood flow (determined using Hoechst 33342). DMXAA also inhibited blood flow--and increased 99mTc-labeled HL-91 uptake--in MDAH-MCa-4 mouse mammary carcinomas and in NZMN10 human melanoma xenografts. Whether 99mTc-labeled HL-91 might also be useful as a biomarker for tumor cell killing was investigated by clonogenic assay of surviving cells 15 h after imaging 99mTc-labeled HL-91 in RIF-1 tumors. Log cell kill in individual tumors showed a statistically significant linear correlation (P < 0.001) with 99mTc-labeled HL-91 uptake after 60 micromol/kg (r2 = 0.79) and 70 micromol/kg (r2 = 0.44) but not at 80 micromol/kg DMXAA. The lack of correlation at high doses presumably reflects the insensitivity of the tumor-averaged 99mTc-labeled HL-91 signal to small regions in which tumor blood flow is preserved (which will limit log cell kill). The results indicate the potential of 99mTc-labeled HL-91 for the noninvasive imaging of tumor blood flow inhibition by antivascular drugs in humans.