Despite intensive research and development of new targeted therapies and radiotherapeutic techniques, prognosis for patients with GBM remains poor indicating the need for new therapeutic approaches. Because of high vascular density of these tumors development of therapies selectively targeting the tumor vasculature may be meaningful. The aim of this study was to identify if the combined treatment with ASA404 and taxol shows synergistic effects in mice bearing U251 human glioblastoma xenografts and if non-invasive 18 F-FDG PET imaging could be used to monitor tumor response early in the course of therapy. Tumors of animals which were treated with ASA404 as a sole agent or in combination with taxol showed significant growth delay in comparison with control or animals which were treated only with taxol. Combined treatment with taxol and ASA404 had a similar effect on tumor growth as ASA404 alone. At the completion of the study, there was only slight but statically significant difference in tumor weight for animals treated with the drug combination as compared to animals receiving ASA404 alone. This difference in statistical significance between the two parameters which were analysed (tumor volume and tumor weight) can be explained by the use of a formula in which one important parameter of the tumor growth – tumor depth, was not considered. The tumor weights were determined only at the end of treatment and in this may represent a more precise parameter to determine an effect of the therapy. On the other hand, the differences among treatment groups were very small. Thus, based on these data, it is difficult to conclude if combined treatment was synergistic. GBM represents one of the most vascularized tumors . Antiangiogenic agents, such as bevacizumab, which is approved for treatment of recurrent GBM patients who have failed previous temozolomide and radiation therapy, inhibit new blood vessel formation from preexisting vasculature . Furthemore, antiangiogenic agents can normalize the tumor vasculature and decrease interstitial fluid pressure, providing an improved drug delivery .
In contrary to antiangiogenic agents, VDAs such a ASA404, have a selective affinity to existing tumor blood vessels inducing their collapse which impairs blood flow, oxygen supply and consequently causes necrosis in tumor tissue .
The synergistic mechanism between ASA404 and taxol seems not to involve potentiation of the vascular disrupting activity of ASA404 or a pharmacokinetic interaction between these two drugs. The synergism might be explained by complementary action against the different subregions of the tumor; ASA404 is more active in the poorly vascularized regions while taxol is active in the well vascularized regions . In addition to its direct cytotoxic effects, taxol can induce the expression of pro-inflammatory cytokines, such as TNF-α and IL-6  which are important mediators of ASA404 activity [7, 8]. Thus, taxol and ASA404 can induce of same cytokines which are responsible for tumor vascular disruption.
It has been reported that chemotherapy drugs should be administered before, or shortly after ASA404 in order to avoid compromised delivery. When taxol was administred 4 h after ASA404, considerable loss of antitumor activity has been observed apparently caused by decreased blood flow which may inhibit taxol distribution in tumor tissue .
As opposed to conventional antineoplastic agents, effects of VDAs such as ASA404 do not result in dramatic changes in tumor volume [22, 24]. This indicates that new approaches are necessary to monitor tumor response to VDAs. We found that 18 F-FDG uptake decreased rapidly after administration of ASA404. This result is consistent with findings that highest increase of intratumoral TNF-α activity is also observed 4 h after tratment with ASA404 . We suppose that the marked reduction in tumor 18 F-FDG uptake is at least partly a reflection of tumor cell death. Moreover, it seems likely that a significant fraction of tumor cells will become necrotic, if perfusion is decreased to such an extent that 18 F-FDG uptake significantly decreases. Preclinical studies in different tumor models have demonstrated that ASA404 directly disrupts the tumor vasculature by selectively inducing apoptosis in tumorvascular endothelial cells [26, 27]. On the other hand, the reason why several of the mice who underwent PET imaging died within several hours after the PET scan is not entirely clear. Because of ASA404 antivascular properties, we presumed that bleeding might cause this effect. Suprisingly, mouse dissection did not show any bleeding source. There is only one report on the use of small animal PET as a biomarker for response to ASA404 . In this case, 18 F-fluromisonidazole (FMISO) PET imaging was conducted and no side effects have been reported. Before application of the 18 F-FMISO tracer, the animals do not have to fasten.
It has been shown that the administration of TNF-α, which is an important mediator of ASA404 antitumoral action, decreases serum glucose levels in mice . For improving tumor visualization, mice have to fasten a 18 F-FDG injection uptake period . We speculate that the combination of ASA404 mediated TNF-α secretion with starving and anesthesia for PET investigation may have caused death of mice.
It has been reported that treatment of nude mice bearing U87 and GL261 orthotopically grown human glioblastoma cells with ASA404 caused statistically significant increase in median survival compared to untreated controls . Measurement with contrast-enhanced magnetic resonance imaging (MRI) and diffusion-weighted MRI which were used to determine tumor blood flow 24 h after treatment with ASA404 clearly demonstrated extravasation and accumulation of the contrast agent in the tumor indicating treatment-induced vascular disruption. Similar effects were observed in fibrosarcoma (MCA205) ectopic and orthotopic tumor models. Remarkably, 3 h after treatment with ASA404 ectopic tumors showed 6-fold greater induction of TNF-α compared to orthotopic tumors . In our glioblastoma U251 model, the rapid mode of action of ASA404 became also apparent by the changes in tumor color 8 h post injection which indicates hemorrhagic necrosis.
It is believed that VDAs are more effective against vessels inside of the tumor. In the periphery of the tumor a characteristic rim of cells will remain viable after treatment [33, 34].
In our experiments, tumors of animals which were treated with ASA404 as a sole agent or in combination with taxol, showed significant growth delay in comparison with control or animals which were treated only with taxol. In both group of ASA404-treated animals, tumor start to regrowth between 11 und 13 days after treatment.
Despite the significant smaller tumors, 20 days after treatment, histopathologic examination revealed that all treated tumors maintained the characteristic growth of glioblastoma without any difference between the different treatment groups (data not shown). No significant increase in the level of cell death occured in tumors of animals treated with ASA404 indicating that surviving tumor cells were able to proliferate causing tumor regrowth. In the Colon 38 adenocarcinoma tumor model , the antitumor effect of ASA404 was schedule-dependent. 100% of the tumors regressed, when the mice were treated with loading (25 mg/kg) and two supplementary (5 mg/kg) doses after 4 and 8 h while in mice treated with a single dose of ASA404 (25 mg/kg) 40% of tumors regressed. We supose that with a prolonged therapy better antitumoral responses might be achieved. The first clinical data obtained from I/II phase clinical trials evaluating ASA404 in combination with taxol and carboplatin in patients with untreated advanced non-small lung cancer were promising  but in one large, randomized phase III placebo controled trial, these results were not confirmed .