A PET imaging study of the vascular disruptive agent OXi4503 to confirm in vivo mechanism of action in a phase I trial

Abstract

e14510 Background: OXi4500, the active product of OXi4503, reversibly binds at the colchicine binding site on the ß subunit of tubulin. This study investigated its mechanism of action using PET molecular imaging Methods: OXi4503 was administered by 3 weekly IV infusions in a phase 1 trial (32 pts). Between 1.6 & 15.4 mg/m2 dose levels, PET imaging was performed with 15O-labelled water (measuring blood perfusion) pretreatment & at 90min & 24hr after first infusion (8pts) & at 28days (1 pt). [18F]fluorodeoxyglucose (FDG) (measuring proliferative activity) was performed pretreatment & at 28 days (9pts) Results: 9 pts (mean age 48.5yrs) were studied: CRC (2), pancreatic (1), MM (6). 2 pts with MM (3.8 & 6.5 mg/m2) achieved stable disease. Toxicity included pyrexia, lethargy & hypertension. There was a dose-dependent linear increase in peak plasma concentrations & AUC of OXi4503 & OXi4500. Sustained tumour blood perfusion reduction (50% at 90min & 24hr) was seen in all metastases in the CRC pt treated at 15.4 mg/m2 after just one dose & resulted in FDG reduction (23% in all 5 mets) & tumor marker response. Doses between 1.9 & 11.0 mg/m2 produced dose independent reductions in tumour perfusion which were more variable & less marked with 3/8 pts showing a reduction in tumour FDG uptake. The 1pt studied at 28 days (8.5 g/m2) showed further sustained reduction in tumour blood flow (40%). Variable dose independent changes in normal tissue perfusion were seen at all dose levels eg in 3/4 pt splenic perfusion showed a dose-independent decrease (20–30%) at 90min with recovery at 24hr Conclusions: OXi4503 shows significant activity at 15.4 mg/m2 in line with its proposed mechanism of action. Characterisation of the mixed tumour perfusion response at lower doses will aid planning & assessing combination studies. The MTD is now being identified. Sponsored by CRUK. [Table: see text]