Purpose: To determine the effects of MRI-assayed vascular leakiness on the delivery of macromolecular therapeutics to tumors.
Materials and methods: MDA-MB 435 tumors, subcutaneously implanted into nude rats were treated with a single dose of bevacizumab at levels of 0.1 mg (n = 5) or 1.0 mg (n = 10) or received saline (control animals, n = 8). After 24 hours, albumin-(Gd-DTPA) (30)-enhanced MRI was performed. Just prior to MRI, the cytotoxic drug vinorelbine was administered intravenously. Upon completion of the MR experiment, tumor vinorelbine concentrations were quantified by high performance liquid chromatography (HPLC). Vascular leakiness (K (PS)) was calculated based on the MRI data using a pharmacokinetic model.
Results: K (PS) was calculated as 3.70 +/- 1.12 (control tumors), 1.95 +/- 0.70 (0.1 mg group) and 0.75 +/- 0.46 microl min (-1)cm (-3) (1.0 mg group). K (PS) was significantly higher in the control group compared to the 1.0 mg bevacizumab group. Vinorelbine concentrations were measured as 409.4 +/- 109.7 (control tumors), 387.5 +/- 47.5 (0.1 mg group) and 250.7 +/- 71.9 (1.0 mg group). These differences were not significant. A moderate and significant correlation was found between K (PS) and Vinorelbine concentrations in tumors (r = 0.49, p < 0.05).
Conclusion: MRI-assayed K (PS) based on dynamic MRI enhanced by albumin-(Gd-DTPA) (30) correlated significantly with vinorelbine accumulation in experimental xenograft tumors under angiogenesis inhibition. Thus, the MRI technique applied in our study could potentially help to predict accumulation of macromolecular cytotoxic drugs and to optimize individual therapeutic regimes in tumors.