Apostolos J. Tsiouris

Bevacizumab is a monoclonal antibody that is now routinely administered intravenously for the treatment of recurrent glioblastomas, and is thought to inhibit vascular endothelial growth factor (VEGF) and therefore angiogenesis. There has been extensive debate in the literature as to the histopathologic effects of bevacizumab on glioblastomas. How to best determine response with imaging is also currently a very hot topic. Neuroradiologists who interpret posttreatment MRI examinations of patients with glioblastoma routinely describe an initial profound decrease in the enhancing component of the tumor as well as the surrounding edema. However, it has been well described that the nonenhancing mass-like component of the tumor remains stable or progresses over short time intervals.

Our group at NYPH-WCMC, led by Dr. John Boockvar of our neurosurgery department, currently has multiple ongoing treatment protocols for the administration of bevacizumab (as well as other chemotherapeutic agents) directly to recurrent glioblastomas utilizing superselective intra-arterial techniques. Like other groups, we noticed a dramatic favorable response on anatomic gadolinium-enhanced MRI after treatment. This prompted our group to try to determine if we were only documenting a profound “pseudoresponse” due to normalization of the blood-brain barrier or if we were witnessing a true tumoricidal effect.

We decided to utilize MRS as an adjunct to anatomic MRI to assess the neurochemical and metabolic changes that may be associated with bevacizumab treatment. In addition, given the directed intra-arterial mechanism utilized for drug delivery, we hypothesized that the effect of bevacizumab would be most profound in this patient population. The results of our first 18 patients treated were very promising, within the limitations of our study. By demonstrating statistically significant improvements in the Cho:NAA ratios within the enhancing portions of tumor, these suggested that the effect of bevacizumab in these patients may represent more than a “pseudoresponse.”

Over the past 2 years, we have treated more than 60 patients with targeted superselective intra-arterial bevacizumab, who have also undergone multivoxel MRS (as well as DSC-MR perfusion). We are in the process of reviewing and analyzing all our data and correlating our MRS (as well as MRP) results with anatomic MRI findings and progression-free survival and overall survival rates. We are also currently applying for NIH funding to directly compare IV and intra-arterial bevacizumab therapies.

Read this article at AJNR.org . . .