MRI in patients with high-grade gliomas treated with bevacizumab and chemotherapy

Pope et al retrospectively review 14 patients with recurrent high-grade gliomas (HGG) treated with chemotherapy and bevacizumab. [1] Fifty percent of patients responded to treatment determined by post-treatment MRI. Several aspects of this report merit commentary. All patients in the study previously failed primary therapy and one prior salvage therapy. The number of patients treated within 3 months of radiotherapy was unclear. Though the protocol specified at least 4 weeks elapsed since administration of radiotherapy, it is recognized that early delayed radiotherapy changes seen on MRI may simulate tumor progression and resolve over time. [2] The performance of MRI to assess response to therapy in patients with recurrent HGG is customarily performed every 2 months and coordinated with chemotherapy cycles. In this report, first post-treatment MRI appears desynchronized with chemotherapy administration. As response rates were based on first post-treatment MRI, response appears not to be bevacizumab dose dependent.

To this author's knowledge, there has not been a single agent trial of bevacizumab for recurrent HGG though trials in other solid cancers indicate a less than 10% single agent response rate. The rationale for combining bevacizumab with CPT-11 was not mentioned in this article. A number of reports evaluating CPT-11 as a single agent for recurrent glioblastoma have concluded the agent has little efficacy. [3] The first report by Stark-Vance indicating response to the combination of bevacizumab and CPT-11 for recurrent HGG was not cited. [4]

The durability of response was also not reported. Clinical trials in recurrent HGG report response duration by overall survival, time to tumor progression and 6-month progression free survival. 15% 6-month progression free survival for patients with recurrent glioblastoma is used as the standard by which new trials are compared with respect to efficacy. [5]

A variety of methods have been utilized to assess response to anti-angiogenic agents including dynamic contrast enhanced MRI. These methods provide proof of principle by quantifying tumor blood volume or perfusion. Pope et al do not report whether the observed response represents an anti-angiogenic effect on tumor vasculature or an effect on blood brain barrier disruption with normalization of peritumoral edema and diminished tumor contrast enhancement.

Pope et al emphasize the complexity of treating recurrent HGG and remind neurologists that the approach to malignant gliomas continues to evolve and will increasingly utilize targeted therapies such as antiangiogenic agents.

References

1. Pope WB, Lai A, Nghiemphu P, Mischel P, Cloughesy TF. MRI in patients with high-grade gliomas treated with bevacizumab and chemotherapy. Neurology 2006;66: 1258-1260.

2. de Wit MCY, de Bruin HG, Eijkenboom W, Sillevis-Smitt PAE, van den Bent MJ. Immediate post-radiotherapy changes in malignant glioma can mimic tumor progression. Neurology 2004;63:535-537.

3. Prados MD, Lamborn K, Yung WKA, et al. A phase two trial of irinotecan (CPT-11) in patients with recurrent malignant glioma: A NABTC study. Neuro Oncol 2006;8: 189-193.

4. Stark-Vance V. Bevacizumab and CPT-11 in the treatment of relapsed malignant glioma (abstract). Neuro-Oncology 2005;7:369.

5. Wong ET, Hess KR, Gleason MJ, et al. Outcomes and prognostic factors in recurrent glioma patients enrolled onto phase II clinical trials. J Clin Oncol 1999;17: 2572-2578.

Disclosure: The author reports no conflicts of interest.

Dr. Chamberlain raises several important issues some of which are directly addressed in our paper. Our study reported the early MR imaging findings of patients with recurrent malignant glioma following chemotherapy/bevacizumab combination treatment. The study was not a clinical trial of the effectiveness of this therapy for recurrent malignant glioma. Therefore, the durability of response as measured by the length of survival, time to tumor progression and 6-month progression free survival was not reported. When a sufficient amount of time has passed to make meaningful conclusions regarding this therapy in relationship to patient survival, we or others will report this data.

The most important question raised by Dr. Chamberlain is whether the response to chemotherapy/bevacizumab therapy could be attributed to resolving radiation changes. We do not believe this is possible. Of the 14 patients reported, all had failed prior radiation therapy. Only one patient initiated treatment within 3 months of completion of radiation therapy. The median time between completion of radiation therapy and initiating chemotherapy/bevacizumab in this patient group was 40 weeks. All patients were scanned within 45 days of treatment. In the article cited by Dr. Chamberlain, [2] only 2 of 32 patients treated for glioma showed spontaneous improvement, suggesting that even if all our patients had been treated within 3 months of radiation therapy, the MR imaging findings of our study would not be explained by radiation change.

Another important question raised by Dr. Chamberlain is whether the observed response represents normalization of the blood brain barrier, a direct effect on tumor, or a combination of the two. We were unable to fully address this issue in our report, due to the space limitations of the journal. We found that for some patients reduction of tumor bulk was clear even on the T2W images, and therefore not simply a result of changes in enhancement due to re-establishment of the blood brain barrier. However, the reduction in edema is likely related to changes in vascular permeability, as VEGF is a potent mediator of increased vessel permeability as noted in our report. We are currently analyzing data obtained from dynamic contrast enhanced MRI to further investigate the effects of chemotherapy/bevacizumab treatment on tumor hemodynamics.

Lastly, we neglected to cite Stark-Vance's abstract [4] on CPT-11 and bevacizumab. We appreciate Dr. Chamberlain's remarks as they allow us to appropriately acknowledge this work.

Disclosure: The authors report no conflicts of interest.