Optimizing Outcomes With Bevacizumab by Better Targeting Patients and Tumors

This article is a review of Bevacizumab in the Treatment of Metastatic Breast Cancer .

Numerous preclinical and clinical studies have demonstrated a role for angiogenesis in the growth and progression of breast cancer. Elevated vascular endothelial growth factor (VEGF) levels have been demonstrated in association with poor outcomes, and thus, this finding is an attractive target for therapeutic intervention.

Bevacizumab (Avastin) is a humanized monoclonal antibody to VEGF that targets the protein, thereby decreasing neovascularization of tumors and "pruning" existing vasculature, with the dual goals of increasing intratumoral delivery of concurrent therapies and curtailing motility and invasion that is the hallmark of metastatic breast cancer.

Theory vs Practice
Traina eloquently reviews the data currently published on the efficacy of bevacizumab(Drug information on bevacizumab) in metastatic breast cancer. To date, these data suggest that bevacizumab as a single agent has modest activity. Can bevacizumab augment the activity of standard cytotoxic therapies? This may or may not be the case. Initial combination with capecitabine(Drug information on capecitabine) (Xeloda) showed no benefit. The Eastern Cooperative Oncology Group (ECOG) 2100 trial of paclitaxel(Drug information on paclitaxel) with or without bevacizumab demonstrated an approximately 6-month prolongation in disease-free survival (DFS), but the AVastin And DOcetaxel(Drug information on docetaxel) (AVADO) trial, examining the benefit of adding bevacizumab to docetaxel (Taxotere), demonstrated a statistically significant improvement in DFS of less than 1 month—the clinical significance of which Dr. Traina appropriately questions. Overall survival data, though limited at this time, do not appear to suggest an overall survival benefit to the addition of bevacizumab to standard therapy in the patient populations studied.

What are we to make of the apparent contradiction between the theoretical benefit of inhibiting angiogenesis in breast cancer and the modest benefits such an approach has had in the clinic? One explanation may be that only some tumors and/or patients are appropriate candidates for this approach. It has become increasingly clear over the past decade that breast tumors are genetically heterogeneous. We are also now coming to appreciate that the tumor host—the patient with breast cancer—also is genetically unique in ways that can significantly impact the host response to tumor or its treatment. Several studies have attempted to identify markers of response to antiangiogenic therapy within both tumor and host that may be critical to the optimal application of this therapy.

Anti-VEGF Therapy
The VEGF gene is located on chromosome 6p21.3 and includes eight exons that, through alternate splicing, produce a family of VEGF proteins. VEGF appears to be regulated by estrogen, hypoxia, growth factors, and cytokines, such as interleukin-6.[1] The gene is highly polymorphic. Several single-nucleotide polymorphisms (SNPs) are frequent in the general population and functional. The -2578CC, -2549del/del, -1154GG, and -634CC have been shown to be associated with higher VEGF production.[2-4] The -634CC and 2578CC SNPs are associated with larger primary breast tumor size and higher tumor grade,[5] leading to a more aggressive tumor phenotype. In a cohort of 1,193 Chinese breast cancer patients, Lu et al demonstrated that the presence of both the -460C and +405G polymorphisms is associated with increased risk of death from the disease.[6]

VEGF polymorphisms may not only be prognostic (ie, associated with aggressive tumor behavior), but may also be predictive of response to anti-VEGF therapy. Historically, approaches aimed at determining tumor changes associated with response, including circulating VEGF levels, K-raf, p53, and microvessel density have been unsuccessful in predicting response to bevacizumab.[7,8]

More recently, however, Schneider et al demonstrated an important link between SNPs in VEGF and response to bevacizumab utilizing the ECOG 2100 study cohort.[9] Of 673 patients from the trial, 363 had paraffin(Drug information on paraffin)-embedded tumor blocks available for genotyping. Three SNPs in the VEGF promoter (-2578C>A, -1498C>T, ‑1154G>A) and two others (634G>C and 935C>T) were evaluated. These SNPs were included based on sufficiently high population frequency and hypotheses regarding functional significance. The authors reported significantly improved overall survival in patients receiving bevacizumab who carried the VEGF-2578 AA genotype (hazard ratio [HR] = 0.58, 95% confidence interval [CI] = 0.36–0.93) as well as those with any -1154A allele, with an additive effect for each allele (HR = 0.62, 95% CI = 0.46–0.83).

Similar studies in other tumor types, including ovarian cancer, have found links between VEGF and VEGF-related genes and response to bevicizumab.[10] While these data require confirmatory studies and mechanistic investigation, such findings suggest that benefits to VEGF inhibition with drugs like bevacizumab may be enhanced in certain patient populations based on genotype.

Conclusions
In summary, initial trials have shown activity for bevacizumab as an adjunct to chemotherapy for metastatic breast cancer. However, treatment of unselected patients has shown only modest improvement in DFS and no overall survival benefit. Recent work demonstrating that functional polymorphisms in the VEGF gene alter outcomes from patients receiving bevacizumab is provocative. If confirmed, these findings provide a rationale and need for novel trial designs based on the selection of appropriate patients for treatment by genotype in future studies. Only by appropriately targeting bevacizumab to sensitive patients will the full potential of this drug in altering the course of metastatic breast cancer be realized.