Is there a role for targeting vascular endothelial growth factor/receptor axis in the treatment of patients with metastatic melanoma?


  • Kevin B. Kim MD

    Corresponding author
    1. Department of Melanoma Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas
    • Department of Melanoma Medical Oncology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030

    Search for more papers by this author
    • Fax: (713) 745-1046

  • See referenced original article on pages 586–92, this issue.


Bevacizumab and other drugs targeting vascular endothelial growth factor and/or its receptor have shown encouraging clinical outcomes in early-phase studies when they are combined with cytotoxic chemotherapy drugs. The success of bevacizumab or other antiangiogenic drugs in the treatment of patients with advanced melanoma will depend on a better understanding of tumoral neovascularization and its effect on tumor stromal tissues, as well as on selecting patients with appropriate relevant biomarkers to yield improved clinical outcomes.

Until recently, immunotherapy and cytotoxic chemotherapy were the mainstays of systemic therapy for patients with metastatic melanoma. Although interleukin (IL)-2 received approval for the treatment of metastatic melanoma from the US Food and Drug Administration in 1998 for its ability to induce a long-term durable response, it has not been widely used because it causes severe toxicity but elicits a meaningful clinical benefit in only a small subset of patients. Among cytotoxic drugs, dacarbazine and temozolomide have been the most commonly used agents; however, the objective response rates for these drugs are low (< 10%), and the median duration of progression-free survival (PFS) is < 2 months.1, 2 The use of the combination regimen of carboplatin and paclitaxel for the treatment of patients with advanced melanoma has gained traction due to its seemingly longer duration of PFS and its tolerability, which have been demonstrated in several multicenter trials,3-5 but to the best of my knowledge this regimen has never been compared with dacarbazine or temozolomide in a randomized study. The response rate for this combination regimen is approximately 16% in the first-line treatment setting, and the median PFS duration is 4 months.3

nab-Paclitaxel is an albumin-bound nanoparticle form of paclitaxel that allows a higher concentration of paclitaxel for binding and avoids the need for the additive polyethoxylated castor oil. On the basis of its improved clinical benefit and more favorable safety profile over paclitaxel in patients with advanced breast cancer,6 nab-paclitaxel has been evaluated in patients with metastatic melanoma. In a multicenter phase 2 study, the response rate of weekly nab-paclitaxel was 22% among 37 chemotherapy-naive patients, and the median PFS duration was 4.5 months.7 In a phase 2 study of the combination of nab-paclitaxel and carboplatin reported by Kottschade et al, the median duration of PFS was 4 months and the median duration of overall survival (OS) was 11 months among chemotherapy-naive patients.8 A large, randomized phase 3 study ( identifier NCT00864253) to compare the clinical efficacy of nab-paclitaxel and dacarbazine has been completed, and the results will be available shortly.

Melanomas are known to be highly vascular tumors, and their progression is associated with the induction of angiogenesis.9 Vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and IL-8 are angiogenic factors that appear to play important roles in the angiogenic stimulation of melanoma progression.10 In addition, VEGF has been implicated in tumor resistance to cytotoxic drugs. In previously reported in vitro experiments, the repeated treatment of melanoma cells with dacarbazine induced drug resistance and the selection of more aggressive clones through upregulation of VEGF expression.11

Bevacizumab, a humanized monoclonal antibody against VEGF, was evaluated in patients with metastatic melanoma, but its single-agent clinical activity was found to be only modest in a phase 2 study.12 This disappointing single-agent activity in melanoma is not surprising, because it is the case for most solid tumors except those whose growth is driven by VEGF signaling, such as renal cell carcinoma. However, bevacizumab has been shown to improve clinical outcomes in various tumors, including non-small cell lung cancer, breast cancer, and colorectal cancer, when it is combined with chemotherapeutic agents.13-15 In patients with melanoma, there appears to be some additive clinical benefit to combining bevacizumab with cytotoxic agents. In a multicenter phase 2 trial, von Moos et al demonstrated that the disease control rate (objective response and disease stabilization rate at week 12) for the combination of temozolomide and bevacizumab was 52%, with a median PFS of 4 months, among 62 previously untreated patients with metastatic melanoma.16 In a randomized phase 2 study of 214 patients, my colleagues and I demonstrated that there was a trend toward a superior objective response rate and PFS and OS durations with the addition of bevacizumab to the combination of carboplatin and paclitaxel, in comparison with carboplatin and paclitaxel alone.17

In the randomized, open-label, phase 2 study published in this issue of Cancer, Kottschade et al evaluated the clinical efficacy of the combination of bevacizumab with temozolomide (TB) or bevacizumab with nab-paclitaxel and carboplatin (ABC) in treatment-naive patients with metastatic melanoma.18 A total of 93 patients were randomized in this study, among whom 42 patients and 51 patients, respectively, were treated with the TB regimen and the ABC regimen. Using Response Evaluation Criteria In Solid Tumors (RECIST; version 1.0), the PFS rates at 6 months and the response rates were evaluated separately in each treatment arm; the study was not designed to compare the clinical efficacy between the 2 treatment arms. The authors demonstrated that for the ABC regimen, the 6-month PFS rate, which was the primary endpoint, was 54.9% and the objective response rate was 33.3%; the median PFS and OS durations were 6.7 months and 13.9 months, respectively. In the TB treatment arm, the 6-month PFS rate was 32.8% and the objective response rate was 23.8%; the median PFS and OS durations were 3.8 months and 12.3 months, respectively.18

The clinical efficacy of the ABC regimen reported by Kottschade et al18 is certainly encouraging. In 2 large multicenter clinical studies, the combination of carboplatin and paclitaxel had a response rate of approximately 16%, a 24-week PFS rate of approximately 37%, a median PFS duration of approximately 4 months, and a median OS duration of 9 to 11 months among chemotherapy-naive patients with advanced melanoma.3, 17 However, it is not clear whether the substitution of nab-paclitaxel for paclitaxel in the ABC regimen contributed to its promising clinical efficacy. In the randomized phase 2 study of carboplatin and paclitaxel with or without bevacizumab,17 the response rate was 26%, the 24-week PFS rate was 50%, the median PFS duration was approximately 6 months, and the median OS duration was 12 months for patients who received bevacizumab in addition to the chemotherapy. When the clinical efficacy parameters of the 23 patients who were treated with more tolerable doses of the ABC regimen (post-Addendum 5) reported by Kottschade et al18 are compared with the study data from the combination of carboplatin, paclitaxel, and bevacizumab, no apparent advantage of weekly nab-paclitaxel compared with paclitaxel is noted.

However, the data from these studies do indicate a possible additive clinical benefit from the addition of bevacizumab to the cytotoxic chemotherapy. Although the results of what to my knowledge is the only randomized trial to date to examine the addition of bevacizumab did not demonstrate statistically significant differences in clinical outcome between the chemotherapy regimen with or without bevacizumab, all of the clinical efficacy metrics were found to be more favorable with the addition of bevacizumab.17 Likewise, in the current study by Kottschade et al, the response, PFS, and OS data in both the TB and ABC treatment arms were more favorable compared with historical data for cytotoxic chemotherapy alone.18 On the basis of the multitude of promising clinical efficacy results from several clinical studies, a large randomized phase 3 study to evaluate the true clinical benefit of adding bevacizumab to chemotherapy in patients with metastatic melanoma is warranted. However, it is still debatable as to which bevacizumab-containing combination regimen and which comparator regimen should be used in the design of the study.

In addition to VEGF-binding drugs such as bevacizumab, there are a large number of small molecules and antibodies that target the VEGF receptors (VEGFRs). Sorafenib is one of the VEGFR inhibitors that have been tested in patients with metastatic melanoma. However, this multikinase inhibitor was found to have no meaningful clinical benefit as a single agent as well as in combination with cytotoxic drugs (carboplatin and paclitaxel).3, 4, 19 The failure to meaningfully improve clinical benefit with the addition of these VEGF-/VEGFR-targeting drugs to chemotherapeutic agent(s) may be related to combining drugs rather empirically without convincing preclinical models demonstrating a drug synergy. In addition, some of these VEGF-/VEGFR-targeting drugs may induce various molecular modulations other than antiangiogenic actions, which can make the contribution of these agents to the anticancer activity more difficult to assess. For example, sorafenib may inhibit the RAF-mediated mitogen-activated protein kinase pathway, leading to tumor growth arrest in a subset of patients with melanoma. In addition, VEGF inhibits the differentiation and maturation of dendritic cells,10 potentially leading to immunosuppression, and accordingly, drugs inhibiting the VEGF axis could improve the immune functions, as demonstrated by bevacizumab enhancing dendritic cell and T lymphocyte proliferation.10

More selective VEGFR inhibitors have recently been developed and tested. A multicenter, phase 2 study of axitinib, an oral and selective inhibitor of VEGFR-1, VEGFR-2, and VEGFR-3, demonstrated encouraging results.20 Among 32 enrolled patients, 6 (19%) had objective clinical responses. The addition of more effective antiangiogenic drugs to conventional chemotherapeutic drugs or to immune or targeted therapy drugs may potentially yield superior tumor regression and control, and such combination regimens, when based on strong scientific rationale and robust preclinical data, should be evaluated.

In the current study, the doses of the TB regimen (200 mg/m2 of oral temozolomide daily for 5 days and 10 mg/kg of intravenous bevacizumab on days 1 and 15, every 28 days) were well tolerated, with only 3 patients requiring treatment discontinuation and 3 others requiring dose reduction during subsequent cycles. However, in the ABC regimen arm, the original doses (100 mg/m2 of intravenous nab-paclitaxel on days 1, 8, and 15; 10 mg/kg of intravenous bevacizumab on days 1 and 15; and an area under the curve of 6 of intravenous carboplatin on day 1, every 28 days) were not easily tolerable, with frequent ≥ grade 3 hematologic toxicities and fatigue (according to Common Terminology Criteria for Adverse Events, version 3.0), and the dose of nab-paclitaxel had to be reduced to 80 mg/m2 after the first 28 patients. Despite this dose reduction (post-Addendum 5), a significant percentage of the patients still experienced severe myelosuppression, especially neutropenia. The safety profile of the ABC regimen is less favorable than that of the combination of paclitaxel, carboplatin, and bevacizumab.17 If we consider both the clinical efficacy and the tolerability of the 2 regimens, the value of nab-paclitaxel over paclitaxel is not convincing.

Kottschade et al should be commended for their effort to identify predictive plasma biomarkers.18 They found that plasma VEGF-A levels were decreased with 1 cycle of either ABC or TB treatment in a majority of the patients; however, there was no correlation found between the changes in VEGF-A or VEGF-D levels and the clinical response. Therefore, the changes in VEGF-A with the treatment was neither a predictive nor a surrogate marker for antitumor effect in the study.18 This lack of correlation underscores the complicated mechanisms of melanoma apoptosis and progression. Although VEGF has been implicated in melanoma progression and in resistance to cytotoxic drugs, it is not likely that the VEGF-VEGF receptor pathway is the major driving force behind melanoma cell survival and proliferation. In addition, there are other angiogenic factors, such as bFGF and IL-8, that play an important role in endothelial proliferation and the resistance of endothelial cells to VEGF blockage.21 Furthermore, the cytotoxic effect of nab-paclitaxel and carboplatin could have resulted in tumor regression, complicating the correlative analysis of the angiogenic proteins in this study. However, a robust tumor and/or blood biomarker analysis is strongly encouraged in future clinical studies evaluating antiangiogenic or targeted drugs.

In light of recent advances in molecular targeting drugs, exemplified by vemurafenib, and checkpoint immunotherapeutic agents, such as ipilimumab, the future of systemic therapy for melanoma is bright. However, long-term melanoma control is still elusive in the majority of patients with advanced disease, and we need more effective treatment strategies. The success of bevacizumab or other antiangiogenic drugs in the treatment of patients with advanced melanoma will depend on a better understanding of tumoral neovascularization and its effect on tumor stromal tissues, as well as on selecting patients with appropriate relevant biomarkers to yield improved clinical outcomes.


No specific funding was disclosed.


Dr. Kim has served on an advisory board for and has received honoraria and research support from Genentech.