Bevacizumab: A Controversial Agent Against High-Grade Gliomas

Tumor Bevacizumab: A Controversial Agent Against High-Grade Gliomas Sussan Salas, MD 1, Miguel Guzman, MD 2, Kevin Judy, MD 1 1 Department of Neurological Surgery, Thomas Jefferson University, Philadelphia, PA 2 Department of Pathology, St. Louis University School of Medicine, St. Louis, MO Introduction Glioblastoma multiforme (GBM) is the most common primary malignant brain tumor in adults. Despite the current regimen of surgical resection with subsequent external beam radiotherapy and temozolomide, mean survival is 14.6 months and 2-year survival is 26%. 1 GBM is a highly vascular tumor, a result of its increased expression of vascular endothelial growth factor (VEGF) compared to other brain tumors. VEGF promotes endothelial cell proliferation, and is thought to have a pivotal role during tumor progression. Multiple treatment modalities have targeted VEGF and VEGF receptors (VEGFRs) due to their essential roles in the regulation of angiogenic processes. 2 Bevacizumab is a recombinant humanized monoclonal antibody that inhibits VEGF. Positive results from Phase II clinical trials with bevacizumab for recurrent GBM led to its U.S. Food and Drug Administration approval. 3,4 Bevacizumab can produce significant decrease in contrast enhancement as early as 1 to 2 days after the beginning of treatment, and often results in radiologic response rates of 25% to 60%. 3,5 However, it has been noted that rapid improvement in radiographic response is not directly correlated to decreased tumor burden or improved survival. Furthermore, there are significant adverse effects associated with this agent that must be considered when tailoring therapy. These topics and current studies evaluating the use of bevacizumab for high-grade gliomas are discussed. Criteria for Radiographic Response Macdonald, et al. published criteria in 1990 for assessment of response of high-grade gliomas to treatment. These criteria were developed in an effort to adopt more uniform and rigorous norms of treatment, and have become the standard for assessing response to therapy in high-grade gliomas. They are based on two-dimensional tumor measurements on contrast-enhanced computed tomographic (CT) or magnetic resonance imaging (MRI) scans, in conjunction with corticosteroid use and clinical assessment. Tumor size is measured as the largest cross-sectional area (largest cross-sectional diameter x largest diameter perpendicular to it). The Macdonald criteria define progression of disease as either a 25% or more increase in size of the contrast enhancing lesion on sequential MRI scans, or clinical deterioration that is unexplained (Table 1). 6 Table 1. Macdonald Criteria: Current response criteria for malignant gliomas Response Complete response Partial response Stable disease Progression Criteria Requires all of the following: complete disappearance of all enhancing measurable and non-measurable disease sustained for at least 4 weeks; no new lesions; no corticosteroids; and stable or improved clinically Requires all of the following: 50% decrease compared with baseline in the sum of products of perpendicular diameters of all measurable enhancing lesions sustained for at least 4 weeks; no new lesions; stable or reduced corticosteroid dose; and stable or improved clinically Requires all of the following: does not qualify for complete response, partial response, or progression; and stable clinically Defined by any of the following: 25% increase in sum of the products of perpendicular diameters of enhancing lesions; any new lesion; or clinical deterioration The Macdonald criteria have a number of important limitations. Contrast enhancement is nonspecific and can represent treatment-related inflammation, seizure activity, postsurgical changes, ischemia, and radiation necrosis, among others. Multiple factors can influence enhancement such as steroid doses, changes in radiologic techniques, and anti-angiogenic agents such as bevacizumab. By normalizing the abnormally permeable tumor vessels, anti-vegf therapy produces a rapid decrease in enhancement that does not correlate with a decrease in tumor size. Furthermore, a significant number of patients treated with bevacizumab who initially experience a decrease in tumor contrast enhancement subsequently develop progressive increase in non-enhancing T2 or FLAIR signals suggestive of infiltrative tumor (Figure 1). 7 It has been shown that a non-enhancing tumor pattern of progression after treatment with bevacizumab is correlated with worse survival. 8 These findings have triggered multiple studies examining the effect of bevacizumab on tumor cells and employing different imaging modalities to accurately assess tumor response. Physiologic effect of bevacizumab and resulting imaging characteristics There is a significant disparity between the unparalleled imaging response rates that bevacizumab produces in recurrent malignant gliomas and the modest survival benefits, if any, that have been reported. 9 These apparent responses to anti-angiogenic therapy may be partly a result of normalization of abnormally permeable tumor vessels and have been demonstrated not to be representative of an anti-glioma effect. By decreasing contrast leakage, anti-vegf therapy produces a rapid decrease in enhancement that does not always correlate with a decrease in tumor size. Therefore, caution needs to be exercised when evaluating imaging studies for response to antiangiogenic agents. 10 Iwamoto, et al. studied the effect of bevacizumab therapy on GBM growth by analyzing the imaging and histological characteristics of patients who discontinued bevacizumab therapy because of tumor progression. At the time bevacizumab was discontinued, the pattern 11

Figure 1 T1 GD of progression was predominantly nonenhancing tumors in 13 of 37 patients (35%). These results differ greatly from the typically observed pattern of increased enhancement at the site of recurrent disease that occurs in 90-95% of patients who do not receive bevacizumab. 11 Lower Karnofsky performance status (KPS) and a non-enhancing pattern of recurrence were associated with shorter overall survival after discontinuing bevacizumab (p values 0.001 and 0.05, respectively). Iwamoto, Flair Progression of disease after bevacizumab treatment. The top row shows no Gd enhancement and minimal increased Flair signal. The middle row shows Gd-enhancing tumor and a bi-frontal pattern of increased Flair signal at the time of tumor recurrence. The bottom row shows resolution of the Gd enhancement with the progression of Flair signal in both frontal lobes presenting a confusing radiographic picture. Left frontal tumor after surgery, XRT and Temozolomide Tumor recurrence at 6 months Tumor after 5 months of Bevacizumab et al. stipulated that the apparent relationship between a non-enhancing pattern and shorter survival may reflect a change in tumor biology, particularly because non-enhancing tumor progression was a negative prognostic factor independent of performance status. 8 Histological analysis before and after treatment demonstrated a significant increase of hypoxia markers after bevacizumab failure. It has been well established in the literature that hypoxia is a promoter of angiogenesis, tumor invasion, and resistance to therapy. 12 These findings strongly suggest that the anti-angiogenic effect of bevacizumab resulted in an increase in tumor hypoxia, leading to more invasive tumor behavior. Other studies contend that bevacizumab and other anti-vegf therapies mediate temporary control of tumor growth but do not achieve a change in overall survival. Norden, et al. reported their analysis of 34 patients with recurrent malignant glioma that underwent treatment with either bevacizumab or cediranib (AZD2171, a pan-vegf receptor inhibitor) and compared them to 18 patients that underwent cytotoxic chemotherapy (with either edotecarin or gimatecan). Median progressionfree survival was 8 weeks in patients treated with cytotoxic therapy, compared to 22 weeks in patients treated with anti-angiogenic therapy (P=0.01). However, median overall survival was nearly identical in the two groups, with 39 weeks in the cytotoxic therapy group and 37 weeks in the anti-angiogenic therapy group. These results suggest that anti-angiogenic therapy does not prolong overall survival in patients with recurrent malignant glioma. 9 The increased progression-free survival observed could reflect the initial impairment of tumor growth that is short-lived (and may be seen in the rapid decrease in enhancement that is observed after the beginning of treatment), with eventual onset of different forms of resistance that inevitably result in progressive disease. 13 An increasing number of studies investigating the mechanisms of tumor resistance to anti-angiogenic agents suggest that anti- VEGF therapy may result in the co-option of normal vasculature, producing an invasive non-enhancing phenotype. 13-15 Rubenstein, et al. analyzed the histologic pattern of growth of human glioblastoma cells that were stereotactically implanted in the striatum of adult athymic rats that were then treated with systemic anti-vegf antibody. Although anti- VEGF treatment resulted in prolongation of survival (median survival of control vs. treated animals was 18.5 vs. 34.5 days, respectively, P<0.0001), tumors adopted a more infiltrative and invasive pattern on histological evaluation (Figure 2). Comparison of sections of control and anti-vegf treated tumors revealed that, instead of growing as a single confluent mass with a distinct border, anti-vegf treated tumors were surrounded by multiple satellite tumors that appeared to extend from the margin of the original tumor. Immunohistochemistry analysis demonstrated that greater than 90% of the satellite tumors were associated with at least one blood vessel, suggesting 12

Tumor that these tumors infiltrated the brain to adopt or co-opt existing vessels in response to anti-angiogenic treatment (Figure 3). 15 A Paez-Ribes, et al. noted a similar pattern when analyzing the effects of anti-vegf treatment on an orthotopic mouse model of glioblastoma. A B highly invasive and qualitatively distinct tumor pattern was observed, with tumor cells found in close proximity to resident normal blood vessels.13 Further preclinical studies are necessary to elucidate the apparent ability of malignant gliomas to adapt to inhibition of angiogenesis by increased infiltration and cooption of host vessels. Diffusion weighted imaging as biomarker for bevacizumab treatment response C D Figure 2 Recurrent glioblastoma after treatment with bevacizumab and CPT-11. The original GBM shows pseudopalisading necrosis (A) and tumor around abnormal vessels (B). Histology after bevacizumab treatment shows malignant glial cells (C) and mesenchymal proliferation around vessels (D). Histological evaluation demonstrated tumor cells tacking along blood vessels, suggesting co-option of host vessels as a response to anti-angiogenic treatment. Note the profound mesenchymal proliferation around the tumor vessels following bevacizumab treatment. A B Imaging criteria that include non-enhancing components of tumors are necessary to assess the response to bevacizumab treatment. However, evaluating FLAIR or T2 signal abnormality can lead to an inaccurate analysis of response to treatment, since infiltrating tumor is difficult to differentiate from edema, gliosis, or treatment related leukoencephalopathy (Figure 4). Jain, et al. addressed this issue by obtaining volumetric measurements of contrast enhancing lesions and non-enhancing lesions of patients treated with bevacizumab alone or concurrent chemotherapy for a follow-up period of 1 year. These regions of interest were co-registered with corresponding diffusion weighted imaging (DWI) and apparent diffusion coefficient (ADC) maps (lower ADC values are suggestive of high cellularity and higher grade of the tumor, whereas the opposite is true for higher ADC values). Imaging analysis demonstrated a progressive decrease in ADC values especially early on (6 weeks and 3 months after treatment) in non-enhancing lesions in patients that had progression of disease, even though C Figure 3 (A) Original GBM histology showing positivity for VEGF immunohistochemistry; (B) In this recurrent GBM, the sarcomatoid vessels remain positive for endothelial marker CD34; (C) However VEGF immunostain is negative in the transformed vessels. In contrast to a glioblastoma without treatment (which shows strong and diffuse VEGF positive staining) this bevacizumab-treated tumor shows no reaction with VEGF immunostaining. 13

T1 GD Flair ADC DWI Recurrent left frontal GBM Left frontal GBM after 6 months of Bevacizumab Figure 4 This patient presented with a recurrent left frontal GBM with enhancing tumor and increased Flair signal of both frontal lobes. After 6 months of bevacizumab there was a dramatic resolution of the enhancing tumor. However, there has been progression of decreased ADC signal and increased DWI signal in the right frontal lobe that represented tumor progression. they showed a progressive decrease in size of contrast enhancing lesions. This suggests that ADC and DWI measurements can be used to determine early treatment failure by demonstrating non-enhancing infiltrative tumor growth. 16 Efforts are currently underway to identify more accurate radiographic means of evaluating the response of tumors to anti-angiogenic treatment. Studies suggest that advanced MRI techniques such as perfusion imaging (dynamic susceptibility MRI), permeability imaging (dynamic contrast-enhanced MRI), magnetic resonance spectroscopy, and [18F]-fluorothymidine and amino acid positron emission tomography may demonstrate tumor response or differentiate non-enhancing tumor from other causes of increased FLAIR signal. 10,17 Adverse Effects Studies have reported adverse effects such as brain hemorrhages, epistaxis, oral cavity bleeding, vaginal bleeding, proteinuria, spontaneous colon perforations, thromboembolic complications, hypertension, and impaired craniotomy wound healing, among others. Vredenburgh, et al. noted that 4 out of 32 patients (12.5%) undergoing treatment with bevacizumab for recurrent malignant glioma had thromboembolic complications, and 2 patients (6.25%) died one from a pulmonary embolus and the other from an arterial ischemic stroke.4 Kreisl, et al. reported thromboembolic events occurring in 6 out of 48 patients (12.5%) receiving bevacizumab therapy after glioblastoma recurrence. Three of these events were pulmonary emboli, and one was a cerebral vascular event; all patients were removed from the study for drug-associated toxicity. One patient had a bowel perforation. 3 Norden et al. also reported a colon perforation that necessitated emergent surgery on a patient undergoing bevacizumab therapy for malignant glioma. Nine out of 55 patients (16.4%) experienced hemorrhage, primarily epistaxis or other mucosal bleeding. Two patients had asymptomatic brain hemorrhages that were detected on routine imaging. 7 Patients with malignant gliomas have been demonstrated to be at a high risk of thromboembolic events regardless of therapy due to a tumor-induced hypercoagulable state; procoagulant an anti-fibrinolytic substances are preferentially secreted by malignant gliomas compared to other brain tumors. 18 Clearly, the role of anti-angiogenic agents in increasing the risk of thromboembolic and other adverse effects needs to be better defined. Conclusion New criteria for radiographic response, along with innovative imaging techniques, are necessary for future clinical trials employing anti-vegf therapy. Further studies elucidating the molecular interaction between tumor cells and anti-angiogenic agents such as bevacizumab are warranted. The incidence of adverse effects when undergoing bevacizumab therapy is significant and should be appropriately considered when tailoring therapy to an individual patient. References 1. Stupp R, Mason WP, van der Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352(10):987-96. 2. Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med 2003;9(6):669-76. 3. Kreisl TN, Kim L, Moore K, et al. Phase II trial of single-agent bevacizumab followed by bevacizumab plus irinotecan at tumor progression in recurrent glioblastoma. J Clin Oncol 2009;27:740-745. 14

Tumor 4. Vredenburgh JJ, Desjardins A, Herndon JE 2nd et al. Bevacizumab plus irinotecan in recurrent glioblastoma multiforme. J Clin Oncol 2007;25(30):4722-9. 5. Vredenburgh JJ, Desjardins A, Herndon JE II, et al. Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 2007;13:1253-59 6. Macdonald DR, Cascino TL, Schold SJ, et al. Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 1990;8(7):1277-80. 7. Norden AD, Young GS, Setayesh K, et al. Bevacizumab for recurrent malignant gliomas. Neurology 2008; 70(10):779-87. 8. Iwamoto FM, Abrey LE, Beal K, et al. Patterns of relapse and prognosis after bevacizumab failure in recurrent glioblastoma. Neurology 2009;73(15):1200-6. 9. Norden AD, Drappatz J, Muzikansky A, et al. An exploratory survival analysis of anti-angiogenic therapy for recurrent malignant glioma. J Neurooncol 2009;92(2):149-55. 10. Wen PY, Macdonald DR, Reardon DA, et al. Updated response assessment criteria for high-grade gliomas: response assessment in neuro-oncology working group. J Clin Oncol 2010;28(11):1963-72. 11. Gaspar LE, Fisher BJ, Macdonald DR, et al. Supratentorial malignant glioma: patterns of recurrence and implications for external bean local treatment. Int J Radiat Oncol Biol Phys 1992;24:55-7. 12. Maxwell PH, Dachs GU, Gleadle JM, et al. Hypoxiainducible factor-1 modulates gene expression in solid tumors and influences both angiogenesis and tumor growth. Proc Natl Acad Sci USA 1997;94:8104-9. 13. Paez-Ribes M, Allen E, Hudock J, et al. Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 2009;15(3):220-31. 14. Bergers G, Hanahan D. Modes of resistance to anti-angiogenic therapy. Nat Rev Cancer 8(8):592-603, 2008. 15. Rubenstein JL, Kim J, Ozawa T, et al. Anti-VEGF antibody treatment of glioblastoma prolongs survival but results in increased vascular cooption. Neoplasia 2000;2(4):306-14. 16. Jain R, Scarpace LM, Ellika S, et al. Imaging response criteria for recurrent gliomas treated with bevacizumab: role of diffusion weighted imaging as an imaging biomarker. J Neurooncol 2010;96(3):423-31. 17. Pope WB, Kim HJ, Huo J, et al. Recurrent glioblastoma multiforme: ADC histogram analysis predicts response to Bevacizumab treatment. Radiology 2009;252(1):182-9. 18. Marras LC, Geerts WH, Perry JR. The risk of venous thromboembolism is increased throughout the course of malignant glioma. Cancer 2000;89(3):640-6. Brain power Jefferson neurosurgeons have helped to write the textbooks and set training standards referenced by other doctors around the world. No wonder our surgeons treat the largest combined volume of tumors, aneurysms and AVMs in the Philadelphia region. The region s only hospital with an experienced, specialized staff dedicated to neuroscience. Offices in Center City Philadelphia, Langhorne and Voorhees. Most insurance plans accepted, including Medicare. Thomas Jefferson University Hospital Jefferson Hospital for Neuroscience Methodist Hospital www.jeffersonhospital.org/neuroscience 1-800-JEFF-NOW 1-800-533-3669 15