Antiangiogenic agents with chemotherapy for recurrent glioblastoma
In the initial investigation in patients with recurrent glioblastoma, bevacizumab was evaluated in combination with concomitant irinotecan [
24]. This combination was supported by the activity of bevacizumab with irinotecan-containing regimens in patients with metastatic colorectal cancer [
25], by the relative lack of single-agent activity of thalidomide in recurrent glioblastoma [
26], and by preclinical evidence, suggesting that antiangiogenic agents enhance intratumoral chemotherapy delivery [
19,
27]. Additionally, antiangiogenic agents may supplement the effect of chemotherapy by inhibiting the activity of a population of SCLGCs that is not as well differentiated (i.e., chemotherapy-resistant) [
20]. The existence of these cells may partially explain tumor resistance to radiotherapy and chemotherapy, and could contribute to the recurrence of glioblastoma.
Use of bevacizumab with chemotherapy
Data from prospective and retrospective studies indicate that regimens combining bevacizumab and chemotherapy produce superior outcomes relative to those with conventional chemotherapy in patients with recurrent glioblastoma [
7]. In the first prospectively designed, phase II trial, patients with recurrent glioblastoma received bevacizumab plus irinotecan in one of two treatment cohorts: the first cohort (n = 23) received bevacizumab 10 mg/kg plus irinotecan q2w in a 6-week cycle, and a second cohort (n = 12) received bevacizumab 15 mg/kg q3w with irinotecan on days 1, 8, 22, and 29 of a 6-week cycle [
28,
29]. In both cohorts, irinotecan was administered at 340 mg/m
2 to 350 mg/m
2 in patients on enzyme-inducing antiepileptic drugs (EIAEDs) and at 125 mg/m
2 in those not receiving EIAEDs. The 6-month PFS rate among all 35 patients was 46%, the 6-month OS rate was 77%, and the median OS was 42 weeks [
29]. In addition, the overall response rate (ORR) was high (57%). Recently, the 4-year survival rate in this trial was reported to be 11% [
30] (Table
1). The toxicity of the combination of bevacizumab and irinotecan was considered to be significant but acceptable, considering the poor prognosis of the population [
29]. Eleven (31%) of the 35 patients discontinued treatment because of adverse events, which included thromboembolic complications (n = 4), grade 2 proteinuria (n = 2), and grade 2 fatigue (n = 4); one patient experienced a CNS hemorrhage.
Table 1
Efficacy outcomes with antiangiogenic agents in recurrent glioblastoma.
Bevacizumab
| | | | | | |
Vredenburgh [ 28], (n = 23 of 32) d | BV + irinotecan | 4 | 57 | 35 | 20 weeks | 30 |
Vredenburgh [ 29], (N = 35) | BV + irinotecan | 57 | N/A | 24 weeks | 46 |
Narayana [ 33], (n = 37 of 61) e | BV + irinotecan or carboplatin | 13 | 60 | 21 | 5 months | N/A |
Friedman [ 31], Cloughesy [ 32], (N = 167) | BV alone (n = 85) BV + irinotecan (n = 82) | 28 38 | N/A N/A | 4.2 months 5.6 months | 43 50 |
Reardon [ 38], (n = 27 of 59) d | BV + etoposide | 4 | 19 | 70 | 18 weeks | 44 |
| BV → BV + irinotecan | 71 (Levin criteria); 35 (MacDonald criteria) | N/A | 16 weeks | 29 |
Gutin [ 92], (n = 20 of 25) d | BV + hypofractionated stereotactic irradiation | 50 | N/A | 7.3 months | 65 |
Aflibercept
| | | | | | |
De Groot [ 53], (n = 32 of 48) d | Aflibercept alone | 0 | 30 | 52 | N/A | N/A |
Cediranib
| | | | | | |
Batchelor [ 112], (N = 31) | Cediranib alone | 57 (volumetric criteria); 27 (MacDonald criteria) | N/A | 117 days | 26 |
Cilengitide
| | | | | | |
| Cilengitide alone (2000 mg/d [n = 40] or 500 mg/d [n = 41]) | 0 | 9 | N/A | 2000 mg/d, 8.1 weeksc 500 mg/d, 7.9 weeksc | 2000 mg/d, 15 500 mg/d, 10 |
CT-322
| | | | | | |
| CT-322 alone (n = 33) CT-322 + irinotecan (n = 18) | 1 (3) 0 | 1 (3) 0 | N/A N/A | N/A N/A | 23 48 |
XL184
| | | | | | |
| XL184 175 mg qd XL 184 125 mg qd | AAT-naive (n = 34), 21 AAT-pretreated (n = 12), 8 AAT-naive (n = 37), 30 AAT-pretreated (n = 22), 0 | N/A | AAT-naive, 16 weeks AAT-naive, 16 weeks AAT-pretreated, 7.9 weeks | AAT-naive, 10 AAT-naive, 25 AAT-pretreated, 0 |
More recently, Friedman and colleagues investigated the use of bevacizumab with or without irinotecan in a randomized noncomparative phase II trial of 167 patients with recurrent glioblastoma-the BRAIN study [
31,
32]. In this trial, patients were randomized to bevacizumab 10 mg/kg q2w alone (n = 85) or in combination with irinotecan (n = 82). For patients treated with bevacizumab and irinotecan, the estimated 6-month PFS rate was 50.3%, the median OS was 8.9 months, and the ORR was 37.8% at the 6-month follow-up. At 27 months of follow-up, the 12-, 18-, 24-, and 30-month survival rates were 38%, 18%, 17%, and 16%, respectively. In the safety population for the combination arm (n = 79), the most common grade ≥ 3 adverse events were convulsion (13.9%), neutropenia (8.9%), and fatigue (8.9%). Adverse events led to treatment discontinuation for 14 (17.7%) patients. Adverse events associated with bevacizumab included grade ≥ 3 arterial thromboembolism (2.5%), grade ≥ 3 wound-healing complications (1.3%), grade ≥ 3 venous thromboembolism (10.1%), grade 3 gastrointestinal perforation (2.5%), serious reversible posterior leukoencephalopathy syndrome (1.3%), and intracranial hemorrhage (3.8%). In addition, there was one death associated with convulsion in patients treated with bevacizumab and irinotecan.
Data from additional phase II studies, retrospective analyses, and case series of consecutive patients have provided further support for the activity of bevacizumab with chemotherapy in patients with recurrent glioblastoma [
33‐
39]. In these studies, 6-month PFS rates have ranged from 6.7% to 64% in patients with recurrent glioblastoma. In general, bevacizumab was shown to be well tolerated in both prospective and retrospective studies, and no unexpected treatment-related adverse events were reported (Table
2). Reported events were typical of those associated with bevacizumab in the treatment of other tumor types. For example, hypertension and proteinuria have been reported as the most frequently occurring treatment-related adverse events in studies of bevacizumab therapy in other solid tumors [
11,
25,
40]. The incidence of thromboembolic complications in patients with recurrent glioblastoma receiving bevacizumab plus chemotherapy ranged from 11.4% to 12.7% in the two prospective studies [
28,
29,
32]. The relation of bevacizumab to these events, however, is unclear because patients with malignant gliomas are already at an increased risk for symptomatic venous thromboembolism. In a retrospective study of 9489 cases of malignant glioma, the 2-year cumulative incidence of venous thromboembolism was relatively high at 7.5% (n = 715 cases) [
41]. Furthermore, a diagnosis of glioblastoma was identified as a specific risk factor for venous thromboembolism (hazard ratio [HR] = 1.7; 95% confidence interval [CI], 1.4-2.1). Overall, the safety profile of bevacizumab with chemotherapy has been within acceptable limits, without any indications of additive toxicities.
Table 2
Safety profile of antiangiogenic agents for recurrent glioblastoma.
Bevacizumab-containing regimens
|
(n = 23 of 32)c | BV + irinotecan | 9 (28.1) | N/A | 0 | 4 (12.5) | 2 (6.3) |
(N = 35) | BV + irinotecan | 11 (31.4) | N/A | 1 (2.9) | 4 (11.4) | N/A |
(n = 37 of 61)c | BV + irinotecan or carboplatin | 16 (26.2) | Bone marrow toxicity, 6 (9.8) | 6 (9.8) | 6 (9.8) | 0 |
Friedman [ 31], Cloughesy [ 32], (N = 167) | BV alone (n = 84) BV + irinotecan (n = 79) | 4 (4.8) 14 (17.7) | All, 43 (51.2) Hypertension, 9 (10.7) Wound-healing complications, 2 (2.4) Proteinuria, 1 (1.2) All, 56 (70.9) Hypertension, 3 (3.8) Wound-healing complications, 1 (1.3) Proteinuria, 3 (3.8) GI perforation, 2 (2.5) | 3 (3.6) 3 (3.8) | ATE, 4 (4.8) VTE, 3 (3.6) ATE, 3 (3.8) VTE, 9 (11.4) | 2 (2.4) 1 (1.3) |
(n = 27 of 59)c | BV + etoposide | 7 (11.9) | Neutropenia, 14 (23.7) Infection, 5 (8.5) Hypertension, 2 (3.4) | CNS hemorrhage, 2 (3.4) | 7 (11.9) | 1 (1.7) |
(N = 48) | BV → BV + irinotecan | 6 (12.5) | Hypertension, 2 (4.2) Hypophosphatemia, 2 (4.2) Bowel perforation, 1 (2.1) | 0 | 6 (12.5) | N/A |
(n = 20 of 25)c | BV + hypofractionated stereotactic irradiation | 3 (12) | Lymphopenia, 9 (36) Hyponatremia, 6 (24) Bowel perforation, 1 (4) Wound-healing complication, 1 (4%) GI bleeding, 1 (4%) | 1 (4) | N/A | N/A |
Aflibercept
| | | | | | |
(n = 32 of 48)c | Aflibercept alone | 12 (25) | CNS ischemia, 1 (2.1) Systemic hemorrhage, 1 (2.1) | N/A | N/A | N/A |
Cediranib
| | | | | | |
(N = 31) | Cediranib alone | 2 (6.5) | Fatigue, 6 (19.4) ALT, 5 (16.1) Hypertension, 4 (12.9) | N/A | 1 (3.2) | 0 |
Cilengitide
| | | | | | |
(N = 81) | Cilengitide alone (2000 mg/d [n = 40] or 500 mg/d [n = 41]) | N/A | Convulsion, 2 (2.5) Lymphopenia, 7 (8.6) Neutropenia, 1 (1.2) | 1 (1.2) | N/A | 5 (6.2) |
CT-322
| | | | | | |
(n = 51) | CT-322 ± irinotecan | 13 (25.5) | Neutropenia, 4 (7.8) Hypertension, 3 (5.9) | CNS hemorrhage, 1 (2.0) | N/A | 1 (2.0) |
XL184
| | | | | | |
(n = 153) | XL184 (175 mg qd [n = 46] or 125 mg qd [n = 107]) | 18 (11.8) | Fatigue, 31 (20.3) Hypertension, 8 (5.2) GI perforation, 3 (2.0) Wound-healing complications, 2 (1.3) | 3 (2.0; grade 3/4) | 17 (11.1) | N/A |
Other antiangiogenic therapies used with chemotherapy for recurrent glioblastoma
Clinical trials have also evaluated the safety and efficacy of other antiangiogenics, specifically thalidomide and vatalanib, in combination with chemotherapy agents. In phase II trials of patients with recurrent glioblastoma, thalidomide-containing regimens produced 6-month PFS rates between 23% and 27% and objective response rates between 6% and 24% [
42‐
45]. Although the findings of two of these studies suggested that combination therapy was more active than either thalidomide or the chemotherapy partner alone, the benefit-to-risk ratio of thalidomide-containing therapy has not been clearly established, particularly when considering that certain combinations are complicated by significant adverse events (e.g., neutropenia and thromboembolism). A phase I/II trial of vatalanib plus temozolomide (n = 37) or lomustine (n = 23) provided evidence of activity in patients with recurrent glioblastoma-patients receiving vatalanib and temozolomide had a median time to progression of 16.1 weeks and a partial response rate of 9% across all dose groups [
46]. However, vatalanib has since been discontinued from further investigation in patients with glioblastoma.
Continued use of antiangiogenic agents after progression
In the event of progression following treatment with an antiangiogenic agent, patients with glioblastoma have very few therapeutic options. For example, in a prospective study by Kreisl and colleagues, a cohort of 19 patients was subsequently treated with bevacizumab plus irinotecan after progression on bevacizumab monotherapy [
49]. None of these patients responded to therapy, and the median PFS was 30 days. In another prospective phase II study of patients with recurrent malignant gliomas treated with daily temozolomide, it was found that patients with prior exposure to bevacizumab fared worse than patients without bevacizumab exposure (6-month PFS rate of 14% vs 36%, p = 0.12; median OS of 4 vs 18 months, p = 0.005) [
55]. Retrospective reviews of patients with glioblastoma treated either with a bevacizumab-containing regimen or bevacizumab alone have also reported that these patients have limited response to a second treatment, regardless of whether it contains bevacizumab [
36,
56‐
59]. One hypothesis for the lack of response after antiangiogenic treatment is that an alteration of the tumor phenotype results in a highly infiltrative compartment that is angiogenic-independent. Further studies are warranted to identify new therapeutic targets and novel agents that could treat patients who have relapsed following antiangiogenic therapy.
One of the concerns with the administration of antiangiogenic agents is the apparent potential for infiltrative or invasive tumor growth upon disease progression [
33,
35,
36,
60‐
62]. Recent reports, however, indicate that antiangiogenic treatments may not significantly alter patterns of relapse in glioblastoma. For example, in a study of distant spread in 44 matched pairs of patients with recurrent glioblastoma treated with or without bevacizumab-containing regimens, distant recurrences were later observed in 22% (10/44) of bevacizumab-treated patients compared with 18% (8/44) of non-bevacizumab-treated patients on T
1-weighted magnetic resonance imaging (MRI) scans, and in 25% (11/44) of bevacizumab-treated patients compared with 18% (8/44) of non-bevacizumab-treated patients on fluid attenuation inversion recovery (FLAIR) MRI sequences (p > 0.05). This proportion of distant recurrences was in line with previous reports, without significant differences between bevacizumab and non-bevacizumab-containing treatments [
63]. Moreover, a subanalysis of the BRAIN study, in which patient MRI scans were compared at baseline (prior to bevacizumab treatment) and at the time of progression, showed that the majority of patients (55/67 in the bevacizumab-alone group) had no shift in the pattern of progression. A shift from local to diffuse disease was seen in 16% (11/67) of patients in the bevacizumab-alone group [
64]. Other investigators have likewise concluded on the basis of retrospective analyses of radiographic patterns of relapse that the majority of disease patterns with glioblastoma are local at diagnosis and remain so after recurrence and treatment with bevacizumab, and that the rate of nonlocal disease (diffuse, distant, or multifocal) does not appear to increase with the use of antiangiogenic agents [
65‐
67]. Reports have also differed regarding the impact of the pattern of radiographic recurrence on survival outcomes [
36,
58,
64,
67]. In cases in which an infiltrative phenotype is observed at diagnosis, it is possible that antiangiogenic therapy in combination with another agent that targets tumor invasion, such as dasatinib [
68], may be an effective therapeutic strategy.
Antiangiogenic agents in combination with radiation
Increased understanding of molecular mechanisms in the tumorigenesis of glioblastomas has led to the evaluation of targeted agents as potential radiosensitizers [
69,
70]. Preclinical models have shown that VEGF is upregulated in response to radiation, and these elevations may contribute to the protection of tumor blood vessels from radiation-mediated cytotoxicity [
70,
71]. The administration of antiangiogenic agents with radiotherapy may counteract VEGF-mediated radioresistance, thereby sensitizing tumors and associated vasculature to the ionizing effects of radiation (Figure
3) [
69,
70,
72]. As an underlying mechanism, the ability of antiangiogenic agents to lower tumor interstitial fluid pressure and improve vascular function and tumor oxygenation may promote enhanced responsiveness to radiotherapy [
73,
74]. Preclinical studies have also demonstrated that antiangiogenic agents uniquely target the radioresistant and highly tumorigenic cancer stem cell niche [
20,
75]. Finally, the success of initial clinical investigations of bevacizumab with chemoradiation in patients with solid tumors also supports the possible synergies of combined modality therapy [
76,
77].
Efficacy of antiangiogenic agents and chemoradiation
The efficacy and safety of bevacizumab with chemotherapy and radiotherapy have been assessed in clinical studies for the treatment of both recurrent and newly diagnosed glioblastoma [
78,
79]. In the frontline setting, the use of bevacizumab plus radiotherapy and temozolomide has been described in two reports. In a phase II pilot study, 10 patients with glioblastoma underwent surgery followed by radiotherapy (30 fractions of 2 Gy per fraction) with bevacizumab 10 mg/kg q2w plus concomitant temozolomide 75 mg/m
2[
78]. Temozolomide therapy was continued until disease progression or for a maximum of 24 cycles, while bevacizumab therapy continued every 2 weeks until progression. At the time of reporting, the median PFS was >8.8 months, but it was too early to establish the median OS. The most commonly occurring, possibly treatment-related adverse events were fatigue, myelotoxicity, wound-healing complications, and venous thromboembolic events. The only unexpected toxicity was the development of presumed radiation-induced optic neuropathy in one patient. The study investigators noted, however, that the observed toxicities were at an acceptable level to continue enrollment toward a target of 70 patients.
In a subsequent feasibility study in a consecutive series of patients, Narayana and colleagues reported outcomes from 15 patients with high-grade glioma, including 12 patients with glioblastoma, who underwent surgery followed by radiotherapy (59.4 Gy over 6.5 weeks) [
79]. Bevacizumab 10 mg/kg was administered on days 14 and 28 along with concomitant temozolomide 75 mg/m
2 daily during radiotherapy. After the completion of radiotherapy, treatment with bevacizumab and temozolomide continued for 12 cycles. At a median follow-up of 12 months (range, 5-21 months), the PFS rate was 59.3% and the OS rate was 86.7%. Nonhematologic toxicities were reported in three patients (20%), and grade 3 or 4 hematologic toxicities were reported in another three patients (20%) [
79]. No intracerebral hemorrhage or treatment-related deaths occurred during the study. Several ongoing clinical trials have also recently reported interim data on the use of bevacizumab with radiotherapy and either temozolomide or irinotecan in patients with previously untreated glioblastoma [
80‐
86]. In two of the trials with longer follow-up, the addition of bevacizumab with or without irinotecan to standard radiotherapy and temozolomide was shown to provide significant benefit in PFS relative to historic controls [
80,
82]. In one trial having a minimum follow-up of 18 months, the regimen incorporating bevacizumab and irinotecan was associated with a median PFS that was approximately double that seen with standard therapy in patients with newly diagnosed glioblastoma (14 vs 6.9 months, respectively) [
82]. In both trials, the incorporation of bevacizumab into standard frontline regimens was considered to be tolerable [
80,
82]. Large phase III studies evaluating bevacizumab-containing regimens in patients with newly diagnosed glioblastoma have recently begun enrolling patients, including a global-based study (AVAglio [NCT00943826]) [
87] and a US-based study (RTOG-0825 [NCT00884741], which is sponsored by the Radiation Therapy Oncology Group).
Results from a phase I/II trial of cilengitide in combination with temozolomide and radiotherapy in patients with newly diagnosed glioblastoma have also demonstrated promising efficacy [
88]. After tumor resection, 52 patients received standard radiotherapy (2 Gy × 30 fractions) and temozolomide 75 mg/m
2, with cilengitide 500 mg twice weekly started 1 week before chemoradiation and given throughout the duration of chemotherapy or until progression. The 6-and 12-month PFS rates were 69% and 33%, respectively; the median PFS was 8.0 months. The 12- and 24-month OS rates were 68% and 35%, respectively, with a median OS of 16.1 months. The authors reported that PFS and OS in patients with O(6)-methylguanine-DNA methyltransferase (MGMT) promoter methylation (13.4 and 23.3 months, respectively) were longer than those in patients without MGMT promoter methylation (3.4 and 13.1 months, respectively). Seven patients (14%) discontinued treatment for adverse events that were possibly treatment-related. The regimen was found to be well tolerated, with no additional toxicities [
88].
Early phase studies have evaluated additional antiangiogenic agents, such as vatalanib, vandetanib, and ABT-510, in combination with temozolomide and radiotherapy for the treatment of patients with newly diagnosed glioblastoma [
89‐
91]. These trials provide further evidence for the feasibility of combining these treatment modalities in the frontline setting.
Recent studies have also reported on the feasibility of using bevacizumab with radiotherapy in patients with recurrent malignant gliomas [
92,
93]. One of these studies reported outcomes in 25 patients (20 patients with glioblastoma and five patients with anaplastic glioma) who received bevacizumab 10 mg/kg q2w until tumor progression, along with hypofractionated stereotactic radiotherapy (30 Gy total as 6 Gy × 5 fractions) after the first cycle of bevacizumab therapy [
92]. In the glioblastoma cohort, the regimen was associated with a 6-month PFS rate of 65% (95% CI, 40%-82%) and a median PFS of 7.3 months (95% CI, 4.4-8.9 months). The median OS was 12.5 months (95% CI, 6.9-22.8 months), the 1-year survival was 54%, and the ORR was 50%. The overall toxicity of the regimen was comparable to that in other clinical trials of bevacizumab in glioblastoma [
28,
29,
31,
78]. Three patients in the overall population experienced a grade 4 adverse event-bowel perforation, wound-healing complication, and gastrointestinal bleeding. Other nonhematologic and hematologic toxicities were transient. No significant adverse events appeared to be attributable to the interaction of bevacizumab with radiation, with the exception of a single instance of wound dehiscence; radiation necrosis was not observed in this previously irradiated population. Overall, the high 6-month PFS rate and improved therapeutic ratio of this combination suggest that it should be investigated in larger trials of patients with recurrent disease and supports ongoing trials of bevacizumab with radiochemotherapy in patients with newly diagnosed glioma.
Other considerations with antiangiogenic therapies
The role of antiangiogenic therapy also requires further evaluation of its potential use in glioblastoma-related conditions. One example is pseudoprogression, which may be visualized on brain scans in patients who have received chemoradiotherapy and temozolomide, resulting from increased cerebral edema. In clinical studies, both bevacizumab and cediranib have shown activity in reducing the need for steroid therapy to treat tumor-associated cerebral edema [
31,
94]. Therefore, these agents may be useful in cases in which pseudoprogression is suspected, as well as in patients with large, inoperable glioblastomas who are dependent on steroid therapy.
Antiangiogenic treatment has also been proposed for the management of radiation necrosis, a process in which endothelial cell dysfunction leads to tissue hypoxia and necrosis, with the concomitant release of vasoactive compounds [
95]. In a small randomized double-blind study, Levin and colleagues reported outcomes in 14 patients who received either placebo or bevacizumab for radiographically-proven or biopsy-proven CNS necrosis. All of the bevacizumab-treated patients (5/5 randomized patients; 7/7 crossover patients), but none of the placebo-treated patients (n = 7), showed improvement in neurologic symptoms or signs and had a reduction in the volume of necrosis on T
2-weighted FLAIR (average reduction of 59% in randomized patients) and T
1-weighted gadolinium-contrast MRI (average reduction of 63% in randomized patients) [
96]. Similar radiographic responses, along with improved or stable clinical outcomes, were also achieved with bevacizumab treatment in a retrospective analysis of eight patients with documented radiation necrosis [
95], as well as a case series of six patients with biopsy-proven radiation necrosis [
97].
In addition to its role in the treatment of glioblastoma, bevacizumab has also been evaluated in other high-grade gliomas. Results from phase II studies and retrospective reviews of bevacizumab for the treatment of anaplastic gliomas have been encouraging. In a phase II study of 33 patients with recurrent grade 3 malignant gliomas (anaplastic astrocytoma, anaplastic oligodendroglioma, and anaplastic oligoastrocytoma), Desjardins and colleagues found the use of bevacizumab and irinotecan to be active (6-month PFS = 55%; 6-month OS = 79%; ORR = 61%) and to have acceptable toxicity, with infrequent significant adverse events [
98]. In a more recent study of 31 patients with recurrent anaplastic glioma, single-agent bevacizumab was associated with a median PFS of 3.7 months, a median OS of 12.4 months, reduced steroid requirements (a 40% reduction, on average, in steroid dose), and improved neurologic symptoms [
99]. The activity and safety of single-agent bevacizumab have also been described in retrospective studies of patients with recurrent alkylator-refractory anaplastic oligodendroglioma and anaplastic astrocytoma [
100,
101]. The NCCN guidelines now include the use of bevacizumab with or without chemotherapy as a treatment option for recurrent anaplastic gliomas [
4].
Another consideration is the impact of antiangiogenic agents on radiographic evaluations of treatment response in malignant gliomas. Some investigators argue that it is challenging to determine disease progression and tumor response to antiangiogenic therapy because of the effect of these agents on vascular permeability, which results in diminished contrast enhancement on computed tomography or MRI scans [
102‐
104]. Because the current standard response criteria (MacDonald criteria) are based on contrast enhancement MRI, there is some debate as to whether these criteria are still adequate in the era of antiangiogenic agents. Proposals for new treatment response assessment criteria have been presented by various authors and also by the Response Assessment in Neuro-Oncology Working Group, and include taking into account T
2/FLAIR (non-contrast enhancing) imaging, favoring the use of Levin criteria, or changing the criterion of response by cross-sectional area of enhancement measurement (e.g., a > 25% decrease vs ≥ 50% decrease) [
35,
99,
105]. Additional imaging techniques and analyses for the assessment or predictors of antiangiogenic treatment response that have been proposed for additional investigation include FLAIR MRI, dynamic contrast-enhanced MRI, diffusion-weighted MRI, pretreatment apparent diffusion coefficient histogram analysis, and perfusion imaging or dynamic susceptibility contrast MRI [
60,
105‐
109]. The breadth of these recommendations further underscores the need for a standardized approach of response assessment.