Antiangiogenic agents in the management of non-small cell lung cancer

Abstract

Several therapies targeting angiogenesis are currently in development for non-small cell lung cancer (NSCLC). This review discusses results of recent clinical trials evaluating chemotherapy plus antiangiogenic therapy for NSCLC. Bevacizumab, an anti-VEGF antibody, is currently approved for the treatment of advanced NSCLC in combination with carboplatin and paclitaxel. Completed phase III trials evaluating bevacizumab plus chemotherapy have shown prolonged progression-free survival; however, not all trials showed significant improvement in overall survival (OS). Phase III trials of the tyrosine kinase inhibitors (TKIs) vandetanib and sorafenib and the vascular disrupting agent ASA404 also failed to improve OS compared with chemotherapy alone. Clinical trials are ongoing involving several new antiangiogenic therapies, including ramucirumab, aflibercept, cediranib, nintedanib (BIBF 1120), sunitinib, pazopanib, brivanib, ABT-869, axitinib, ABT-751 and NPI-2358; several of these agents have shown promising phase I/II results. Results from recently completed and ongoing phase III trials will determine if these newer antiangiogenic agents will be incorporated into clinical practice.

Introduction

The 5-y survival rate for non-small cell lung cancer (NSCLC) remains low at approximately 20%.¹ In an effort to improve patient outcomes, recent clinical research has focused on evaluating combinations of chemotherapeutic agents with targeted therapies. Because angiogenesis is required for larger tumors to obtain nutrients and oxygen, inhibition of angiogenesis is one approach that is being actively pursued.² Signaling via vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF) and fibroblast growth factor (FGF) receptors are important proangiogenic pathways.³ Several agents are currently in clinical development in NSCLC to target angiogenic signaling pathways (Fig. 1A). These include agents that bind the VEGF ligand [e.g., bevacizumab (Avastin, Genentech) and aflibercept (Regeneron)], and agents that inhibit the receptor tyrosine kinases activated by VEGF, FGF and PDGF, such as the anti-VEGF receptor (VEGFR) antibody ramucirumab (IMC-1121B; ImClone Systems Inc.) and small molecule tyrosine kinase inhibitors [TKIs; e.g., vandetanib (Zactima, AstraZeneca), sorafenib (Nexavar, Bayer), sunitinib (Sutent, Pfizer), and nintedanib (BIBF 1120, Boehringer Ingelheim)]. Blocking these receptors may inhibit activation of important pro-angiogenic signaling pathways (Fig. 1B). In addition, vascular disrupting agents (VDAs) are compounds designed to disrupt existing tumor vasculature, with the hope that this will result in tumor ischemia and necrosis.⁴

Figure 1.

Mechanisms of action of approved and investigational antiangiogenic agents. (A) Diagram depicting inhibition of tumor cell receptors (1) and endothelial cell and pericyte receptors (3 and 4) by TKIs and ramucirumab, targeting of VEGF (2) molecules by bevacizumab and VEGFR trap (aflibercept) and (5) disruption of vascular integrity by VDAs. (B) Receptor tyrosine kinase-activated signaling pathways involved in angiogenesis. [Part (A) reprinted with permission ^©2006 Springer; part (B) reproduced with permission of Alphamed Press, Inc. in the format Journal via Copyright Clearance Center.]

The currently approved antiangiogenic therapy for NSCLC is bevacizumab. Bevacizumab is a monoclonal antibody that binds VEGF ligand and thereby inhibits activation of VEGFRs.⁵ It is indicated for non-squamous histology and contraindicated for patients with a history of clinically significant hemorrhage or hemoptysis because of an increased risk of bleeding episodes.⁶ While intracranial hemorrhage has been reported by a small proportion of patients,^6–8 several recent studies suggest that bevacizumab does not markedly increase the risk of severe intracranial hemorrhage in patients with treated brain metastases.^9–11 Additionally, bevacizumab does not increase the risk of bleeding when given concurrently with therapeutic anticoagulation in a phase III randomized trial.¹²

Patients who receive bevacizumab eventually become resistant to it. Animal and tumor cell models suggest that the redundancy in signaling pathways or synergistic interactions between VEGF, PDGF, and/or FGF signaling will lead to resistance against therapeutics directed solely against VEGF/VEGFR.^13,14 In addition, increased PDGF receptor (PDGFR) expression has been correlated with resistance against VEGF-targeted therapeutics in tumor cell lines.^15,16 This review discusses the results of recent clinical trials evaluating combined treatments with cytotoxic and antiangiogenic agents with diverse mechanisms of action, including antibodies that target VEGF and VEGFR, VEGF trap, multi-kinase inhibitors and VDAs.

Monoclonal Antibodies

Bevacizumab.

Trials of bevacizumab combined with chemotherapy in NSCLC. The randomized phase III Eastern Cooperative Oncology Group (ECOG) 4599 trial (Table 1) evaluated carboplatin/paclitaxel with or without bevacizumab 15 mg/kg in 878 chemotherapy-naive patients with non-squamous NSCLC with an ECOG performance status of ≤1.¹⁷ The primary endpoint was overall survival (OS) and secondary endpoints included response rate (RR) and progression-free survival (PFS). RR was higher in the bevacizumab arm (35 vs. 15%; p < 0.001) and bevacizumab increased PFS [6.2 vs. 4.5 mo; hazard ratio (HR), 0.66; p < 0.001] and OS (12.3 vs. 10.3 mo; HR, 0.79; p = 0.003; Fig. 2A and Table 1). Fifteen treatment-related deaths (five due to hemorrhage) occurred among patients receiving bevacizumab, compared with two deaths in the carboplatin/paclitaxel only arm (p < 0.001). Significantly higher rates of grade ≥3 adverse events (AEs) with paclitaxel/carboplatin/bevacizumab were noted vs. paclitaxel/carboplatin alone and are summarized in Table 2. In a retrospective subset analysis of 224 elderly (age ≥70 y) patients, grade 3–5 toxicities were significantly higher in the bevacizumab arm (87 vs. 61%; p < 0.001), without a survival benefit for either PFS or OS vs. chemotherapy alone.¹⁸ Elderly patients also had higher grade ≥ 3 neutropenia, bleeding and proteinuria compared with younger patients on the bevacizumab combination.

Table 1.

Phase II and III clinical trials evaluating approved and investigational antiangiogenic antibodies in combination with chemotherapy for NSCLC

Trial	Study design	Endpoints	PFS (mo)	OS (mo)	TTP (mo)	RR (%)	Common grade ≥3 AEs (%)
Bevacizumab (first-line)
ECOG 4599; phase III; N = 878; completed17	A: Carbo/pac (n = 444) B: Carbo/pac + bev (n = 434)	OS	A: 4.5 B: 6.2 HR, 0.66; p < 0.001	A: 10.3 B: 12.3 HR, 0.79; p = 0.003	NR	A: 15 B: 35	Bleeding events (A, 0.7; B, 4.4), febrile neutropenia (2.0; 5.2), hyponatremia (1.1; 3.5), hypertension (0.7; 7.0), headache (0.5; 3.0), rash (0.5; 2.3)
AVAiL*; phase III; N = 1,043; completed19,20	A: Gem/cis + bev 7.5 mg/kg (n = 345) B: Gem/cis + bev 15 mg/kg (n = 351) C: Gem/cis + placebo (n = 347)	Primary: PFS Secondary: OS	A: 6.7 B: 6.5 C: 6.1 A vs. C: HR, 0.75; p = 0.003 B vs. C: HR, 0.82; p = 0.03	A: 13.6 B: 13.4 C: 13.1 A vs. C: HR, 0.93; p = 0.420 B vs. C: HR, 1.03; p = 0.761	NR	A: 34.1 B: 30.4 C: 20.1	Neutropenia (A, 40; B, 36; C, 32), thrombocytopenia (27; 23; 23), anemia (10; 10; 13)
Patel et al.; phase II; n = 50; completed22	Pem/carbo + bev (up to 6 cycles) followed by pem + bev maintenance	RR, PFS, OS	7.8	14.1	NR	55	Infection (10), fatigue (8), diverticulitis (8), anemia (6), thrombocytopenia (6), venous thrombosis (6), neutropenia (4)
SAiL; phase IV; N = 2,212; completed11	Bev + standard chemotherapy (up to 6 cycles) followed by bev maintenance	Primary: Safety	NR	14.6	7.8		Thromboembolism (8), hypertension (6), overall bleeding (4), proteinuria (3), gastrointestinal disorders (2), vascular disorders (2), pulmonary hemorrhage (1)
ARIES; OCS; phase IV; N = 1,970; completed26	Bev + chemotherapy (62% of patients received carbo/pac)	Efficacy and safety in NSCLC patient subpopulations	6.7	13.6	NR	NR	Pulmonary hemorrhage (0.8), bleeding excluding pulmonary hemorrhage (3.0), CNS bleeding (0.1)
PointBreak; phase III; N = 900; recruiting27	A: Carbo/pem + bev induction, followed by pem + bev maintenance B: Carbo/pac + bev induction, followed by bev maintenance	Primary: OS Secondary: ORR, DCR, PFS, TTP, safety/toxicity	NR	NR	NR	NR	NR
ECOG 5508; NCT01107626; phase III; N = 1,282; recruiting28	A: Bev maintenance B: Pem maintenance C: Bev + pem maintenance After carbo/pac/bev induction	Primary: OS Secondary: PFS, objective RR, toxicity, pharmacokinetics	NR	NR	NR	NR	NR
NCT00976456; phase III; N = 250; recruiting	A: Bev + pem B: Bev + pem/carbo	Primary: noninferiority of PFS with monochemotherapy Secondary: OS, overall RR, safety, QoL	NR	NR	NR	NR	NR
NCT00948675; phase III; N = 360; recruiting29	A: Pem/carbo followed by maintenance pem (n = 180) B: Pac/carbo + bev followed by maintenance bev (n = 180)	Primary: PFS without grade 4 toxicity; Secondary: OS, PFS, RR, DCR	NR	NR	NR	NR	NR
Bevacizumab (second-line)
BeTa; phase III; N = 636; completed25	A: Erlotinib B: Erlotinib + bev	Primary: OS Secondary: PFS, objective RR	A: 1.7 B: 3.4	A: 9.2 B: 9.3 HR, 0.97; p = 0.7583	NR	A: 6 B: 13 objective	Rash (A, 6; B, 16), hypertension (1; 5), hemorrhage (2; 3 [grade 5, 1]), arterial thromboembolic event (< 1; 3 [grade 5, 1]), interstitial lung disease-like events (< 1 [grade 5, < 1]; < 1 [grade 5, < 1])
Bevacizumab (adjuvant)
ECOG E1505; phase III; N = 1,500; recruiting; NCT0032480530	1 of 4 chemotherapies (vin/cis, doc/cis, gem/cis or pem/cis) with or without bev	Primary: OS Secondary: PFS, safety/toxicity, biomarker analysis	NR	NR	NR	NR	Interim; bev + CT vs. CT alone: hypertension (19.6 vs. 2.0), proteinuria (3.2 vs. 0.7), abdominal pain (4.6 vs. 0.3)
Bevacizumab (maintenance)
ATLAS; phase III; N = 768; active, no longer recruiting23,24	Following platinum-based doublet chemotherapy: A: Bev + erlotinib B: Bev + placebo	Primary: PFS Secondary: OS	A: 4.8 B: 3.7 HR, 0.722; p = 0.0012	A: 15.9 B: 13.9 HR, 0.90; p = 0.2686	NR	NR	NR
Ramucirumab (first-line)
NCT00735696; phase II; N = 40; active, no longer recruiting32	Ramucirumab + carbo/pac (single-arm)	Primary: PFS at 6 mo Secondary: Safety/toxicity, PFS, objective RR, OS, OS at 1 y, DOR	5.7	NR	NR	67 objective	NR
NCT01160744; phase II; N = 44; recruiting	Randomized (4-arm): pem + carbo or cis with ramucirumab or placebo; gem + carbo or cis with ramucirumab	Primary: PFS Secondary: Objective RR, OS, duration of response, pharmacokinetics	NR	NR	NR	NR	NR
Ramucirumab (second-line)
NCT01168973; phase III; N = 1,156; recruiting	Randomized (2-arm): ramucirumab + doc vs. placebo vs. doc	Primary: OS Secondary: PFS, objective RR, DCR, improvement of symptoms, pharmacokinetics	NR	NR	NR	NR	NR

AEs, adverse events; Bev, bevacizumab; Carbo, carboplatin; Cis, cisplatin; CNS, central nervous system; CT, chemotherapy; DCR, disease control rate; Doc, docetaxel; DOR, duration of response; Gem, gemcitabine; NR, not reported; NSCLC, non-small cell lung cancer; OCS, observational cohort study; OS, overall survival; Pac, paclitaxel; Pem, pemetrexed; PFS, progression-free survival; QoL, quality of life; RR, response rate; TTP, time to disease progression; Vin, vinorelbine.

^*First-or-second-line.

Figure 2.

Efficacy outcomes from phase III trials evaluating bevacizumab in combination with chemo-therapy in patients with NSCLC. OS and PFS curves from ECOG 4599 (A) and AVAiL (B).¹⁹,²⁰ [(A) Reprinted with permission ^©2006 Massachusetts Medical Society. (B) Reprinted by permission of ^©2011 Oxford University Press and ^©2008 American Society of Clinical Oncology. All rights reserved.]

Table 2.

Significantly higher rates of grade ≥ 3 adverse events with bevacizumab plus chemotherapy vs. chemotherapy alone (ECOG 4599 trial)¹⁷

Adverse event, %	Paclitaxel/carboplatin (n = 440)	Paclitaxel/carboplatin + bevacizumab (n = 427)	p value
Grade ≥ 3 bleeding	0.7	4.4	<0.001
Grade 4 neutropenia	16.8	25.5	0.002
Grade 4 thrombocytopenia	0.2	1.6	0.04
Grade 3 febrile neutropenia	1.8	4.0	0.02
Grade 3 hyponatremia	0.9	2.6	0.02
Grade 3 hypertension	0.5	6.8	<0.001
Grade 3 proteinuria	0	2.6	<0.001
Grade 3 headache	0.5	3.0	0.003
Grade 3 rash	0.5	2.3	0.02

In a similarly designed randomized phase III trial (AVAiL) (Table 1), patients with advanced or recurrent non-squamous NSCLC were randomized to receive cisplatin/gemcitabine with or without 7.5 or 15 mg/kg bevacizumab.¹⁹ The primary endpoint was PFS, with a secondary endpoint of OS. Addition of bevacizumab was associated with an improvement in PFS at both doses, 6.7 mo for bevacizumab 7.5 mg/kg vs. 6.1 mo for placebo (HR, 0.75; p = 0.003) and 6.5 mo for bevacizumab 15 mg/kg (HR, 0.82; p = 0.03 vs. placebo). RR was also significantly higher with bevacizumab 7.5 mg/kg (34.1%; p < 0.0001 vs. placebo) and 15 mg/kg (30.4%; p = 0.0023 vs. placebo) than with placebo (20.1%). However, bevacizumab had no significant effect on OS [bevacizumab 7.5 mg/kg vs. placebo: HR, 0.93; 95% confidence interval (CI), 0.78–1.11; p = 0.420; bevacizumab 15 mg/kg vs. placebo: HR, 1.03; 95% CI, 0.86–1.23; p = 0.761; Fig. 2B and Table 1].²⁰ The most common grade ≥ 3 AEs were neutropenia, thrombocytopenia and anemia (Table 1).¹⁹ In a retrospective subgroup analysis of 304 elderly (age ≥ 65 y) patients, PFS benefits were observed with both doses of bevacizumab, unlike what was observed in the ECOG 4599 elderly subset, perhaps due to different age cut-offs used for the subset analysis. The safety profile was generally consistent with that of the overall population.²¹

Bevacizumab 15 mg/kg was evaluated in combination with carboplatin/pemetrexed as first-line therapy in a multicenter, single-arm, phase II trial in 50 patients with chemotherapy-naive stage IIIB or stage IV non-squamous NSCLC²² (Table 1). Objective RR was 55%, median PFS was 7.8 mo and median OS was 14.1 mo. Hemorrhagic and hypertensive events exceeding grade 2 were not observed in this trial.²²

Bevacizumab has also been evaluated in combination with the epidermal growth factor receptor (EGFR) inhibitor erlotinib (Tarceva, Genentech) in phase III trials (ATLAS, BeTa; Table 1).^23–25 In the ATLAS trial, patients received four cycles of bevacizumab plus chemotherapy as first-line therapy followed by maintenance bevacizumab plus erlotinib or bevacizumab plus placebo.^23,24 The trial met its primary endpoint (PFS) and was stopped at the second planned interim efficacy analysis. Median PFS was significantly longer with bevacizumab plus erlotinib vs. bevacizumab alone (4.8 vs. 3.7 mo; HR, 0.722; 95% CI, 0.592–0.881; p = 0.0012).²³ However, the recently reported post hoc analysis of OS (as of June 2009) showed no significant differences between groups (HR, 0.90; 95% CI, 0.74–1.09; p = 0.2686).²⁴ Detailed safety data has not yet been reported, but it was noted that the safety profile for the combination of bevacizumab and erlotinib was consistent with known safety profiles for these agents.²³ The BeTa trial evaluated bevacizumab plus erlotinib vs. placebo plus erlotinib in the second-line setting.²⁵ While OS did not differ between groups (HR, 0.97; 95% CI, 0.80–1.18; p = 0.7583), PFS appeared longer with bevacizumab plus erlotinib vs. erlotinib alone (3.4 vs. 1.7 mo) and objective RR was improved (13 vs. 6%), but statistical testing was not performed on secondary endpoints. Patients receiving bevacizumab had a higher proportion of grade 5 events.

Phase IV trials evaluating safety and efficacy of bevacizumab with chemotherapy. The phase IV SAiL observational report outlined the toxicities seen (Table 1) in previously untreated patients with locally advanced, metastatic or recurrent non-squamous NSCLC treated with a bevacizumab (7.5 or 15 mg/kg) containing regimen.¹¹ Patients received bevacizumab in combination with standard chemotherapy for up to six cycles, followed by bevacizumab maintenance therapy until disease progression or discontinuation. The most common grade ≥ 3 AEs observed were thromboembolism (8%), hypertension (6%), bleeding (4%) and proteinuria (3%).

Another phase IV observational cohort study, ARIES, collected data from patients with locally advanced non-squamous NSCLC (N = 1,970) who received first-line chemotherapy plus bevacizumab.²⁶ Overall incidences of severe pulmonary hemorrhage, grade ≥ 3 bleeding (excluding pulmonary hemorrhage) and grade ≥ 3 central nervous system (CNS) bleeding were 0.8, 3.0 and 0.1%, respectively. In patients ≥ 70 y of age (N = 650), incidences of severe pulmonary hemorrhage, grade ≥ 3 bleeding and grade ≥ 3 CNS bleeding were 0.3, 3.0 and 0%, respectively. Of note, no grade ≥ 3 CNS bleeding events were reported in patients with CNS metastases (n = 150).

Ongoing studies and advanced stage trials of bevacizumab plus chemotherapy. The “PointBreak” study is currently recruiting patients and is a phase III study of pemetrexed/carboplatin/bevacizumab induction followed by pemetrexed/bevacizumab maintenance compared with paclitaxel/carboplatin/bevacizumab induction followed by bevacizumab maintenance in patients with advanced non-squamous NSCLC (Table 1).²⁷ Approximately 900 patients (450 per treatment arm) will receive four cycles of induction therapy followed by maintenance therapy until disease progression or treatment discontinuation. The primary endpoint is OS.

Three other studies of bevacizumab combined with chemotherapy are also currently recruiting patients with advanced or recurrent non-squamous NSCLC (Table 1). In the ECOG 5508 study (NCT01107626), patients will receive bevacizumab alone, pemetrexed alone or combined bevacizumab and pemetrexed after induction therapy with carboplatin/paclitaxel/bevacizumab.²⁸ The primary endpoint is OS; PFS and objective RR are secondary outcome measures. In NCT00976456, elderly patients will receive bevacizumab plus pemetrexed or bevacizumab plus pemetrexed/carboplatin; the primary endpoint is noninferiority of bevacizumab/pemetrexed compared with bevacizumab/pemetrexed/carboplatin based on PFS. NCT00948675 will evaluate chemotherapy with pemetrexed/carboplatin followed by maintenance therapy with pemetrexed compared with bevacizumab plus paclitaxel/carboplatin followed by maintenance therapy with bevacizumab.²⁹ The primary endpoint is PFS without grade 4 toxicity.

Adjuvant treatment with bevacizumab. The ECOG E1505 study (NCT00324805; Table 1) is currently recruiting patients with completely resected stage IB-IIIA NSCLC to evaluate OS in patients treated with an adjuvant chemotherapy regimen of vinorelbine/cisplatin, docetaxel/cisplatin, gemcitabine/cisplatin or pemetrexed/cisplatin with or without bevacizumab. The primary endpoint is OS; disease-free survival and toxicity are secondary outcome measures. Interim safety data showed significantly increased rates of grade 3/4 hypertension (19.6 vs. 2.0%; p < 0.001), proteinuria (3.2 vs. 0.7%; p = 0.03), abdominal pain (4.6 vs. 0.3%; p = 0.001), and overall grade 3/4 toxicity (84.0 vs. 68.0%; p < 0.001) with bevacizumab plus chemotherapy vs. chemotherapy alone, respectively.³⁰

Ramucirumab.

Ramucirumab is an investigational monoclonal antibody that binds to VEGFR-2 and blocks ligand binding and activation.³¹ A phase II open-label study (Table 1) is currently evaluating ramucirumab as first-line NSCLC therapy in combination with carboplatin/paclitaxel, with preliminary results from the first 15 patients reporting an overall RR of 67%.³² Another phase II trial is recruiting patients with previously untreated NSCLC to examine ramucirumab in combination with four different chemotherapeutic regimens as first-line therapy (NCT01160744; Table 1), and a phase III trial is recruiting patients with NSCLC to test ramucirumab in combination with docetaxel as second-line therapy after failure of platinum-based therapy (NCT01168973; Table 1).

VEGF Trap

Aflibercept.

Aflibercept (Regeneron) is an investigational recombinant protein composed of epitopes of the extracellular domains of human VEGFR fused to the constant region (Fc) of human IgG1 with potential antiangiogenic activity. Aflibercept, functioning as a soluble decoy receptor, binds to pro-angiogenic VEGFs and prevents them from binding to their endogenous receptors; this may result in the inhibition of tumor angiogenesis, metastasis and ultimately tumor regression.³³ A phase II trial showed minimal efficacy with single-agent aflibercept in 98 patients with lung adenocarcinoma.³⁴ Currently, a phase III trial (VITAL; NCT00532155) is testing aflibercept plus docetaxel as a second-line NSCLC treatment (Table 3), but based on recently presented data (N = 913), the study did not meet its primary endpoint of OS (10.05 vs. 10.41 mo for aflibercept plus docetaxel vs. placebo plus docetaxel, respectively; HR, 1.01; 95.1% CI, 0.87–1.17; p = 0.898). However, the addition of aflibercept showed benefit in terms of PFS (HR, 0.82; 95% CI, 0.72–0.94; p = 0.0035) and RR (23.3 vs. 8.9%; p < 0.0001).³⁵ AEs with > 10% higher incidence with the combination vs. docetaxel alone were stomatitis, loss of weight, hypertension, epistaxis and dysphonia.

Table 3.

Phase II and III clinical trials testing aflibercept (VEGF trap) and ASA404 in combination with chemotherapy for NSCLC

Trial	Study design	Endpoints	PFS (mo)	OS (mo)	TTP (mo)	RR (%)	SD (%)	Common grade ≥ 3 AEs (%)
Aflibercept (first-line)
NCT00794417; phase I/II; N = 100; active, no longer recruiting	Aflibercept + pem/cis	Primary: ORR, PFS at 6 mo Secondary: Safety, pharmacokinetics	NR	NR	NR	NR	NR	NR
Aflibercept (second-line)
VITAL; NCT00532155; phase III; N = 900; active, no longer recruiting35	A: Docetaxel + aflibercept B: Docetaxel	Primary: OS; Secondary: PFS, TR, QoL	NR	A: 10.05 B: 10.41	NR	A: 23.3 B: 8.9	NR	NR
ASA404 (first- or second-line)
NCT00832494; phase I/II; N = 105; completed	A: Carbo/pac + ASA404 B: Carbo/pac	Primary: TR, TTP, DOR, safety/toxicity	NR	NR	NR	NR	NR	NR
ASA404 (first-line)
McKeage et al. 2009; phase II; N = 30; completed80	ASA404 + carbo/pac	Primary: Best overall TR, SD rate, TTP; Secondary: PFS, OS, safety, pharmacokinetics	5.5	14.9	5.5	37.9 PR	48.3	Neutropenia (81), leukopenia (39), infection (13), thrombocytopenia (13), nervous system disorders (19)
McKeage et al. 2008; phase II; N = 73; completed79	A: Carbo/pac + ASA404 (n = 36) B: Carbo/pac (n = 37)	Primary: OS, TTP, TR	NR	A: 14.0 B: 8.8	A: 5.4 B: 4.4	A: 31 B: 22 TR	A: 43.8 B: 32.3	Neutropenia (A, 62; B, 39), leukopenia (27; 28), alopecia (22; 28), hyperglycemia (14; 25), infection (11; 3), thrombocytopenia (11; 6), cardiac disorders (11; 3)
ATTRACT-1 NCT00662597; phase III; N = 1,299; terminated83,91	A: Carbo/pac + ASA404 (n = 649) B: Carbo/pac (n = 650)	Primary: OS; Secondary: OS in patients with squamous vs. non-squamous histology	A: 5.5 B: 5.5	A: 13.4 B: 12.7	NR	A: 24.7% B: 24.6%	A: 39.6 B: 39.5	Grade 4 neutropenia (A, 27; B, 19)
ASA404 (second-line)
ATTRACT-2 NCT00738387; phase III; N = 900; terminated	A: Docetaxel + ASA404 B: Docetaxel	Primary: OS Secondary: PFS, QoL, overall RR, biomarker analysis, pharmacokinetics	NR	NSD	NR	NR	NR	NR

AEs, adverse events; Carbo, carboplatin; Cis, cisplatin; DOR, duration of response; NR, not reported; NSCLC, non-small cell lung cancer; NSD, no significant difference; ORR, objective response rate; OS, overall survival; Pac, paclitaxel; Pem, pemetrexed; PFS, progression-free survival; PR, partial response; QoL, quality of life; RR, response rate; SD, stable disease; TR, tumor response; TTP, time to disease progression; VEGF, vascular endothelial growth factor.

Investigational Antiangiogenic Tyrosine Kinase Inhibitors

Vandetanib.

Vandetanib is a TKI that inhibits VEGFR-1 and -2, rearranged during transfection (RET) and EGFR.^36,37 Initial phase II results (N = 168) showed that vandetanib increased PFS (primary endpoint; HR, 0.69; 95% CI, 0.50–0.96; p = 0.013) in patients with NSCLC compared with the EGFR inhibitor gefitinib (Iressa, AstraZeneca).³⁸ However, phase III studies (Table 4) that tested vandetanib monotherapy vs. placebo after failure of chemotherapy and EGFR TKI treatment (ZEPHYR³⁹), second-line docetaxel with or without vandetanib (ZODIAC⁴⁰) and second-line vandetanib combined with pemetrexed (ZEAL⁴¹) or compared with erlotinib (ZEST⁴²) failed to show any improvement in OS with vandetanib. Based on the results of these four phase III trials, the application for Food and Drug Administration approval for vandetanib use in NSCLC has been withdrawn.⁴³ Vandetanib was also tested vs. placebo as maintenance therapy for advanced NSCLC in a phase II study (N = 162) and showed no significant PFS or OS benefit (Table 4).⁴⁴

Table 4.

Phase II and III clinical trials testing investigational TKIs in combination with chemotherapy for NSCLC

Trial	Study design	Endpoints	PFS (mo)	OS (mo)	RR (%)	SD (%)	Common grade ≥ 3 AEs (%)
Vandetanib (second-line)
ZEAL; phase III; N = 534; completed41	A: Pem + vandetanib B: Pem + placebo	Primary: PFS Secondary: OS, objective RR, TTP, safety	A: 17.6 wks B: 11.9 wks	A: 10.5 B: 9.2	A: 19 B: 8 objective	NR	Rash (A, 6; B, 3), diarrhea (4; 2), anemia (1; 6)
ZEST; phase III; N = 1,240; completed42	A: Erlotinib + vandetanib B: Erlotinib + placebo	Primary: PFS Secondary: OS, objective RR, TTP, safety	A: 2.6 B: 2.0	A: 6.9 B: 7.8	A: 12 B: 12 objective	NR	Diarrhea (A, 5; B, 3), rash (3; 4), fatigue (4; 4), dyspnea (4; 6)
ZODIAC; phase III; N = 1,391; completed40	A: Doc + vandetanib B: Doc + placebo	Primary: PFS Secondary: OS, objective RR, DCR, TTP, safety	A: 4.0 B: 3.2	A: 10.6 B: 10.0	NR	NR	Rash (A, 9; B, 1), neutropenia (29; 24), leukopenia (14; 11), febrile neutropenia (9; 7)
Vandetanib (second-line or third-line)
ZEPHYR; phase III; N = 924; completed39	A: Vandetanib B: Placebo	Primary: OS Secondary: PFS, objective RR, DCR at 8 weeks	NR	A: 8.5 B: 7.8	A: 2.6 B: 0.7 objective	NR	NR
Vandetanib (maintenance)
NCT00687297; phase II; N = 162; completed44	A: Doc/carbo/vandetanib followed by maintenance vandetanib B: Doc/carbo/vandetanib followed by maintenance placebo	Primary: PFS Secondary: objective RR, safety	A: 4.5 B: 4.2	A: 9.8 B: 9.4	NR	NR	Rash (A, n = 5; B, n = 2), diarrhea (A, n = 3; B, n = 0), hypertension (A, n = 2; B, n = 0)a
Cediranib (first-line)
BR24; phase II/III; N = 296; active, no longer recruiting47	A: Carbo/pac + cediranib 30 mg B: Carbo/pac + placebo	Primary: PFS Secondary: OS, toxicity, QoL, biomarker analysis	NR	A: 10.5 B: 10.1	A: 38 B: 16	NR	Hypertension (A, 19; B, 2), diarrhea (15; 2), anorexia (6; 3), stomatitis (6; 0), fatigue (29; 19), dyspnea (10; 13)
BR29; phase III; N = 750; active, no longer recruiting; NCT00795340	A: Carbo/pac + cediranib 20 mg B: Carbo/pac + placebo	Primary: OS Secondary: PFS, objective RR, DOR, QoL, cost utility ratios, biomarker analysis	NR	NR	NR	NR	NR
NCT00326599; phase II; N = 102; active, no longer recruiting	A: Gem/carbo + cediranib B: Gem/carbo + placebo	Primary: Objective TR Secondary: DOR, PFS at 6 mo, TTP, OS at 1 y, safety/toxicity	NR	NR	NR	NR	NR
Cediranib (second-line or third-line)
NCT00410904; phase II; N = 74; recruiting49	A: Cediranib + pem (patients who have not received prior bev) B: Cediranib + pem (patients with prior bev)	Primary: RR Secondary: PFS, OS	A: 5.6	A: 11	A: 29 (n = 38)	NR	A: fatigue (22), neutropenia (14), diarrhea (14), infection (8), febrile neutropenia (5)
Nintedanib (first-line or second-line)
NCT00979576; phase II; n = 132; recruiting	Nintedanib + pem	Primary: safety, PFS Secondary: TR, OS, clinical improvement, QoL, pharmacokinetics	NR	NR	NR	NR	NR
Nintedanib (second-line)
LUME-Lung 1; NCT00805194; phase III; N = 1,300; active, no longer recruiting	A: Doc + Nintedanib B: Doc + placebo	Primary: PFS Secondary: OS, TR, safety, pharmacokinetics, QoL	NR	NR	NR	NR	NR
LUME-Lung 2; NCT00806819; phase III; N = 1,302; active, no longer recruiting	A: Pem + folic acid + Nintedanib B: Pem + folic acid + placebo	Primary: PFS Secondary: OS, TR, safety, QoL, clinical improvement, pharmacokinetics	NR	NR	NR	NR	NR
Sorafenib (first-line)
ESCAPE; phase III; N = 926; halted55	A: Carbo/pac + sorafenib B: Carbo/pac + placebo	Primary: OS Secondary: PFS	A: 4.6 B: 5.4	A: 10.7 B: 10.6	NR	NR	Rash (A, 9; B, 1), diarrhea (4; 3), hand-foot disease (8; 1)
NExUS; NCT00449033; phase III; N = 904b; completed56	A: Gem/cis + sorafenib followed by maintenance sorafenib B: Gem/cis + placebo followed by maintenance placebo	Primary: OS Secondary: PFS, safety, PROs	A: 183 d B: 168 d	A: 376 d B: 379 d	A: 28 B: 26	NR	Thrombocytopenia (A, 9.9; B, 6.2), hand-foot skin reaction (8.6; 0.3), fatigue (7.3; 3.6), rash/desquamation (5.7; 0.5), neutropenia (5.4; 6.2)
NCT00300885; phase III; N = 900; terminated	A: Carbo/pac + sorafenib B: Carbo/pac + placebo	Primary: OS Secondary: PFS, TR, DOR, QoL, PROs, biomarker analysis	NR	NR	NR	NR	NR
NCT00449033; phase III; N = 906; completed	A: Gem/cis + sorafenib B: Gem/cis + placebo	Primary: OS Secondary: PFS, safety/toxicity, PROs	NR	NR	NR	NR	NR
NCT00801801; phase II; N = 43; active, no longer recruiting	Doc + sorafenib	Primary: PFS at 2 mo Secondary: objective RR	NR	NR	NR	NR	NR
Sorafenib (second-line)
NCT00454194; phase II; N = 100; active, no longer recruiting58	A: Pem + sorafenib B: Pem + placebo	Primary: PFS Secondary: TTP, OS, safety/toxicity, TTP, DOR, biomarker analysis	A: 3.4 B: 4.1	A: 9.4 B: 9.7	NR	NR	NR
Sunitinib (first-line)
SABRE-L; phase II; N = 56; completed62	A: Carbo/pac + bev + sunitinib B: Carbo/pac + bev	Primary: TR Secondary: OS, PFS, safety	A: 3.8 B: 4.5	A: 6.6 B: NR	PR A: 8 B: 26	A: 80 B: 68	Neutropenia (A, 65.5; B, 50.0), thrombocytopenia (37.9; 19.2), leukopenia (27.6; 11.5), febrile neutropenia (13.8; 7.7), hypertension (10.3; 3.8)
NCT00695292; phase II; N = 35; active, no longer recruiting	Irinotecan + carbo + sunitinib	Primary: OS at 1 y Secondary: objective RR, TTP, safety/toxicity, OS	NR	NR	NR	NR	NR
Sunitinib (second-line)
CALGB 30704; NCT00698815; phase II; N = 225; recruiting	A: Sunitinib B: Pem C: Sunitinib + pem	Primary: PFS at 18 weeks Secondary: PFS, RR, duration of response, SD, OS, safety/toxicity, TR	NR	NR	NR	NR	NR
Sunitinib (salvage)
NCT01019798; phase II; n = 16; recruiting	Doc/cis + sunitinib	Primary: RR Secondary: TTP, OS, safety/toxicity	NR	NR	NR	NR	NR
Sunitinib (maintenance)
NCT00693992; phase III; N = 244; recruiting	A: Combination chemo + sunitinib maintenance B: Combination chemo + placebo maintenance	Primary: PFS Secondary: RR, OS, safety/toxicity	NR	NR	NR	NR	NR
Pazopanib (first-line)
NCT01179269; phase II; N = 53; recruiting	Pazopanib + pac	Objective RR	NR	NR	NR	NR	NR
NCT00871403; phase II; N = 107; completed	A: Pazopanib + pem B: Pem/cis	Primary: PFS Secondary: OS, best overall response, safety/toxicity	NR	NR	NR	NR	NR
NCT00866528; phase II; N = 180; active, no longer recruiting	A: Pazopanib + pac B: Carbo/pac	Primary: safety/tolerability, PFS Secondary: pharmacokinetics, clinical activity, OS, best overall response	NR	NR	NR	NR	NR
Pazopanib (second-line)
NCT01107652; phase II; N = 72; recruiting	A: Pazopanib + pem B: Pazopanib	Primary: DCR Secondary: CR + PR rate, PFS, OS, safety/toxicity, biomarker analysis	NR	NR	NR	NR	NR
Pazopanib (maintenance)
NCT01313663; phase II; N = 200; recruiting	A: Pazopanib B: Pemetrexed	Primary: PFS Secondary: OS, best overall response, safety	NR	NR	NR	NR	NR
Linifanib (first-line)
NCT00716534; phase II; N = 120; recruiting	A: Carbo/pac B: Carbo/pac + linifanib 7.5 mg C: Carbo/pac + linifanib 12.5 mg	Primary: PFS Secondary: OS, OS at 1 y, TTP, objective RR, best RR, TR	NR	NR	NR	NR	NR
Axitinib (first-line)
NCT00735904; phase II; N = 38; active, no longer recruiting	(squamous NSCLC) Axitinib + gem/cis	Primary: objective RR Secondary: OS at 2.5 y, PFS, DOR, safety	NR	NR	NR	NR	NR
NCT00768755; phase I/II; N = 180; active, no longer recruiting	Axitinib + pem/cis	Primary: efficacy Secondary: safety/toxicity	NR	NR	NR	NR	NR
NCT00600821; phase II; N = 119; active, no longer recruiting	A: Carbo/pac + axitinib B: Carbo/pac + bev	Primary: PFS Secondary: pharmacokinetics, OS, DOR, safety, biomarker analysis	NR	NR	NR	NR	NR
Motesanib (first-line)
NCT00369070; phase II; N = 186; active, no longer recruiting77	A: Carbo/pac + motesanib 125 mg once daily B: Carbo/pac + motesanib 75 mg twice daily C: Carbo/pac + bevacizumab	Primary: objective RR Secondary: pharmacokinetics, PFS, OS, safety	A: 7.7 B: 5.8 C: 8.3	A: 14.0 B: 12.8 C: 14.0	A: 30 B: 23 C: 37 objective	A: 35 B: 50 C: 42	Diarrhea (A, 19; B, 11; C, 3), fatigue (A, 17; B, 5; C, 8), dehydration (A, 17; B, 11; C, 3), anorexia (A, 12; B, 2; C, 3)
MONET1; NCT00460317; phase III; N = 1,090; active, no longer recruiting78	A: Carbo/pac + motesanib B: Carbo/pac + placebo	Primary: OS Secondary: PFS, safety, objective RR	A: 5.6 B: 5.4	A: 13.0 B: 11.0	A: 40 B: 26 objective	NR	Neutropenia (A, 22; B, 15), diarrhea (A, 9; B, 1), hypertension (A, 7; B, 1), cholecystitis (A, 3; B, 0)

AEs, adverse events; Bev, bevacizumab; Carbo, carboplatin; Cis, cisplatin; CR, complete response; DCR, disease control rate; Doc, docetaxel; DOR, duration of response; Gem, gemcitabine; NR, not reported; NSCLC, non-small cell lung cancer; OS, overall survival; Pac, paclitaxel; Pem, pemetrexed; PFS, progression-free survival; PR, partial response; PROs, patient-reported outcomes; QoL, quality of life; RR, response rate; SD, stable disease; TKI, tyrosine kinase inhibitor; TR, tumor response; TTP, time to progression.

^aDuring maintenance phase.

^bOf 904 patients randomized, 132 were excluded from primary analyses (squamous cell histology) leaving 772 patients evaluable for efficacy and 769 patients evaluable for safety.

Cediranib.

Cediranib (AZD2171; Recentin, AstraZeneca) inhibits VEGFR, PDGFR and FGF receptor (FGFR) isoforms, as well as the stem-cell factor receptor (c-kit).^45,46 Cediranib was initially tested as a first-line therapy in combination with carboplatin/paclitaxel in a phase II/III placebo-controlled trial (BR24; N = 296) for advanced NSCLC; the primary endpoint was PFS. Despite longer PFS (HR, 0.77; 95% CI, 0.56–1.08) and higher RR (38 vs. 16% with placebo; p < 0.0001) in patients receiving cediranib 30 mg/day (Table 4), the study was halted to review imbalances in assigned causes of death in the two study arms.⁴⁷ Patients in the cediranib arm exhibited more grade ≥ 3 toxicities, and there were 10 fatal AEs in this arm. Median OS was 10.5 mo with cediranib vs. 10.1 mo with placebo (HR, 0.78; 95% CI, 0.57–1.06; p = 0.11).

A similar trial (BR29; Table 4) of patients with NSCLC was initiated to test a lower dose of cediranib (20 mg); however it has recently closed to further accrual and further information is awaited.⁴⁸ In addition, a phase II trial (NCT00326599) is currently testing cediranib with gemcitabine/carboplatin as first-line NSCLC therapy and another phase II trial (NCT00410904) is recruiting patients to evaluate cediranib plus pemetrexed as second- or third-line NSCLC therapy with an expected enrollment of 74 patients (Table 4).⁴⁹ The preliminary RR (primary endpoint) for patients with no prior bevacizumab (Cohort A) was 29% (n = 38). Disease control rate [response plus stable disease (SD)] was 74%, median PFS was 5.6 mo, and median survival was 11 mo. One patient each developed pulmonary hemorrhage, cardiac ischemia and cerebrovascular event.

Nintedanib (BIBF 1120).

Nintedanib inhibits VEGFR-1, -2 and -3, PDGFR-α/β, and FGFR-1, -2 and -3, and also has activity against members of the v-src sarcoma viral oncogene homolog (Src) family and fms-like tyrosine kinase 3 (flt-3).⁵⁰ In a phase I trial in patients with NSCLC, when nintedanib was combined with pemetrexed, SD was achieved in 13 of 26 patients (50%).⁵¹ The most common nintedanib-related grade ≥ 3 AEs, occurring in ≥10% of patients were fatigue (23.1%) and reversible liver enzyme elevations (11.5%); grade 1 or 2 drug-related bleeding events (no grade ≥ 3) were observed below the maximum tolerated dose (MTD) in 33.3% (4 of 12) of patients and by no patients at or above the MTD.⁵¹

A subsequent phase II trial (N = 73) evaluated nintedanib monotherapy in patients with NSCLC with an ECOG performance status of 0–2 and who had previously failed one or two lines of chemotherapy.⁵² The primary endpoints were PFS and objective tumor response. In all patients, median PFS was 6.9 weeks, median OS was 21.9 weeks and 48% of patients exhibited SD. In patients with an ECOG performance status of 0–1 (N = 56), median PFS was 11.6 weeks and median OS was 37.7 weeks. The most common drug-related grade ≥ 3 AEs included reversible alanine transaminase (ALT) elevation (9.6%), diarrhea (8.2%), nausea (6.8%), gamma glutamyl transpeptidase increase (4.1%), vomiting (2.7%) and abdominal pain (2.7%).

A phase I/II trial (NCT00979576; Table 4) being conducted in Japan is currently recruiting patients with NSCLC to evaluate the efficacy and safety of nintedanib plus pemetrexed after failure of first-line chemotherapy; endpoints are MTD (phase I) and PFS (phase II). The LUME-Lung 1 phase III trial (NCT00805194; Table 4) is currently evaluating nintedanib plus docetaxel vs. placebo plus docetaxel as second-line therapy for NSCLC, with an estimated enrollment of 1,300 patients; the primary endpoint is PFS. The LUME-Lung 2 phase III trial (NCT00806819; Table 4) has also been initiated to evaluate nintedanib plus pemetrexed vs. placebo plus pemetrexed as second-line NSCLC treatment, with an estimated enrollment of 1,302 patients; the primary endpoint is PFS.

Sorafenib.

Sorafenib inhibits VEGFR-2 and -3, PDGFR-β, c-kit, v-raf 1 murine leukemia viral oncogene homolog 1 (Raf) and flt-3.⁵³ Several phase II and III trials evaluating sorafenib in combination with chemotherapeutic drugs have been completed and several more are currently ongoing (Table 4). A phase II trial compared sorafenib monotherapy with placebo for NSCLC, with disease control 2 mo after randomization being the primary endpoint. Patients receiving sorafenib (n = 56) had longer PFS compared with placebo (N = 41; 3.6 vs. 1.9 mo; p = 0.01) and a higher rate of SD (29 vs. 5%; p = 0.002).⁵⁴

A phase III study (ESCAPE; N = 926) evaluated sorafenib in combination with carboplatin/paclitaxel vs. carboplatin/paclitaxel alone in patients with advanced untreated NSCLC.⁵⁵ ESCAPE was halted because no significant OS (primary endpoint) or PFS (secondary endpoint) benefit was observed with sorafenib (Table 4). OS was 10.7 mo with sorafenib and 10.6 mo without sorafenib (HR, 1.15; p = 0.915). PFS was 4.6 mo with sorafenib vs. 5.4 mo without sorafenib (HR, 0.99; p = 0.433). Patients with squamous histology exhibited a higher mortality rate when receiving sorafenib; OS was 8.9 mo with sorafenib plus carboplatin/paclitaxel vs. 13.7 mo with carboplatin/paclitaxel alone (HR, 1.85). Patients with squamous histology also had a higher incidence of fatal hemorrhagic episodes when receiving sorafenib (1.8 vs. 0.3% for patients with non-squamous histology). The most common grade ≥ 3 sorafenib-related AEs were rash (8.4%), hand-foot skin reactions (7.8%) and diarrhea (3.5%). Another phase III study (NExUS; NCT00449033; Table 4) evaluated sorafenib with gemcitabine/cisplatin but did not meet the primary endpoint of OS.⁵⁶ Patients with squamous histology were excluded from this study based on the results of ESCAPE. In a phase II study (N = 168) of sorafenib as second-line or third-line therapy in combination with erlotinib in advanced NSCLC (both non-squamous and squamous histology), the addition of sorafenib did not significantly improve PFS (3.38 vs. 1.94 mo with erlotinib alone; HR, 0.86; p = 0.196) or OS (7.62 vs. 7.23 mo with erlotinib alone; HR, 0.89; p = 0.290).⁵⁷ The most common grade ≥ 3 AEs in the combination arm were diarrhea (15 vs. 0% with placebo), fatigue (14 vs. 9%) and anorexia (10 vs. 0%). Preliminary results of a phase II trial of second-line sorafenib plus pemetrexed vs. pemetrexed alone in patients with non-squamous advanced NSCLC were recently presented, and PFS and OS were not significantly different between the two groups (Table 4); a significantly increased rate of grade 3 nonhematologic AEs was observed with sorafenib/pemetrexed compared with pemetrexed alone (76 vs. 39%; p < 0.001).⁵⁸

Sunitinib.

Sunitinib inhibits VEGFR, PDGFR, c-kit and flt-3.⁵⁹ Sunitinib was investigated in a phase II trial (N = 63) as a second-line or third-line monotherapy for NSCLC. Results from this trial⁶⁰ showed an objective RR (primary endpoint) of 11.1% (95% CI, 4.6–21.6%), OS of 23.4 weeks (95% CI, 17.0–28.3), PFS of 12 weeks (95% CI, 10.0–16.1) and a SD (≥ 8 weeks) rate of 28.6%. Fatigue/asthenia (29%), lymphopenia (25%), pain/myalgia (17%), dyspnea (11%) and nausea/vomiting (10%) were the most common grade ≥ 3 AEs.

Another phase II study (n = 47) evaluated single-agent sunitinib as NSCLC monotherapy.⁶¹ The objective RR (primary endpoint) was 2.1% (95% CI, 0.1–11.3), 23.4% of patients achieved SD (≥8 weeks), median PFS was 11.9 weeks (95% CI, 8.6–14.1), and OS was 37.1 weeks (95% CI, 31.1–69.7). Fatigue was the most common grade 3 or 4 AE (17.0%); other common grade ≥ 3 AEs were neutropenia (8.7%) and hypertension (8.5%).

A randomized phase II study (SABRE-L) evaluated patients with previously untreated non-squamous NSCLC.⁶² Patients received either carboplatin/paclitaxel/bevacizumab or carboplatin/paclitaxel/bevacizumab plus sunitinib (25 mg/day on a 2 weeks on/1 week off schedule). A second phase of the study was to include a cohort receiving 37.5 mg sunitinib, but this was never initiated because of poor tolerability of the 25-mg dose (Table 4). Twenty-six patients received carboplatin/paclitaxel/bevacizumab and 30 patients received this combination plus sunitinib. The median treatment duration was 6.0 weeks in patients receiving sunitinib and 10.3 weeks in patients not receiving sunitinib. Of patients receiving sunitinib, 35% required a dose reduction, 52% required treatment interruption and 59% discontinued treatment because of AEs. The primary endpoint was best tumor response, which was partial response (PR) in both study arms (8% PR rate with sunitinib vs. 26% without). SD was observed in 80% of patients receiving sunitinib vs. 68% of patients receiving carboplatin/paclitaxel/bevacizumab. Of 29 patients evaluated in the sunitinib arm, the most common grade ≥ 3 AEs included neutropenia (65.5%), thrombocytopenia (37.9%), leukopenia (27.6%), febrile neutropenia (13.8%) and hypertension (10.3%).

While results of combination treatment with chemotherapy plus sunitinib alone have not yet been reported, phase II and III trials are currently ongoing or recruiting patients with NSCLC to examine sunitinib in combination with irinotecan/carboplatin, pemetrexed or docetaxel/cisplatin, as well as for maintenance therapy after chemotherapy (Table 4). Sunitinib has been evaluated in a phase III trial in combination with erlotinib vs. placebo/erlotinib as second- or third-line therapy (N = 960); PFS was significantly longer with the combination (15.5 vs. 8.7 weeks; HR, 0.807; 95% CI, 0.695–0.937; p = 0.0023), but OS was similar between groups (9.0 vs. 8.5 mo; HR, 0.922; 95% CI, 0.797–1.067; p = 0.1388).⁶³ The most common grade ≥ 3 AEs were diarrhea and rash.

Pazopanib.

Pazopanib (GW786034; GlaxoSmithKline) inhibits VEGFR, FGFR, PDGFR and c-kit.^64,65 A phase II trial (N = 35) evaluated pazopanib monotherapy in patients with resectable NSCLC, with RR as the primary endpoint. The PR rate was 8.6%. AEs were generally grade ≤ 2 and included hypertension (43%), diarrhea (37%), fatigue (37%), nausea (34%), ALT elevation (23%; 6% grade 3) and headache (23%).⁶⁶ A number of phase II trials are currently ongoing or recruiting patients to test pazopanib in NSCLC in a variety of settings; these include first-line therapy in combination with paclitaxel (NCT00866528), second-line therapy in combination with pemetrexed (NCT01107652) and maintenance therapy (NCT013136663; Table 4). In the latter, pazopanib is being compared with pemetrexed in patients with stage IV NSCLC who did not progress (SD, PR or complete response) after induction therapy containing platinum and pemetrexed.

Brivanib.

Brivanib (Bristol-Myers Squibb) is an inhibitor of both VEGFR and FGFR.^67,68 Brivanib was tested for safety in humans with advanced solid tumors.^67,69 The most common AE observed in patients treated with brivanib (N = 4) was grade 1/2 fatigue (75%).⁶⁹ A phase I trial is evaluating the safety of brivanib in combination with chemotherapy for patients with advanced solid tumors (NCT00798252).

Linifanib (ABT-869).

Linifanib (ABT-869; Abbott) inhibits all three VEGFR isoforms, as well as PDGFR.^70,71 A randomized phase II trial of 139 patients with advanced NSCLC that had progressed after previous therapy reported an objective RR of 5.0%, median PFS of 3.6 mo and median OS of 9.0 mo.⁷² The most common grade 3/4 linifanib-related event was hypertension (14%). A phase II trial (NCT00716534) is ongoing to test two doses of linifanib with carboplatin/paclitaxel as a first-line treatment for patients with advanced non-squamous NSCLC (Table 4).

Axitinib.

Axitinib (AG-013736; Pfizer) inhibits VEGFRs and PDGFR-β.⁷³ A single-arm phase II study (N = 32) evaluated axitinib in patients with NSCLC;⁷⁴ the primary endpoint was confirmed objective RR. Three (9%) confirmed PRs were observed. PFS was 4.9 mo (95% CI, 3.6–7.0) and OS was 14.8 mo (95% CI, 10.7 not estimable). Grade 3 AEs included fatigue (22%), hyponatremia (9%) and hypertension (9%). Phase II trials are ongoing to test axitinib in combination with cisplatin/gemcitabine (NCT00735904), pemetrexed/cisplatin (NCT00768755) and carboplatin/paclitaxel doublet combinations (NCT00600821; Table 4).

Motesanib.

Motesanib (Amgen) inhibits VEGFR-1, -2 and -3, PDGFR-β, c-kit and RET.^75,76 Motesanib was evaluated in the first-line setting in combination with carboplatin/paclitaxel in a phase II trial (N = 186) and showed comparable efficacy to that observed with bevacizumab plus carboplatin/paclitaxel (Table 4).⁷⁷ In the phase III MONET1 trial (N = 1,090), motesanib was evaluated as first-line therapy in combination with carboplatin/paclitaxel in advanced non-squamous NSCLC; preliminary results reported that neither PFS (5.6 vs. 5.4 mo; HR, 0.785; 95% CI, 0.684–0.901; p = 0.006) nor OS (13.0 vs. 11.0 mo; HR, 0.897; 95% CI, 0.776–1.035; p = 0.137) was significantly improved with motesanib compared with chemotherapy alone. Grade ≥ 3 AEs were more frequent with motesanib (73%) vs. carboplatin/paclitaxel alone (59%) and included neutropenia (22 vs. 15%), diarrhea (9 vs. 1%) and hypertension (7 vs. 1%).⁷⁸

Vascular Disrupting Agents

ASA404 (Novartis), also called vadimezan, is a fused tricyclic analog of flavone acetic acid with potential antineoplastic activity. Vadimezan induces cytokines that lead to hemorrhagic necrosis and a decrease in angiogenesis. This investigational agent also stimulates the antitumor activity of tumor-associated macrophages.³³ Initial phase II results in patients with NSCLC suggested that ASA404 may have potential clinical benefit when added to paclitaxel/carboplatin in chemotherapy-naive patients with NSCLC.^79,80 However, two subsequent phase III trials (ATTRACT-1 and ATTRACT-2; Table 3) that tested ASA404 in combination with paclitaxel/carboplatin and docetaxel, respectively, were terminated because interim analyses showed no increased OS benefit (primary endpoint).^81–83

BNC105P is an investigational benzofuran-based VDA prodrug with potential antivascular and antineoplastic activities. Upon administration, BNC105P, the disodium phosphate ester of BNC105, is rapidly converted to BNC105. In activated endothelial cells, BNC105 binds to tubulin and inhibits its polymerization, which may cause cytotoxic effects and result in a blockage of mitotic spindle formation, cell cycle arrest and disruption of the tumor vasculature leading to hypoxic conditions and tumor cell apoptosis. In addition to its VDA activity, this agent has a direct cytotoxic effect on tumor cells by inhibiting tubulin polymerization. BNC105 is not a substrate for the multidrug-resistance P-glycoprotein transporter.³³ BNC105P is currently being evaluated with everolimus (Afinitor, Novartis) in renal cell carcinoma. However, two other tubulin-binding VDAs are being investigated in phase I/II trials for NSCLC. ABT-751 (Abbott) was investigated in a phase I/II trial in combination with pemetrexed vs. placebo plus pemetrexed and showed similar PFS (2.3 vs. 1.9 mo, respectively; p = 0.819) in the overall population.⁸⁴ However, in a subgroup of patients with squamous histology, the combination numerically improved PFS (2.8 vs. 1.4 mo; HR, 0.58; 95% CI, 0.30–1.15; p = 0.112) and significantly prolonged OS (8.1 vs. 3.3 mo; HR, 0.47; 95% CI, 0.23–0.96; p = 0.034). The incidence of grade 3/4 treatment-related AEs was similar in both groups. NPI-2358 (Nereus Pharmaceuticals, Inc.)⁸⁵ is being evaluated in a phase II trial in combination with docetaxel (NCT00630110). Ombrabulin (AVE8062; Sanofi-Aventis) is being evaluated as first-line therapy for metastatic NSCLC in a phase II trial in combination with docetaxel followed by cisplatin or paclitaxel followed by carboplatin (NCT01263886).

Potential Biomarkers of Response to Antiangiogenic Therapy

Several efforts have been made to identify potential predictive biomarkers for response to antiangiogenic agents in NSCLC. The phase II BATTLE study (N = 255) screened heavily pretreated patients with NSCLC for 11 different biomarkers related to four molecular pathways in NSCLC (ranked in order of deemed importance): EGFR, Kirsten rat sarcoma viral oncogene homolog (KRAS), v-Raf murine sarcoma viral oncogene homolog B1 (BRAF), VEGF and RXR/Cyclin D1. Patients with multiple biomarkers present were assigned to one group based on the highest ranked marker that was positive; patients with inadequate tissue/no biomarkers were assigned to a fifth group. Based on these biomarker groupings, patients were randomized equally and then adaptively into four treatment groups: erlotinib, sorafenib, vandetanib and erlotinib/bexarotene. While this study did not investigate a specific angiogenesis-related biomarker, it did evaluate the role of predictive biomarkers in guiding treatment selection. The primary endpoint of the study was disease control at 8 weeks, and secondary endpoints included OS and PFS. Absence of EGFR mutations (p = 0.012) or high EGFR polysomy (p = 0.048) were associated with a higher 8-week disease control rate among patients treated with sorafenib. A higher disease control rate was also observed with vandetanib in patients who were positive for VEGFR-2 expression (p = 0.05). Patients with mutations in the KRAS gene had a higher disease control rate with sorafenib vs. the other treatments (p = 0.11), but the Cys amino acid substitution was associated with shortened PFS in the sorafenib arm compared with all other KRAS mutations (p = 0.013).⁸⁶

Hypertension has also been evaluated as a potential marker of response with bevacizumab. In a sub-study of the aforementioned ECOG 4599 trial (Table 1) of bevacizumab in combination with carboplatin and paclitaxel, patients who experienced hypertension had numerically improved OS (15.9 vs. 11.5 mo) and PFS (7.0 vs. 5.5 mo) compared with those who did not experience hypertension; however, formal comparison failed to reach significance because the analysis was underpowered.⁸⁷ Conflicting results were found in a meta-analysis of phase III clinical trials that evaluated bevacizumab in patients with several types of metastatic cancers (including NSCLC as well as colorectal, breast and renal cancer). In five of the six trials analyzed, the incidence of hypertension was not predictive of improved OS or PFS.⁸⁸ In addition, an analysis of patients from the BR24 clinical trial of cediranib in combination with carboplatin/paclitaxel showed that hypertension was not a predictor of outcome with cediranib; however, the development of hypertension was suggested as a potential prognostic indicator in both arms.⁸⁹ Another study evaluated a panel of cytokine and angiogenic factors and found that several markers, including VEGFR-2 and some interleukins, correlated with tumor shrinkage with pazopanib in early-stage NSCLC.⁹⁰

Conclusions

Antiangiogenic therapies are currently being evaluated in a variety of treatment settings in NSCLC. Bevacizumab is now approved for the first-line treatment of advanced NSCLC in combination with carboplatin and paclitaxel. Its role in the maintenance and adjuvant settings is currently being defined. Several other novel strategies of disrupting the angiogenesis pathway have been investigated in NSCLC. However, many of the recently reported trials have failed to impact overall survival. There are multiple key reasons for this. First, without a validated biomarker, specific subgroups of patients who are more likely to respond cannot be selected. Second, the redundancy in the angiogenic pathways leads to primary and secondary resistance to an agent that targets a specific angiogenic pathway; as a result, agents that target multiple angiogenic pathways are currently under investigation. Finally, it is unlikely that an antiangiogenic agent could achieve complete inhibition of its target(s), which may result in reduced but not completely abrogated signaling. Results of ongoing phase II and phase III trials of antiangiogenic agents will further clarify its role in the therapeutic landscape of NSCLC.

Abbreviations

AE

adverse event

Akt

protein kinase B

ALT

alanine transaminase

AVAiL

avastin in lung study

Bev

bevacizumab

BRAF

v-Raf murine sarcoma viral oncogene homolog B1

CG

cisplatin/gemcitabine

CI

confidence interval

CNS

central nervous system

DAG

diacylglycerol

ECOG

Eastern Cooperative Oncology Group

EGFR

epidermal growth factor receptor

FAK

focal adhesion kinase

FGF

fibroblast growth factor

FGFR

fibroblast growth factor receptor

HR

hazard ratio

IP3

inositol trisphosphate

KRAS

Kirsten rat sarcoma viral oncogene homolog

MAPK

mitogen-activated protein kinase

MTD

maximum tolerated dose

NO

nitric oxide

NOS

nitric oxide synthase

NSCLC

non-small cell lung cancer

OS

overall survival

PI3K

phosphatidyl inositol-3-kinase

PC

paclitaxel/carboplatin

PCB

paclitaxel/carboplatin/bevacizumab

PDGF

platelet-derived growth factor

PDGFR

platelet-derived growth factor receptor

PFS

progression-free survival

PKC

protein kinase C

PLCγ

phospholipase Cγ

PR

partial response

PROs

patient-reported outcomes

Raf

v-raf 1 murine leukemia viral oncogene homolog 1

Ras

retrovirus-associated DNA sequences

RET

rearranged during transfection

RR

response rate

SD

stable disease

SHC

Src homology 2 domain containing transforming protein

TGFβ

transforming growth factorβ

TKI

tyrosine kinase inhibitor

VDA

vascular disrupting agent

VEGF

vascular endothelial growth factor

VEGFR

vascular endothelial growth factor receptor

Disclosure of Potential Conflicts of Interest

G.R.S., N.S. and C.A. have no potential conflicts of interest to disclose.