Bevacizumab is a monoclonal antibody directed towards VEGF-A, and is the only anti-angiogenic agent to date, that has been shown to improve overall survival in lung cancer [ 17]. Toxicity concerns, including venous thromboembolic disease and fatal hemorrhage were observed in the initial randomized phase II trial [18]. This limited further development of bevacizumab in NSCLC to good performance status patients with non-squamous histology and no history of thrombosis, bleeding, gross hemoptysis, or brain metastases.

Five randomized trials have evaluated the addition of bevacizumab to standard platinum-based chemotherapy as first-line therapy for advanced and metastatic NSCLC (Table 2) [17–22]. The Eastern Cooperative Oncology Group (ECOG) 4599 trial randomized patients to carboplatin, paclitaxel with or without bevacizumab 15 mg/kg [17]. Greater efficacy was observed for patients receiving chemotherapy plus bevacizumab, including higher response rates (RR; 35 % vs. 15 %; p < 0.001), longer progression free survival (PFS; 6.2 vs. 4.5 months; hazard ratio [HR], 0.66; 95 % confidence interval [CI], 0.57–0.77) and improved overall survival (OS; 12.3 vs. 10.3 months; HR, 0.79; 95 % CI, 0.67–0.92). The Avastin in Lung Cancer trial (AVAiL) , evaluated the addition of two dose levels of bevacizumab (7.5 and 15 mg/kg) to cisplatin and gemcitabine chemotherapy. Significant improvements were observed in response rates and PFS for patients randomized to either dose of bevacizumab, compared with cisplatin and gemcitabine alone [20]. However, no improvement was observed in OS for patients randomized to bevacizumab 7.5 mg/kg (HR, 0.93; 95 % CI, 0.78–1.11), or 15 mg/kg (HR, 1.03; 95 % CI, 0.86–1.23) [21]. Two additional trials done in Japan and China provide confirmatory data regarding the efficacy of bevacizumab in combination with carboplatin and paclitaxel [19, 22]. Both demonstrate improvements in RR and PFS.

In all of these trials, bevacizumab was administered as maintenance therapy until disease progression. This strategy has not been formally evaluated, although retrospective review of data from both the ECOG 4599 trial and the US Oncology network suggests that patients continuing bevacizumab until disease progression experienced longer PFS and overall survival [23, 24].

Two meta-analyses of platinum-based chemotherapy with or without bevacizumab both report a significant improvement in OS for patients randomized to receive bevacizumab [25, 26]. Soria et al, report a 10 % reduction in the risk of death from the addition of bevacizumab to chemotherapy (HR 0.90, 95 % CI 0.81–0.99) [25]. Significant improvements were also observed in response rate and PFS [26]. However, the addition of bevacizumab to platinum-based chemotherapy is associated with increased toxicities. In the ECOG 4599 trial, patients randomized to chemotherapy plus bevacizumab experienced more grade 3–5 neutropenia (p = 0.002), thrombocytopenia (p = 0.04), febrile neutropenia (p = 0.02), hyponatremia (p = 0.02), hypertension (p < 0.001), proteinuria (p < 0.001), headache (p = 0.003), rash/desquamation (p = 0.02), and bleeding events (p < 0.001). There were 15 treatment-related deaths, including 5 due to pulmonary hemorrhage [17]. A similar profile of adverse events was observed in the AVAiL trial.

Given the toxicity concerns with bevacizumab, a variety of sub group analyses have been performed on these trials to identify subgroups of patients who derive greater benefit from bevacizumab therapy. In the ECOG 4599 trial, a larger improvement in overall survival was observed in patients with adenocarcinoma histology [27]. The median survival was 14.2 months compared with 10.3 months in the adenocarcinoma subgroup (HR 0.69, 95 %CI 0.58–0.83). Subgroup analyses of Asian patients in the AVAiL trial suggest that the addition of bevacizumab may improve survival in this group of patients [28]. Biomarkers including serum VEGF, intracellular adhesion molecule (ICAM), βFGF have all been evaluated as predictive biomarkers for bevacizumab. While ICAM was prognostic for improved survival and both ICAM and VEGF were predictive for response rates, no biomarker to date has been shown to predict a differential effect on OS for patients receiving bevacizumab. The onset of hypertension, a known side effect of bevacizumab, has also been evaluated as a predictive biomarker of efficacy [29]. However, a combined analysis of seven trials including disease sites other than lung cancer, did not support hypertension as a predictive biomarker. Therefore none of the biomarkers evaluated to date, are predictive for an improvement in overall survival from bevacizumab [30]. Therefore, it is not possible to further define subgroups of patients to receive bevacizumab therapy.

Competing therapeutic strategies to bevacizumab have emerged over recent years. Maintenance therapy with pemetrexed following initial platinum-doublet chemotherapy has been shown to improve survival in advanced NSCLC [31, 32]. Therefore, additional trials have tried to evaluate the role of pemetrexed, in bevacizumab eligible patients. Two trials have incorporated maintenance pemetrexed therapy in patients receiving bevacizumab. The AVAPERL trial randomized patients to maintenance therapy with bevacizumab alone (Table 2), or in combination with pemetrexed, following induction therapy with cisplatin, pemetrexed plus bevacizumab [33]. Three hundred and seventy six patients were registered and 253 randomized to maintenance therapy. PFS, the primary outcome, was prolonged in patients receiving maintenance bevacizumab plus pemetrexed compared with bevacizumab alone (7.4 m vs. 3.7 m HR 0.48, 95 %CI 0.35–0.66). AVAPERL failed to show a significant difference in overall survival as a secondary endpoint (HR 0.75, 95 %CI 0.47–1.19). Global quality of life (QoL), assessed by the EORTC QLQ-30, was similar between the two arms, although role function, fatigue and appetite favoured the combination of bevacizumab and pemetrexed [34]. Similarly, the Point Break trial evaluated the addition of pemetrexed to maintenance bevacizumab [35]. Patients were randomized to carboplatin, paclitaxel, bevacizumab then maintenance bevacizumab, versus carboplatin, pemetrexed, bevacizumab then maintenance pemetrexed plus bevacizumab. There was a modest improvement in PFS (6 m vs. 5.6 m, HR 0.83, 95 %CI 0.71–0.96), but no improvement in overall survival (13.4 m vs. 12.6 m, HR 1.0, 95 %CI 0.86–1.16). Maintenance therapy with pemetrexed in addition to bevacizumab should not be considered the standard of care at this time in the absence of improved overall survival.

One additional trial (Pronounce) compared carboplatin, paclitaxel plus bevacizumab with a non-bevacizumab regimen of carboplatin, pemetrexed then maintenance pemetrexed [36]. The primary outcome was a composite of progression free survival without grade 4 toxicity. Similar response rates, PFS and overall survival were observed between the two arms, suggesting that bevacizumab may not add to the efficacy of more effective chemotherapy. Currently, bevacizumab is widely used in combination with platinum-based chemotherapy and has been incorporated into NSCLC treatment algorithms [37]. It does have significant incremental toxicity and data from the Pronounce trial suggest that cisplatin or carboplatin plus pemetrexed followed by maintenance pemetrexed provides an alternative to bevacizumab based therapy.

There has been some interest in continuation of bevacizumab beyond progression, however, there are no data from randomized trials to support this. There is an ongoing trial (AVaALL) evaluating this strategy [38]. However, two trials have evaluated the addition of bevacizumab to erlotinib in the second-line setting [39, 40]. Herbst et al. [40], conducted a randomized phase II trial comparing the addition of bevacizumab to either second-line chemotherapy, or erlotinib. Response rates were highest with the combination of erlotinib plus bevacizumab (Table 2) and improvements in PFS and OS were seen with both bevacizumab containing arms. Subsequently, the phase III BeTa trial randomized patients to erlotinib plus bevacizumab versus erlotinib alone [39]. Significant improvements were observed in RR (13 % vs. 6 %) and PFS (3.4 m vs. 1.7 m, HR 0.62, 95 %CI 0.52–0.75). However, there was no improvement in overall survival (9.3 m vs. 9.2 m, HR 0.97, 95 %CI 0.80–1.18). As a result, there are no data currently supporting the use of bevacizumab as second-line therapy in combination with either erlotinib or chemotherapy.

Ramucirumab is an IGG1 humanized monoclonal antibody targeting the VEGFR-2 receptor. It is being evaluated in a number of disease sites including NSCLC. A phase II trial of ramucirumab in combination with carboplatin and paclitaxel was reported at the American Society of Clinical Oncology meeting in 2010 (Table 3) [41]. Data were reported on 31 patients with NSCLC (all histologies were allowed). There were 10 responses in 15 evaluable patients (RR 67 %). Detailed toxicity information was not available.

Doebele et al, reported the results of a randomized phase II trial of cisplatin or carboplatin in combination with pemetrexed, with or without ramucirumab for four to six cycles [42]. Pemetrexed was continued as maintenance therapy either alone or with ramucirumab. A total of 140 patients were randomized. The addition of ramucirumab appeared to improve the efficacy of a platinum agent plus pemetrexed. Higher response rates were observed (49.3 % vs. 38 %). There were trends towards improved PFS (HR 0.75, 90 %CI 0.55–1.03) and OS (HR 0.83, 90 %CI 0.56–1.22), although these did not achieve statistical significance. Patients randomized to ramucirumab experienced more grade 3 neutropenia, anemia, thrombocytopenia, nausea, fatigue, hypertension, and back pain.

At this point ramucirumab remains an investigational agent with some promise. There are ongoing randomized phase II trials in both first and second-line therapy for NSCLC.

Aflibercept is a human fusion protein designed to block biding of VEGF-A and VEGF-B, as well as human placental growth factor (commonly referred to as VEGF-trap). It has evidence of minor single agent activity in heavily pretreated patients [43]. A phase II trial of first-line therapy with aflibercept in combination with cisplatin and pemetrexed showed modest activity [44]. The RR was 26.3 % with median PFS of 5 months. However, the trial was stopped early because of three cases of reversible posterior leukoencephalopahy syndrome (RPLS). A randomized phase III trial of docetaxel plus or minus aflibercept as second-line therapy, showed some evidence of activity for the addition of aflibercept [45]. The addition of aflibercept to docetaxel improved RR (23.3 % vs. 8.9 %, p < 0.001) and PFS (HR 0.82, 95 %CI 0.72–0.94). However, the study failed to achieve its primary outcome of improved overall survival. In addition there were no improvements in quality of life as measured by the Lung Cancer Symptom Scale (LCSS). There are multiple ongoing trials of aflibercept in other cancers. However, data to date do not support the use of aflibercept in NSCLC.

Vadimezan is another VEGF-Trap compound [46]. Activity was observed in early phase clinical trials. A randomized phase II trial of carboplatin and paclitaxel plus/or minus vadimezan suggested that vadimezan may improve the effectiveness of platinum-based chemotherapy [47]. However, a phase III trial of the same regimen failed to improve on treatment outcomes [48]. There were no significant differences in response rates, PFS and overall survival. Additionally, quality of life between the two groups was similar. Therefore, vadimezan does not improve the therapeutic options for patients with advanced NSCLC.

Sorafenib is a multitargeted TKI active against VEGFR, PDGFR, Raf, c-Kit and FLT-3 [49]. Preliminary data from heavily pretreated NSCLC patients showed disease stabilization in 59 % of patients [50]. As a result, sorafenib underwent further development in NSCLC. A phase I/II trial of sorafenib in combination with standard doses of carboplatin and paclitaxel demonstrated sorafenib could be administered in full doses with chemotherapy [51]. Subsequently, two phase III trials were conducted [52, 53]. The ESCAPE trial randomized 926 patients to carboplatin and paclitaxel plus sorafenib or placebo [53]. Patients were all good performance status (ECOG 0-1) and all NSCLC histologies were included in the trial. The trial was stopped early following an interim analysis that met criteria for futility. At the time of the final analysis, no differences were observed in OS between the two groups (median OS 10.7 m vs. 10.3 m, HR 1.15, 95 %CI 0.94–1.41). Additionally, there were no differences observed in the secondary outcomes (RR or PFS) and quality of life was not assessed in the trial. Interestingly a planned analysis according to histology demonstrated that patients with squamous cancer randomized to sorafenib had worse OS than patients in the placebo group (8.9 m vs. 13.6 m, HR 1.85, 95 %CI 1.22–2.81). Patients randomized to sorafenib experienced more thrombocytopenia, rash / desquamation, hand-foot reaction, hypertension and pruritis than in the placebo group. However, patients with squamous cancer did not experience incrementally worse toxicity.

A similarly designed trial (NEXUS), evaluated the addition of sorafenib or placebo, to cisplatin and gemcitabine in 772 patients [52]. Recruitment of patients with squamous cancer was halted following analysis of the ESCAPE trial. Data have been reported on the non-squamous histology population. No improvement in OS was observed for patients randomized to sorafenib (12.4 m vs. 12.5 m, HR 0.98, 95 %CI 0.83–1.16). A similar toxicity profile was observed to that seen in the ESCAPE trial. In summary there is no evidence to support the addition of sorafenib to first-line platinum-based chemotherapy.