The Safety and Effectiveness of Bevacizumab in the Treatment of Nonsquamous Non-Small-Cell Lung Cancer: A Meta-Analysis of Randomized Controlled Trials

Objective Bevacizumab was currently available for nonsquamous non-small-cell lung cancer (NSqNSCLC) patients and has been studied in several randomized controlled trials (RCTs) for treatment of these patients. This meta-analysis summarizes the most up-to-date evidences regarding the effects and adverse reactions of bevacizumab in the treatment of NSqNSCLC patients. Methods The authors searched for RCTs from electronic database including PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials. Experimental arm was defined as the bevacizumab-containing group and the control arm as the bevacizumab-free group. Data of objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and adverse reactions were synthetically extracted. A protocol for this meta-analysis has been registered on PROSPERO (http://www.crd.york.ac.uk/prospero). Results Ten RCTs that involved a total of 3134 patients were included. The experimental group was associated with significant superior ORR (RR 1.63, 95% CI 1.24 to 2.14, P < 0.001), OS (HR 0.90, 95% CI 0.82 to 0.99, P < 0.001), and prolonged PFS (HR 0.68, 95% CI 0.62 to 0.74, P < 0.001) compared to the control. No significant difference was observed regarding DCR (RR 1.13, 95% CI 0.99 to 1.30, P = 0.08). The experimental group showed higher rate of hypertension (RR 6.91, 95% CI 4.62 to 10.35, P < 0.00001) and hemorrhagic events (RR 3.07, 95% CI 1.78 to 5.30, P < 0.0001) than the control group. The experimental group showed lower rate of anemia (RR 0.72, 95% CI 0.55 to 0.96, P = 0.02) than the control group. No significant difference was observed regarding treatment-related adverse event grade 3-5 (TRAE3-5) (RR 1.23, 95% CI 0.99 to 1.53, P = 0.06), thrombocytopenia (RR 1.11, 95% CI 0.92 to 1.33, P = 0.29), and neutropenia (RR 1.11, 95% CI 0.88 to 1.40, P = 0.36). Conclusion This meta-analysis showed that bevacizumab could increase ORR, OS, and prolonged PFS for treatment of NSqNSCLC patients. However, no significant improvement in DCR was observed and bevacizumab could increase the rate of hypertension and hemorrhagic events. Bevacizumab was an acceptable option for NSqNSCLC patients. This trial is registered with PROSPERO registration number: CRD42021226790.


Introduction
As the leading cause of cancer death worldwide, lung cancer accounts for 18.4% of the total cancer deaths [1]. Non-smallcell lung cancer (NSCLC) accounts for approximately 80-85% of all lung cancer cases and usually allocated to advanced stage at their first diagnosis [2]. The NCCN guideline suggested that systemic palliative chemotherapy and/or radiotherapy remained the standard care for these locally advanced or metastatic NSCLC patients [3].
To inhibit angiogenesis is another treatment option, because tumor angiogenesis is critical for the process of primary tumor growth, proliferation, differentiation, and metastasis and has been identified as an important therapeutic target for tumor in recent decades [4,5]. Antiangiogenic therapy has been used for cancer treatment, which inhibits the delivery of oxygen and nutrients to cancer cells [6]. As a key mediator of angiogenesis, vascular endothelial growth factor (VEGF) and its receptors are considered to be the pivotal pathway in angiogenesis-related molecular mechanisms which have been well studied [7,8].
Bevacizumab is a VEGF monoclonal antibody, which inhibits angiogenesis to suppress tumor growth by restricting oxygen and nutrient supply to tumors [9]. Increasing numbers of clinical trials have been conducted with bevacizumab for the treatment of patients with advanced NSCLC since it was approved. Previous meta-analysis found that bevacizumab used in combination with paclitaxel and carboplatin did increase objective response rate (ORR), overall survival (OS), and prolonged progression-free survival (PFS) compared with paclitaxel and carboplatin for NSCLC [10][11][12][13][14]. However, the important outcome disease control rate (DCR) was not analyzed and only five RCTs were included. Whether bevacizumab containing could improve ORR, DCR, and OS and increase adverse reactions for treatment of nonsquamous non-small-cell lung cancer (NSqNSCLC) patients is still controversial. In consideration of these controversial results, we carried out this meta-analysis. Our meta-analysis includes 10 randomized controlled trials (RCTs) and identifies the precise effect of bevacizumab containing for NSqNSCLC patients on outcomes of ORR, DCR, OS, PFS, and treatmentrelated adverse event compared with bevacizumab free.

Method
2.1. Protocol and Registration. This meta-analysis was performed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) recommendations. This study was not a human or animal experiment; thus, ethical approval was not necessary. A protocol for this meta-analysis has been registered on PROSPERO (http://www.crd.york.ac.uk/prospero), and the registration number is CRD42021226790.

Search Strategy.
Databases including PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials were searched with a combination of the terms "non-small-cell lung cancer (NSCLC) or nonsquamous non-small-cell lung cancer (NSqNSCLC)" and "angiogenesis inhibitors or bevacizumab" within the restriction limit of "randomized controlled trial (RCT)." In addition, reference lists of the included studies were manually checked for potentially eligible studies and Google Scholar search engines were used to find additional references. The last search was performed on December 8, 2020, without any restriction to language of publication.

Inclusion and Exclusion
Criteria. Inclusion criteria are as follows: (1) research types: RCTs publicly published at home and abroad; (2) research objects: adult patients with confirmed locally advanced or metastatic NSqNSCLC; and (3) intervention measures: the experimental group using bevacizumab plus standard chemotherapy regimen and the control group using standard chemotherapy regimen alone. Exclusion criteria are as follows: exclude articles that do not meet the inclusion criteria, cannot obtain the main indicators in the article, and have not received a response through contacting the author, and republished articles.

Quality Assessment and Data
Extraction. The Cochrane risk of bias tool was used to evaluate the quality of each study by two reviewers, and the following 7 categories were assessed: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of the outcome assessment, incomplete outcome data, selective outcome reporting, and other biases [15]. The overall methodologic quality of each included study was assessed as "low risk of bias," "high risk of bias," or "unclear risk of bias." A third reviewer would be invited if there were any dispute.
Two authors independently completed the data extraction. The extracted general data included author, year, and country of publication and sample size. The primary endpoint was OS, and the secondary end-points contained ORR, DCR, PFS, grade 3-5 of treatment-related adverse event (TRAE3-5), hypertension neutropenia, thrombocytopenia, anemia, and hemorrhagic events.

Statistical
Analysis. Outcomes were estimated by calculating the pooled risk ratio (RR) (95% confidence intervals [CIs]) for ORR, DCR, and TRAE by RevMan software (version 5.1; Cochrane Collaboration, Copenhagen, Denmark), and hazard ratio was pooled for survival outcomes (OS and PFS) by STATA version 12.0 (StataCorp, College Station, TX). A P < 0:05 was considered statistically significant. Heterogeneity was assessed by visual inspection of the forest plot combined with the results of the test for heterogeneity and the I 2 test. Fixed-effects model would be employed for outcomes with low heterogeneity (I 2 < 50%). Otherwise, the random-effects model of DerSimonian and Laird [16] would be selected. Sensitivity analysis would be conducted by omission of each single study to evaluate stability of the results if heterogeneous studies existed.

Search
Results and Characteristics of Included Studies. 551 potential articles were initially identified through database searches on 8 December 2020. Two hundred and thirty-three studies were considered potentially eligible for further assessment after duplicates were removed. Finally, 10 RCTs [17][18][19][20][21][22][23][24][25][26] that involved a total of 3134 patients published between 2006 and 2020 met the inclusion criteria and were included in this meta-analysis after a full-text review. Figure 1 shows the literature selection process. Table 1 summarizes the details of both the included studies and agents.

Risk of Bias.
All included RCTs were assessed by two authors independently according to Cochrane risk of bias tool. Detailed information can be found in Figure 2.

Discussion
Interference with VEGFR functions has been an alternative approach for the treatment of NSCLC [27]. Bevacizumab, a novel targeted therapeutic, differs in their modes of action and tolerability profiles from those of cytotoxic agents and can be combined with traditional chemotherapy to offer greater clinical benefits [25]. It has been approved for use in combination with the standard platinum-based chemotherapy or as a maintenance therapy after chemotherapy  We conducted this meta-analysis with 10 RCTs included 3134 advanced NSqNSCLC patients to compare therapeutic efficacy and adverse reactions of bevacizumab containing and bevacizumab free for NSqNSCLC patients. According to the current outcomes, treatment regimens containing bevacizumab had significant improvements for ORR, OS, and PFS outcomes, when compared with the treatment regimens without bevacizumab. However, significant outcome was not observed in DCR. It is indicated that significant       [10]. However, the important outcome disease control rate (DCR) was not analyzed and only five RCTs were included. Five RCTs including 1852 patients, 7 RCTs including 2671 patients, and 8 RCTs including 2897 patients indicated that the bevacizumabcontaining group and bevacizumab-free group had a similar rate of TRAE3-5, thrombocytopenia, and neutropenia. All 10 RCTs reported hypertension and 7 RCTs reported hemorrhagic events, and the bevacizumab-containing group showed a higher rate of hypertension and hemorrhagic events compared to the bevacizumab-free group. Seven RCTs reported anemia and the bevacizumab-containing group showed a lower rate of anemia compared to the bevacizumab-free group.
Though more RCTs were included in this meta-analysis, improvement was not observed with respect to DCR in the