Safety and Efficacy of Treatment with/without Ramucirumab in Advanced or Metastatic Cancer: A Meta-Analysis of 11 Global, Double-Blind, Phase 3 Randomized Controlled Trials

Ramucirumab, as a vascular endothelial growth factor receptor-2 inhibitor, was first approved in 2014 for treated advanced or metastatic gastric/gastroesophageal junction (GEJ) adenocarcinoma. This study deeply analyzed the efficacy and safety of advanced or metastatic cancer treated with ramucirumab, which included 11 global, double-blind, phase 3 randomized controlled trials with a total of 7410 patients. Subgroup analysis based on different cancer types showed that standard regimens plus ramucirumab significantly increased progression-free survival and overall survival compared with placebo groups in patients with advanced non-small-cell lung cancer (NSCLC), hepatocellular carcinoma, gastric cancer, or GEJ adenocarcinoma. Although a higher proportion of patients achieved overall response and disease control than those treated with placebo, the overall response was not statistically significant between the two groups in advanced NSCLC. Grade 3 or worse treatment-emergent adverse events (TEAEs) that occurred in at least 5% of patients were neutropenia (30.5% in the ramucirumab group vs. 23.5% in the placebo group), leucopenia (14.8% vs. 9.2%), weight decreased (14.2% vs. 8.0%), myalgia (11.7% vs. 7.7%), fatigue (10.9% vs. 7.7%), hypertension (9.2% vs. 2.3%), and anaemia (6.2% vs. 7.7%). In the TEAEs of special interest, the ramucirumab group had a significantly higher incidence of bleeding (mainly grade 1-2 epistaxis and gastrointestinal bleeding), hypertension, proteinuria, liver injury/failure (grade 1-2), venous thromboembolism (grade 1-2), and gastrointestinal perforation (grade ≧3) than the control group.


Introduction
Cancer is a serious global public health problem and is expected to become the leading cause of death in countries around the world in the 21st century, which may bring great harm to human health [1]. In 2021, there was an estimation of 1,898,160 new cancer cases and 608,570 cancer-related deaths in the United States [2]. In the special period of the COVID-19 global pandemic, the delay in diagnosis and treatment of early cancer may further increase the incidence and mortality of advanced cancer due to the scarcity of medical resources. Tis phenomenon inevitably leads to a heavier economic burden on families and society. Traditional treatments for cancer include surgery, chemotherapy, and radiation. However, due to the untimely early diagnosis, most patients lose the opportunity for surgical treatment. Chemotherapy or radiotherapy also primer its acquired drug resistance and serious side efects, forcing people to seek other treatments. Terefore, the combination of targeted therapies, immunotherapy, and gene therapy has developed rapidly in recent years [3].
Te formation of blood vessels can provide oxygen and nutrients to tissues and play an important role in the growth, invasion, and metastasis of tumor cells [4]. Tis hypothesis was frst proposed by Folkman in 1971, who believed that the continuous growth of solid tumors larger than 1-2 mm 3 requires new blood vessels to promote, and gave some evidence to support this view in subsequent studies [5,6]. In the past 50 years, a large number of studies have reported the efects of a variety of regulatory factors on tumor angiogenesis, such as vascular endothelial growth factor (VEGF), transforming growth factor, and hypoxia-inducible factor [7][8][9][10]. VEGF is one of the targets of anticancer drugs. Ligands VEGF-A, -C, and -D can specifcally bind to receptor VEGFR-2, increase angiogenesis and vascular permeability, and provide sufcient oxygen and nutrients to cancer cells [11]. Terefore, limiting the binding of VEGF ligands and receptors may delay the progression of cancer. An abnormal expression of VEGF has been found in various types of cancer and is associated with poor prognosis. For example, digestive cancers included esophageal cancer [12], gastric cancer [13], colorectal cancer [14], intrahepatic cholangiocarcinoma [15], and pancreatic cancer [12], as well as nondigestive cancers, included lung cancer [16], breast [17], cervical cancer [18], ovarian cancer [19], osteosarcoma [20], prostate cancer [21], and renal cancer [22]. VEGF is therefore a promising therapeutic target in advanced or metastatic cancers. Te VEGFR-2 inhibitor ramucirumab, a fully human immunoglobulin G1 monoclonal antibody, was frst approved in the United States in 2014 for the treatment of advanced or metastatic gastric/ gastroesophageal junction (GEJ) adenocarcinoma [23][24][25]. Today, results from 11 worldwide, phase 3, double-blind randomized controlled trials (RCTs) investigating the efcacy and safety of RAM in diferent types of cancer have been published, and a number of Phase 1/2 RCTs is also underway. Based on the data reported in these 11 worldwide, phase 3 RCTs, the present study's purpose is to investigate whether adding ramucirumab to standard regimens could improve the treatment outcomes in advanced or metastatic cancer, including overall survival (OS), progression-free survival (PFS), objective response rate (ORR), disease control rate (DCR), treatment-emergent adverse events (TEAEs), or of special interest, and to evaluate the efcacy and safety of combination treatment.

Search Strategy.
Te literature search included standard regimens with/without ramucirumab treatment for advanced or metastatic cancer by PubMed, Web of Science, and the Cochrane Library databases, as well as three international conferences including the American Society of Clinical Oncology and European Cancer Conference without language restrictions. Te deadline was on January 31, 2022. Table S1 shows the search strategy of the PubMed database; other databases' search strategy is similar to this. Besides, we manually searched the references of relevant articles to fnd other articles that meet the research conditions.

Inclusion and Exclusion
Criteria. Te inclusion criteria were as follows: (1) patients who were diagnosed with advanced or metastatic cancer cytologically or histologically, (2) published phase 3 RCTs, (3) the experimental group was treated by standard regimens (chemotherapy, radiotherapy, targeted therapy, and best supportive care) plus ramucirumab, and the control group was standard regimens plus placebo, and (4) at least one outcome was reported (PFS, OS, ORR, DCR, TEAEs, or of special interest). Articles or abstracts that do not meet any of the abovementioned criteria were excluded.

Data Extraction.
Te titles and abstracts of published articles were frst selected according to the keywords. After preliminary evaluation, the full texts that meet the collection requirements were downloaded for in-depth reading. Second, the detailed data of qualifed studies, including frst author, title, year of publication, treatment measures, sample size, disease stage, and primary and secondary outcomes, were independently extracted into two spreadsheets by two authors, and the authenticity of these data was carefully checked by the other two authors to avoid possible evaluation bias caused by researchers. If a study has more than one publication, the authors aggregated the latest results for data analysis. Te authors did not extract data for all of the adverse events mentioned in the study. Instead, only treatment-emergent adverse events reported in more than two studies were analyzed. Any diferences of opinion were resolved by discussion with the principal investigator or arbitration by a third examiner.

Quality Evaluation.
Te risk of bias for each article was assessed using the Cochrane Collaboration tool, and the following evaluation domains were assessed accordingly: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases [26]. Review Manager (version 5.4), a statistical software provided by the Cochrane Collaboration, was used to draw the "Risk of bias graph" ( Figure S1) and "Risk of bias summary" ( Figure S2).

Statistical
Analysis. Data analysis and fgures production were completed by the Review Manager. PFS and OS were analyzed by the hazard ratio (HR) and 95% confdence interval (CI). ORR, DCR, and adverse events were analyzed by the odds ratio (OR) and 95%CI, p < 0.05 means a statistical signifcance. Te heterogeneity of studies meeting inclusion criteria were assessed by I-square and chisquare tests. When I 2 > 50% or p≦0.1, the random-efect model was adopted; if not, the fxed-efect model was used.
the full text, 11 RCTs were fnally included in this metaanalysis ( Figure 1). Te main characteristics and dosage administered of each RCTs are shown in Table 1, including RAINBOW-Asia [27], RELAY [28], REACH-2 [29], RANGE [30], RAINFALL [31], RAINBOW [32], RAISE [33], REVEL [34], REACH [35], REGARD [36], and ROSE/TRIO-012 [37]. Tere were 7 RCTs compared the chemotherapy plus ramucirumab with chemotherapy plus placebo, 3 RCTs compared best supportive care (BSC) plus ramucirumab after chemotherapy with BSC plus placebo after chemotherapy, and 1 RCT compared EGFR-TKI plus ramucirumab with EGFR-TKI plus placebo. Te total population was 7410, with 4078 (55%) in the ramucirumab group and 3332 (45%) in the control groups. Te remaining characteristics are shown in Tables 2 and S2. 3.1. PFS and OS. PFS is defned as the time of disease progression or death from any cause since randomization, and OS is defned as the time from randomization to the date of death from any cause. Meta-analysis of PFS ( Figure 2) and OS ( Figure 3)  and urothelial carcinoma respectively, so we could not conduct a meta-analysis on them. In general, the abovementioned results indicated that the standard regimen combined with ramucirumab could achieve more benefcial PFS and OS than the standard regimen plus placebo in patients with advanced NSCLC, HCC, gastric cancer, or GEJ adenocarcinoma.

ORR and DCR.
ORR is defned as complete response rate plus partial response rate. DCR is defned as complete response rate plus partial response rate and disease stability rate. Similarly, a meta-analysis of ORR ( Figure 4) and DCR ( Figure 5) was also conducted according to diferent cancer types.

TEAEs of Special
Interest. Te merged data of grade 1-2 and grade≧3 TEAEs of special interest were analyzed, respectively (Table 3). Te number of events in each RCT and the raw data for the meta-analysis were provided in the supplementary material as Table S3. 3.3.1. Bleeding/Haemorrhage. 10 RCTs reported the risk of bleeding/haemorrhage events, including 1304 (34.5%) patients in the ramucirumab group and 503 (16.6%) in the control group with grade 1-2 events, 107 (2.8%) in the ramucirumab group and 87 (2.9%) in the control group with grade≧3. Although the results showed a higher incidence of bleeding/haemorrhage events in the ramucirumab group at grade 1-2 (OR: 2.72, 95% CI: 2.31-3.09, p < 0.00001), there was no statistical signifcance between the two groups at grade≧3 (OR: 1.28, 95%CI: 0.82-1.99, p � 0.27). Since pulmonary haemorrhage is a problem of NSCLC receiving antiangiogenic drugs, a separate metaanalysis of two studies (REVEL and RELAY) in advanced NSCLC was performed, and the results showed there were no statistically signifcant diferences between the two groups in grade 1-2 (OR: 2.04, 95% CI: 0.45-9.33, p � 0.36) or grade≧3 (OR: 0.88, 95% CI: 0.35-2.24, p � 0.79) pulmonary haemorrhage events.

Hypertension.
Hypertension was defned as newly diagnosed arterial hypertension during treatment or a previously diagnosed hypertension that worsened during treatment. Hypertension (hypertension and increased blood pressure) data were provided in 10 RCTs, included 600 (15.9%) patients in the ramucirumab group and 190 (6.3%) in the control group with grade 1-2 events, 379 (10.0%) in the ramucirumab group and 88 (2.9%) in the control group with grade≧3. Te results of diferent analysis between two groups in the occurrence of grade 1-2 and grade≧3 events were (OR: 2.57, 95%CI: 2.16-3.095, p < 0.00001) and (OR: 3.86, 95%CI: 3.04-4.89, p < 0.00001), respectively. In general, the incidence of hypertension in the ramucirumab group was higher than in the control group, regardless of grade 1-2 or grade≧3.

Proteinuria.
Tere were 486 (13.0%) patients in the ramucirumab group and 153 (5.0%) in the control group with grade 1-2 events, 55 (1.5%) in the ramucirumab group, and 4 (0.1%) in the control group with grade≧3. Moreover, the meta-analysis showed that the incidence of proteinuria in the ramucirumab group was higher than in the control

TEAEs.
Te merged data of grade 1-2 and gra de≧3 TEAEs were analyzed, respectively (Table 4). Te number of events in each RCT and the raw data for the meta-analysis were provided in the supplementary material as Table S4. In the total incidence of TEAEs, fatigue (55.2%) and neutropenia (52.3%) were the most common, followed by diarrhoea (33.5%), nausea (33.0%), alopecia (29.1%), stomatitis (28.3%), decreased appetite (27.8%), leucopenia (26.7%), anaemia (25.1%), and the incidence rate of other TEAEs was 10%-25%. In TEAEs, the incidence of grade 1-2 and grade≧3 fatigue, hypertension, decreased appetite, stomatitis, neutropenia, and thrombocytopenia in the  Journal of Oncology ramucirumab group were higher than in the control group. Te incidence of grade 1-2 nausea, epistaxis, peripheral oedema, pyrexia, proteinuria, cough, hypoalbuminaemia, ascites, headache, leucopenia events, or grade≧3 neuropathy, myalgia events were also higher in the ramucirumab group than in the control group. However, abdominal pain, alopecia, diarrhoea, dyspnoea, vomiting, back pain, weight decreased, and rash were not statistically diferent between the two groups. It is worth noting that the incidence of grade≧3 constipation events and all grade anaemia events were higher in the control group.

Subgroup Analysis Based on Diferent Treatment
Regimens. Subgroup meta-analyses of TEAEs of special interest based on diferent treatment regimens also have   10 Journal of Oncology been performed (Table 5). Te incidence of all grades of hypertensive events, grade 1-2 proteinuria, and bleeding/ haemorrhage events was higher in the ramucirumab group than the control group, regardless of what treatment regimens were combined. Besides, the addition of ramucirumab also increased the incidence of grade 1-2 liver injury/failure events and grade≧3 proteinuria events in BSC. However, the incidence of grade≧3 VTE events was higher in the control group. Furthermore, the addition of ramucirumab increased the incidence of grade 1-2 liver injury/failure events in PTX, and all grades of VTE events in DOC.

Discussion
To our knowledge, ramucirumab plus BSC, targeted therapy, or chemotherapy signifcantly increased OS compared with placebo groups in patients with advanced NSCLC, HCC, gastric cancer, or GEJ adenocarcinoma. Tese patients treated with ramucirumab also had signifcantly longer PFS. Although a higher proportion of patients achieved overall response and disease control than those treated with a placebo, the ORR was not statistically signifcant between the two groups in advanced NSCLC.
In the TEAEs that were statistically diferent between the two groups, fatigue, diarrhoea (grade 3), nausea (grade 1-2), decreased appetite, and stomatitis were some of the most frequently reported nonhaematological events. Neutropenia, leucopenia (grade 1-2), and thrombocytopenia were the most frequently reported haematological TEAEs. Moreover, the incidence of TEAEs reported above was higher in the ramucirumab group. In addition, the TEAEs of special interest such as bleeding/haemorrhage (mainly grade 1-2 epistaxis and gastrointestinal haemorrhage), hypertension, liver injury/failure (grade 1-2), proteinuria, VTE (grade 1-2), and GI perforation (grade 3) were also more common in the ramucirumab group, as expected, consistent with the incidence of adverse events associated with most antiangiogenic therapies (38)(39)(40). Most of grade 4 or 5 TEAEs of special interest occurred with a very low incidence in both groups. Tere were only 0.07% of patients (3/4040) reported grade 4 hypertensive crisis, and 0.03% of patients (1/3782) reported grade 4 proteinuria in the ramucirumab group (0% in control), with no grade 5 events. Grade 4 and grade 5 GI perforation events were 0.36% (10/2783) and 0.22% (6/2783) in the ramucirumab group (only 1 case in each control group). Fortunately, in the reported cases of RCTs, these events were all within the safe range that could be controlled  VEGF has been proved to be essential for the occurrence, development, and metastasis of cancer, and its mediated carcinogenesis is mainly completed by afecting the formation of new blood vessels and vascular permeability. In the past few decades, many anticancer drugs that target VEGF receptors or subtypes or signaling pathways have been approved by the United States Food and Drug Administration. Tese targeted drugs can be divided into the following categories according to the mechanism of action: (1) block the binding of ligands to receptors, such as bevacizumab, which targets VEGF-A, (2) block signaling through VEGFR, such as ramucirumab, which targets VEGFR2, and (3) tyrosine kinase inhibitors that block VEGFR1, VEGFR2, and VEGFR3 kinase activity, such as pazopanib and lenvatinib [41,42]. Te frst approved antiangiogenic drug, bevacizumab, was used in clinical practice as early as 15 years ago. Although bevacizumab in combination with chemotherapy was well tolerated in multiple cancers and resulted in more benefcial PFS, no OS beneft was reported. Terefore, a large number of clinical trials of ramucirumab have been carried out, and its efcacy and safety have been reported in 11 phase 3 global RCTs. Two previous meta-analyses of 6 global trials have been published [43,44]. However, these two studies did not report the results of PFS, OS, ORR, or DCR. Te present study was a comprehensive update and meta-analysis of 11 global, double-blind, phase 3 RCTs, which were treated with standard regimens plus ramucirumab or placebo for advanced or metastatic cancer. Tis study represented one of the largest meta-analysis of antiangiogenic therapy for advanced cancer. A total of 4078 patients treated with ramucirumab and 3332 patients treated with placebo were included in this evaluation, which was a great strength and provided high-quality evidence for the safety and efcacy of ramucirumab therapy. A potential limitation of this metaanalysis is that 3 of these RCTs only reported once on diferent types of cancer. Encouragingly, the global RCTs on ramucirumab are still ongoing. Although the current RCTs have confrmed the important role of the VEGFR-2 signaling pathway as a therapeutic target in advanced cancer; furthermore, studies also need to explore the potential predictive biomarkers of ramucirumab and provide guidance for clinical treatment.

Conclusion
In conclusion, the meta-analysis showed that standard regimens combined with ramucirumab resulted in more benefcial PFS and OS outcomes in advanced NSCLC, HCC, gastric cancer, or GEJ adenocarcinoma. Te importance of the results for clinical practice was that the incidence of hypertension, proteinuria, bleeding, liver injury/failure, VTE, and GI perforation were consistent with reported TEAEs of special interest associated with the angiogenesis inhibitor class. However, the authors do not observe an increased incidence associated with ramucirumab for highgrade bleeding, ATE, IRR, renal failure, fstula, or healing complication across these RCTs. Also, these adverse events were largely controllable after dose adjustment and supportive treatment.

Data Availability
Te original contributions presented in the study are included in the article.

Conflicts of Interest
Te authors declare that they have no conficts of interest.