Risks of hypertension and thromboembolism in patients receiving bevacizumab with chemotherapy for colorectal cancer: A systematic review and meta‐analysis

Abstract Background Guidelines show that for metastatic colorectal cancer (mCRC), a combination of three‐drug regimens, fluorouracil, leucovorin, and oxaliplatin and bevacizumab (BVZ), is one of the first‐line standard therapies. BVZ is generally well tolerated; however, it is associated with infrequent, life‐threatening side effects such as severe hypertension (HTN) (5%–18%), Grade ≥3 arterial thromboembolism (ATE) (2.6%), Grade ≥3 hemorrhagic events (1.2%–4.6%), and gastrointestinal perforation (0.3%–2.4%). This meta‐analysis aims to evaluate the additive risk of BVZ‐induced severe HTN and thromboembolism when BVZ is combined with a standard chemotherapy regime in patients with mCRC. Methods Our search was conducted from January 29, 2022, to February 22, 2022, through databases of PubMed, clinicaltrial.gov, EMBASE, Web of Science, and Cochrane Library. Data analysis from randomized controlled trials (RCTs) and clinical trials was conducted using Review Manager V.5.4, comparing BVZ‐chemotherapy to chemotherapy only, focusing on cardiovascular AE such as HTN and arterial and venous thromboembolism. Results The analysis from 26 clinical trials and RCTs showed that the odds of HTN were about four times higher, and ATE subgroup analysis of 11 studies showed over two times higher odds of ATE in patients being treated with BVZ compared to the chemotherapy‐only group. Conclusion BVZ, when added to the standard chemotherapy regimen for mCRC, was associated with higher odds of developing HTN and thromboembolism, specifically ATE, than the chemotherapy‐only group. Our findings are significant as they provide vital information in analyzing the risk–benefit ratio of adding BVZ to the standard chemotherapy regime in patients with mCRC, especially in patients with vascular comorbidities.


| INTRODUCTION
Colorectal cancer (CRC) is the second leading cause of death related to cancer (53,200) and the fourth most frequently diagnosed cancer (147,950) in 2020, per the American Cancer Society in the United States. 7,11,59evacizumab (BVZ), a recombinant humanized monoclonal IgG antibody directed against the human vascular endothelial growth factor (VEGF), is commonly used in frontline to later lines of treatment in advanced or metastatic colorectal cancer (mCRC). 1,6BVZ inhibits endothelial cell proliferation and neovascularization by inhibiting the interaction of VEGF with its receptors (Flt-1 and KDR) situated on the surface of endothelial cells and has a long half-life of 3-4 weeks. 2 The connection between the vascularization of tumors and metastasis has made VEGF the most important of several growth factors to regulate angiogenesis and a target for antineoplastic agents. 8VZ was first approved by the FDA (Food and Drug Administration) to be used in combination with 5-fluorouracil (5-FU) based chemotherapy in 2004. 3ubsequently, multiple trials have also shown that BVZ combined with capecitabine, irinotecan, or oxaliplatinbased chemotherapy improved progression-free survival and overall survival in patients with previously untreated mCRC. 1,3,7,12Adding BVZ to oxaliplatin, 5-FU, and leucovorin for patients with mCRC previously treated with chemotherapy improves survival. 13Patients with mCRC, on average, are likely to be exposed to BVZ for a long duration in their overall treatment course.
Treatment guidelines for advanced or mCRC show that a cytotoxic combination of fluorouracil, leucovorin, and oxaliplatin (FOLFOX) and a biologic, BVZ, is a standard first-line choice of therapy.Other valuable firstline chemotherapy combinations may include irinotecan such as FOLFOXIRI (folinic acid, fluorouracil, oxaliplatin, and irinotecan) or FOLFIRI (folinic acid, fluorouracil, and irinotecan) and are still often given together with BVZ. 4,5,60Subsequently, maintenance therapy with fluoropyrimidine and third-line therapy with trifluridine/tipiracil often also include BVZ. 14,15ide effects of BVZ are often manageable; however, it is associated with occasional, life-threatening side effects such as severe hypertension (HTN) (5%-18%), Grade ≥3 arterial thromboembolism (ATE) (2.6%), Grade ≥3 hemorrhagic events (1.2%-4.6%),delayed wound healing, and gastrointestinal perforation (0.3%-2.4%). 1,16The risk of ATE is higher in patients with preexisting cardiovascular risk factors or cardiovascular diseases, including myocardial infarction, chronic heart failure, or stroke.The mechanism of BVZ's increased risk of ATE is not well studied, but some studies suggest that BVZ increases the risk by promoting inflammation and atherosclerotic instability, vasoconstriction by decreasing nitrous oxide production, and platelet aggregation and adhesion to vascular endothelium by lowering endothelial cell renewal capacity. 10 HTN of all grades has been reported in up to 36% of patients being treated with BVZ.In contrast, the reported incidence of high-grade HTN ranges from 1.8% to 22%, with up to 1% of events being grade 4 HTN.Unmanaged HTN can lead to cardiovascular complications, subarachnoid hemorrhage, and encephalopathy.The prevailing hypothesis for the mechanism of BVZ-induced HTN is the inhibition of VEGF-mediated vasodilation leading to an increase in vascular tone. 9s we observe an upward trend in using BVZ in the first-line treatment for patients with mCRC, it is important to analyze the side effects and potential toxicities that this drug can cause in this population.Earlier randomized controlled trials (RCTs) could not estimate the risk of thromboembolic toxicity with high precision due to low thromboembolic event rates.This meta-analysis aims to evaluate and assess the risk of BVZ-induced severe HTN and the odds of thromboembolic toxicity of BVZ, including subgroup analysis of FOLFOX subgroup and arterial and venous thromboembolic events with BVZ in combination with most frequently used chemotherapy regimens compared to chemotherapy without BVZ in patients with mCRC.

| METHODS
This meta-analysis of RCTs and clinical trials compared BVZ-chemotherapy to chemotherapy only, focusing on cardiovascular AE such as HTN and arterial and venous thromboembolism.The review was registered on Prospero under registration ID CRD42021246788.

| Search criteria
A systematic search was conducted using meta-analyses of observational studies in epidemiology guidelines 1  We included RCTs and other clinical trials from 1980 to March 2022 with BVZ added to cytotoxic chemotherapy backbone for patients with CRC if adverse event reporting included cardiovascular side effects.Both adjuvant and metastatic settings were included.We excluded all studies where patients did not receive treatment with BVZ for CRC.We excluded review articles, meta-analyses, observational studies, or case series.Additionally, non-English literature, animal studies, and non-full text were excluded.
Our search was conducted from January 29, 2022, to February 22, 2022, through databases of PubMed, clini caltr ial.gov, EMBASE, Web of Science, and Cochrane Library.Abstracts were reviewed, after which full articles were checked for the availability of data on the cardiovascular side effects of BVZ in patients with CRC.Studies included in this meta-analysis used BVZ to treat CRC and reported side effects, including cardiovascular side effects.AD and DM independently screened all studies identified and simultaneously assessed full texts to check eligibility.Any disagreements were resolved through discussion with another reviewer (AC).A flow diagram depicting the literature search and study selection process has been described in Figure 1.

| Data extraction
We manually extracted the following variables from the included studies using prespecified data collection forms by two authors (NK and AK) with a common consensus on the disagreement of another author (AD).Quantitative data were extracted from collection forms to a Google spreadsheet.We collected the following characteristics in our study, including the first author's last name, date of publication, country, sample size, median age, sex, the dosage of BVZ, chemotherapy regimen, the incidence of grade 3 or 4 HTN, arterial and venous thromboembolism in BVZ treatment, and chemotherapy-only control groups.Separate subgroup analyses were performed to calculate the OR of HTN in the FOLFOX subgroup, arterial and venous thromboembolism.
Before inclusion in the review, two independent reviewers (AD and DM) assessed quantitative papers selected for retrieval for methodological validity, utilizing standardized critical appraisal instruments by the Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument.Any reviewer disagreements were resolved with discussion or with a third reviewer (AC).In case, the study has more than one outcome comparison.In that case, data from the most severe outcome in the analysis were used to minimize the study's overall selection bias.

| Statistical analysis
Dichotomous variables were calculated using the Mantel-Haenszel formula to obtain ORs and their 95% CIs.It was used to analyze the relationship between BVZ and cardiovascular side effects (grade 3 or 4 HTN, arterial and venous thromboembolism) compared to chemotherapy alone.Regardless of heterogeneity, random-effect models

Identification of studies via databases and registers
Records not relevant to our review, removed after screening the abstract or removed for other reasons (n = 454) were used to estimate the combined effect and its precision to give a better conservative estimate of the ORs and 95% CI.
Chi-square and I 2 tests evaluated heterogeneities across individual studies.The I 2 statistic was utilized to analyze statistical heterogeneity.The I 2 statistic of >50% was deemed significant heterogeneity.p-value < 0.05 was considered significant.
Data analysis was performed using Review Manager V.5.4 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).To evaluate the effect of publication bias and heterogeneity by excluding outlying studies on the funnel plot, sensitivity analysis was performed.Funnel plots were visually assessed for publication bias. 18

| RESULTS
Of the 589 studies published as of March 2022, 31 RCTs and clinical trials were selected after excluding duplicates and irrelevant studies, including 17,599 patients in our study.All studies were deemed to be of moderate quality using the Cochrane risk-of-bias (RoB 2) tool.A total of 9609 patients received adjuvant chemotherapy plus BVZ therapy, and 7990 received adjuvant chemotherapy alone.
T A B L E Baseline characteristics of all patients included in the analysis.Patients in this study had a mean age of 65 years.The proportion of the male population was 55%.The entire analysis control group had several guideline-directed chemotherapies, and the "FOLFOX subgroup analysis" control group focused on FOLFOX therapy alone.The baseline characteristics of all patients included in the analysis are listed in Table 1.

PMID
The analysis (Figure 2) pooled from 26 clinical trials, and RCTs showed that the odds of HTN were about four times higher (OR 3.82, 95% CI 3.35-4.36,p-value < 0.00001, I 2 = 78%) in 9789 patients treated with BVZ (12.3%) than the 8018 patients in the control group (4%).Of the 26 studies included in this analysis, 24 suggest higher odds of HTN, while 2 suggested lesser odds of HTN in the BVZ group than in the chemotherapy-only group.
Further, we did a subgroup analysis to study the odds of ATE and VTE in the BVZ+ chemotherapy group versus chemotherapy only.In the ATE subgroup analysis (Figure 5), we included 11 studies comprising 5125 patients in the BVZ + Chemotherapy group and 3602 in the control group.Our analysis showed over two times higher odds of ATE (OR 2.14, 95% CI 1.45-3.15,p-value < 0.00001, I 2 = 0%) in patients being treated with BVZ (2%) as compared to the chemotherapy-only group (1%).
The third subgroup analysis (Figure 6) to study the odds of VTE included 18 studies comprising 6963 patients in the BVZ + Chemotherapy group and 5108 in the chemoonly control group.We calculated that the odds of VTE in patients being treated with BVZ + Chemotherapy (6.55%) were around 1.3 times higher (OR 1.30, 95% CI 1.11-1.51,p-value < 0.0009, I 2 = 31%) than the chemotherapy-only group (5.2%).
Heterogeneity among study results can be attributed to all stages of CRC being considered and comorbidities such as HTN, diabetes, ischemic heart disease, and previous cerebrovascular accident/TIA.Respective funnel plots have been added as supplementary data to assess the symmetry and potential risk of bias (Figures 7-11).added toxicities.Cardiovascular causes are the number one cause of mortality across the globe, and we need to prevent severe HTN and ATE-associated mortality with BVZ.Our meta-analysis is an up-to-date study of cardiovascular side effects of BVZ with the largest cohort of patients in the treatment and control groups, increasing the power of the analysis.We analyzed patients across multiple subgroups, taking into account potential confounding factors, to obtain a more refined odds ratio for the given subgroup of patients.

| DISCUSSION
We demonstrated that 12.3% of patients develop HTN, 8.2% develop thromboembolism, and 2% develop ATE with BVZ when added to the standard chemotherapy regimen for mCRC.The critical findings in our systematic review and meta-analysis are as follows.First, BVZ was associated with about four times higher risk of developing HTN than the chemotherapy-only group.The FOLFOX subgroup analysis, based on the regime used, demonstrated about five times higher odds of developing HTN (Grade 3 or more) in the BVZ + FOLFOX group compared with the FOLFOX-only group.1][22] The proposed mechanism for BVZ-induced HTN, a well-known side-effect of the drug, is increased vascular resistance due to the inhibition of vasodilation mediated by vascular endothelial growth factor.However, it has yet to be established as the sole factor for this effect. 95][26] The current AHA guidelines recommend initiation or up-titration of treatment for secondary HTN due to angiogenesis inhibition by BVZ. 23An earlier study has shown that temporary discontinuation of BVZ can bring the blood pressure level to pretreatment levels, 28 and such a temporary hold is recommended in cases of refractory HTN on the maximum tolerated antihypertensive regime and hypertensive emergency. 27Prevention and management of BVZ-specific HTN is an important aspect of best supportive care during the active treatment phase for mCRC.
Second, the odds of thromboembolism in patients managed with BVZ were over 30% higher than in the control group.Our meta-analysis demonstrates a significant arterial thromboembolic risk of BVZ when added to the advanced CRC treatment standard regime.We found that the odds of getting ATE with BVZ were over two times higher than the chemotherapy-only control groups.These results affirm the manufacturer's ATE claims and are an area of further research. 27Previously, studies have tried to evaluate the prophylactic role of aspirin in preventing ATE in patients receiving BVZ but could not conclude. 28stablishing guidelines to anti-coagulate patients getting BVZ should be studied.Preventing ATE is crucial in improving the morbidity and mortality associated with BVZ treatment.
Our study had several limitations: Firstly, RCTs keep strict inclusion and exclusion criteria.Only patients with appropriate major organ functions are included in these trials; therefore, actual patients may not be represented by the results of this meta-analysis, and our results may not be applicable to the general population in daily practice.Second, although all the studies included RCT, there was a significant difference in the timing and the primary endpoint.The follow-up time is different from study to study, and there might be unreported events of HTN and thromboembolism after the study was concluded.Third, the studies included in our analysis were heterogeneous regarding patient characteristics across the studies.Fourth, potential differences among the trials may exist as HTN events are collected for each individual trial, such as various international institutions and administration schedules of BVZ increasing the clinical heterogeneity and difficulty in interpreting the analysis results.Finally, the treatment designs varied, and the individual patient data were not focused on in the study.Compared with individual patient data analyses, meta-analyses based on published data tend to overestimate the treatment effects. 29Future prospective studies reporting individual subgroup AE data with recurrent, advanced, or mCRC in adjuvant and metastatic settings will help focus on specific subgroups and formulate customized treatment guidelines.
There is a significant risk of publication bias based on the funnel plot obtained regarding the review process.Secondly, the incidence of BVZ-associated HTN, because of the definition of the Common Terminology Criteria for Adverse Events (CTCAE), may have been underestimated.Per the CTCAE, blood pressure greater than 150/100 mmHg or diastolic pressure increased by more than 20 mmHg is considered hypertensive.In some studies, these strict criteria would have reduced the number CRC is a leading cause of cancer-related death in the United States.Most patients with mCRC are likely to receive BVZ with adjuvant chemotherapy.Our findings are significant as they provide vital information in assessing the risk-benefit ratio of adding BVZ to the standard chemotherapy in patients with mCRC, especially in patients with vascular comorbidities.Cardiovascular AEs are crucial in determining all-cause mortality in patients with mCRC.
The magnitude of severe HTN, thromboembolic events, and cardiovascular morbidity demonstrated from our meta-analysis informs oncology clinicians to regularly inform patients and monitor patients with CRC receiving chemotherapy with BVZ.Prevention and management of BVZ-specific HTN and the ATE episodes in the first place can be essential in mitigating the morbidity and mortality associated with BVZ treatment and is an area of future research in this patient population.Establishing evidence-based guidelines for blood pressure management and anti-thrombotic medications will be important for both clinicians and researchers in the future.

FUNDING INFORMATION
None.

CONFLICT OF INTEREST STATEMENT
I, Akshit Chitkara, on behalf of all coauthors, declare that I have no financial or personal conflicts of interest related to this work or its publication.My involvement in this work was solely for academic or professional purposes.I did not receive any financial or other benefits that could be perceived as a conflict of interest.Furthermore, I affirm that I have disclosed any financial or personal relationships with individuals or organizations that could have influenced my work on this project.If any potential conflicts of interest arise in the future, I will promptly disclose them and take appropriate actions to mitigate any potential influence they may have on the work.
BVZ has been used to treat more than 1.56 million patients with mCRC worldwide.19Previous RCTs have shown that BVZ, in combination with a standard cytotoxic regime, is safe to administer and without any significant risk of F I G U R E 2 Forest plot showing odds of HTN in the BVZ group versus chemo-only group.BVZ, bevacizumab; HTN, hypertension.

F I G U R E 3
Forest plot showing odds of HTN in the BVZ group versus FOLFOX-only subgroup.BVZ, bevacizumab; FOLFOX, fluorouracil, leucovorin, and oxaliplatin; HTN, hypertension.F I G U R E 4 Forest plot showing odds of thromboembolism in the BVZ group versus chemo-only group.BVZ, bevacizumab.

F I G U R E 5
Forest plot showing odds of ATE in the BVZ group versus chemo-only group.ATE, arterial thromboembolism; BVZ, bevacizumab.F I G U R E 6 Forest plot showing odds of VTE in the BVZ group versus chemo-only group.BVZ, bevacizumab.

F I G U R E 7
Funnel plot showing OR of HTN in the BVZ group versus chemo-only group with asymmetrical distribution.BVZ, bevacizumab; HTN, hypertension.F I G U R E 8 Funnel plot showing OR of HTN in the BVZ group versus FOLFOX-only group with asymmetrical distribution.BVZ, bevacizumab; FOLFOX, fluorouracil, leucovorin, and oxaliplatin; HTN, hypertension.

F I G U R E 9
Funnel plot showing OR of thromboembolism in the BVZ group versus chemo-only group with symmetrical distribution.BVZ, bevacizumab.F I G U R E 1 0 Funnel plot showing OR of ATE in the BVZ group versus chemo-only group with asymmetrical distribution.ATE, arterial thromboembolism; BVZ, bevacizumab. of hypertensive patients noted compared with the clinical criteria for the diagnosis of HTN (140/90 mmHg).

F I G U R E 1 1
Funnel plot showing OR of VTE in the BVZ group versus chemo-only group with asymmetrical distribution.BVZ, bevacizumab.