Hypertension Induced by Combination Therapy of Cancer: A Systematic Review and Meta-Analysis of Global Clinical Trials

Background: Nowadays, due to the limitation of single therapy, combination therapy for cancer treatments has become important strategy. With the advancement of research on cardiotoxicities induced by anti-cancer treatment, among which cancer treatment-induced hypertension is the most frequent case. However, due to the small sample size and the absence of comparison (single-arm study alone), these studies have limitations to produce a feasible conclusion. Therefore, it is necessary to carry out a meta-analysis focusing on hypertension caused by cancer combination therapy. Methods: We systematically searched PubMed, Embase, Cochrane Library, Web of Science, and CNKI, from database inception to November 31, 2020, with randomized controlled trials (RCTs) associated with hypertension induced by cancer combination drugs. The main endpoint of which was to assess the difference in the incidence of hypertension in cancer patients with monotherapy or combination therapy. We calculated the corresponding 95% confidence interval (95% CIs) according to the random effect model and evaluated the heterogeneity between different groups. Results: According to the preset specific inclusion and exclusion criteria, a total of 23 eligible RCTs have been included in the present meta-analysis, including 6,241 patients (Among them, 2872 patients were the control group and 3369 patients were the experimental group). The results showed that cancer patients with combination therapy led to a higher risk of hypertension (All-grade: RR 2.85, 95% CI 2.52∼3.22; 1∼2 grade: RR 2.43, 95% CI 2.10∼2.81; 3∼4 grade: RR 4.37, 95% CI 3.33∼5.72). Furthermore, compared with the control group who received or did not receive a placebo, there was a higher risk of grade 3-4 hypertension caused by cancer combination treatment. Conclusion: The present meta-analysis carries out a comprehensive analysis on the risk of patients suffering from hypertension in the process of multiple cancer combination therapies. Findings in our study support that the risk of hypertension may increase significantly in cancer patients with multiple cancer combination therapies. The outcomes of this meta-analysis may provide a reference value for clinical practice and may supply insights in reducing the incidence of hypertension caused by cancer combined treatment.

Background: Nowadays, due to the limitation of single therapy, combination therapy for cancer treatments has become important strategy. With the advancement of research on cardiotoxicities induced by anti-cancer treatment, among which cancer treatment-induced hypertension is the most frequent case. However, due to the small sample size and the absence of comparison (single-arm study alone), these studies have limitations to produce a feasible conclusion. Therefore, it is necessary to carry out a meta-analysis focusing on hypertension caused by cancer combination therapy.
Methods: We systematically searched PubMed, Embase, Cochrane Library, Web of Science, and CNKI, from database inception to November 31, 2020, with randomized controlled trials (RCTs) associated with hypertension induced by cancer combination drugs. The main endpoint of which was to assess the difference in the incidence of hypertension in cancer patients with monotherapy or combination therapy. We calculated the corresponding 95% confidence interval (95% CIs) according to the random effect model and evaluated the heterogeneity between different groups.
Results: According to the preset specific inclusion and exclusion criteria, a total of 23 eligible RCTs have been included in the present meta-analysis, including 6,241 patients (Among them, 2872 patients were the control group and 3369 patients were the experimental group). The results showed that cancer patients with combination therapy led to a higher risk of hypertension (All-grade: RR 2.85, 95% CI 2.52∼3.22; 1∼2 grade: RR 2.43, 95% CI 2.10∼2.81; 3∼4 grade: RR 4.37, 95% CI 3.33∼5.72). Furthermore, compared with the control group who received or did not receive a placebo, there was a higher risk of grade 3-4 hypertension caused by cancer combination treatment.

INTRODUCTION
Hypertension has been recognized as the most common comorbidity among various types of cancers, which directly affects the prognosis of cancer patients, and is one of the high-risk factors for cancer survivors suffering from the comorbidity of heart diseases (Jain and Townsend, 2007). In the early stage of diagnosis, there is generally a similar probability of developing hypertension. However, with different cancer treatment patterns, patients may experience significantly altered incidence of hypertension, especially those receiving chemotherapy, which can reach 38% (Piccirillo et al., 2004;Maitland et al., 2010). In addition, novel cancer therapies, such as targeted therapy, which is a type of cancer treatment that targets proteins controlling cancer cells' growth, division, and spreading, are also associated with the incidence of hypertension. Cardio-Oncology is an evolving discipline which aims to analyze the relationship between cancer treatment and cardiotoxicity (Lenneman et al., 2016;Barac, 2020). Cardiovascular toxicity in cancer treatment refers to the occurrence of cardiovascular disease during the disturbance or elimination of cancer cells in patients in vivo. Significantly, cardiovascular disease is the second leading cause of the morbidity and mortality of cancer survivors. According to previous studies, the probability of all-grade hypertension is between 15 and 67% during the treatment by using small molecule vascular endothelial growth factor tyrosinase inhibitors (e.g., sunitinib, sorafenib, pazopanib, etc.), and the rate would be higher with the use of inhibitors with higher efficiency (e.g., axitinib) (Brinda et al., 2016). The incidence of hypertension induced by tyrosinase inhibitors ranges from 5 to 80% in a dose-dependent manner (Agarwal et al., 2018). In addition, some patients may have a history of hypertension before the diagnosis of cancer. However, some patients develop hypertension due to anti-cancer treatment, and hypertension may be the direct result of cancer treatment under this circumstance.
The progress of cancer treatment has promoted the development of multiple new treatment strategies. Combination therapies means combining two or more therapies for cancer patients and the effectiveness may be excellent than single therapy. However, most programs will be accompanied by a series of cardiovascular adverse reactions, especially the existed high correlation of some new drugs with hypertension. In addition, the use of some chemotherapy drugs can also induce hypertension.
Generally, angiogenesis is a necessary process of tumorigenesis, growth, and metastasis. Vascular endothelial growth factor (VEGF) is an angiogenic growth factor. Angiogenesis inhibitor is a classic drug highly associated with the occurrence of hypertension (Hamnvik et al., 2015), primarily including monoclonal antibodies and small-molecule drugs. It has been documented that the proposed highly specific drugs are important inhibitors of angiogenesis, which play a role by blocking the signaling pathways necessary for angiogenesis, such as blocking Vascular Endothelial Growth Factor Receptor (VEGFR), Epidermal Growth Factor Receptor (EGFR), basic Fibroblast Growth Factor (bFGF), Platelet-derived Growth Factor Receptor (PDGFR), etc. (Folkman, 2007). To be specific, VEGF is the main growth factor that controls angiogenesis. Epidermal growth factor (EGF) is responsible for differentiation and apoptosis. bFGF can regulate the proliferation and differentiation of specific types of cells and has an effective effect on angiogenesis. Platelet-derived growth factor (PDGF) involves significantly cell growth, cell division, and angiogenesis (Wilkins et al., 2014;Agarwal et al., 2018).
With the emergence of various novel approaches to cancer treatment, the survival of cancer patients is becoming higher, which, however, is accompanied by an increasingly more obvious change in cardiotoxicity. Given the differences in cancer tissue types, therapeutic drugs, and drug doses, a systematic review and meta-analysis were carried out on hypertension caused by cancer treatment (Said et al., 2017), which aimed to clarify the incidence and risk of hypertension in cancer patients treated with combination therapy. At present, there is incomplete knowledge of hypertension caused by cancer combination therapy. Besides, there is few systematic reviews or metaanalyses in this aspect based on the comprehensive analysis of previous literature. Accordingly, through comprehensive literature analysis, it is expected to analyze and elaborate the risk factors of hypertension caused by cancer combination therapy, to provide a certain reference value for clinical treatment.

METHODS
The present systematic review and meta-analysis were conducted following PRISMA guidelines (Moher et al., 2009). The protocol has been registered in PROSPERO with the registration number CRD42021220923.

Data Sources and Searches
A comprehensive literature search was made in databases [PubMed, embase, Cochrane Library, Web of Science, and CNKI] since November 31, 2020, to identify all articles related to the subject. In addition to the above databases, the clinical trial registration website (https://clinicaltrials.gov/) was searched to obtain information about registered prospective trials.
The keywords used in PubMed were listed as follows:  (9) The final selected literatures were checked and reviewed separately to include the latest and most complete clinical trial reports in the case of repeated publications. All the search results were incorporated into the management tool of Endnote.

Study Selection and Data Extraction
The major objective of our study was to determine the incidence of hypertension associated with combination therapy for cancer and to establish a relationship between combination therapy and the risk of hypertension. Therefore, eligible studies were those evaluating the combination of drugs with hypertension induced in cancer patients. Phase I trial was excluded considering the multi-dose level and limited sample size. In addition, phase II, III, and IV randomized controlled trials (RCTs) in combination therapy were enrolled in the analysis compared with those without combination therapy.
The eligible studies met the inclusion criteria: 1) Phase II, III, and IV trials involving cancer patients; 2) RCTs for cancer treatment; 3) Intervention group: combination therapy (including targeted therapy and chemotherapy); 4) Control group: monotherapy or placebo treatment; 5) Studies with available data on hypertension events or incidence and sample size.
The exclusion criteria: 1) Review articles 2) Not randomized control trial 3) Reports from same study sample 4) Not report associate with hypertension 5) Not report associate with cancer combination therapy 6) No usable data 7) No comparable trial 8) Republished literature Two investigators (G.X and Q.X) extracted data independently, and any disagreements between the two reviewers were resolved by consensus. Online studies before publication were also eligible, but not including reviews, Conference reviews, studies published only in abstract form, quality of life research, non-randomized trials, and studies that could not determine the toxicity of combination therapy. Data extraction covered author, year of publication, research institution, journal name, trial phase, cancer tissue type, combination therapy, number of patients, age of patients, administration schedule and drug dose, size of control group, number of patients with hypertension, with the data of hypertension at all grades extracted.

Data Synthesis and Analysis
Statistical analysis of this study was performed by using the Cochrane Review Manager (RevMan 5.3) software provided by the Cochrane Library Collaboration Network.
The proportion of patients with hypertension in each study was calculated by dividing the number of patients with hypertension caused by combination therapy extracted from eligible clinical trials by the total number of patients receiving combination therapy in each study. We refer to all levels of hypertension events as "All-grade," "1-2 grade" is combined the grade of 1 or 2 hypertension events, and "3-4 grade" which is the sum of the level of 3 or 4 hypertension events.
For each study enrolled in this analysis, the relative risk (RR) and 95% confidence interval (95% CI) of the incidence of events between the intervention group and the control group were calculated according to the number of reported events and sample size. The I2 index and Q-statistics were used to evaluate the heterogeneity among studies, among which the Q-test is widely used at present (Zintzaras and Ioannidis, 2005). p < 0.05 of the Q-test indicated the existence of heterogeneity (Zhang et al., 2019), and p < 0.05 meant the existence of statistical significance. If p > 0.05, the results of the independent studies might be homogeneous, suggesting the use of the fixed-effect model; On the contrary, the random-effect model should be used and/or consider the clinical suitability of combination therapy when there was heterogeneity with p < 0.05. I 2 can quantify the heterogeneity among studies, which is calculated generally based on χ 2 test. It describes the percentage of variation among studies in total variation, which may indicate a higher heterogeneity with the increase of the value of I 2 (Huedo- Medina et al., 2006). I 2 > 25, 50, and 75% suggest that there may be low, moderate, and high heterogeneity among studies. Besides, it is generally believed that there is substantial heterogeneity when I 2 > 50%.
In all eligible studies, the average age of patients ranged from 18 to 89 years old. Among the eligible research articles, papers published in the United States accounted for the majority, with 8 articles, followed by China with 5 articles, Canada with 3 articles, Japan with 3 articles, Britain with 1 article, France with 1 article, Germany with 1 article and Switzerland with 1 article. Meanwhile, 8 articles were published in "Journal of Clinical  Oncology,", 4 in "The Lancet Oncology,", and 3 in "European Journal of Cancer."

Evaluation of Included Studies
The Modified Jadad Scores scale (Jadad et al., 1996) was used to evaluate the quality of the 23 eligible articles. Following the evaluation based on the Randomization, Concealment of Allocation, Double Blinding, Withdrawals, and Dropouts, etc., there were 15 articles in 7 points, 5 articles in 5 points, 3 articles in 4 points, and 1 article in 3 points, as shown in Table 3.

Relative Risk of Hypertension
A total of 3,369 patients received cancer combination therapy, as well as 2,872 patients received cancer single therapy and/or placebo, which was available for comparative analysis. The incidence of grade 1-2 hypertension events ranged from 0 to 75%, and cediranib combined with mFOLFOX6 for the treatment of Colorectal Cancer had the highest probability of inducing hypertension (Kato et al., 2012). However, no events were observed in grade 1-2 hypertension in one trial (Sinn et al., 2020). Using the random-effect model, the RR in all patients developing grade 1-2 hypertension was 2.43 [95% CI 2.10-2.81, p < 0.001, Figure 2A]. Furthermore, the probability of grade 3-4 hypertension in all patients ranged from 0 to 40.9%, among which cediranib combined with Olaparib in treating Ovarian Cancer showed the highest probability of developing hypertension events (Liu et al., 2014;Liu et al., 2019). However, no grade 3-4 hypertension events were observed in the use of Oxaliplatin combined with oxaliplatin and Tiggio in the treatment of Gastric Cancer (Yan et al., 2017). Based on the random-effect model, the RR in all patients developing grade 3-4 hypertension was 4.37 [95% CI 3.33-5.72, p < 0.001, Figure 2B]. In addition, the incidence of all-grade hypertension ranged from 5.26 to 85.71%, and the highest incidence of hypertension was observed in the use of cediranib combined with mFOLFOX6 for the treatment of Colorectal Cancer (Kato et al., 2012). In the random-effect model, the RR in all patients developing grade 3-4 hypertension was 2.85 [95% CI 2.52-3.22, p < 0.001, Figure 2C].

Heterogeneity and Bias of Included Studies
As presented in Figure 2, there was moderate heterogeneity in grade 1-2 hypertension (I 2 55%, p < 0.001), low heterogeneity in grade 3-4 hypertension (I 2 5%, p 0.39), and moderate heterogeneity in all grades of hypertension (I 2 68%, p < 0.001) caused by cancer combination therapy, with the presence of statistical significance. Using the risk-of-bias assessment tool (Higgins et al., 2011), the results of the Cochrane risk-of-bias assessment of the enrolled 23 RCTs are shown in Table 4 and Figures 7-9 showed that the funnel plot indicated evidence of heterogeneities and publication bias in the studies included in the meta-analysis with scatters beyond 95% CI and asymmetry display (p < 0.00001).

DISCUSSION
To our knowledge, the present meta-analysis for the first time evaluated the potential risk of hypertension in cancer patients treated with combination therapy. As a "silent killer,", hypertension has been reported to have a doubled prevalence in the past 40 years, with 7.6 million people dying of  hypertension annually in the world (Arima et al., 2011). Despite no significant direct influence, long-term hypertension may result in damage of the heart and blood vessels, and cerebral artery vasospasm as well.
In the field of Cardio-Oncology, cancer combination therapy may produce the effective outcome in killing cancer cells and controlling the deterioration of cancer. Nevertheless, there is an inevitable adverse effect of heart disease, especially the occurrence of hypertension. In this regard, there is an urgent need for medical staff to adjust the therapeutic schemes of patients, timely prevent and alleviate side effects during and after cancer treatment, to ensure the life safety of patients.
Current anti-hypertensive therapeutics included Selective α1 adrenoceptor antagonist, non-selective α1 and α2-antagonists, β-adrenoceptor antagonists, angiotensin II receptor blockers, calcium channel blockers, ACE inhibitors, renin inhibitors, direct vasodilators, loop diuretics (Kumar et al., 2020). However, we should pay more attention to the related complications which they are accompanied, such as organ damage, hypotension and so on (Kumar et al., 2020).
In our meta-analysis, based on the collection of all relevant data from retrospective clinical trials, the final objects of study were a total of 23 clinical trials involving 6,241 patients. The combination therapy of cancer patients resulted in a higher risk of developing hypertension (All-grade: RR 2.85, 95% CI 2.52-3.22; 1-2 grade: RR 2.43, 95% CI 2.10-2.81; 3-4 grade: RR 4.37, 95% CI 3.33-5.72). According to the results, the risk of grade 3-4 hypertension induced by cancer combination therapy was higher than that of the control group with or without placebo therapy.
There may exist different mechanisms of increase in blood pressure under different anti-cancer therapeutic schemes. The mechanism of elevated blood pressure by using anti-cancer drugs may exhibit a direct association with its anti-cancer mechanism. The mechanism of hypertension induced by cancer combination therapy may be explained by the following reasons. To be specific, monoclonal antibodies (for example, bevacizumab) may reduce the number of capillaries in microcirculation, competitively inhibit the binding of EGFR with other ligands, and block the interaction between VEGF and endothelial cell surface receptors, resulting in inhibit the signal pathway of VEGF, reduce the activity of endothelial nitric oxide synthase and the production of NO and PGI 2 by vascular endothelial cells, decrease vascular permeability and vasodilation, increased peripheral vascular resistance and blood flow, and finally lead to hypertension (Chen et al., 2011;Mayer et al., 2011;Mourad and Levy, 2011;Campia et al., 2019). Besides, it has been reported that reducing the activity of eNOS will lead to expression of uncoupling protein of eNOS, produces a large amount of reactive oxygen species and then decrease the level of NO (Kumar et al., 2020). Meanwhile, NO is involved in maintaining the steady state of sodium ions and participating in tubuloglomerular feedback to regulate renal blood flow and glomerular filtration, which can increase systemic blood pressure (Lankhorst et al., 2017). Another possible mechanism of hypertension caused by inhibiting other VEGF pathways is that angiogenesis inhibitors may reduce the number of blood vessels and lead to hypertension owing to the thinning of peripheral microvessels ). In addition, additional research also reveals that the increase in blood pressure may be related to the inhibition of VEGFR-2 (Kamba and McDonald, 2007). Also, Small molecular targeted drugs (such as sunitinib) can upregulate endothelin-1, increase salt sensitivity, and further increase in blood pressure owing to thrombotic glomerular injury (Kidoguchi et al., 2021). In addition, some novel targeted drugs (e.g., brutinib) may increase the risk of hypertension by inhibiting PI3K/Akt or reducing the level of NO (Dickerson et al., 2019). (Figure 10) With respect to the above, there is necessary to adopt targeted treatment of hypertension. Before the treatment of cancer patients, it is recommended to adopt a comprehensive risk assessment, including blood pressure measurement and examination of known risk factors. For cancer patients with existed cardiovascular diseases, it is necessary to consider carefully whether to use anti-cancer drugs that may lead to cardiotoxicity or not. In the field of Cardio-Oncology, further consideration of the overall health status of patients is required for doctors to make a prudent decision in patients with a high risk of hypertension and those with hypertension prior to the use of anti-cancer drugs. Moreover, in case of poor control of cancer development by monotherapy, the better therapeutic outcome may be produced by combination therapy, However, it should be noted that combination therapy may also lead to a higher risk of hypertension.
So far, there is still no systematic analysis of hypertension caused by cancer combination therapy. Data in our study fully supports that cancer combination therapy has a high risk of inducing hypertension. Findings in this meta-analysis suggest that much attention shall be paid constantly to the adverse reactions of combined use of drugs, with in-time prevention required simultaneously. However, there are limitations in this study. For example, due to the absence of experimental data, relevant experiments are needed in the future to fully clarify the pathophysiological basis of hypertension caused by the combination of drugs and to increase the credibility of the results of our study.

CONCLUSION
The accuracy of meta-analysis research is high, but there is also a certain degree of publication bias, and risk of bias is low. It is worth mentioned that the reliability of meta-analysis results as well as the suitability in clinical practice might still requires critical thinking and objective judgments.
To sum up, the present meta-analysis carries out a comprehensive analysis on the risk of patients suffering from hypertension in the process of multiple cancer combination therapies. Findings in our study support that the risk of hypertension may increase significantly in cancer patients with multiple cancer combination therapies. The outcomes of this meta-analysis may provide a reference value for clinical practice and may supply insights in