Cardiovascular toxicities of molecularly targeted antiangiogenic agents

INTRODUCTION — Antiangiogenic agents (systemic medications that inhibit angiogenesis) are used in cancer therapy because angiogenesis is required for tumor growth [1].

Several classes of antiangiogenic agents are available that block the vascular endothelial growth factor (VEGF) pathway (figure 1), including:

●Ligand inhibitors – These bind to and inhibit ligand binding to the VEGF receptor, thereby preventing activation of the receptor. Examples include bevacizumab, ramucirumab, and aflibercept.

●Receptor tyrosine kinase inhibitors (TKIs) – These block the enzymatic activity of the intracellular domain of the VEGF receptor (VEGFR). Examples include sunitinib, sorafenib, pazopanib, vandetanib, cabozantinib, axitinib, ponatinib, lenvatinib, regorafenib, tivozanib, and fruquintinib.

Antiangiogenic agents are associated with a wide spectrum of toxicities, which may be fatal in rare cases [2,3]. While some adverse effects are shared with chemotherapy, many are unique to antiangiogenic agents. Toxicities of antiangiogenic agents may be specific to the drug class or to the agent itself.

This topic will discuss the cardiovascular toxicities of antiangiogenic agents.

Separate topics discuss:

●Non-cardiovascular toxicities of antiangiogenic agents – (See "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents" and "Multiple myeloma: Prevention of venous thromboembolism".)

●Kidney toxicities of antiangiogenic agents – (See "Nephrotoxicity of molecularly targeted agents and immunotherapy", section on 'Antiangiogenic agents'.)

●Infusion reactions to monoclonal antibodies – (See "Infusion-related reactions to therapeutic monoclonal antibodies used for cancer therapy".)

●The basic science of angiogenesis inhibitors – (See "Overview of angiogenesis inhibitors".)

PATHOPHYSIOLOGY OF CARDIOVASCULAR TOXICITIES — The cardiovascular toxicities of angiogenesis agents appear to be generated by negative effects on multiple levels, including myocardial cells, endothelial cells, and pericytes [4]. Thus, the pathophysiology of the cardiovascular toxicities of vascular endothelial growth factor (VEGF) inhibitors is multifactorial and impacted by preexisting comorbidities that reduce vascular reserve [5].

Many of the data are derived from animal models:

●In animal models, sunitinib increases the vasoconstriction mediated by endothelins, resulting in hypertension and increased cardiac afterload [6]. Reduced myocardial perfusion from this and other VEGF inhibitors can reduce myocardial contractility and lead to both ischemia and heart failure.

●In animal models, sorafenib induces myocyte necrosis, with surviving myocytes undergoing hypertrophy [7]. Moreover, inhibition of KIT positive stem cells by sorafenib exacerbates damage by decreasing cardiac repair.

●Sunitinib can potentially compromise myocardial energy metabolism by enhancing glycolysis and reducing oxidative phosphorylation [8].

●Animal models have demonstrated acceleration of atherosclerosis from a VEGF receptor inhibitor, potentially explained by disruption of endothelial cell homeostasis by a dose-dependent increase in mitochondrial superoxide generation, resulting in reduced nitric oxide availability and decreased proliferation [9]. The disruption of VEGF-dependent events may lead to downstream endothelial cell apoptosis, plaque erosions, and acute arterial thrombotic events [10].

●A preclinical study described sunitinib-induced pericyte depletion and coronary microvascular dysfunction mediated by targeting the platelet-derived growth factor receptor (PDGFR) [11].

●In patients treated with bevacizumab, systemic capillary rarefaction has been observed, which may increase afterload [12]. The afterload increase, coupled with reduced cardiac contractility and myocyte hypertrophy, leads to increased oxygen demand and upregulates the VEGF axis.

●As anticipated, preexisting coronary artery disease and hypertension have been identified as predictors for the development of heart failure with sunitinib [13,14].

●Other indirect mechanisms that may contribute to cardiovascular toxicities include hypothyroidism resulting from VEGF receptor inhibitors, which may reduce cardiac contractility and preload. Moreover, concurrent or recent use of cardiotoxic chemotherapeutic agents may exacerbate the cardiovascular toxicities of angiogenesis inhibitors.

RISK OF FATALITY — Meta-analyses have demonstrated a small risk of fatal adverse events (approximately 1.5 to 2.5 percent, relative risk [RR] 1.5 to 2.2) with both antiangiogenic tyrosine kinase inhibitors (TKIs) and bevacizumab [2,3,15]. In one analysis, bevacizumab was associated with an increased risk of fatal events when used in combination with taxanes or platinum agents (RR 3.49) but not in combination with other agents (RR 0.85) [2].

In two meta-analyses, hemorrhage was the most common fatal adverse event with both classes of agents; other causes of treatment-related death were cardiac, gastrointestinal tract perforation, hepatic dysfunction, infection, and cerebrovascular events [2,3]. (See "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Bleeding'.)

In another meta-analysis examining fatal events with antiangiogenic TKIs, fatalities from heart failure, pulmonary emboli, hepatic failure, intestinal perforation, and pneumonia/respiratory failure were numerically higher on the TKI treatment arms [15]. In this study, the increased RR for death with the TKIs was statistically significant for patients with renal cell carcinoma (RCC), but not for those with lung cancer. However, the increased risk seen in patients treated for RCC may be ascribed partly to increased exposure time to the antiangiogenic TKI in patients with RCC relative to those with lung cancer.

CLASS SIDE EFFECTS OF ALL VEGF INHIBITORS

Hypertension — Hypertension is a class side effect that is associated with all antiangiogenic agents. Vascular endothelial growth factor (VEGF) plays a key role in the maintenance of vascular homeostasis via mediation of the production of the vasodilator nitric oxide, and decreased vascular resistance through the generation of new blood vessels [16-20]. Both of these influences are associated with decreases in blood pressure. Thus, it is not surprising that inhibition of VEGF signaling might be associated with hypertension [21,22]. Because of this, all trials evaluating inhibitors of angiogenesis have restricted eligibility to patients with controlled blood pressure at baseline.

Incidence and characteristics — The following key studies have addressed the incidence of hypertension in patients treated with angiogenesis inhibitors:

●In a systematic review and meta-analysis including 77 phase III trials, angiogenesis inhibitors were associated with a higher risk of hypertension (odds ratio [OR] 5.28, 95% CI 4.53-6.15) and severe hypertension (OR 5.59, 95% CI 4.67-6.69) [23]. The subgroup analyses examining VEGF ligand inhibitors versus small molecule agents showed no significant differences.

●In a meta-analysis of 29,252 patients from 71 randomized controlled trials, hypertension (relative risk [RR] 3.78, 95% CI 3.15-4.54) was reported with VEGF tyrosine kinase inhibitors (TKIs) [24].

●In a meta-analysis of 72 randomized controlled trials including 30,013 patients, the incidence of high-grade and all-grade hypertensive events with VEGF receptor (VEGFR) TKIs was 23.0 percent (95% CI 20.1-26.0 percent) and 4.4 percent (95% CI 3.7-5.0 percent), respectively [25]. VEGFR TKIs increased the risk of developing high-grade (RR 4.60, 95% CI 3.92-5.40) and all-grade (RR 3.85, 95% CI 3.37-4.40) hypertension.

●In a meta-analysis of 12,949 patients treated with or without bevacizumab for advanced solid tumors, the RR of developing "significantly raised" blood pressure (defined as more than one drug needed for treatment, more intensive treatment needed than used previously, or life-threatening consequences such as hypertensive crisis; or grade 3 or 4 hypertension according to the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events CTCAE (table 1)) among patients receiving bevacizumab was 5.38 (95% CI 3.63-7.97) [26]. Among patients receiving bevacizumab, the overall incidence of all raised blood pressure events was 24 percent (95% CI 20-29 percent), while the incidence of significantly raised blood pressure was 8 percent (95% CI 6-10 percent). Risk was dose dependent; the RRs of severe hypertension at doses of 5 and 2.5 mg/kg per week were 7.17 (95% CI 3.91-13.13) and 4.11 (95% CI 2.49-6.78), respectively.

●In a 2017 meta-analysis of individual patient safety data from six randomized placebo-controlled trials, rates of severe hypertension with ramucirumab were 9 percent, versus 2.5 percent for placebo [27]. For all grades of hypertension, the risk was 21 versus 9 percent.

●Another meta-analysis analyzed the incidence of hypertension in 13 prospective studies with 4999 patients who were treated with sunitinib for renal cell carcinoma (RCC) or other malignancies [28]. Among patients receiving sunitinib, the overall incidence of all-grade hypertension was 22 percent, with severe hypertension in 7 percent.

●Almost identical findings were noted in a systematic review that analyzed the incidence of hypertension in 4599 patients treated with sorafenib in nine prospective studies [29].

●Among the VEGFR TKIs, it is unclear if any one agent causes hypertension with greater frequency; in one meta-analysis, pazopanib was associated with higher rates of all-grade hypertension than sorafenib or sunitinib (36 versus 23 and 22 percent, respectively), but similar rates of severe hypertension (6.5 versus 5.7 and 6.8 percent, respectively) [30]. In phase III trials directly comparing TKIs in advanced RCC, pazopanib and sunitinib were associated with similar rates of all-grade hypertension (46 percent, 41 percent), and axitinib was associated with somewhat more hypertension compared with sorafenib (40 versus 29 percent) [31,32].

On the other hand, higher rates of both all grade (68 percent) and severe (42 percent) hypertension are reported with lenvatinib [33].

●The incidence of severe hypertension may be higher with aflibercept than with other VEGF-targeted therapies, but no direct comparisons of aflibercept with other VEGF inhibitors exist (19.1 percent for aflibercept compared with 1.5 percent for placebo when both were administered in combination with chemotherapy in patients with colorectal cancer [34]).

Preexisting hypertension, age ≥60 years, and body mass index (BMI) ≥25 kg/m² have been independently associated with a higher risk for anti-VEGF therapy-induced blood pressure elevation [35]. Dramatic increases in both diastolic (up to 27 mmHg [36]) and systolic (up to 29 mmHg [36]) blood pressure may be seen as early as the first week of therapy [36,37]. It is not possible to predict which patients will have this magnitude of blood pressure elevation [38]. There is some evidence that genetic predisposition may play a role, but no reliable markers are available to predict risk that can be used in clinical practice [39,40].

Because serious adverse events have been associated with unmanaged hypertension, all patients receiving therapy with an angiogenesis inhibitor should have their blood pressure actively monitored during treatment, with more frequent measurement in the first several weeks of therapy. (See 'Monitoring and management' below.)

Association with antitumor efficacy — Several reports note an association between improved antitumor efficacy and the development of systolic or diastolic hypertension during therapy with both bevacizumab and the antiangiogenic TKIs [41-49]; however, this has not been seen in all studies [50-52]. The following represents the range of findings:

●In a combined series analyzing 544 patients enrolled on four prospective trials of sunitinib for advanced RCC, antitumor efficacy was significantly better in patients with one or more systolic blood pressure readings ≥140 mmHg (median overall survival, median progression-free survival, and objective response rates were 30.9 versus 7.2 months, 12.5 versus 2.5 months, and 55 versus 9 percent, respectively) [47]. Similar results were seen using a criterion of diastolic blood pressure ≥90 mmHg. Multivariate analysis confirmed that treatment with sunitinib was an independent factor for survival after incorporating other known risk factors (hazard ratio [HR] 0.28, 95% CI 0.22-0.37). Secondary analyses found that the improvement in efficacy was maintained in patients who were treated for hypertension or had a dose reduction of sunitinib [53].

●On the other hand, a lack of association between early hypertension and clinical benefit from bevacizumab was suggested in an analysis of seven randomized trials of bevacizumab versus no bevacizumab in patients with RCC, colorectal, breast, non-small cell lung, and pancreatic cancer [51]. In six of the seven studies, early hypertension (defined as an increase in systolic blood pressure of >20 mmHg or of diastolic blood pressure >10 mmHg within the first 60 days of treatment) was not predictive of clinical benefit from bevacizumab.

●Other data suggest that the apparent relationship between efficacy and toxicity may be related to increased drug exposure. In a report that systematically analyzed pharmacokinetic and pharmacodynamic data from six clinical studies of sunitinib, sunitinib dose intensity and cumulative weekly dose correlated with both improved clinical outcomes and maximum blood pressure [54].

Hence, further studies are warranted to validate hypertension as a predictive biomarker of efficacy in patients treated with angiogenesis inhibitors.

Monitoring and management — Although minimalist approaches to managing hypertension for patients with incurable diseases might be favored in some circumstances, management of comorbidities, including hypertension, may actually improve survival [55]. Furthermore, active control of hypertension should allow patients to tolerate the highest effective doses of therapy, for the longest period of time [38].

The Investigational Drug Steering Committee of the NCI formed a Cardiovascular Toxicities Panel, joining members of its Angiogenesis Task Force with experts in the management of hypertension in cancer patients to generate consensus recommendations for risk assessment, monitoring, and safe administration of angiogenesis inhibitors [38]. Their recommendations included the following four components (table 2A-B):

●Perform a pretreatment evaluation and screening, including formal risk assessment for potential cardiovascular complications.

●Identify and treat preexisting hypertension before using these agents.

●Actively monitor blood pressure during treatment, with more frequent measurement in the first several weeks of therapy.

●Manage blood pressure during therapy, with an ideal goal, when possible, of keeping blood pressure less than 130/80 mmHg for most patients, lower in those with specific preexisting cardiovascular risk factors, such as diabetes or chronic kidney disease. (See "Goal blood pressure in adults with hypertension".)

Continuous ambulatory blood pressure monitoring has been proposed for monitoring of blood pressure and management of hypertension in patients with advanced RCC receiving sunitinib [56]. However, this approach is not in widespread use, at least in the United States.

Patients who develop hypertension during treatment (defined as blood pressure ≥140/90 mmHg or a 20 mm increase in diastolic blood pressure over baseline) should be treated with antihypertensives. The choice of agent must take into consideration the severity of the hypertension and the urgency of blood pressure control. (See "Management of severe asymptomatic hypertension (hypertensive urgencies) in adults".)

In addition, several other considerations may influence the choice of antihypertensive therapy. As examples:

●Optimal anti-hypertensive agents in this context have not been defined. However, sorafenib and sunitinib undergo partial metabolism via cytochrome P450, a system inhibited by some antihypertensive agents (eg, verapamil, diltiazem) [21]. Therefore, these agents should probably be avoided in patients who develop hypertension while receiving sorafenib or sunitinib.

●Organ dysfunction and/or adverse effects on organ function may influence the choice and dose of the antihypertensive agent [57]. (See "Choice of drug therapy in primary (essential) hypertension".)

●For patients with renal cell cancer who develop hypertension while receiving treatment with an antiangiogenic agent, treatment with angiotensin system inhibitors (ASIs; eg, angiotensin converting enzyme inhibitors [ACEIs], angiotensin receptor blockers [ARBs]) may be preferred over other agents. (See "Antiangiogenic and molecularly targeted therapy for advanced or metastatic clear cell renal carcinoma", section on 'Toxicities of VEGF inhibition (hypertension)'.)

Arterial and venous thromboembolism — An increased risk for arterial thromboembolic events (ATEs) has been linked to the use of bevacizumab, aflibercept, and a number of the antiangiogenic TKIs [34,58]. As an example, in a systematic review and meta-analysis including 77 phase III trials, angiogenesis inhibitors were associated with a higher risk of arterial thromboembolism (OR 1.52, 95% CI 1.17-1.98), but subgroup analyses did not reveal differences between VEGF ligand inhibitors (antibodies or decoy receptors) and small molecule agents [23].

On the other hand, whether the risk of venous thromboembolic events (VTEs) is increased in these patients is uncertain; the data are mixed. Notably, patients with prior cardiovascular or venous thrombotic events within 6 to 12 months were not enrolled on clinical trials of these agents in a variety of malignancy types.

Pathophysiology — The pathophysiology underlying the increased risk of thromboembolism in patients treated with inhibitors of angiogenesis remains unresolved [59]. The main hypothesis is that perturbation of tumor-associated endothelial cells can switch the endothelium from a naturally anticoagulant surface to a prothrombotic surface, thus mediating the activation of systemic coagulation in cancer patients, who are already more susceptible to thromboembolism due to their underlying disease. (See "Cancer-associated hypercoagulable state: Causes and mechanisms".)

The VEGF pathway has been reported to protect and regulate endothelial cell function via pathways that inhibit apoptosis and inflammation. VEGF-induced nitric oxide production by endothelial cells is associated with several vascular protective effects, including inhibition of proliferation of vascular smooth muscle cells, antiplatelet actions, and inhibition of leucocyte adhesion [60,61]. As a consequence, endothelial cell dysfunction may expose pro-thrombotic phospholipids and underlying stroma [62].

Bevacizumab and aflibercept

Arterial thromboembolic events

Incidence — An increased incidence of potentially fatal ATEs (including transient ischemic attacks, strokes, angina, and myocardial infarction) was initially reported with the use of a bevacizumab-containing chemotherapy regimen for advanced colorectal cancer [63]. In randomized trials, the 4 to 5 percent incidence of serious ATEs in patients receiving bevacizumab in combination with a short-term infusional fluorouracil (FU)-based chemotherapy regimen represented an approximately two- to threefold higher incidence than in the control groups receiving the same chemotherapy regimen without bevacizumab [64].

The risk of ATEs in patients treated with bevacizumab plus chemotherapy compared with chemotherapy alone in a variety of malignancies has been further clarified by the findings from at least four meta-analyses:

●In an individual patient level meta-analysis of data from five trials in three different tumor entities, the use of bevacizumab with concomitant chemotherapy increased the risk of an ATE twofold (RR 2.0, 95% CI 1.05-3.75) compared with chemotherapy without bevacizumab [65]. Patients over 65 years of age (hazard ratio [HR] 2.1) and those with a prior history of ATEs (HR 4.18) appeared to be at highest risk for an ATE, particularly if they had both risk factors combined (HR 7.6).

●The risk may also vary according to the type of malignancy. In a trial-level meta-analyses of over 13,000 patients treated on 20 randomized trials, the RR increase for ATEs in patients receiving bevacizumab for a variety of advanced malignancies was 1.46 (95% CI 1.11-1.93); this translated into an incidence of 2.6 percent (95% CI 2-3.5) [66]. The highest ATE incidence was seen in patients treated for mCRC (3.2 percent, 95% CI 1.9-5.4), and the lowest was in patients treated for breast cancer (0.7, 95% CI 0.1-3.6), a diagnosis for which bevacizumab is no longer approved. Incidence rates for patients treated for non-small cell lung cancer (NSCLC) and RCC were 2.5 percent (95% CI 1.8-3.7) and 2.3 percent (95% CI 1.4-3.7), respectively.

●A dose-response effect has been shown in only one [67] of three trial-level meta-analyses [66-68].

Fewer data are available for aflibercept. ATEs were not observed in phase I and II trials [69,70]. In a phase III trial of chemotherapy with and without aflibercept in patients with mCRC, the rate of all-grade ATEs in the aflibercept group was 2.6 versus 1.5 percent with chemotherapy alone, but the rate of a grade 3 or 4 event was 1.8 versus 0.5 percent [34].

Management — We do not consider that a prior history of an ATE represents an absolute contraindication to use of antiangiogenic agents. We suggest caution and maintaining a low index of suspicion for drug-related ATEs in patients with a prior history of thromboembolic events. A discussion of the use of bevacizumab in older adult patients who have a history of an ATE within the past 6 to 12 months can be found elsewhere. (See "Therapy for metastatic colorectal cancer in older adult patients and those with a poor performance status", section on 'Bevacizumab and biosimilars'.)

The United States Prescribing Information indicates that bevacizumab should be discontinued for any severe ATE, but do not define "severe" [71]. The labeling information for aflibercept recommends discontinuation of the drug in patients who experience an ATE [72]. We discontinue both drugs for all grade 3 or higher (table 3) new or worsened arterial thrombotic events during therapy, unless the clinical benefits considerably outweigh the risks [73]. Continuation with concurrent anticoagulation might be considered in selected patients who are clearly benefiting from antiangiogenic therapy and who are willing to accept the hypothetical risk from continued use of bevacizumab in this setting.

Venous thromboembolism

Incidence — The use of bevacizumab has been inconsistently associated with an increased risk of VTEs in cancer patients [65,74,75]:

●Two individual patient level meta-analyses failed to identify a significant increase in risk of VTE in patients treated with bevacizumab plus chemotherapy versus chemotherapy alone [65,75]. In one of these meta-analyses (n = 1745 patients), the HR for VTE was 0.89, and in the second study (n = 6055 patients), the OR was 1.14, but neither result was statistically significant.

●On the other hand, a significant risk of VTE in patients treated with bevacizumab was shown in two other trial-level meta-analyses [67,74]. In one analysis of 7956 patients, the risk was significantly elevated; RR was 1.33, but it disappeared when examining these events per unit of time [74]. In the second analysis of 20,050 patients, the risk for venous adverse events with bevacizumab was 1.29 (95% CI 1.12-1.47) [67].

Thus, the association between bevacizumab and VTE remains uncertain.

The available data on aflibercept are limited. In a phase III trial comparing chemotherapy with and without aflibercept in patients with mCRC, the risk of all-grade VTE was 9.3 versus 7.3 percent with chemotherapy alone, and for grade 3 or 4 VTE (table 4), it was 7.8 versus 6.2 percent [34].

Prevention and management — There is not enough information available to recommend for or against the use of routine anticoagulation to prevent VTE in ambulatory cancer patients receiving, or about to receive, therapy with bevacizumab or aflibercept. We agree with the Clinical Practice Guidelines in Oncology from the National Comprehensive Cancer Network (NCCN) [76], which suggest stratifying recommendations for aspirin or anticoagulant prophylaxis in patients with cancer according to the presence or absence of risk factors for VTE. The NCCN does not consider the use of angiogenesis inhibitors to represent a risk factor for VTE in cancer patients, and we largely agree with this. (See "Risk and prevention of venous thromboembolism in adults with cancer", section on 'Primary prevention'.)

Prompt diagnosis and management is essential for patients who develop any grade of VTE during treatment (table 4). Issues related to diagnosis and treatment of VTE in patients with malignancy are presented elsewhere. (See "Epidemiology and pathogenesis of acute pulmonary embolism in adults" and "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)

The United States Prescribing Information gives no recommendations for management of bevacizumab or aflibercept in patients who develop VTE while receiving the drug [71,72]. The decision to discontinue the angiogenesis inhibitor in such cases must be individualized. Continuation of bevacizumab or aflibercept with concurrent anticoagulation is a reasonable approach for patients who are clearly benefiting from the drug.

Angiogenesis inhibitors can be associated with an increased risk of hemorrhage, raising concern about the safety of therapeutic anticoagulation in patients who are being treated with these agents. (See "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Bleeding'.)

However, the concurrent administration of bevacizumab and therapeutic doses of warfarin appeared to be safe in a large retrospective pooled analysis (risk of bleeding 1.9 versus 1.2 percent in bevacizumab and control patients, respectively) [75]. Moreover, the majority of bleeds were minor epistaxis episodes, and the rates of severe (grade ≥3) bleeds were similar and infrequent in both groups (0.2 percent).

A similarly low rate of bleeding events with continued bevacizumab and full-dose anticoagulation was shown in another analysis of three studies, two conducted in patients with mCRC and one in advanced NSCLC [77]. There are no data on this approach for patients receiving aflibercept.

Low molecular weight heparins are generally preferred for the treatment of VTE in patients with cancer; their use has been associated with a lower recurrence risk than with warfarin, without significantly increasing the rate of serious bleeding. This subject is discussed in detail elsewhere. (See "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy", section on 'Principles of therapy'.)

Ramucirumab — Ramucirumab may be distinct among antiangiogenic agents in terms of risk for ATE and VTE. In a 2017 meta-analysis of individual patient safety data from six randomized placebo-controlled trials in a variety of malignancies, there was no definite increased risk with ramucirumab of either ATE (all-grade, RR 0.8, 95% CI 0.5-1.3; grade ≥3, RR 0.9, 95% CI 0.5-1.7) or VTE (all-grade, RR 0.7, 95% CI 0.5-1.1; grade ≥3, RR 0.7, 95% CI 0.4-1.2) [27]. Nevertheless, the United States Prescribing Information recommends discontinuation of the drug in the event of a serious ATE [78].

VEGFR tyrosine kinase inhibitors — An increased risk of ATEs has also been shown in patients treated with antiangiogenic TKIs that prevent vascular endothelial growth factor receptor (VEGFR) signaling (relative to no treatment with a TKI), but as with bevacizumab, the relationship to VTEs remains uncertain.

Arterial thromboembolic events

Incidence — The following data are available regarding the risk of ATEs in patients receiving antiangiogenic TKIs:

●In a systematic review and meta-analysis that included 10,255 patients (87 percent with RCC) treated with sunitinib or sorafenib in 10 studies (which included phase II and III trials as well as data from expanded access programs), the incidence of ATEs was 1.4 percent (RR 3.0 compared with controls) [79]. There were no differential impacts based on the underlying malignancy (RCC versus others) and specific TKI (sunitinib versus sorafenib).

●In a placebo-controlled phase III trial of 903 patients treated for advanced RCC, 2.9 percent of patients receiving sorafenib developed myocardial ischemia or infarction compared with 0.4 percent of patients receiving placebo [80].

●Similarly, in a placebo-controlled randomized trial of pazopanib in advanced RCC, ATEs occurred in 3 percent of pazopanib-treated patients versus none in the placebo group [81].

●Likewise, in a placebo-controlled trial of lenvatinib, ATEs occurred in 5 percent of the treated group versus 2 percent with placebo [82].

Management — Both prevention and prompt management of an arterial thromboembolic event are critical. Prior to instituting therapy with antiangiogenic TKIs, predisposing cardiovascular risk factors (ie, hypertension, hyperlipidemia, and diabetes) should be aggressively managed. Baseline blood pressure should be controlled and the drug preferably not administered to patients with serious cardiovascular events within 6 to 12 months. Low-dose aspirin prophylaxis is reasonable in high-risk patients such as those with a prior ATE [83,84].

Patients who develop an ATE of any grade while receiving an antiangiogenic TKI should discontinue the antiangiogenic treatment and the ATE should be managed according to usual standards of care. Reintroduction of the agent may be considered cautiously in select cases for patients who are clearly benefiting from therapy and who are willing to accept the uncertainty of risk from continued use of the TKI in this setting.

Venous thromboembolic events

Risk — The association between antiangiogenic TKIs and VTE is uncertain. Semaxanib (SU5416), an investigational TKI that inhibits VEGF signaling, demonstrated a potential increase in VTEs in castration-resistant prostate cancer (11 percent), multiple myeloma (15 percent), and soft tissue sarcomas (15 percent) [85-87]. However, a meta-analysis of 7441 patients from nine phase III and eight phase II trials failed to demonstrate an increase in VTE in patients with solid tumors treated with sunitinib, sorafenib, pazopanib, vandetanib, or axitinib [88]. The summary RR for all-grade VTEs was 1.10 (95% CI 0.73-1.66) compared with no TKI controls, and for high-grade VTEs, it was 0.85 (95% CI 0.58-1.25). There was no difference in risk based on tumor type or age.

Similar conclusions were reached in a second meta-analysis of 14 trials of pazopanib, sunitinib, sorafenib, and vandetanib [89].

Management — There is not enough information available to recommend for or against the use of routine anticoagulation to prevent VTE in ambulatory cancer patients receiving, or about to receive, therapy with antiangiogenic TKIs. We agree with the Clinical Practice Guidelines in Oncology from the National Comprehensive Cancer Network (NCCN) [76], which suggest stratifying recommendations for aspirin or anticoagulant prophylaxis in patients with cancer according to the presence or absence of risk factors for VTE. The NCCN does not consider the use of angiogenesis inhibitors to represent a risk factor for VTE in cancer patients, and we largely agree with this. (See "Risk and prevention of venous thromboembolism in adults with cancer", section on 'Primary prevention'.)

If a patient undergoing therapy with an antiangiogenic TKI develops a VTE, prompt diagnosis and treatment are essential. Issues related to diagnosis and treatment of VTE in patients with malignancy are presented elsewhere. (See "Epidemiology and pathogenesis of acute pulmonary embolism in adults" and "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)

A brief withdrawal of the TKI may be prudent, while anticoagulation is being attained. Resumption of the TKI once anticoagulation is established is reasonable assuming the patient is receiving antitumor benefit. However, the United States Prescribing Information for tivozanib recommends that the drug be discontinued in patients who develop severe or life-threatening VTE events [90].

Left ventricular dysfunction and myocardial ischemia

Risk — Declines in left ventricular function can be seen in patients treated with any of the VEGF-targeted therapies:

●In a systematic review and meta-analysis including 77 phase III trials, angiogenesis inhibitors were associated with a higher risk of cardiac ischemia (OR 2.83, 95% CI 1.72-4.65) and cardiac dysfunction (OR 1.35, 95% CI 1.06-1.70), but subgroup analyses did not reveal differences between VEGF ligand inhibitors and small molecule agents [23].

●A total of 29,252 patients from 71 randomized controlled trials were included in a meta-analysis of VEGFR TKIs, which demonstrated a higher cardiac ischemia RR (RR 1.69, 95% CI 1.12-2.57), with the highest risks observed for sorafenib and patients with renal cancer [24]. Left ventricular systolic dysfunction was also increased after VEGFR TKIs (RR 2.53, 95% CI 1.79-3.57), with the highest risks observed for sunitinib and for those patients with hepatocellular cancer (HCC).

●A total of 10,553 patients from 36 clinical trials were included in another meta-analysis of VEGF TKIs, which reported an overall incidence of all-grade and high-grade congestive heart failure (CHF) of 3.2 percent (95% CI 1.8-5.8 percent) and 1.4 percent (95% CI 0.9-2.3 percent), respectively, and the risk of developing all-grade (OR 2.37, 95% CI 1.76-3.20) and high-grade (OR 3.51, 95% CI 1.74-7.05) CHF also increased [91]. Interestingly, meta-regression suggested that CHF occurs early in treatment with VEGFR TKIs.

●In a systematic review and meta-analysis of 10,647 patients from 21 randomized trials of a variety of approved multitargeted VEGFR TKIs (no trials included cabozantinib or regorafenib), heart failure of any grade developed in 138 of 5752 patients receiving VEGFR TKIs compared with 37 of 4895 in the non-TKI group (2.39 versus 0.75 percent). Serious (grade 3 or worse (table 5)) heart failure occurred in 17 of 1426 patients receiving VEGFR TKIs compared with 8 of 1232 patients in the non-TKI group (1.19 versus 0.65 percent). The RR of all-grade and high-grade heart failure for the TKI versus no TKI arms was 2.69 (95% CI 1.86-3.87) and 1.65 (95% CI 0.73-3.70), respectively. In this analysis, there was no difference in the risk of heart failure with the relatively specific TKI axitinib as compared with nonspecific TKIs such as sunitinib, sorafenib, vandetanib, and pazopanib [92]. However, cardiac failure, potentially fatal, has been observed in patients treated with axitinib [90].

Bevacizumab — Heart failure associated with bevacizumab has been sporadically reported in several trials of bevacizumab given in conjunction with anthracyclines or paclitaxel in women with metastatic breast cancer, an indication for which bevacizumab is no longer approved [93-95]. It has not been reported in patients treated with bevacizumab for other advanced cancers.

In contrast with heart failure, ischemic cardiac events are increased in patients treated with bevacizumab. (See 'Arterial thromboembolic events' above.)

The specific incidence of cardiac ischemia was addressed in a meta-analysis of five controlled trials conducted in patients with mCRC or advanced RCC that separated out risks of cardiac events and strokes [68]. The summary RR of high-grade cardiac ischemia in patients receiving bevacizumab was 2.14 (95% CI 1.12-4.08), and the overall incidence was 1.5 percent (95% CI 1-2.1 percent).

Specific VEGFR tyrosine kinase inhibitors

●Axitinib – In a randomized phase III trial, heart failure was reported in 6 of 339 patient treated with axitinib (2 percent, two fatal), versus 3 of 355 of those treated with sorafenib (1 percent, one fatal) [96]. Major adverse cardiovascular events occurred in 7 percent of patients with advanced renal cell cancer treated with axitinib plus avelumab in the JAVELIN Renal 101 trial; these events included death due to cardiac events (1.4 percent), grade 3 or 4 myocardial infarction (2.8 percent), and grade 3 or 4 heart failure (1.8 percent) [97].

●Lenvatinib – In a placebo-controlled trial in patients with advanced thyroid cancer, cardiac dysfunction (decreased ejection fraction, cardiac failure, or pulmonary edema) was reported in 7 percent versus 2 percent in the placebo group [82]. The majority of cases were findings of decreased ejection fraction by echocardiography.

●Pazopanib – In trials of pazopanib, cardiac dysfunction (decreased LVEF and clinical heart failure) has been observed. As an example, in a trial in advanced soft tissue sarcoma, myocardial dysfunction (symptomatic, or ≥15 percent absolute decline in LVEF compared with baseline, or decrease in LVEF to ≥5 percent below the lower limit of normal) was observed in 6.7 percent of patients treated with pazopanib compared with 2.4 percent of placebo-treated patients [98].

●Regorafenib – In a placebo-controlled clinical trial in patients with metastatic colorectal cancer, regorafenib was associated with an increased incidence of myocardial ischemia and infarction (1.2 versus 0.4 percent with placebo) [99]. Heart failure has not been reported.

●Ripretinib – Ripretinib is an inhibitor of KIT and PDGFRA kinases that is approved as a fourth-line agent for treatment of advanced gastrointestinal stromal tumors; in vitro, it also inhibits VEGFR2. (See "Tyrosine kinase inhibitor therapy for advanced gastrointestinal stromal tumors", section on 'Ripretinib'.)

Heart failure has occurred in approximately 1 percent of treated patients, and grade 3 decreased ejection fraction (table 6) has been seen in 2 to 3 percent [100,101]. Whether these changes reflect inhibition of VEGFR2, KIT, PDGFRA, or a combination of these targets remains unclear. (See "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines", section on 'Ripretinib'.)

●Sunitinib and sorafenib – In retrospective series and clinical trials, sunitinib has been associated with a decline in left ventricular ejection fraction (LVEF) in up to 28 percent of treated patients and clinical heart failure in 3 to 15 percent [13,14,102-104].

In the pivotal randomized trial that led to approval of sunitinib for advanced renal cell cancer, 21 percent of patients experienced a decline in LVEF, but this was symptomatic in only 10 percent [103]. All cases were reversible and not associated with an adverse clinical outcome. In an early study of patients receiving sunitinib for a gastrointestinal stromal tumor (GIST), left ventricular dysfunction developed in 10 versus 3 percent of the sunitinib and placebo-treated patients, respectively [104].

In a meta-analysis of the published literature that included 6936 patients receiving sunitinib for a variety of oncologic indications and who had regular cardiac function monitoring, the summary incidence of all grades of heart failure was 4.1 percent (95% CI 1.5-10.6 percent) and of grade 3 or 4 heart failure (table 5 and table 7 and table 8) was 1.5 percent (95% CI 0.8-3.0 percent) [105]. There were no differences in subgroups of patients receiving sunitinib for renal cell versus nonrenal cell cancer indications. Both hypertension and a history of coronary artery disease are thought to increase the risk of sunitinib cardiotoxicity.

However, these data from clinical trials probably underestimate the risk of LVEF decline and clinical heart failure [106]. It is now thought that the vast majority of patients treated with sunitinib (29 of 36 in one analysis [13]) experience an LVEF drop during therapy that approximately one in five have LVEF reductions of 15 percent or more, that 11 percent of patients receiving repeated cycles of sunitinib a cardiovascular event, mostly heart failure, and that preexisting hypertension and coronary artery disease are risk factors for cardiac toxicity [13].

Patients who have had a cardiovascular event (eg, myocardial infarction, severe/unstable angina, coronary/peripheral artery bypass graft, symptomatic heart failure, cerebrovascular accident or transient ischemic attack, or pulmonary embolism) within 12 months prior to sunitinib administration have been excluded from the sunitinib studies. If sunitinib is administered to such patients, they should be carefully monitored for clinical signs and symptoms of heart failure, and undergo periodic reevaluation of LVEF during therapy.

Less data on cardiotoxicity are available with sorafenib, but the risk seems to be lower than with sunitinib:

•In one phase III trial of patients treated for advanced renal cell cancer, 2.9 percent of patients receiving sorafenib developed myocardial ischemia or infarction compared with 0.4 percent of patients receiving placebo [80].

•In a second trial of sorafenib in advanced HCC, the incidence of cardiac ischemia or infarction with sorafenib was 2.7 versus 1.3 percent in the placebo group [107].

A major problem with defining the precise rate of cardiotoxicity associated with both sunitinib and sorafenib and its reversibility is that phase III trials have not pursued cardiac endpoints, and the identification of cardiac side effects with both drugs has predominantly been based on the occurrence of clinical symptoms. Potentially, the most comprehensive data on the range and frequency of cardiotoxicity with these drugs come from an observational series of 86 consecutive patients with renal cell cancer who were treated with either sunitinib or sorafenib at a single Austrian center over a 15-month period [108]. During treatment, 25 of 74 assessable patients (34 percent, including 11 receiving sunitinib and 14 treated with sorafenib) experienced a cardiovascular event, as defined by new elevation in cardiac enzymes, symptomatic arrhythmia requiring treatment, new left ventricular dysfunction, or acute coronary syndrome. Reported electrocardiogram (ECG) changes included changes in rhythm, conduction disturbance, change in axis or QRS amplitude, ST or T wave changes, and corrected QT (QTc) prolongation (which has been reported by others with sunitinib). (See "Acquired long QT syndrome: Definitions, pathophysiology, and causes".)

All symptomatic patients discontinued use of the TKI and began medical cardiovascular treatment. All recovered, although three had persisting abnormal cardiac enzymes through the entire TKI treatment period.

However, the high rate of cardiotoxicity in this series could have been related to the broad (and unconventional) definition of a cardiac event. In a preliminary report of cardiac toxicities with sunitinib or sorafenib use for the adjuvant therapy of RCC (ASSURE phase III trial), cardiac function in patients starting with normal LVEF was not impaired significantly over placebo [109]. Ischemic events were uncommon and not clearly associated with treatment. However, it should be noted that cardiac function was assessed at the six month time point and, therefore, excluded many patients who stopped treatment earlier due to side effects.

The nature of myocardial damage from antiangiogenic TKI treatment has not been extensively investigated, although hypotheses as to an underlying off-target pathophysiologic mechanism of cardiotoxicity caused by sunitinib and sorafenib have been proposed [110-112].

Monitoring and management — Baseline assessment of LVEF by echocardiogram or MUGA scan and an ECG are suggested variably by the manufacturers of the different agents. Some have suggested that patients receiving these drugs be treated as "stage A" heart failure patients (ie, at risk for heart failure), but without structural heart disease or symptoms [113]. Year 2013 guidelines for management of stage A heart failure from the American Heart Association suggest that it may be reasonable to evaluate those who are receiving potentially cardiotoxic agents for left ventricular dysfunction [114].

However, obtaining a baseline assessment of LVEF in all patients receiving these drugs is not supported by compelling data. We obtain a baseline LVEF evaluation only in older adult patients and those with a history of prior cardiovascular disease or prior anthracycline exposure. Caution and close serial monitoring of LVEF is warranted during therapy with bevacizumab and VEGFR TKIs for older adult patients and for those with a history of poorly controlled hypertension, heart disease, or anthracycline exposure. However, in our view, routine serial monitoring of ECGs and cardiac enzymes during therapy in the absence of symptoms is not warranted for other patients. The clinical significance of asymptomatic increases in cardiac enzymes or ECG changes to detect myocardial ischemia remains undefined. Clinicians should maintain a low threshold for cardiac evaluation in symptomatic patients.

The United States Prescribing Information offers specific interventions for patients receiving the following antiangiogenic agents who develop LV dysfunction or myocardial ischemia [90]:

●Sunitinib – Discontinue therapy in the presence of clinical manifestations of heart failure. Interrupt therapy and/or dose reduce in patients without clinical evidence of heart failure but with an ejection fraction <50 percent but >20 percent below baseline or below the lower limit of normal if baseline ejection fraction was not obtained.

●Sorafenib – Temporarily or permanently discontinue therapy in any patient who develops cardiac ischemia and/or infarction.

●Lenvatinib – Withhold therapy for any grade 3 cardiac dysfunction. Discontinue for grade 4 cardiac dysfunction (table 8 and table 7 and table 5).

●Regorafenib – Withhold for new or acute onset cardiac ischemia or infarction. Resume only after resolution of acute cardiac ischemic events and if the potential benefits outweigh the risks of further cardiac ischemia.

●Ripretinib – Permanently discontinue for grade 3 or 4 LV dysfunction.

●Other antiangiogenic agents – Guidance is not available for the other agents of this class. However, in general we hold the TKI for cardiotoxicity, and if the event was not major and the patient recovered, we (and others [115]) would consider restarting therapy if there was evidence of clinical benefit. Some patients who recover from left ventricular dysfunction may go on to tolerate re-exposure for long periods of time [115]. For patients who develop heart failure during treatment, it is also reasonable to switch to an alternative, less cardiotoxic anti-VEGF agent, if one is available (eg, pazopanib or bevacizumab for patients with RCC who develop heart failure while receiving sunitinib).

Aortic dissections and aneurysms — Accumulating case reports link the use of VEGFR TKIs and bevacizumab with abnormal structural changes of artery walls, resulting in aortic dissections and aneurysms [116-127]. The largest series is a pharmacovigilance study derived from VigiBase, the World Health Organization's centralized database of medication-related adverse drug reactions (ADRs) spontaneously reported by patients and clinicians from July 18, 2005 to January 19, 2019 [127]. Of the 217,664 ADRs potentially related to receipt of one of 14 antiangiogenic drugs, 494 (0.23 percent) were artery dissections or aneurysms. The antiangiogenic agent was bevacizumab in 45 percent, a VEGFR TKI in 36 percent, a mechanistic (previously called mammalian) target of rapamycin (mTOR) inhibitor in 12 percent, and cabozantinib in 2 percent. Of the VEGFR TKIs, sunitinib was the most common agent reported, while regorafenib was associated with the fewest cases (14.4 versus 1.4 percent). The time to onset of the ADR varied widely (median 89 days, interquartile interval 27 to 212 days). The dissection or aneurysm involved the aorta in 42 percent of cases, the cerebral arteries in 35 percent, and other/nonspecified arteries in the remainder. Overall, 24 percent of cases were fatal and 18 percent were life-threatening.

Although cases are described in the absence of hypertension, treatment-related hypertension may increase the risk by contributing to a damaged endothelium. In a report from the Japanese Adverse Drug Event Report database that included over 16,000 patients treated with VEGF pathway inhibitors and 59 patients with aortic dissection, the likelihood of aortic dissection with a VEGF pathway inhibitor after adjustment for concomitant hypertension remained high (adjusted OR 19.4, 95% CI 10.2-40.8) [117]. By comparison, the adjusted OR for hypertension alone (without a VEGF pathway inhibitor) was only 3.8 (95% CI 2.3-6.4). These reports led Health Canada to issue a safety alert to inform health care professionals about this risk [128]; a similar alert was recommended to be added to product labeling by the European Medicines Agency.

This observation emphasizes the need to control hypertension in patients on VEGFR TKIs and to have a high index of suspicion for aortic dissection in patients presenting with otherwise unexplained chest or abdominal pain. (See "Clinical features and diagnosis of acute aortic dissection", section on 'Symptoms and signs'.)

Bleeding — Bleeding associated with antiangiogenic therapy is discussed in detail separately. (See "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Bleeding'.)

CLASS SIDE EFFECTS OF VEGF TYROSINE KINASE INHIBITORS

Prolongation of the QTc interval and cardiac arrhythmias

Risk — Many drugs delay cardiac repolarization, an effect that is reflected on the surface electrocardiogram (ECG) by a prolonged heart rate corrected QT (QTc) interval. Although a prolonged QTc interval is not immediately harmful, it can be associated with potentially fatal cardiac arrhythmias. The ventricular tachyarrhythmia most typically triggered is of a unique form, known as torsades de pointes, which is most often transient, but when sustained, can give rise of symptoms of impaired cerebral circulation or degenerate into ventricular fibrillation, usually with a fatal outcome. (See "Acquired long QT syndrome: Definitions, pathophysiology, and causes" and "Congenital long QT syndrome: Pathophysiology and genetics", section on 'Pathophysiology'.)

The risk of QTc prolongation with vascular endothelial growth factor (VEGF)-targeted tyrosine kinase inhibitors (TKIs) has been addressed in the following studies:

●In a meta-analysis of 29,252 patients from 71 randomized controlled trials, QTc prolongation (relative risk [RR] 6.25, 95% CI 3.44-11.38) was reported with VEGF TKIs [24].

●Additional data are available from a trial-level meta-analysis of 6548 patients from 18 eligible randomized controlled trials in which one of the arms was a US Food and Drug Administration (FDA)-approved VEGF receptor TKI (there were no trials of sorafenib, cabozantinib, or regorafenib) [129]. Overall, 4.4 and 0.83 percent of patients exposed to a VEGF receptor TKI had all-grade and high-grade QTc prolongation, respectively. The RR for all-grade and high-grade QTc prolongation for the TKI versus no TKI arms was 8.66 (95% CI 4.92-15.2) and 2.69 (95% CI 1.33-5.44), respectively, with most of the events being asymptomatic QTc prolongation. On subgroup analysis, sunitinib and vandetanib were both associated with a statistically significant risk of QTc prolongation, while the increased RR seen with pazopanib and axitinib was not statistically significant. Higher doses of vandetanib were associated with a greater risk (RR 12.2 versus 3.6 for lower doses). The rate of serious arrhythmias including torsades de pointes did not seem to be higher in patients who developed high-grade QTc prolongation. The risk of QTc prolongation was independent of the duration of therapy.

Specific agents — Of the VEGF receptor TKIs, lenvatinib, vandetanib, and sunitinib have been most convincingly associated with QTc prolongation, while the risk with other VEGF receptor TKIs, including sorafenib, is less certain [115].

●Lenvatinib – In a placebo-controlled trial, approximately 9 percent of lenvatinib-treated patients developed QTc interval prolongation versus 2 percent in the control group; the incidence of grade 3 or greater QTc prolongation was 2 percent versus non in the placebo group. The United States Prescribing Information recommends monitoring ECGs in patients with congenital long QT syndrome, heart failure, bradyarrhythmias, or those taking drugs known to prolong the QT interval (table 9).

●Sunitinib – Sunitinib also has a dose-dependent effect on the QTc interval [108,115,130,131]. In contrast, the effect of sorafenib on changes in the QTc interval appears modest and unlikely to be of clinical significance [115,132].

Specific guidelines for monitoring with ECGs during sunitinib therapy and managing the dose in patients who develop QTc prolongation are lacking. In general, we obtain a baseline ECG in patients treated with sunitinib and monitor with ECGs during therapy only if the patient is also receiving therapy with other drugs that have the potential to prolong the QTc interval (table 9).

●Vandetanib – In clinical trials, vandetanib has been associated with prolongation of the QTc interval (waveform 1), torsades de pointes, and sudden death [115,133,134]. The magnitude of risk can be illustrated by a meta-analysis of nine phase II or III trials of vandetanib in a variety of malignancies [134]. The overall incidence of all-grade (>0.45 seconds) and high-grade (>0.5 seconds or symptomatic and requiring treatment) QTc interval prolongation was 16.4 and 3.7 percent, respectively, among non-thyroid cancer patients (predominantly breast and lung cancer), and 18 and 12 percent, respectively, among thyroid cancer patients who received treatment for a substantially longer period of time (median 18.8 versus 1.8 to 3 months) [134].

Largely because of its cardiovascular risk, vandetanib is only available through a restricted distribution program (the Caprelsa Risk Evaluation and Mitigation Strategy [REMS] program). (See "Medullary thyroid cancer: Systemic therapy and immunotherapy", section on 'Vandetanib'.)

Vandetanib should not be initiated in a patient with a QTc interval >450 milliseconds (ms) [135]. Because of the risk of cardiotoxicity, the United States Prescribing Information includes a black box warning to correct hypocalcemia, hypokalemia, and/or hypomagnesemia prior to drug administration. In addition, given the long half-life of the drug (19 days), ECGs are recommended to monitor the QT interval at baseline, at two to four weeks, and 8 to 12 weeks after starting treatment, and every three months thereafter. Monitoring of serum potassium, calcium, and magnesium levels as well as TSH is recommended on the same schedule. Concurrent administration of drugs known to prolong the QTc interval should be avoided (table 9). (See 'Monitoring and management' below.)

Patients who develop a QTc interval greater than 500 ms during treatment should stop taking the drug until the QTc returns to <450 ms; dosing should be resumed at a reduced dose.

Monitoring and management — Specific guidelines for assessing and monitoring the QTc interval and recommendations for managing the drug based on the grade of toxicity (table 10) are available for patients receiving vandetanib and lenvatinib [90]. Formal guidelines are not available for any of the other antiangiogenic TKIs.

The optimal formula to use to calculate the QTc interval in patients receiving chemotherapy drugs that have the potential to alter the QTc interval is debated [136-138]. The United States Prescribing Information for all of the VEGF receptor TKIs does not specify use of any one formula over another. (See "Acquired long QT syndrome: Clinical manifestations, diagnosis, and management", section on 'ECG findings'.)

However, some have recommended preferential use of the Framingham or Fridericia formulae (calculator 1) [136,138].

For any patient receiving therapy with an antiangiogenic TKI, careful review of concomitant medications is warranted, especially drugs that are associated with increased QTc (table 9). Patients with a history of QT interval prolongation, those who are taking antiarrhythmics, and those with relevant preexisting cardiac disease, bradycardia, or electrolyte disturbances may be more prone to developing a prolonged QTc. Concomitant treatment with strong cytochrome P450 3A4 (CYP3A4) inhibitors (table 11), which may increase plasma concentrations of antiangiogenic TKIs, may require a dose reduction of the TKI.

Management of the patient who develops an arrhythmia is discussed separately. (See "Acquired long QT syndrome: Clinical manifestations, diagnosis, and management".)

AGENT-SPECIFIC EFFECTS

Thrombotic microangiopathy — Rarely, cases of drug-induced thrombotic microangiopathy (DITMA) have been reported with specific antiangiogenic agents (eg, bevacizumab, sorafenib, sunitinib). Further details are discussed separately. (See "Nephrotoxicity of molecularly targeted agents and immunotherapy", section on 'Antiangiogenic agents' and "Drug-induced thrombotic microangiopathy (DITMA)".)

SUMMARY AND RECOMMENDATIONS

●Classification of antiangiogenic agents – Antiangiogenic agents (systemic medications that inhibit angiogenesis) have unique toxicities that differ from those of cytotoxic chemotherapy. Toxicities may be specific to the drug class or to the agent itself. Classes of antiangiogenic agents include:

•Ligand inhibitors – These agents bind to and inhibit ligand binding to the vascular endothelial growth factor (VEGF) receptor, thereby preventing activation of the receptor. Examples include bevacizumab, ramucirumab, and aflibercept.

•Receptor tyrosine kinase inhibitors (TKIs) – These agents block the intracellular domain of the VEGF receptor (VEGFR). Examples include sunitinib, sorafenib, pazopanib, vandetanib, cabozantinib, axitinib, ponatinib, lenvatinib, regorafenib, tivozanib, and fruquintinib.

●Cardiovascular effects of antiangiogenic agents – Cardiovascular effects of antiangiogenic agents may occur early and may be independent of dose. Some can be serious and potentially fatal. A multidisciplinary approach that includes the treating oncologist and cardiologist (preferably with expertise in cardio-oncology) and careful selection of patients for therapy with angiogenesis inhibitors, as well as close monitoring and prompt intervention, is necessary to reduce the risks posed by these agents. (See 'Introduction' above and 'Risk of fatality' above.)

•Hypertension – All commercially available angiogenesis inhibitors have been implicated in the development of hypertension. The overall incidence of hypertension is approximately 22 to 25 percent, with severe hypertension in 7 to 8 percent. Systolic or diastolic hypertension may represent a surrogate marker for response or improved outcomes with both bevacizumab and the antiangiogenic TKIs. (See 'Incidence and characteristics' above and 'Association with antitumor efficacy' above.)

Recommendations for monitoring and management of hypertension from the National Cancer Institute (NCI) Cardiovascular Toxicities Panel include (see 'Monitoring and management' above):

-Perform a pretreatment evaluation, including formal risk assessment for potential cardiovascular complications.

-Identify and treat preexisting hypertension before using these agents.

-Actively monitor blood pressure during treatment, with more frequent measurement in the first several weeks of therapy.

-Treat hypertension, with a goal blood pressure of 130/80 mmHg for most patients, lower in those with specific preexisting cardiovascular risk factors, such as diabetes or chronic kidney disease.

•Arterial thrombotic events – An increased risk for arterial thromboembolic events (ATEs) has been linked to the use of bevacizumab, aflibercept, and a number of antiangiogenic TKIs. The risk appears to be highest in patients with a prior history of an ATE and, in some reports, in those over age 65; bevacizumab should be used with caution in this group. The drug should be discontinued for grade 3 or higher (table 4) new or worsened ATEs during bevacizumab therapy. Continuation with concurrent anticoagulation might be considered in selected patients who are clearly benefiting from bevacizumab and who are willing to accept the uncertainty of risk from continued use of bevacizumab in this setting. (See 'Arterial thromboembolic events' above.)

For patients being considered for an antiangiogenic TKI, low-dose aspirin prophylaxis is reasonable for high-risk patients (eg, a prior ATE). Patients who develop an ATE should discontinue the TKI, and the ATE should be managed according to current standards of care. Reintroduction of the TKI may be considered cautiously in selected cases if the patient was deriving benefit from the drug. (See 'Management' above.)

•Venous thromboembolism (VTE) – The aggregate of data suggests no significant increase in the risk of VTE with VEGF-targeted therapy, with the possible exception of bevacizumab. (See 'Venous thromboembolism' above and 'Venous thromboembolic events' above.)

The decision to continue bevacizumab in a patient who develops a VTE during therapy must be individualized. Continuation of bevacizumab with concurrent anticoagulation is a safe approach for patients who are clearly benefiting from bevacizumab (see 'Prevention and management' above). If a patient undergoing therapy with an antiangiogenic TKI develops a VTE, a brief withdrawal of the TKI may be prudent while anticoagulation is being attained. (See 'Management' above.)

•LV dysfunction – Declines in left ventricular ejection fraction (LVEF) can be seen in patients treated with VEGF-targeted therapies; predominantly in patients treated with sunitinib. Myocardial ischemia has been seen in patients treated with bevacizumab, sorafenib, and regorafenib.

Baseline assessment of LVEF prior to treatment is warranted in older adults and in those with prior cardiovascular disease or anthracycline exposure. Caution and close serial monitoring of LVEF are warranted during therapy with bevacizumab and antiangiogenic TKIs in older adults and those with prior hypertension, heart disease, or anthracycline exposure. (See 'Left ventricular dysfunction and myocardial ischemia' above.)

•Aortic dissection and aneurysm – Accumulating case reports link VEGF receptor (VEGFR) TKIs and bevacizumab with aortic dissections and aneurysms; uncontrolled hypertension may increase this risk. This observation emphasizes the need to control hypertension in patients on VEGFR TKIs and to have a high index of suspicion for aortic dissection in patients presenting with otherwise unexplained chest or abdominal pain. (See 'Aortic dissections and aneurysms' above.)

•QTc prolongation – Prolongation of the corrected QT (QTc) interval has been described with vandetanib, lenvatinib, and sunitinib, but not with other antiangiogenic TKIs. Specific guidelines for managing QTc prolongation during vandetanib and lenvatinib therapy are available. (See 'Prolongation of the QTc interval and cardiac arrhythmias' above.)

For any patient receiving therapy with an antiangiogenic TKI, careful review of concomitant medications is warranted, especially drugs that are associated with increased QTc (table 9). Patients with a history of QTc interval prolongation, those taking antiarrhythmics, and those with preexisting cardiac disease, bradycardia, or electrolyte disturbances may be more prone to developing a prolonged QTc. Concomitant treatment with strong cytochrome P450 3A4 (CYP3A4) inhibitors (table 11), which may increase concentrations of the antiangiogenic TKI, may require a dose reduction of the TKI.

●Other adverse effects – Non-cardiovascular toxicities and nephrotoxicity of antiangiogenic agents are presented separately. (See "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents" and "Nephrotoxicity of molecularly targeted agents and immunotherapy", section on 'Antiangiogenic agents'.)