The initial treatment of established VTE in cancer patients is generally patterned after therapeutic approaches in other non-cancer settings. However, the duration of therapy to prevent early recurrence is often extended in cancer patients with persistent disease or continuing on cancer treatment [47]. Current recommendations call for low molecular weight heparin (LMWH) for the initial 5–10 days of anticoagulation in cancer patients with established VTE, as well as for secondary prevention of recurrence for at least 6 months. Patients with unsuspected or incidental VTE should be treated the same as symptomatic VTE. High-risk patients on systemic therapy for persistent malignancy should be considered for extended anticoagulation to prevent VTE recurrence. The development of a number of new oral and parenteral antithrombotic agents is likely to have future application to patients with cancer [48, 49].

Although reported rates vary considerably, VTE is a common cause of death in hospitalized cancer patients [2, 4, 50–53]. Although three large randomized controlled trials (RCTs) have demonstrated that thromboprophylaxis reduces VTE risk in hospitalized patients with acute medical illness, cancer represented only a small proportion of the study populations [54–57]. Nevertheless, due to the increased risk of VTE associated with malignancy, prophylactic anticoagulation of most hospitalized patients with major medical illnesses including cancer or reduced mobility is recommended in the absence of a serious bleeding risk with anticoagulation.

Likewise, patients undergoing major cancer surgery are at an increased risk for VTE as well as for bleeding complications [58]. Patients undergoing major surgical procedures for cancer should receive thromboprophylaxis unless contraindicated, while combined mechanical prophylaxis and anticoagulation may be considered in high-risk patients [59]. Prophylactic anticoagulation should be initiated preoperatively when possible and continued for at least 7–10 days. Extended prophylaxis for up to 4 weeks postoperatively should also be considered in high-risk patients including those with restricted mobility, obesity, or a history of VTE. Of note, there remain inadequate data to support routine thromboprophylaxis in those with short admissions for chemotherapy or for minor procedures [46, 60].

The risk of VTE in ambulatory cancer patients varies widely with the type of cancer and treatment and associated comorbid conditions. As discussed further in this chapter, the emergence of more aggressive interventions and a number of new cancer therapies and supportive care agents with an increased risk of VTE has resulted in increased interest in the potential value of VTE prophylaxis in this setting [8, 61–72]. Several RCTs of thromboprophylaxis in ambulatory cancer patients have been reported including nine with LMWHs. The greatest impact on the absolute risk of VTE has been observed in patients with advanced pancreatic cancer receiving specified chemotherapy [73, 74]. A meta-analysis has estimated a relative risk for symptomatic VTE across studies of 0.47 (0.36–0.61; P < 0.001) but with only an absolute reduction in VTE risk of 2.8 % (1.8–3.7 %; P < 0.001) [75]. Due to the small incremental benefit observed in most trials of ambulatory cancer patients, routine thromboprophylaxis is not recommended with the exception of patients with multiple myeloma receiving thalidomide or lenalidomide with chemotherapy and dexamethasone where the risk of VTE is very high. However, real-world studies in unselected ambulatory cancer patients receiving cancer chemotherapy have suggested rates of VTE twofold to threefold greater than those reported in selected patients in reported RCTs (Fig. 9.1) [76]. Therefore, thromboprophylaxis may be considered on an individual basis in selected high-risk patients with solid tumors receiving chemotherapy balancing the potential benefits and harms [46, 77].

VEGF is essential for tumor angiogenesis and increased VEGF expression is associated with increased tumor invasiveness, metastatic ability, and recurrence [87, 88]. There are three receptors (VEGFR-1/Flt-1, VEGFR-2/Flk-1/KDR, VEGFR-3/Flt-4). The ligand VEGF-A, generally referred to as VEGF, binds to VEGFR-2 on endothelial cells leading to pro-angiogenic effects [89]. By binding to the VEGF receptor, bevacizumab disrupts downstream VEGF signaling, preventing tumor angiogenesis and increasing the delivery of cancer therapy to tumor cells [90]. It is hypothesized that the vascular toxicities of bevacizumab, which include both thrombosis and bleeding, occur through similar mechanisms and are the result of an imbalance in the tightly regulated hemostatic system. This system includes a balance of pro- and anticoagulant proteins, platelet-activating and platelet-inhibiting factors, and pro- and antifibrinolytic products [12, 90]. Inhibition of VEGF results in decreased endothelial cell survival, platelet aggregation and thrombosis, increased platelet reactivity, and downregulation of several factors. The vasculature is susceptible to damage induced by trauma [12, 90], and disruption of the endothelial barrier results in exposure of the subendothelial von Willebrand factor and tissue factor, platelet aggregation, and thrombus formation [86, 91]. The increase in platelet-endothelial interactions may further precipitate thrombosis [92]. Finally, downregulation of several factors that are modulated by VEGF including nitric oxide, prostacyclin, and thrombolytic serine proteases (u-PA and t-PA) contributes to increased vascular toxicity such as vasoconstriction, endothelial cell apoptosis, and increased arterial stiffness [86, 90]. Several other mechanisms have been proposed as contributing to VEGF inhibitor-induced vascular toxicity. Drug-induced hypertension generating high shear stress at plaque sites may accelerate atherothrombosis [88, 91]. Associated inflammation and cell lysis have also been shown to increase thrombogenicity, and apoptotic cell blebs may sustain vascular inflammation and activate the complement cascade [88, 91]. Anti-angiogenic drugs are known to hinder the recognized insulin anti-atherogenic actions, including glucose uptake, lipogenesis, and antilipolysis, ultimately producing a thrombophilic hyperglycemic , atherogenic lipoprotein prone to thrombosis [9]. Finally, VEGF inhibition may result in lack of protective growth factors, potentially resulting in plaque instability and thrombosis [93]. Figure 9.2 illustrates the proposed mechanisms for the vascular toxic effects of bevacizumab.

The major cardiovascular toxicities of bevacizumab are hypertension, hemorrhage, perforation, and thrombosis [90]. Thrombotic events include thrombosis and thromboembolism, generally in the form of acute coronary syndrome, stroke, and peripheral vascular disease, although coronary ischemia is the most frequent arterial thrombotic event (ATE) [94]. While initial trials of bevacizumab only reported an insignificant increase in thrombotic or bleeding events, multiple subsequent larger clinical studies have demonstrated an elevated risk of vascular events [87]. A meta-analysis of 1745 patients from five randomized trials showed combination treatment with bevacizumab and chemotherapy, compared with chemotherapy alone, was associated with an increased risk of ATE (HR 2.0, 95 % CI 1.05–3.75, P = 0.031), but not venous thromboembolism [10]. A meta-analysis published in 2010 to evaluate the incidence of arterial thromboembolic events associated with bevacizumab included over 12,500 patients with various advanced solid tumors and found a 3.3 % incidence of ATE (95 % CI 2.0–5.6) and a 2.0 % incidence of high-grade arterial thrombotic events (95 % CI 1.7–2.5), defined as acute coronary syndrome, transient ischemic attack, stroke, life-threatening peripheral arterial thrombosis or requiring surgery, and death [94]. Compared to controls, bevacizumab was associated with a relative risk of ATE of 1.44 (95 % CI 1.08–1.91, p < 0.013). In particular, this meta-analysis found that bevacizumab was associated with a significantly increased risk of cardiac ischemia (RR 2.14, 95 % CI 1.12–4.08, p < 0.021) [94]. Vascular events have been noted early after starting VEGF inhibitor therapy, with a median time of event occurrence after drug initiation of 7 months (1–12 months is general range) [88, 95]. Moreover, the risk of ATE associated with bevacizumab may differ by tumor type. One meta-analysis found a significantly higher risk of high-grade ATE in patients with renal cell carcinoma (RR 5.14 95 % CI 1.35–19.64) [94]. Patients with pre-existing cardiovascular disease are at an increased high risk for thrombotic complications [96, 97]. Specifically, patients older than 65 years of age, with diabetes, with known atherosclerosis, or with a history of a prior cardiovascular event, appear to be at a higher risk for bevacizumab-associated ATE [97].

Traditionally arterial thrombotic events have been associated with grave prognosis in cancer patients [98]. Once an arterial thrombotic event develops in a patient receiving VEGF inhibitor therapy, it is generally recommended therapy be permanently stopped and the ATE treated as per guidelines for the particular event, such as the American College of Cardiology/American Heart Association (ACC/AHA) recommendations for acute coronary events [99, 100]. Some clinicians advocate for avoiding anti-VEGF therapies in patients with a history of coronary or peripheral vascular disease. Others recommend prophylactic antiplatelet therapy such as aspirin or clopidogrel [86]. Aspirin therapy has been shown to be associated with a significant improvement in short-term cardiovascular outcomes [10, 88]. Theoretically, statins and angiotensin-converting-enzyme inhibitors (ACE-inhibitors) may also exert antioxidant and anti-inflammatory effects resulting in a decrease in atherothrombotic risk [10]. However, the use of anticoagulants and antiplatelet agents in patients receiving VEGF inhibitors remains controversial, especially in the setting of thromboembolic events given their association with bleeding.

Despite the increased frequency of thromboembolic events in cancer patients, there are limited data on the use of bevacizumab concurrently in patients being treated with therapeutic anticoagulation . A post hoc analysis of three phase III clinical trials of bevacizumab compared to control evaluated the incidence of thrombotic and bleeding in patients receiving therapeutic anticoagulation (warfarin and low molecular weight heparin) [95]. The incidence of thrombotic adverse events for patients receiving bevacizumab compared to the placebo group was primarily venous and on the order of 9.6–17.3 %. The overall rates of severe bleeding for all patients in the control group were 2.5 % versus 3.3 % in the bevacizumab group. The authors concluded that combining bevacizumab with therapeutic anticoagulation did not substantially increase the risk of bleeding beyond the risk of bleeding expected from therapeutic anticoagulation alone [95]. Similarly, a prospective observational cohort study including 1953 patients to assess the safety of bevacizumab for treatment of metastatic colorectal cancer (BRiTE study) showed serious bleeding events were comparable among patients on prophylactic anticoagulation therapy versus not [101]. In the absence of formal guidelines, it is generally recommended that patients be treated for thrombotic events with anticoagulation albeit cautiously with close monitoring for adverse bleeding events. There are insufficient data and therefore no standard recommendations to use prophylactic anticoagulation in this setting. Of course, an individual approach for patients should be conducted based on comorbidities and prior thrombotic and bleeding events. Furthermore, a cardiovascular risk assessment is recommended in all patients undergoing such therapy, with consideration for additional testing in certain patient populations.

Similar to bevacizumab, these small molecule TKIs work by blocking the intracellular domain of the VEGF receptor leading to inhibition of the VEGF pathway. Moreover, these TKIs affect multiple additional pathways. Sunitinib has activity against all three VEGFRs, platelet-derived growth factor receptor (PDGF) α and β, stem cell factor receptor (KIT), and fms-like kinase receptor 3 (FLT3) . It is approved for treatment of advanced renal cell carcinoma, gastrointestinal stromal tumors (GIST) , and advanced pancreatic neuroendocrine tumors [86, 88, 97]. Sorafenib has activity against all VEGFRs, PDGF-β, KIT, FLT3, and RET, as well as the intracellular kinases CRAF, BRAF, and mutant BRAF. It has been approved for the treatment of advanced hepatocellular carcinoma, thyroid cancer, and advanced renal cell carcinoma [86, 88, 97]. Pazopanib also has activity against VEGFRs, PDGFRs, fibroblast growth factor receptors (FGFRs) 1 and 3. It has been approved for the treatment of metastatic renal cell carcinoma and advanced soft tissue sarcomas that have received prior chemotherapy [86, 88, 97].

The mechanisms for vascular toxicity in these TKIs with VEGF pathway inhibiting characteristics may be similar to those proposed for bevacizumab. Individual response to VEGF inhibition and variation in response to the growth factor pathway blocking lends to variation in vascular complications and individual vulnerability to off-target effects [12, 86, 88, 97]. Figure 9.2 illustrates proposed mechanisms for vascular toxic effects of VEGF signaling pathway inhibitors.

The major cardiac vascular toxicities of small molecule tyrosine kinase inhibitors with VEGF-inhibiting properties include hypertension, hemorrhage, perforation, and thrombosis [90]. As it relates to thrombosis , there is an increased incidence of acute coronary syndrome, cerebrovascular accidents, peripheral vascular disease, and bleeding observed with these anti-angiogenic TKIs.

Cardiac ischemia and related coronary artery disease has been a common manifestation of ATE associated with TKIs (Fig. 9.3). One observational study showed that among patients treated with sorafenib or sunitinib, there was an extremely high rate of cardiac events (33.8 %) leading to interruption or discontinuation of TKI therapy in 60 % of those patients [102]. Of the patients who experienced cardiac events, 52 % were symptomatic with angina, dyspnea, and dizziness while 48 % of the patients were asymptomatic with cardiac biomarker elevation or ECG changes [102]. In a large meta-analyses of over 38,000 patients evaluating the vascular complications of VEGF inhibitors, including both TKIs and bevacizumab , there was an increased risk of myocardial infarction , hypertension , and arterial thromboembolism [103]. Compared to the control group, recipients of VEGF inhibitors had significantly higher risk of myocardial infarction (RR 3.54, 95 % CI 1.61–7.80, I ² = 0 %, tau2 = 0), arterial thrombotic events (RR 1.80, 95 % CI 1.24–2.59, I ² = 0 %, tau2 = 0), and hypertension (RR 3.46, 95 % CI 2.89–4.15, I ² = 58 %, tau2 = 0.16) [103]. In 2010, Choueiri et al. published a meta-analysis of clinical trials with sorafenib or sunitinib that included over 10,000 patients [91]. The rate of arterial thrombotic events was increased threefold (2 % absolute risk) with these drugs, and this finding was independent of type of malignancy or TKI [91]. Again, the most common type of ATE was coronary ischemia, followed by stroke [91].

Interestingly, a newer agent under investigation semaxanib, a potent and selective inhibitor of VEGFR-2/Flk-1/KDR, with activity against VEGFR-1, KIT, and FLT3, was aborted because of a remarkably high thrombosis rate of 42 % during a phase I trial [12, 88, 104]. The authors concluded that semaxanib likely causes endothelial cell activation, and in the setting of chemotherapy-induced triggered coagulation cascade, high thrombosis rates can occur. However, it is speculated that this high rate of thrombosis (both arterial and venous) is most likely related to the combination of semaxanib with cisplatin and gemcitabine , which are also associated with vascular events independently [104, 105].

As with bevacizumab, there are no standardized recommendations on specifically how to manage these patients within the realm of cardio-oncology. Withdrawal of TKI therapy and treatment of arterial thrombosis as per oncology and cardiology guidelines is currently advised [99, 100]. The role of screening and prophylactic anticoagulation is not defined. Furthermore, large studies assessing the role of various risk stratification tools to evaluate patients vulnerable to the atherothrombotic effects of TKIs are lacking [12, 86].

The mutated fusion protein, Bcr-Abl, is expressed only on malignant cells, and thus Bcr-Abl-inhibiting TKIs are theoretically specific to leukemic cells. The TKIs bind to the amino acids of the Bcr-Abl tyrosine kinase ATP-binding site and stabilize the inactive form of the protein, preventing tyrosine autophosphorylation and downstream phosphorylation of its substrates [108]. Second- and third-generation Bcr-Abl-inhibiting TKIs target the genetic mutations leading to imatinib resistance. Ponatinib is an example and has activity against the specific T3151 mutation in Bcr-Abl implicated in many imatinib-resistant cases [107]. The mechanisms underlying the association between the second- and third-generation Bcr-Abl-inhibiting TKIs and vascular complications are largely unknown. Unlike other anti-angiogenic therapies , these small molecule TKIs do not have established VEGF receptor or pathway inhibiting properties. However, they are often referred to as “accidental” angiogenesis inhibitors as they likely still affect the angiogenesis pathway [88, 96, 109]. There is some speculation that the vascular toxicity associated with these agents may be related to unrecognized kinase targets being inhibited or at least preferentially blocked by the second- and third-generation TKIs which have demonstrated acute atherothrombosis and accelerated atherosclerosis [110]. These TKIs have been demonstrated to have effects on the discoidin domain receptor 1 (DDR1), which has been implicated in plaque formation in atherosclerosis [110]. Additional targets of these TKIs, KIT and PDGFR, appear to be involved in regulation of various vascular and perivascular cells, and thus disruption may lead to vascular events [110]. Some small prospective trials show evidence that nilotinib may also result in metabolic derangements including glucose intolerance and even diabetes, altered lipid profiles, and atherogenesis [110–112].