SEARCH

SEARCH BY CITATION

Keywords:

  • cardiovascular disease;
  • endothelial dysfunction;
  • inflammation;
  • nuclear factor kappa-B;
  • radiotherapy/adverse effects;
  • radiation injuries

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References

Abstract.  Halle M, Hall P, Tornvall P (Karolinska Institutet, Stockholm, Sweden). Cardiovascular disease associated with radiotherapy: activation of nuclear factor kappa-B (Review). J Intern Med 2011; 269: 469–477.

There have been several recent reports of an increased risk of cardiovascular disease after radiotherapy. Hence, with an increasing number of cancer survivors, the incidence of cardiovascular disease caused by radiotherapy will increase. The existence of a type of vascular disease, or vasculopathy, induced by radiotherapy has been known for decades. It is important to identify and understand the molecular causes of this vasculopathy to determine preventive strategies. Recently, a chronic inflammation with similarities to atherosclerosis has been observed, with activation of the transcription factor nuclear factor kappa-B (NF-κB) as a possible cause. However, the trigger for NF-κB activation is unclear although it may be that reactive oxygen species or direct DNA damage is involved. To minimize the risk of cardiovascular disease in vulnerable patients, careful selection of patients, radiation dose and fractionation are important, together with the development of new techniques that reduce radiation dose to the blood vessels. In the light of the finding of an interaction between risk factors for cardiovascular disease and radiotherapy, it is reasonable to modify these factors including diabetes mellitus, hyperlipidaemia, hypertension and smoking. We belive that preventive strategies focusing on NF-κB can reduce the risk of future adverse cardiovascular events.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References

Today, an estimated 50 million cancer survivors have been treated with radiotherapy worldwide [1]. A type of vascular disease, or vasculopathy, which is induced by radiotherapy, has been known for decades [2]. The characteristics of this vasculopathy are not well defined and the consequences regarding the future morbidity and mortality in cancer survivors have not been fully determined. Recently, a number of studies, which are reviewed below, have shown that cardiovascular disease can develop after radiotherapy to the heart, neck and brain. The clinical manifestations vary from myocardial infarction to heart failure and stroke. The pathology of these cardiovascular manifestations is largely unknown but they clearly progress slowly as the time from radiotherapy to clinical symptoms often exceeds 10 years [3, 4]. The pathology of this vasculopathy mimics that of atherosclerosis in many respects. Here, we will review not only the clinical manifestations of radiotherapy-induced vasculopathy but also the possible molecular mechanisms. We will also discuss the possible treatment options to ameliorate this condition.

Patient studies

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References

Many studies, in a range of clinical disciplines, have provided evidence of cardiovascular disease in cancer survivors many years after radiotherapy [3–8]. This disease is presumably caused by a vasculopathy induced by radiotherapy. It has been described as a progressive disease with neovascular proliferation together with gradual stenoses of conduit arteries, giving rise to a variety of adverse clinical outcomes, dependent on the target, dose and fractionation. However, differences between clinical settings and sites of irradiation make interpretation of findings difficult and causality of associations unclear.

Results from epidemiological studies suggest that the cause of radiotherapy-induced vasculopathy is induction or acceleration of atherosclerosis in conduit arteries located in the irradiated field [9], but observational biological data from human studies are scarce. Until recently, there has been a lack of data in humans regarding the sequence of events, from endothelial dysfunction to manifest cardiovascular disease (Fig. 1).

image

Figure 1. Key references for studies to investigate the sequence of events from endothelial dysfunction to clinical cardiovascular manifestations [4, 10, 11, 13, 18, 26]. NO, nitric oxide; FMD, flow-mediated dilatation; IMT, intima-media thickness; MI, myocardial infarction.

Download figure to PowerPoint

Sugihara et al. [10] showed, ex vivo, that nitric oxide (NO)-mediated endothelial-dependent relaxation is impaired in human cervical arteries, 4–6 weeks after radiotherapy, and Beckman et al. [11] demonstrated in vivo that radiotherapy impairs flow-mediated dilatation, i.e. endothelium-dependent vasodilatation, of conduit arteries. Both these studies demonstrated that radiotherapy-induced vasculopathy is associated with reduced endothelial NO bioavailability, which is generally accepted to be an early sign of atherosclerosis [12].

Strong support for a vascular response to radiotherapy with progressive vascular disease is also provided by the results of a study by Dorresteijn et al. [4], who showed an increased intima-media thickening 10 years after exposure of the carotid artery. This was confirmed in a recent study by Russel et al. [13], using immunohistochemical analysis of human conduit arteries at a mean of 4 years after radiotherapy for head and neck cancer. Following the rapid development of high-resolution imaging techniques, a growing body of morphological evidence has also contributed to an increased knowledge of vascular changes in previously irradiated tissues. After both cranial and midfacial radiotherapy, late vascular occlusion could be observed together with parallel networks of vascular proliferation [5, 14, 15].

Vascular brachytherapy serves as a good model for radiotherapy-induced vasculopathy. Brachytherapy has been used for the prevention of restenosis after percutaneous coronary intervention. Despite initial positive results, long-term follow-up has shown a progressive loss of benefit in clinical outcome after intracoronary radiotherapy. The results of a randomized study by Ferrero et al. [16] confirmed a delaye and progressive restenotic process after stent implantation and radiotherapy in de novo lesions. A delayed prothrombotic effect of radiotherapy may also be involved as late stent thrombosis has been described years after intracoronary brachytherapy [17]. Taken together, radiotherapy-induced vasculopathy might result in several different clinical manifestations depending on the site of irradiation (Table 1).

Table 1. Studies of the association between radiation therapy and cardiovascular disease
Type of CVDType of cancerRadiotherapy doseReferences
  1. CVD, cardiovascular disease; CHD, coronary heart disease; PVD, peripheral vascular disease; BC, breast cancer; CNST, central nervous system tumours; HL, Hodgkin’s lymphoma; H&N, head and neck cancer; PC, pelvic cancer. Other heart diseases; heart failure, valvular heart disease and pericardial disease.

CHDBC, HL>15 Gy3, 18, 20, 22
Other heart diseaseBC, HL>15 Gy18–22
StrokeCNST, HL, H&N>40 Gy23–28
PVDBC, PC8, 31, 32

Heart disease

One recent study of 14 000 5-year survivors of cancer who were diagnosed under the age of 21 showed that radiotherapy increased the risk of myocardial infarction, heart failure, valvular heart disease and pericardial disease [18]. These results were confirmed, with regard to cardiovascular mortality, in a recent but smaller study [19]. In both studies, doses of radiotherapy delivered to the heart were estimated. The threshold for the risk of myocardial infarction, heart failure, valvular heart disease and pericardial disease [18] was 15 Gy, whereas the risk of cardiovascular death was increased from 5 Gy [19]. Symptoms of coronary heart disease (CHD) occur 10–15 years after radiotherapy, and its incidence is increased in patients with ‘classical’ risk factors, such as smoking, hypertension and obesity [20]. Radiotherapy for breast cancer, at least with some of the older radiotherapy regimens, has been associated with a significant excess of nonbreast cancer mortality, mainly from cardiovascular disease [21, 22]. Further strong support for late, progressive disease is provided by a study by Darby et al. [3] who demonstrated that left- compared to right-sided breast cancer was associated with CHD, but not until more than 10 years after radiation exposure. It is interesting that radiotherapy not only affects the coronary arteries but also the heart muscle, the endocardium (including the valves) and the pericardium, possibly because of general radiation fibrosis, as seen in other tissues.

Stroke

An increased risk of stroke has been reported after cranial radiotherapy for brain tumours in childhood or adolescence [23, 24] as well as after neck radiotherapy for Hodgkin’s disease and head and neck cancer treatment in adults [25–28]. In general, these studies have been small, making it impossible to establish associations between the radiation dose and risk of stroke.

In addition, modern imaging technologies have been able to demonstrate occlusion of medium-sized cerebral vessels in previously irradiated fields, often accompanied by a fine collateral network of new vessels [5, 14, 15, 29]. Postmortem histological findings of irradiated cerebral blood vessels have shown prominent features such as fibrous thickening of the intima with foam cells in the absence of inflammatory vasculitis [30].

Other vascular manifestations

There have been several case reports of other vascular manifestations possibly related to radiotherapy, such as arterial and venous stenoses affecting the axillary and inguinal regions after radiotherapy of breast and pelvic cancer, respectively [8, 31, 32].

Radiation retinopathy is another complication of radiotherapy that occurs in patients irradiated for cranial tumours [33]. It has been shown that radiotherapy induces both an acute transudative and a slowly progressive occlusive vasculopathy together with parallel networks of dysfunctional neovascularization [34], eventually leading to blindness.

Furthermore, it has been demonstrated that radiation induces microvascular dysfunction leading to an increased risk of defective wound healing [35]. Schultze-Mosgau et al. [36] demonstrated that both number and diameter of capillaries were reduced in the irradiated graft bed tissue and that vascularization of the graft bed decreased continuously as a function of the total dose and time after radiotherapy [37]. Occlusion of microvascular anastomoses during free tissue transfers is also more common when coupled to previously irradiated recipient vessels [38], possibly because of endothelial dysfunction and an activated prothrombotic response [39].

Molecular mechanisms

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References

Based on the timing of various clinical manifestations, a slowly progressive vasculopathy is likely to be involved. Other types of tissues appear to undergo chronic inflammatory changes years after radiotherapy [40, 41], but such changes are poorly described for the vasculature. These findings suggest that a chronic inflammatory response may be involved in the progressive vasculopathy. During the last two decades, knowledge of vascular inflammation in atherosclerosis has increased [42]. A plausible assumption would be that a radiation-induced progressive inflammatory response may share some similarities with atherosclerosis such as intimal hyperplasia (Fig. 2).

image

Figure 2. Histopathology specimens from an irradiated cervical artery showing signs of intimal hyperplasia with NF-κB activation indicated by p65 staining (b) compared to a healthy nonirradiated radial artery (a) from the same patient. Arrowheads indicate internal elastic lamina. Scale bar = 50 μm.

Download figure to PowerPoint

Experimental studies of radiotherapy-induced vasculopathy have so far focused on acute effects and mainly been performed in cell culture [43–46] and animal models [47–51]. The high radiation sensitivity of the vasculature has previously been linked mainly to endothelial dysfunction [43, 45, 48]. This may be explained by the fact that irradiated tissues suffer from chronic oxidative stress with increased production of reactive oxygen species (ROS) [40]. It is interesting that overproduction of ROS is regarded as an integral part of atherosclerosis [52]. Also, most classical CHD risk factors contribute to this oxidative stress, which causes a disruption in the balance between NO and ROS, resulting in a relative decrease in bioavailable endothelium-derived NO [53].

The results of previous in vitro studies have suggested that radiation induces endothelial activation [54] characterized by activation of the transcription factor nuclear factor κB (NF-κB) [55, 56], resulting in alterations in vascular adhesion molecule expression [55, 57] and chemokine and cytokine production [44]. The activated endothelium is prone to atherosclerosis and is prothrombotic as a result of promoting leucocyte–endothelial cell or platelet–endothelial cell adherence [57, 58], leucocyte infiltration into tissue [55, 59] and thrombus formation [60]. Deiner et al. showed in an in vivo porcine restenosis model that intracoronary radiotherapy initially inhibits cell proliferation, but eventually cellular and molecular inflammatory processes are enhanced within the arterial wall by activation of NF-κB. It has been suggested that this proinflammatory effect of radiotherapy is responsible for the observed delayed proliferation in animals and the resulting late lumen loss seen also in humans, as described above [61]. By comparing irradiated with nonirradiated arteries from the same patient, harvested during cancer reconstruction with free tissue transfer, it has been possible to confirm NF-κB activation by radiotherapy in humans [62] (Fig. 3). Thus, the evidence from in vitro and in vivo studies, including the recent data in humans, suggests that the transcription factor NF-κB is crucial for the development of radiotherapy-induced vasculopathy, a process with many similarities to atherosclerosis.

image

Figure 3. Immunofluorescence staining of an irradiated cervical artery showing co-localization of p65 and cell nuclei (in purple; indicating the presence of activated NF-κB) (b) compared to a healthy nonirradiated radial artery, from the same patient, showing only staining for cell nuclei (in blue) (a). Scale bar = 20 μm.

Download figure to PowerPoint

Activation of NF-κB is regarded as one of the most important and early events of endothelial activation [63]. Endothelial activation has long been known as a key event for the onset of inflammatory disease processes, including atherosclerosis [64]. An activated endothelium is characterized by the expression of leucocyte adhesion molecules and E-selectin [65–67]. Selectin-mediated leucocyte rolling is required for the consecutive activation of leucocytes by endothelial cell-bound chemokines. As a consequence of activation, leucocyte integrins (another class of adhesion molecules) mediate cellular adhesion to the endothelial wall and, finally, leucocytes transmigrate through the endothelium into the underlying tissue. The majority of transcriptionally regulated genes expressed in the endothelium in response to inflammatory mediators contain an NF-κB-binding site in their promoters [65, 67]. Activated NF-κB is present in human atherosclerotic plaques but not in normal vessels devoid of atherosclerosis [68]. Inhibition of NF-κB activation results in highly efficient inhibition of endothelial cell activation [69, 70]. NF-κB is also expressed in arterial smooth muscle cells after balloon injury and is responsible for the expression of several genes encoding for adhesion molecules and chemokines that promote infiltration of monocytes [71], which are central to vascular remodelling in response to injury [72]. Monocyte-/macrophage-derived cytokines and growth factors, amongst them tumour necrosis factor, will further affect the integrity of the vascular wall by directly or indirectly stimulating smooth muscle cell proliferation and migration [73].

Results from immunohistochemical staining of human irradiated arteries indicate that NF-κB is activated in virtually every cell within the arterial wall [62]. One possible explanation for this may be the ‘radiation-induced bystander effect’, which refers to interactions between physically separated cells through two main routes: direct cell–cell communication through gap junctions or cytokine signalling through extracellular matrix [74]. It appears that different tissues possess different radiotherapy sensitivity towards the activation of NF-κB in vivo [75]. Activation of NF-κB by radiotherapy can be affected by tissue type, and distinct molecular components and signalling events have been implicated under different experimental settings. These components and pathways include protein tyrosine kinases, protein kinase C, ROS, the G-protein Ras, ataxia–telangiectasia-mutated gene, DNA-dependent protein kinase, IκB kinase and the ubiquitin–proteasome pathway [75]. Of particular interest with regard to the mechanism of activation of NF-κB by radiotherapy is the possibility that nuclear DNA damage may be directly involved in the activation of the cytoplasmic NF-κB/IκB complex, which is a novel nuclear-to-cytoplasmic, retrograde signalling pathway. Huang et al. [76] provided multiple lines of evidence for the existence of such a signalling pathway induced by the anticancer agent camptothecin. This agent, like radiotherapy, is known to cause double-strand brakes (DSBs) with subsequent activation of nuclear kinases [77, 78]. By reviewing published observations, it is apparent that there is no consensus on the activation mechanism for the effects of radiotherapy on NF-κB, especially at clinically relevant doses. It is unclear whether radiation-induced DSBs represent the initiating signal event or whether other stressors, such as ROS, mediate this activation (Fig. 4).

image

Figure 4. Activation of the transcription factor NF-κB in blood vessels by radiotherapy leads to endothelial dysfunction and inflammatory alterations. Following irradiation, the activation may be initiated by either DNA double-strand breaks or reactive oxygen species.

Download figure to PowerPoint

A complication is the undetectable nature of NF-κB activation in many normal tissues after whole-body radiotherapy, including at relatively high doses (up to 8.5 Gy). Many human normal diploid cell types are refractory to NF-κB activation following radiotherapy in vitro, although tumour necrosis factor can efficiently activate NF-κB in these cells. Also, different radiation doses seem to invoke distinct signalling events to cause activation of NF-κB. Therefore, it has been suggested that further in-depth analyses of signalling pathways and differential radiotherapy sensitivity of human normal versus cancerous cell types for activation of NF-κB are warranted [75].

Future perspectives and treatment options

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References

The US National Cancer Institute has identified long-term survival from cancer as one of the new areas of public health emphasis, particularly ‘studying adverse long-term or late effects of cancer and its treatment’ [41]. There is growing evidence of dose–effect and dose–volume relationships for radiation-associated cardiovascular disease [79]. It is essential to realize that epidemiological findings involve patients treated in the past with radiation volumes, doses and techniques that are no longer used, although one recent study from Sweden showed that there has not been a dramatic decrease in cardiac doses in patients with breast cancer over time [80]. However, the results are important for the understanding of late radiation-induced effects to adapt treatment strategies for the future [81]. In this context, it is extremely important to reduce the risk of adverse effects of radiotherapy. Regarding cardiovascular disease, the aim should be to minimize the amount of radiation to the heart and blood vessels as there is a clear dose–response relationship.

Vasculopathy induced by radiotherapy has many similarities with atherosclerosis, as discussed earlier. It is likely that risk factors for atherosclerosis will potentiate the risk of manifest cardiovascular disease in patients treated with radiotherapy. Therefore, a primary aim should be to reduce risk factors for CHD, such as diabetes, hyperlipidaemia, hypertension and smoking, to prevent progression of radiotherapy-induced vasculopathy.

There are several possible new therapeutic options to ameliorate the adverse cardiovascular effects caused by radiotherapy. As noted earlier, radiation retinopathy is characterized by endothelial cell damage, transudation, occlusion and neovascularization [34]. In this condition, a humanized monoclonal antibody to vascular endothelial growth factor (VEGF), bevacizumab (Avastin®; Genetech, San Francisco, CA, USA), has recently been found to successfully decrease vascular permeability and inhibit neovascularization following radiation. Reduction in permeability of the retinal arteries was associated with resolution of oedema, haemorrhage and exudation, with improvement in visual acuity [82]. The authors suggested that anti-VEGF medication should be investigated for use in other types of radiation vasculopathy. This is interesting because VEGF seems to play a pivotal role in disruption of the integrity of the vessel wall during inflammation [83], and NF-κB activates VEGF expression [83, 84] whereas inhibition of NF-κB decreases levels of VEGF [85].

Another therapeutic option are statins, as some of these agents have been found to have anti-inflammatory effects by inhibiting NF-κB gene expression [86, 87]. Statins are well-established drugs with comparably few side effects and have already been shown to reduce endothelial activation both in vitro and in vivo [88, 89]. However, to our knowledge, there have been no studies in humans to test the hypothesis that statins can decrease the risk of radiotherapy-induced vasculopathy.

A third approach, which has already been used in an attempt to decrease healthy tissue injury caused by radiotherapy, is treatment with adult stem cells. This can be in the form of bone marrow transfusion (i.e. haematopoietic plus mesenchymal stem cells), mesenchymal stem cells or the mobilization of autologous stem cells into the circulation by growth factors such as granulocyte colony-stimulating factor. These strategies have already been tested in preclinical models of radiation injury in skin, salivary glands, intestine and oral mucosa [90], but may be extended to include the vasculature, as mobilization of endothelial progenitor cells has been shown to restore endothelial function [91]. This hypothetical role of endothelial progenitor cells in endothelial radiation injury is supported by the radiation response that is seen in skin and gut, in which there is a compensatory proliferation for cell death, caused by radiotherapy, within local stem cell compartments.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References

With an increasing number of cancer survivors and the prevalent use of radiotherapy, cardiovascular disease caused by radiotherapy will increase. It is important to understand the molecular mechanisms behind radiotherapy-induced vasculopathy to identify preventive strategies. Risk factors in common for cancer and CHD, such as diabetes, hyperlipidaemia, hypertension and smoking, increase the risk for CHD after radiotherapy and measures should be taken to reduce these factors. Such a strategy together with careful radiotherapy planning to minimize radiation to the heart and blood vessels is likely to decrease the risk of CHD in vulnerable patients. With an increased knowledge of the molecular mechanisms behind vasculopathy induced by radiotherapy, new more selective preventive strategies can be developed to minimize the risk of future adverse CHD. In summary, there are several therapeutic possibilities that may decrease the adverse effects of radiotherapy, and targeting NF-κB seems to be of particular interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patient studies
  5. Molecular mechanisms
  6. Future perspectives and treatment options
  7. Conclusion
  8. Conflict of interest statement
  9. References