Clinical applications of PET imaging for evaluating treatment-induced vascular toxicity in oncology

Abstract

Purpose of review

Vascular toxicity resulting from cancer and its treatment remains a largely uncharacterized and minimally treated pathology. This review highlights ongoing developments and applications of PET imaging for noninvasive characterization of cancer-associated vascular toxicity and discusses the potential prognostic value of PET imaging for predicting adverse cardiovascular outcomes in patients with cancer.

Recent findings

Numerous clinical investigations have used PET imaging to evaluate vascular inflammation/toxicity in patients undergoing or completing cancer therapies, including chemotherapy, immune checkpoint inhibitors, and radiation therapy. The most widely utilized PET radionuclide for noninvasively detecting vascular toxicity in clinical oncology has been fluorine-18-fluorodeoxyglucose, which has demonstrated promising associations with disease severity and clinical outcomes. Emerging radionuclides continue to be developed for targeting immune cells and may increase the sensitivity and specificity of PET imaging for detecting vascular toxicity associated with cancers and their treatment.

Summary

The use of PET imaging for noninvasive detection and quantification of cancer treatment-associated vascular toxicity continues to evolve and could provide a unique approach for predicting risk of adverse cardiovascular outcomes in various forms of cancer and treatment.

INTRODUCTION

The impact of anticancer therapies on the development of cardiotoxicity has been well characterized using imaging modalities such as echocardiography [1]; however, there remains a lack of comprehensive reports on the use of noninvasive imaging techniques to specifically detect, quantify, and predict vascular toxicity induced by cancer and therapeutic strategies for cancer. PET is the primary nuclear medicine imaging modality used for cancer diagnosis and monitoring and allows for quantitative assessment of various physiological parameters such as organ perfusion and tumor metabolism through the use and imaging of targeted radionuclides [2]. This review provides an overview of the processes underlying healthy vascular function, describes the mechanisms by which cancer therapies induce vascular toxicity and related complications such as venous thromboembolism (VTE), and summarizes the clinical studies to date that have incorporated PET imaging for noninvasive assessment of vascular toxicity in patients with cancer. 

Box 1.

no caption available

PATHOGENESIS AND MANIFESTATIONS OF CANCER THERAPY ASSOCIATED VASCULAR TOXICITY

Many cancer-targeting chemotherapies have been known to induce vascular dysfunction in patients, leading to impaired vascular function and clinical manifestations such as VTE, hypertension, vasospastic events, and peripheral ischemia [2,3]. Different subclasses of chemotherapies have various risks, including cardiac dysfunction with the use of anthracyclines, alkylating agents and antimetabolites, ischemia with antimetabolites and microtubule inhibitors, and thromboembolic events with alkylating agents [5–7].

Mechanistically, anticancer therapies induce direct endothelial damage, including DNA damage, disrupted signaling, oxidative stress and altered gene expression, and indirect systemic effects such as cell damage, inflammation, and immune system activation [5]. The vascular endothelium regulates vasodilation through the synthesis of nitric oxide by endothelial nitric oxide synthase, which in turn activates smooth muscle cells and promotes vasorelaxation [5,8–11]. When the endothelium is damaged, reactive oxygen species are upregulated, thereby reducing nitric oxide bioavailability and leading to impaired vasodilation [12].

Several mechanisms involved in thrombosis may also be attributed to the endothelium. Thrombus formation is largely driven by platelets and plasma proteins such as fibrinogen and von Willebrand factor (VWF). Nitric oxide released by endothelial cells inhibits platelet aggregation and adhesion, two key steps in thrombus formation, by diffusing into platelets and preventing the release of intracellular calcium, thereby inhibiting platelet activation [13]. Additionally, during inflammation, the release of cytokines such as tumor necrosis factor and interleukin-1 triggers increases in plasma concentrations of procoagulant proteins, namely, VWF, E-selectin, platelet activating factors, and tissue factor. Inflammatory cytokines also raise plasma concentrations of soluble anticoagulant thrombomodulin, which implies the loss of this protein from the endothelial surface and encourages thrombus adhesion to the vessel wall [14–16].

In an attempt to repair the damaged endothelium, monocytes in the blood interact with endothelial cells to facilitate inflammation, angiogenesis, and tissue remodeling [4]. Specifically, when endothelial damage occurs, the local vasculature becomes highly permeable and adhesive, and increased expression of chemokines recruits leukocytes, neutrophils, and macrophages to the injured site. The high permeability allows recruited immune cells to infiltrate into injured sites to phagocytose necrotic cells and pathogens [17–19]. The infiltrated macrophages further facilitate tissue remodeling by releasing factors such as matrix metalloproteinase to reconstruct the extracellular matrix and stimulate smooth muscle cell and endothelial cell proliferation [20].

The cellular and molecular pathologies in the vasculature associated with chemotherapy can clinically manifest as conditions, including VTE (i.e., deep vein thrombosis and pulmonary embolism), hypertension, and heart failure [3]. Recent studies have found that anthracyclines, including doxorubicin, induce cardiovascular dysfunction at cumulative doses lower than 300 mg/m²[21–24] and promote reductions in brachial artery flow-mediated dilation, a surrogate marker of peripheral vascular endothelial function [25,26]. Further supporting this point, a recent longitudinal study in breast cancer patients who were undergoing anthracycline-based chemotherapy revealed that over a 15-month treatment period pulse wave velocity (i.e., vascular stiffness) increased by 10% and flow-mediated dilation (i.e., endothelial function) declined by over 25% [27]. Beyond impairment of both cardiac and vascular function, anticancer therapies, including radiation therapy, have also been shown to accelerate the process of atherosclerosis through the promotion of reactive oxygen species and dyslipidemia, thereby reducing endothelial function, stimulating lipid accumulation in arterial walls, increasing systemic vascular inflammation, and increasing deposition of calcium-rich arterial plaques [28–30].

In addition to disruption of vascular function and acceleration of atherosclerotic disease processes, chemotherapy-induced VTE also remains a critical concern for patients with cancer, with VTE representing a leading cause of death in patients undergoing cancer treatment [31,32]. Among the commonly reported thrombotic events that may occur during cancer treatment [29], patients are at particularly elevated risk for cerebrovascular events such as cerebral venous thrombosis, intracranial hemorrhage, and ischemic stroke [33–50]. With an established risk of cardiovascular disease and adverse vascular events in patients with cancer, there remains an urgent need for clinical practice guidelines that describe appropriate dosing and timing of prophylactic anticoagulation to prevent future VTE events [51], in addition to creation of robust prognostic models and long-term monitoring strategies to mitigate risk of future cardiovascular events [28,52]. To this end, there is a growing body of clinical research that has begun to leverage PET-based molecular imaging strategies to evaluate the occurrence of vascular toxicity and predict risk of vascular events in patients with cancer.

PET IMAGING OF VASCULAR INFLAMMATION

In recent decades, hybrid PET/computed tomography (CT) imaging with ¹⁸F-fluorodeoxyglucose (FDG) has emerged as an approach for evaluating vascular inflammation induced by various cancers and their treatment. ¹⁸F-FDG is a glucose analog that accumulates in metabolically active cells, and due to cancer cells possessing high metabolic activity, ¹⁸F-FDG uptake has traditionally been utilized to distinguish benign from malignant tumors and therefore become an indispensable tool in the diagnosis, staging, and monitoring of many cancers [53]. In addition to being a radionuclide for metabolic imaging in cancer, ¹⁸F-FDG has also been shown to accumulate in metabolically-active macrophages and monocytes that have infiltrated the walls of injured or inflamed vessels [54], thereby making ¹⁸F-FDG PET/CT imaging a promising complementary biomarker for assessing endothelial dysfunction and predicting risk of cardiovascular events in patients with cancer [55].

¹⁸F-FDG PET/CT imaging has demonstrated promise as an approach for evaluating vascular inflammation/toxicity in several cancer types as well as in response to various therapeutic interventions for cancer treatment, which are summarized in Table 1. Although most clinical investigations to date using ¹⁸F-FDG PET/CT imaging have assessed cancer-associated vascular inflammation in large arteries such as the aorta and carotid arteries, recent work by Beall and colleagues [56] alternatively quantified serial changes in ¹⁸F-FDG uptake in the femoral and popliteal veins of pediatric, adolescent and young adult patients during lymphoma treatment. This work demonstrated both qualitative and quantitative changes in peripheral vein ¹⁸F-FDG uptake in response to the first round of cancer treatment, with ¹⁸F-FDG uptake patterns significantly differing between patients who experienced VTE events and those who remained VTE-free in the first year after lymphoma diagnosis, suggesting that ¹⁸F-FDG PET/CT imaging of venous inflammation may provide prognostic value for predicting risk of VTE in young patients undergoing cancer treatment (Fig. 1).

Table 1.

PET imaging studies evaluating cancer treatment-induced vascular toxicity

References	Vascular target	Clinical condition	Therapeutic intervention	Radionuclide
Beall et al. [56]	Femoral and popliteal vein	Pediatric, adolescent, and young adult lymphoma	Not defined	18F-FDG
Chen et al. [59]	Carotid artery	Head and neck cancer	Radiation therapy	18F-FDG
Xie et al. [57▪]	Aortic wall	Diffuse large B-cell lymphoma	Rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP)	18F-FDG
Evanson et al. [62]	Ascending and descending aorta, and aortic arch	Nonsmall cell lung cancer	Proton and photon radiation therapy and concurrent chemotherapy	18F-FDG
Ripa et al. [61]	Carotid and iliac arteries	Lymphoma	Radiation therapy	18F-FDG
Lawal et al. [63]	Carotid and femoral artery, ascending and abdominal aorta	Hodgkin lymphoma	Doxorubicin, bleomycin, vinblastine, and dacarbazine	18F-FDG
Liriano et al. [58▪]	Ascending aorta	Lung cancer, melanoma, genitourinary cancer, esophageal cancer, head and neck cancer, breast cancer	Pembrolizumab (immune checkpoint inhibitor)	18F-FDG
Rankin et al. [64▪]	Thoracic aorta	Diffuse large B-cell lymphoma	Doxorubicin	18F-FDG
Vlachopoulos et al. [65]	Aorta	Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL)	HL: doxorubicin, bleomycin, vinblastine, dacarbazine; NHL: rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine, prednisolone	18F-FDG
Borja et al. [60]	Carotid artery and aortic arch	Head and neck cancer	Radiation therapy	18F-FDG
Raynor et al. [54]	Aorta and coronary artery	Prostate cancer	Not defined	18F-NaF

FIGURE 1.

¹⁸F-FDG PET/CT imaging of chemotherapy-induced venous toxicity in pediatric lymphoma. Pictured is (a) a 7-year-old patient who demonstrated a high inflammatory response to the first cycle of lymphoma treatment (i.e., 11% increase in femoral and 49% increase in popliteal vein) and experienced VTE of the brachiocephalic vein 45 days after diagnosis, and (b) a 7-year-old patient who alternatively demonstrated a negative inflammatory response (i.e., 2% decrease in femoral and 13% decrease in popliteal vein) and remained VTE event-free for 1 year after lymphoma diagnosis. Veins of interest are denoted by white arrows. ¹⁸F-FDG, fluorine-18-fluorodeoxyglucose; PET/CT, PET/computed tomography; VTE, venous thromboembolism. Adapted from [56].

In clinical investigations focused on evaluating treatment-induced changes in arterial inflammation with ¹⁸F-FDG PET/CT imaging, Xie et al. [57^▪] analyzed PET/CT data from patients with diffuse large B-cell lymphoma who received six cycles of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone combination therapy. In comparison to a control cohort, aortic uptake of ¹⁸F-FDG was found to be significantly higher pretreatment in patients with lymphoma. Additionally, aortic uptake of ¹⁸F-FDG was found to be significantly higher post versus pretreatment. Aside from lymphoma treatment, Liriano et al. [58^▪] analyzed the effect of pembrolizumab, an immune checkpoint inhibitor, and demonstrated increases in arterial inflammation in response to therapy using ¹⁸F-FDG PET/CT imaging. In the evaluation of potential radiation therapy-induced changes in arterial inflammation in patients with head and neck cancer, Chen et al. [59] found significantly higher uptake of ¹⁸F-FDG in the carotid arteries three months after completion of chemoradiation therapy, while Borja et al. [60] showed a marked increase in ¹⁸F-FDG uptake in the carotid artery and aortic arch postradiation therapy. As an extension of these studies in head and neck cancers, Ripa et al. [61] found that arterial uptake of ¹⁸F-FDG uptake was associated with intima media thickness multiple years after completion of radiation therapy in survivors of lymphoma, suggesting that ¹⁸F-FDG PET imaging may detect radiation-induced changes in vascular remodeling. Further supporting this notion, Evanson et al. [62] found that photon radiotherapy resulted in significant increases in ¹⁸F-FDG uptake in the aorta and aortic arch of patients with nonsmall cell lung cancer who underwent treatment with concurrent chemotherapy. Collectively, all aforementioned studies demonstrated that ¹⁸F-FDG uptake may be indicative of radiation therapy-related vasculitis [59–62]. These similar observational findings of increasing arterial uptake of ¹⁸F-FDG in patients undergoing various forms of treatment also highlights how ¹⁸F-FDG PET/CT imaging may serve as a broadly translatable approach for monitoring treatment-induced vascular toxicity across a variety of cancers.

In contrast to other studies that have revealed increasing patterns of ¹⁸F-FDG uptake with cancer treatment, investigations by Lawal et al. [63] and Rankin et al. [64^▪] alternatively found no significant change in arterial uptake of ¹⁸F-FDG when PET/CT imaging was performed at 65 weeks or one month after completion of chemotherapy regimens, thus indicating that timing of PET scans may be critical for detecting therapy-induced changes in arterial inflammation. In agreement with these findings, work by Vlachopoulos et al. [65] also revealed that patients with Hodgkin lymphoma exhibited decreases in arterial uptake of ¹⁸F-FDG after completion of chemotherapy. While a decrease in arterial inflammation may contradict the findings of most cancer studies, one proposed hypothesis is that this may be due to many patients undergoing cancer treatment receiving prednisone, a steroid that could have anti-inflammatory properties on the vasculature.

Unlike the previously described studies that have utilized ¹⁸F-FDG for evaluating vascular inflammation, a recent clinical investigation by Raynor et al. [54] evaluated the utility of ¹⁸F-sodium fluoride (NaF), a radionuclide with a high binding affinity to hydroxyapatite, for noninvasively detecting and quantifying increasing levels of active arterial microcalcification in the coronary arteries and aortas of patients with prostate cancer. This study demonstrated increased uptake of ¹⁸F-NaF in arteries of interest when compared to healthy control subjects and provides supporting evidence for an accelerated atherosclerotic disease process in the setting of cancer. Additionally, the findings of this study showing increased arterial uptake of ¹⁸F-NaF suggests that there may be a need for more targeted radionuclides as well as more robust biological targets beyond glucose metabolism and inflammation to better understand and quantify cancer-associated vascular pathologies with higher precision and accuracy.

NOVEL PET RADIONUCLIDES FOR IMAGING VASCULAR INFLAMMATION

¹⁸F-FDG PET/CT imaging is a widely used tool in clinical oncology for diagnosis and staging, but due to the limited specificity of ¹⁸F-FDG for targeting inflammation, novel radionuclides are continually in development and testing for molecular imaging that may provide a more targeted approach in the future evaluation of cancer-associated vascular toxicity [66]. One such target of interest is CD45, a surface antigen expressed by all white blood cells [67]. CD45 nanobodies (small antibody-based constructs) have recently been conjugated to the radioisotope zirconium-89 (⁸⁹Zr) for targeting of CD45⁺ cells [67] and have demonstrated potential in preclinical models of acute lung injury and inflammatory bowel disease [68^▪▪]. Another promising target is CD20, a marker of B lymphocytes. Rituximab, an antibody used as a therapeutic agent against CD20, has also been radiolabeled with ⁸⁹Zr and has shown promise for targeting of inflammatory responses to rheumatoid arthritis treatments and diagnosis of ocular inflammatory diseases [69,70]. In the context of vascular inflammation, gallium-68 (⁶⁸Ga)-DOTATATE has revealed potential for targeting atherosclerosis-induced arterial inflammation due to its high binding affinity to somatostatin receptor 2 (SST₂), which co-localizes with CD68⁺ macrophages [71]. In addition to ⁶⁸Ga-DOTATATE PET imaging of macrophage activity, copper 64-labeled CD163-binding peptides have also demonstrated utility for targeted in-vivo imaging of macrophages in preclinical models of atherosclerosis and ex-vivo imaging of atherosclerotic human artery specimens [72]. Beyond imaging of macrophages, several additional targets that could have promise for PET imaging of vascular inflammation are actively being explored, including C-C chemokine receptor 2, T and B lymphocytes, and fibroblast-activating protein alpha [66]. Although many of the discussed emerging radionuclides for PET imaging have not been explored in the context of vascular inflammation, the continued development of such targeted tracers provides unique opportunities for future imaging of cancer-associated vascular toxicity.

FUTURE DIRECTIONS

The early use of PET imaging for identifying and monitoring the progression of vascular toxicity and predicting risk of adverse events in patients with cancer has demonstrated promise. ¹⁸F-FDG is a ubiquitous and well understood radionuclide for cancer diagnosis and staging, but more research is warranted to validate the utility of ¹⁸F-FDG as a potential biomarker of cancer-associated vascular pathologies. To support the initial promising clinical evidence of PET/CT imaging in oncology populations, additional preclinical studies are warranted to confirm co-localization of ¹⁸F-FDG with infiltrating immune cells in the setting of cancer-associated vascular toxicity. Further, to date, the bulk of the clinical studies evaluating the utility of PET/CT imaging for detecting vascular toxicity have involved patients undergoing treatment with a variety of combination therapies and anti-inflammatory drugs. Future work should identify the relative contributions of different drug classes to the development of vascular toxicity. Finally, many PET radionuclides have emerged in recent years for targeted imaging of specific immune cells. Additional investigation of these radionuclides in the setting of vascular toxicity would be beneficial for facilitating a mechanistic understanding of vascular toxicity during cancer treatment.

CONCLUSION

Anticancer therapies are known to induce damage to the vasculature that can lead to adverse outcomes such as stroke and VTE; however, there remains a limited number of clinical investigations implementing molecular imaging strategies to identify and monitor the development of cancer-associated vascular toxicity. With the recent clinical exploration of ¹⁸F-FDG PET/CT imaging as a potential approach for quantifying vascular inflammation and continual development of novel radionuclides for targeted imaging of immune cells, a noninvasive and sensitive imaging tool may be on the horizon for oncologists that can improve the state-of-the-art detection and monitoring of vascular toxicity.

Acknowledgements

None.

Financial support and sponsorship

This work was supported by National Institutes of Health award R01 HL171715 (M.R. Stacy).

Conflicts of interest

Drs. Jayaraman and Stacy have no relevant conflicts of interest to declare for this work.

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

• ▪ of special interest

• ▪▪ of outstanding interest