Abstract
Aim: The aim of the present study was to investigate myocardial perfusion in relation to disease history and laboratory parameters of atherosclerosis risk in asymptomatic patients with breast carcinoma. Patients and Methods: One-hundred and eighty-one patients with breast carcinoma were studied. Myocardial perfusion was assessed using single-photon emission computed tomography (SPECT) with 99mtechnetium sestamibi. Results: Perfusion defects were detected in 12 patients (7%). Higher body-mass index, increased concentrations of D-dimers, C-reactive protein, fibrinogen, glucose, triglycerides, and urinary albumin, a history of hypertension and of radiotherapy to the left chest wall were all associated with increased risk of perfusion defects. In a multivariate stepwise selection logistic regression model, body mass index, albuminuria and radiotherapy to the left hemithorax were significantly associated with the presence of perfusion defects. Conclusion: In addition to other factors, treatment history may be associated with the presence of perfusion defects in patients with breast cancer.
- Atherosclerosis
- breast cancer
- single-photon emission computed tomography
Substantial progress accomplished in cancer therapy has been translated into improved survival of patients with different types of primary tumors. As the result of improvement of survival or even cure of patients with advanced cancer, some long-term sequelae have emerged, including metastases in unusual sites (1), second primary cancer (2), or atherosclerosis and associated disorders (3-5). In fact, in many patients, comorbidity rather than cancer is the ultimate cause of death (6). Advanced age, smoking, obesity and oxidative stress are all linked with increased risk of both atherosclerosis and malignant tumors (7). Consequently, cardiovascular disorders currently represent an important issue in cancer survivors. Moreover, the toxicity of anticancer therapy may result in progression of atherosclerosis (8, 9). Clinical data from retrospective series indicate increased incidence of cardiovascular disorders in survivors of childhood cancers or germ-cell tumors to be associated with a history of chemotherapy (3-5). The information on the prevalence of complications of atherosclerosis is limited for most of the common cancer types (7, 10-12). For patients with breast carcinoma, different reports have indicated both increased and decreased incidence of complications of atherosclerosis (13, 14).
Much effort has been devoted in the past several decades to identify biomarkers of risk of atherosclerosis, and many laboratory parameters, e.g. cholesterol, homocysteine or C-reactive protein (CRP) (15), have been shown to predict cardiovascular events associated with atherosclerosis. Strategies have also been developed for early diagnosis of asymptomatic cardiovascular disease. The identification of patients at increased risk or early diagnosis of asymptomatic disease opens the way for interventions that could prevent cardiovascular events. Stress/rest myocardial perfusion scan using single-photon emission computed tomography (SPECT), and measurement of carotid intima media thickness have been the most widely used diagnostic tests in asymptomatic individuals. Stress/rest myocardial perfusion has been demonstrated to predict cardiovascular events in asymptomatic patients with different risk of coronary artery disease (16, 17).
In an earlier investigation, we reported that in patients with breast cancer, the intima media thickness, an indicator of the presence of atherosclerosis, is associated not only with clinical and laboratory risk factors, but also with the history of anticancer therapy (18). The aim of the present study was to assess the prevalence of perfusion defects, assessed by SPECT, in asymptomatic patients with breast carcinoma in relation to disease history and laboratory risk factors of atherosclerosis in part of the same cohort of patients.
Patients and Methods
One-hundred and eighty-one female patients with histologically-verified carcinoma and no history of cardiac disorder, aged (mean±standard deviation) 53±11 (range=28-76) years, were included in the present study. The investigations were approved on April 14, 2005 by the Institutional Ethical Committee (file number 200504 S14P), and the patients signed informed consent.
Body mass index (BMI) was calculated with the formula: weight (kg)/height (m)2. The menopausal status, history of breast cancer, including the time from diagnosis, earlier radiotherapy, chemotherapy, or the presence of distant metastases, history of smoking, cardiac disorder, hypertension, diabetes, disorders of lipid metabolism and thyroid disorders were recorded. Blood samples were drawn from a peripheral vein after overnight fast and processed for the determination of hemoglobin, leukocyte and platelet counts, fibrinogen, antithrombin, D-dimers, glycosylated hemoglobin, serum CRP, lipoprotein (a), serum cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, homocysteine, glucose, creatinine, magnesium, uric acid, albumin, alpha-tocopherol and retinol as described (19). Early urine samples were collected and the measurement of urinary neopterin, creatinine, N-acetyl-β-D-glucosaminidase (NAG) and albumin was performed as described earlier (19, 20). One-day stress (bicycle exercise) and rest electrocardiogram gated SPECT myocardial 99mtechnetium sestamibi perfusion scintigraphy was performed as described in an earlier report (19).
Differences between groups of patients were analyzed by the Mann–Whitney U-test. Categorical data were analyzed with univariate logistic regression. Multivariate analysis was performed using stepwise selection logistic regression. The decision on statistical significance was based on p=0.05 level. The analyses were performed using SAS software (Version 9.2; SAS Institute, Cary, NC, USA).
Results
The median time from diagnosis for the present cohort was 18 months (range=0-96 months). Perfusion defects were detected in only 12 patients (7%), including reversible ischemia in 11 cases and scar in one patient. Patients with perfusion defects had significantly higher BMI, fibrinogen, glucose and urinary albumin concentrations (Table I). The continuous clinical and laboratory parameters were then dichotomized based on the upper limit of normal [D-dimers, fibrinogen, glucose, uric acid, total cholesterol, LDL cholesterol, lipoprotein (a) and CRP] or lower limit of normal (serum albumin), or, in the case of parameters where fewer than 10% of values were above (or below) the normal range or the normal range was difficult to define, above the third quartile (age, BMI, leukocyte and platelet counts, antithrombin, creatinine, triglycerides, homocysteine, glycosylated hemoglobin, retinol and alpha-tocopherol, urinary NAG, albumin and neopterin) and below the first quartile (hemoglobin, magnesium and HDL cholesterol).
After this dichotomization, in addition to increased BMI, fibrinogen, glucose and urinary albumin, increased D-dimers, CRP and triglycerides were also associated with increased risk of perfusion defects (Table II). The decreased risk associated with higher HDL cholesterol concentrations was of borderline significance. The distribution of perfusion defects in subgroups of patients defined by categorical variables is shown in Table III. Fifty-eight patients had previous radiotherapy of the left hemithorax, and perfusion defects were detected more frequently in these patients. Perfusion defects were also more common in patients with history of hypertension (Table III).
Because the significance of some parameters in the univariate analysis was borderline around p=0.05, the threshold p-value in a multivariate stepwise selection logistic regression model (where the choice of predictive variables was carried out by an automatic procedure from all variables) was set at p=0.07. In this model, BMI ≥29.76 mg/m2, albuminuria ≥0.88 g/mol creatinine and radiotherapy to the left hemithorax were significantly associated with the presence of perfusion defects (p=0.005, p=0.034, and p=0.031, respectively), while the significance of alpha-tocopherol ≥21.18 μmol/l was of borderline statistical significance (p=0.068; Table IV).
Discussion
Present data indicate a relatively low prevalence of myocardial perfusion defects in asymptomatic patients with breast cancer during the first decade after diagnosis. It has been demonstrated that SPECT myocardial perfusion predicts cardiac events in patients with cancer (21). The presence of perfusion defects in patients with breast cancer after anticancer therapy has been well-documented (22-25). Considering the fact that a substantial proportion of patients in the present cohort were pre-treated with systemic chemotherapy and irradiation of the left chest wall, the incidence of myocardial perfusion defects may seem to be rather low. In previous studies, perfusion defects were reported in up to 24% of patients treated with chemotherapy and left-sided irradiation (23), and in other series, the rate of perfusion defects was even higher (24, 25). Perfusion defects have also been reported in patients with other primary tumors treated with the combination of chemotherapy and thoracic radiation therapy (26). The lower incidence of perfusion defects in the present cohort is probably due to the short duration of follow-up after therapy. Left-sided irradiation has been shown to result in an increased rate of cardiovascular events (27, 28). However, these events manifest late, and maximum increased risk is observed after 20 years (28). From this perspective, the duration of follow-up in the present cohort may be too short.
Because of the small number of patients with perfusion defects in the present study, only limited information may be derived regarding the risk factors for cardiovascular disease in this population. In addition to the risk factors of atherosclerosis that are common in the general population, in patients with history of breast cancer the parameters associated with therapy may also determine the presence of perfusion defects. As expected, the BMI was significantly associated with the presence of perfusion defects in multivariate analysis. The presence of perfusion defects also seems to be associated with previous irradiation of the left chest wall. Albuminuria was also an independent predictor of the presence of perfusion defects in the present study. Both BMI and albuminuria have been reported to increase in cancer survivors as a result of prior treatment, including chemotherapy (8, 29).
Most of the data on increased incidence of cardiovascular events associated with atherosclerosis in patients with cancer were reported for pediatric or young adult patients surviving long after the cure to allow for the manifestation of these chronic complications (3-5, 8). The data on atherosclerosis in survivors of more common adult solid tumors, including breast cancer, are limited and sometimes even conflicting. While the risk of stroke was reported to be increased (14), the risk of myocardial infarction or coronary artery disease was significantly lower in patients with early breast cancer treated with radiation therapy (13).
In a review on the topic of coronary artery disease in patients with cancer published in 1978, Kopelson and Herwig noted that coronary artery disease is rare in patients with cancer (10). Obviously, this review covered an era when no metastatic cancer could be cured (10). Moreover, in earlier studies, some complications of atherosclerosis, e.g. coronary artery disease, might have appeared to be less common in patients with cancer due to the bias in selecting the control group. However, an effect of chest radiotherapy, notably when combined with chemotherapy, was noted even in early reports. Ogawa et al. in an autopsy series of 1,642 patients with cancer found an incidence of myocardial infarction of 6.5% (7). Compared to other patients with cancer, significantly higher incidence of myocardial infarction was found in patients with lung carcinoma, head and neck carcinoma and urothelial cancer. Risk factors associated with coronary artery disease in this population were smoking, hypertension and hyperlipidemia. Increased incidence of acute myocardial infarction in patients with lung carcinoma compared to other tumors was also reported by Fujiwara et al. (11). Pehrsson et al. took an opposite approach and examined cancer incidence in patients treated for acute myocardial infarction, angina pectoris and intermittent claudication (12). The incidence of cancer was slightly but significantly increased in this population, mostly due to increased incidence of tobacco-related cancer.
In addition to hypertension or hypercholesterolemia, anticancer therapy or advanced cancer itself may affect some more recently discovered laboratory parameters associated with the progression of atherosclerosis, including systemic inflammatory response. Atherosclerosis is currently considered to be an inflammatory disorder (30). Systemic inflammatory response resulting from chronic infections, reflected in increased serum concentrations of CRP, has been postulated to play an important role in the progression of atherosclerosis. CRP is increased in patients with advanced cancer as well as in patients with atherosclerosis. It has been demonstrated in numerous studies that parameters of systemic inflammatory response, e.g. serum CRP, are predictive of the risk of cardiovascular events (15, 31-33). Neopterin, a biomarker of systemic immune response produced by activated macrophages is increased in patients with acute myocardial infarction (34), and increased neopterin concentrations are associated with cardiovascular and all-cause mortality (35, 36). Moreover, high neopterin concentrations are linked to mortality in the elderly (36). In the present study, increased CRP was associated with the risk of perfusion defects in univariate, but not multivariate analyses. In contrast, no association between urinary neopterin concentrations and perfusion defects was observed.
In conclusion, perfusion defects suggestive of myocardial ischemia are rare in asymptomatic patients with breast cancer during the first decade after diagnosis. The presence of perfusion defects may be associated with higher BMI, albuminuria and history of radiotherapy to the left hemithorax.
Acknowledgements
This study was supported by grants of the Internal Grant Agency of the Czech Republic NR9096, NT/13564, and project Biomedreg CZ.1.05/2.1.00/01.0030.
- Received February 28, 2014.
- Revision received May 14, 2014.
- Accepted May 16, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved