Practical Radiation Oncology (2013) 3, e149–e155
www.practicalradonc.org
Original Report
Women at increased risk for cardiac toxicity following chemoradiation therapy for esophageal carcinoma Lauren M. Tait MD a,⁎, Joshua E. Meyer MD b , Erin McSpadden MPH a , Jonathan D. Cheng MD c , Frank A. Baciewicz MD d , Neal J. Meropol MD e , Steven J. Cohen MD c , Antoinette J. Wozniak MD f , Minsig Choi MD f , Andre A. Konski MD a a
Department of Radiation Oncology, Wayne State University/Barbara A. Karmanos Cancer Center, Detroit, Michigan Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania c Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania d Thoracic Surgical Oncology, Wayne State University/Barbara A. Karmanos Cancer Center, Detroit, Michigan e Division of Hematology and Oncology, University Hospitals/Case Comprehensive Cancer Center, Cleveland, Ohio f Division of Hematology and Oncology, Wayne State University/Barbara A. Karmanos Cancer Center, Detroit, Michigan b
Received 8 June 2012; revised 23 January 2013; accepted 6 February 2013
Abstract Purpose: The purpose of this study was to identify factors associated with cardiac toxicity in patients treated with chemoradiation therapy (CRT) for esophageal carcinoma. Methods and Materials: One hundred twenty-seven patients with adenocarcinoma or squamous cell carcinoma of the esophagus treated from July 2002 to June 2011 at 2 academic institutions with preoperative or definitive CRT were retrospectively reviewed. Association of cardiac toxicity with a number of variables was investigated, including heart disease, cardiac bypass and angioplasty, diabetes, insulin use, smoking, chemotherapy regimen, and tumor location. T test assessed risk of cardiac toxicity secondary to age. Dose volume histograms (DVH) were evaluated for percentage of heart volume receiving N 20, 30, 40, and 50 Gy (V20-V50). The Fisher exact test analyzed for an association between dose volume parameters and cardiac toxicity. Results: Patient population included 100 men and 27 women with a mean age of 64 years. Median follow-up was 12.7 months (range, 0.3-99.6 months). Any cardiac toxicity occurred in 28 patients, the majority of which were pericardial effusion (23/28). Odds ratio for toxicity in women was 4.15 (95% confidence interval [CI], 1.63-10.50; P = .0017) and time to cardiac toxicity by sex was significant (P = .0003). Patients above the median cutoff for V20, V30, and V40 had increased odds of developing cardiac toxicity (P = .03, .008, .002). There was 4.0 increased odds of developing cardiac toxicity with V40 N 57% (95% CI, 1.5-10.3, P = .002). On multivariable logistic regression analysis, sex was the only variable associated with any cardiac toxicity and pericardial effusion (P = .0016, P = .0038). None of the other investigated variables were associated with increased risk of cardiac toxicity. Abstract presented at American Society of Clinical Oncology Gastrointestinal Symposium, January 19-21, 2012, San Francisco, CA. Conflicts of interest: None. ⁎ Corresponding author. Department of Radiation Oncology, 4501 X Street, Suite G-140, Sacramento, CA 95816. E-mail address: [email protected] (L.M. Tait). 1879-8500/$ – see front matter © 2013 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.prro.2013.02.001
e150
L.M. Tait et al
Practical Radiation Oncology: October-December 2013
Conclusions: Female patients and dose greater than the median for V20-V40 were associated with the development of cardiac toxicity, specifically pericardial effusion. These data suggest exercising increased care when designing radiation fields in women undergoing CRT for esophageal carcinoma, as pericardial effusion may be a long-term complication. © 2013 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.
Introduction Radiation therapy in patients treated for breast cancer and Hodgkin lymphoma has been validated as a significant contributing factor for increased risk of cardiac toxicity in numerous previous studies. 1–13 Potential for risk of cardiac injury has been correlated with dose, fraction size, and volume of heart irradiated. 2,7,9,10,12 These findings have contributed to awareness of potential cardiac toxicity following radiation to the thorax for other malignancies, particularly esophageal carcinoma. An estimate of cardiac dose is achievable through the analysis of dose-volume histograms (DVHs). Radiation dose to the heart extrapolated from DVHs has been associated with cardiac toxicity in patients treated for breast cancer and lymphoma. 4,7,9 Milano et al 14 arrived at normal tissue radiation dose constraints for the heart based upon published literature from this collection of patients. An analysis of patients with Hodgkin lymphoma treated with radiation revealed there was an increased risk of developing pericarditis when either the total dose to the heart exceeded 41 Gy or when the radiation dose per fraction was greater than 3 Gy. 9 Martel et al 15 were the first to validate similar findings in a retrospective review of 57 patients with esophageal cancer. Their study evaluated patients treated with nonstandard radiation fractionation schedules and reported 5/57 patients developed nonmalignant pericardial effusions after treatment with chemoradiation therapy (CRT). After correcting for fraction size, a mean and maximum dose to the heart of 27.1 and 47.0 Gy, respectively, was predictive of pericarditis (P = .014). 15 More recently there have been several retrospective analyses of esophageal carcinoma patients treated with CRT. These studies have conflicting outcomes with regard to correlation of heart dose and treatment-associated cardiac toxicity. In a study by Wei et al, 16 patients treated definitively using standard fractionation showed the rate of pericardial effusion to be 73% in a subset of patients who had greater than 46% of their heart volume receiving 30 Gy (V30). On the contrary, a study by Tripp et al 17 did not observe any statistical or clinical significance associated between radiation-dose to the heart, left ventricle, or left anterior descending artery. Tripp and colleagues did observe a short-term decrease in cardiac ejection fraction; however, the magnitude of change was not significant. A small retrospective review by Mukherjee et al 18 supports the relevance of cardiac function preservation in patients who undergo CRT in the preoperative setting. They evaluated posttreatment cardiac ejection fraction and
determined a decreased ejection fraction in 12 of 15 patients assessed (P = .003), but their small sample size was a crucial limiting factor. The discordance in the published data has yet to isolate a critical dose-volume constraint for the heart or patient characteristics associated with an increased risk for cardiac injury with esophageal CRT. The purpose of this study is to identify factors associated with cardiac toxicity in the largest cohort of patients treated with CRT for esophageal carcinoma. Our aims are to characterize the frequency of cardiac toxicity in patients with esophageal cancer treated with standard fractionation CRT and explore the relationships between dose and volume of heart irradiated with cardiac toxicity in this population.
Methods and materials The charts of 144 patients were retrospectively reviewed from 2 academic institutions. Patients evaluated had a pathologically confirmed diagnosis of esophageal carcinoma and received neoadjuvant, definitive, or adjuvant radiation therapy in either institution from July 2002 through June 2011. Of these patients, 127 were eligible for our study; ineligible patients had incomplete DVH data, received a palliative radiation dose, or had previous thoracic radiation. Institutional review board approval was received from both academic centers for this study. Patients underwent a complete history and physical examination prior to treatment. Pretreatment diagnoses of cardiac comorbidities, including hypertension, dyslipidemia, heart failure, myocardial infarction, arrhythmia, cardiomyopathy, or valvular disease were recorded. Incidence of diabetes mellitus, insulin dependence, and smoking history was also recorded. All patients had a pathologic diagnosis of primary squamous cell carcinoma or adenocarcinoma of the esophagus and staging was performed clinically or surgically.
Treatment The majority of patients received combined modality therapy with concurrent CRT. Chemotherapy regimen was administered at the discretion of the treating medical oncologist. Three-dimensional computed tomography (CT) scans were utilized in all radiation treatment plans. Patients were simulated in the supine position, with their arms above their head and were secured with an
Practical Radiation Oncology: October-December 2013
immobilization device. CT scans were performed with normal free breathing in 3-mm slices from the mandible to the upper border of the pelvis. The treating radiation oncologist determined the gross esophageal tumor on the planning CT scan from pretreatment endoscopic ultrasound, diagnostic CT scan, and positron emission tomography scan. A 1-cm circumferential and 3.5-cm superior and inferior expansion were applied, forming the clinical target volume (CTV). Finally, to account for daily setup error, a 1.5-cm expansion from the CTV in all directions formed the planning target volume (PTV). Any involved locoregional lymph nodes were also included in the CTV and a 1- to 2-cm expansion formed the PTV. A commercial treatment planning system was used to develop 3-dimensional conformal or intensity modulated radiation therapy treatment plans, which restricted the maximum spinal cord dose to 45 Gy. Radiation treatment dose was prescribed by the treating physician and ranged from 45 to 59.4 Gy; the median dose was 50.4 Gy in 1.8 Gy fractions.
Dosimetric analysis A single radiation oncologist was responsible for contouring the external heart border on acquired treatment CT scans to control measured cardiac dose. The dose distributions and DVHs of the heart were calculated using the treatment planning system. The volume of the heart receiving at least 20, 30, 40, and 50 Gy (V20, V30, V40, and V50) was recorded for each patient.
Toxicity The primary investigator and co-investigators prospectively recorded the incidence of cardiac toxicity and associated symptoms during patients’ follow-up; if toxicity was reported on an imaging report it was verified on the corresponding CT scan or x-ray. The Radiation Therapy Oncology Group (RTOG) Late Radiation Morbidity Scoring Scheme and the Common Toxicity Criteria Adverse Events (CTCAE), version 3.0, were used to measure cardiac toxicity. Association of cardiac toxicity with a number of clinical variables was investigated, including history of heart disease, cardiac bypass or angioplasty, diabetes mellitus, insulin use, smoking, and tumor location.
Statistical analysis Standard descriptive statistics (with 95% 2-sided confidence intervals [CI]) were used to characterize the study population. We calculated the proportion of study subjects who developed cardiac events (symptomatic or asymptomatic) after the completion of CRT or radiation therapy alone. The χ 2 and Fisher exact tests were used to
Cardiac toxicity and esophageal chemo-TR
e151
test for an association between dose-volume parameters and cardiac toxicity; the median volume for V20 through V50 was used as a dividing point for our dose-volume parameters. Furthermore, the χ 2 and Fisher exact tests evaluated whether individual patient characteristics (gender, heart disease, cardiac bypass or angioplasty, diabetes, insulin use, smoking, chemotherapy regimen, and tumor location) were associated with cardiac toxicity. Logistic regression analyses and a multivariable logistic regression analysis were performed for patient characteristics and DVH parameters that were indicative of cardiac toxicity. These explanatory variables were also separately analyzed with a multivariable logistic regression for pericardial effusion, which was the most frequently observed toxicity. The time to the development of cardiac events and survival was estimated in this population using Kaplan-Meier Table 1
Patient and tumor characteristics
Characteristic Sex Male Female Histology Adenocarcinoma Squamous cell carcinoma Unknown Tumor location Middle Lower Gastroesophageal junction Unknown Preexisting heart disease Yes No Unknown Diabetes Yes No Unknown Insulin dependent Yes No Unknown Current or past smoker Yes No Unknown Previous cardiac bypass Yes No Unknown Age Mean Median Mean male Mean female
Frequency n (%) 100 (79) 27 (21) 97 (76) 23 (18) 7 (6) 23 (18) 37 (29) 61 (48) 6 (5) 71 (56) 54 (43) 2 (1) 24 (19) 101 (80) 2 (1) 5 (4) 120 (95) 2 (1) 87 (69) 38 (30) 2 (1) 15 (12) 110 (87) 2 (1) 64 y 64 y (37-87) 63.5 y (37-85) 64.5 y (42-87)
e152
L.M. Tait et al
Practical Radiation Oncology: October-December 2013
Table 2 Cardiac toxicity frequency and grade according to the Radiation Therapy Oncology Group (RTOG) Late Radiation Morbidity Scoring Scheme and the Common Toxicity Criteria Adverse Events (CTCAE), version 3.0 Cardiac toxicity
N (%)
RTOG (N)
CTCAE (N)
Sympt
Asymp
Men
Women
Angina Cardiomegaly Heart failure Myocardial infarction Pericardial effusion Sick sinus syndrome Total
1 (3.6) 1 (3.6) 1 (3.6) 1 (3.6) 23 (82.1) 1 (3.6) 28
3 1 3 3 3 3 1 3 4
3 1 2 3 1 3 1 2 3 4
1 0 1 1 5 1 9
0 1 0 0 18 0 19
1 0 0 1 13 1 16
0 1 1 0 10 0 12
(1) (1) (1) (1) (20), 4 (3) (1) (1) (24) (3)
(1) (1) (1) (1) (18), 3 (4), 4 (1) (1) (19) (1) (7) (1)
Asymp, asymptomatic; Sympt, symptomatic.
analysis. All statistical analysis was performed with SAS 9.2 software (SAS Institute, Cary, NC).
Results One hundred forty-four patients were included as part of our retrospective review; of this cohort, 127 patients met inclusion criteria. This group included 100 males and 27 females with a mean age of 64 years. The mean follow-up was 24.8 months and the median patient follow-up was 12.7 months (range, 0.3-99.6 months). Patient descriptive factors are provided in Table 1. Cardiac toxicity occurred in 28 patients (crude rate, 22%), with pericardial effusion representing 23 of these. Other toxicities included angina, cardiomegaly, heart failure, myocardial infarction, and sick sinus syndrome (Table 2). As graded by the RTOG scoring system, there were 24 grade 3 and 4 grade 4 cardiac toxicities. When characterized per the CTCAE, version 3.0, system, 7 grade 3 and a single grade 4 toxicity occurred. Median time to
cardiac event was 7.8 months, (range, 0-28.9 months). Cardiac toxicity was documented in 16 men and 12 women. In an analysis of men versus women, median time to cardiac toxicity was 6.4 and 4.1 months, respectively, and was statistically significant on Kaplan-Meier analysis (95% CI, 1.74-7.89; P = .0003 [Fig 1]). Patient demographics were analyzed for correlation to cardiac events. The Fisher exact test, comparing sex and toxicity, demonstrated female patients were 4.15 times more likely to have cardiac toxicity (95% CI, 1.63-10.50; P = .0017). The mean age of patients who had an event was 62.9 years, with a median of 63 years. There was no correlation between age and toxicity (P = .33). Other variables, including histology, tumor location, smoking history, diabetes, insulin dependence, preexisting cardiac disease, and previous cardiac bypass or angioplasty were not correlated to toxicity. Heart and left ventricle dosimetric volumes were evaluated for predictive value of cardiac toxicity. Left ventricle DVH data did not predict for toxicity and therefore was not included in our analysis (data not shown). Analysis of dosimetric factors demonstrated that a significant difference in V20, V30, and V40 existed between patients with cardiac toxicity and those without (Table 3). Patients above the median for V20 (71%) had Table 3 Dose-volume histogram parameters and cardiac toxicity analysis with the Fisher exact test analyzing the significance of the relationship between heart dose-volume histogram data and the development of cardiac events; variables assessed included volume of heart receiving 20-50 Gy (V20, V30, V40, V50)
Figure 1 Kaplan-Meier analysis of time to cardiac toxicity by sex (95% confidence interval, 1.74-7.89; P = .0003). The table beneath the graph depicts the number at risk by sex.
Variable
Median
P value
Odds ratio (OR)
Heart Heart Heart Heart
71 64.5 57 10
.03 (SS) .008 (SS) .002 (SS) .39
2.59 3.39 4.0 1.4
V20 V30 V40 V50
OR 95% confidence interval 1.1-6.3 1.3-8.7 1.5-10.3 0.62-3.4
Practical Radiation Oncology: October-December 2013 Table 4
Cardiac toxicity and esophageal chemo-TR
e153
Multivariable logistic regression analysis for statistically significant variables associated with any cardiac toxicity
Parameter
Standard estimate
Wald error
χ2
P value
Point estimate
95% Wald confidence interval
Intercept V20 V30 V40 Sex
− 2.7544 − 0.5343 0.9386 1.2043 1.6416
0.5338 0.9364 1.1607 0.8015 0.5195
26.6204 0.3255 0.6539 2.2579 9.9842
b .0001 .5683 .4187 .1329 .0016 (SS)
0.586 2.556 3.334 5.164
0.094-3.673 0.263-24.869 0.693-16.040 1.865-14.295
2.59 increased odds of developing cardiac toxicity (95% CI, 1.1-6.3; P = .03). Likewise, median V30 of 64.5% had a significant odds ratio (OR) of 3.39 for cardiac toxicity (95% CI, 1.3-8.7; P = .008). The greatest correlation with toxicity was demonstrated in which 57% of the heart received a dose of at least 40 Gy (OR, 4.0; 95% CI, 1.5-10.3; P = .002). There was no significant difference in the volume of heart irradiated when comparing V20, V30, V40, and V50 between men and women (P = .52, .65, .48, and .47). The mean and median maximum dose to the heart, or Dmax, was 52 and 53 Gy, respectively, ranging from 47 to 62 Gy. When evaluated by sex, the median Dmax in both men and women was 53 Gy as well. A multivariable logistic regression analysis was performed to investigate whether the increased odds ratio for women was a reflection of the relationships between the dose-volume effects and cardiac toxicity. Explanatory variables in the multivariable analysis included sex and DVH values associated with increased cardiac toxicity (V20, V30, and V40). Table 4 summarizes each variable and confirms that sex was independently associated with any cardiac toxicity (P = .0016). Multivariable logistic regression was also analyzed for pericardial effusion. Analyzed variables are summarized in Table 5 and demonstrate an independent likelihood of pericardial effusion associated with sex (P = .0038). Details of chemotherapy regimens are listed in Table 6; most patients received concurrent cisplatin and 5-fluorouracil (5-FU) (n = 81, 64%). Cisplatin/5-FU and paclitaxel-containing chemotherapy regimens did not significantly differ in relation to cardiac toxicity (95% CI, 0.17-2.35; P = .48). Sixty-five patients underwent surgical resection as a component of their multimodality treatment of esophageal carcinoma. Sixty-three patients had surgery following treatment with chemoradiation, 2
Table 5
patients had surgery followed by adjuvant chemoradiation, and 62 patients did not undergo surgery. Mean overall survival for all patients was observed to be 24.8 months with a median of 15 months and a range of 0-100.8 months. Mean survival for patients who experienced cardiac toxicity was 31 months. Comparison of survival by sex did not demonstrate a difference between women versus men despite an increased risk of cardiac toxicity (95% CI, 0.598-2.206; P = .89 [Fig 2]).
Discussion In this study we demonstrated that sex and cardiac dose were 2 factors associated with the development of cardiac toxicity following chemoradiation or radiation alone to the esophagus. Twenty-two percent of patients who were treated with standard fractionation radiation therapy developed some form of cardiac event, which for the majority was pericardial effusion. Additionally, cardiac toxicity was approximately 4 times more likely to occur in women. Women did not demonstrate significant variation in amount of dose or volume of heart receiving radiation and patient sex was confirmed to have an independent relationship with cardiac toxicity, specifically pericardial effusion, on multivariable analysis. Moreover, none of the other clinical variables, cardiac comorbidities, or differences in chemotherapy regimens were associated with cardiac injury. Dosimetric analysis verified patients above the median cutoff for V20, V30, and V40 had increased odds of developing cardiac toxicity. Heart volume greater than 57% receiving 40 Gy had the greatest correlation with development of cardiac toxicity. The only parameter that did not correlate with increased cardiac complications was V50. The median V50 was 10%, which was considerably lower in comparison with the median V20-V40 DVH
Multivariable logistic regression analysis for statistically significant variables associated with pericardial effusion
Parameter
Standard estimate
Wald error
χ2
P value
Point estimate
95% Wald confidence interval
Intercept V20 V30 V40 Sex
− 3.1728 − 0.7604 1.1161 1.3787 1.6724
0.6348 0.9776 1.2244 0.8848 0.5770
24.9798 0.6051 0.8308 2.4282 8.4001
b .0001 .4366 .3620 .1192 .0038 (SS)
0.467 3.053 3.970 5.325
0.069-3.176 0.277-33.648 0.701-22.484 1.718-16.501
e154
L.M. Tait et al
Practical Radiation Oncology: October-December 2013
Table 6 Chemotherapy regimens for patients with esophageal carcinoma treated with concurrent chemoradiation therapy Chemotherapy regimen
n (%)
Cisplatin, 5-FU 5-FU Paclitaxel, carboplatin Paclitaxel, cisplatin, 5-FU Paclitaxel, cisplatin 5-FU, oxaliplatin, Leucovorin Other Induction a No chemotherapy Unknown
81 10 8 7 3 2 6 3 3 7
(64) (8) (6) (5.5) (2) (1.5) (5) (2) (2) (5.5)
FU, fluorouracil. a Three patients had induction chemotherapy in addition to concurrent chemoradiation.
values (V20 71%, V30 64.5%, V40 57%). This may be a reflection of the fact that many patients were treated to 45 or 50.4 Gy and the V50 was relatively low if not zero. Therefore, it is possible that a small subset of patients had a V50 greater than the median explaining why the V50 did not demonstrate a correlation with cardiac toxicity. Previous studies conducted in esophageal CRT correlating dose-volume parameters and risk of cardiotoxicity have significant variation in dose, fractionation, radiation technique, and definition of cardiac volumes for radiation dosimetry. Given the paucity of homogenous data, the Quantitative Analyses of Normal Tissue Effects in the Clinics recommended dose constraint for a mean pericardial dose less than 26 Gy and V30 less than
Figure 2 Survival curves comparing outcomes of women versus men with a diagnosis of esophageal carcinoma treated with radiation therapy. There was no significant difference in survival by Kaplan-Meier analysis (95% confidence interval, .598-2.206; P = .89). The table beneath the graph depicts the number at risk by sex.
46% is based solely on the results from Wei et al. 16,19 Our results parallel several of their findings, including a significant median V30 associated with cardiac risk, and 5-month versus 7.8-month median time to onset of cardiac toxicity in their and our experiences, respectively. In addition, a similar crude rate of cardiac toxicity was observed; 27% versus 22% in our experience. In contrast, they investigated both dose to the pericardium and to the whole heart as individual organs at risk. They concluded that pericardial DVHs more precisely predicted pericardial effusion, and whole heart volume was unreliable. One could argue the reproducibility of contouring pericardial volumes and the practicality of repeatedly exercising this in the clinical setting is extremely unlikely. Additionally, their results were complicated by the fact that many of their dosimetric parameters were highly correlated with each other and numerous dosimetric parameters were indicative of pericardial effusion risk. 16,19 Limitations of our study include a short median follow-up of 12.7 months, which is further complicated by the overall short median survival of patients with esophageal cancer; 15 months in this cohort. Therefore, long-term cardiac complications demonstrated in breast cancer and lymphoma patients, including coronary artery sclerosis, myocardial infarction, cardiomyopathy, valvular disease, and conduction abnormalities, would typically not be observed in this narrow interval. This limitation can partially be attributed to the overall poor prognosis of esophageal cancer and suggests that future studies should more accurately analyze early toxicities such as pericardial effusion, which are more likely to occur in this short follow-up. The definition of “cardiac toxicity” in this study, as mentioned previously, was a heterogeneous group of clinical conditions. The lack of selecting for a specific toxicity obscures the etiology of radiation-induced affects as there are different radiobiologic mechanisms in myocardial infarction versus pericarditis. This may be a specification worth addressing in future studies. In addition, women were only 21% of our population study, which was also a limiting factor. However, esophageal carcinoma is 3 to 4 times more common among men than among women and it is likely that this is an accurate representation of the general patient population. Finally, the vast majority of patients received concurrent chemoradiation therapy (CCT) in our study, which may be a confounding factor attributing to cardiac toxicity. Both induction chemotherapy and CCT were identified risk factors for cardiac injury in breast cancer irradiation trials. 20,21 There have been little data to support this finding in radiation-related cardiac toxicity in patients with esophageal carcinoma receiving CRT. We did not correlate a difference in cardiac risk between cisplatin/ 5-FU and Taxol (Bristol-Meyers Squibb, New York, NY) based regimens. Likewise, other studies did not identify
Practical Radiation Oncology: October-December 2013
significant increased toxicity with induction chemotherapy or CCT. 16,17 Future studies should be directed to determine the mechanism behind the significant amount of women experiencing cardiac toxicity, specifically pericardial effusion, following CRT for esophageal carcinoma. Based on our analysis, this specific finding was independent of the dose-volume relationship with toxicity and women did not have increased comorbidities nor did they demonstrate a significant difference in the volume or dose of heart irradiated compared with men. Potential analyses should investigate the rationale behind the lower cardiac dose tolerance in women versus men. Long-term complications following chemoradiation are still a trivial setback in the bigger picture of the poor overall survival in patients with esophageal carcinoma. In our study, patients who exhibited cardiac toxicity generally had a longer survival suggesting that toxicity likely occurs in patients who live long enough to develop these complications. Nonetheless, treatment-induced toxicities are significant; with the growing utilization of CRT in the preoperative setting and improved survival with this treatment modality, reduction of long-term potential toxicities will become of utter importance. Our data confirm the necessity of exercising increased care when designing radiation fields in the treatment of esophageal carcinoma. Standard dose constraints to the volume of heart irradiated should be considered during treatment planning for patients, especially women. Studies evaluating sophisticated treatment planning with intensity modulated radiation therapy or breath-holding techniques may provide evidence demonstrating less heart dose. Additionally, further investigation is still needed to reproduce the increased morbidity demonstrated in women as well as clarify the biologic mechanism linked with this phenomenon.
References 1. Cuzick J, Stewart H, Rutqvist L, et al. Cause-specific mortality in long-term survivors of breast cancer who participated in trials of radiotherapy. J Clin Oncol. 1994;12:447-453. 2. Gagliardi G, Lax I, Ottolenghi A, Rutqvist LE. Long-term cardiac mortality after radiotherapy of breast cancer—application of the relative seriality model. Br J Radiol. 1996;69:839-846. 3. Paszat LF, Mackillop WJ, Groome PA, Schulze K, Holowaty E. Mortality from myocardial infarction following postlumpectomy radiotherapy for breast cancer: A population-based study in Ontario, Canada. Int J Radiat Oncol Biol Phys. 1999;43:755-762. 4. Clarke M, Collins R, Darby S, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local
Cardiac toxicity and esophageal chemo-TR
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
e155
recurrence and 15-year survival: an over- view of the randomised trials. Lancet. 2005;366:2087-2106. Darby SC, McGale P, Taylor CW, Peto R. Long-term mortality from heart disease and lung cancer after radiotherapy for early breast cancer: prospective cohort study of about 300,000 women in US SEER cancer registries. Lancet Oncol. 2005;6:557-565. Bouillon K, Haddy N, Delaloge S, et al. Long-term cardiovascular mortality after radiotherapy for breast cancer. J Am Coll Cardiol. 2011;57:445-452. Carmel RJ, Kaplan HS. Mantle irradiation in Hodgkin's disease. An analysis of technique, tumour eradication, and complications. Cancer. 1976;37:2813-2825. Gustavsson A, Eskilsson J, Landberg T, et al. Late cardiac effects after mantle radiotherapy in patients with Hodgkin's disease. Ann Oncol. 1990;1:355-363. Cosset JM, Henry-Amar M, Pellae-Cosset B, et al. Pericarditis and myocardial infarctions after Hodgkin's disease therapy. Int J Radiat Oncol Biol Phys. 1991;21:447-449. Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin's disease. J Am Med Assoc. 1993;270:1949-1955. Glanzmann C, Kaufmann P, Jenni R, Hess OM, Huguenin P. Cardiac risk after mediastinal irradiation for Hodgkin's disease. Radiother Oncol. 1998;46:51-62. Eriksson F, Gagliardi G, Liedberg A, et al. Long-term cardiac mortality following radiation therapy for Hodgkin's disease: analysis with the relative seriality model. Radiother Oncol. 2000;55:153-162. Aleman BM, van den Belt-Dusebout AW, Klokman WJ, Van't Veer MB, Bartelink H, van Leeuwen FE. Long-term cause-specific mortality of patients treated for Hodgkin's disease. J Clin Oncol. 2003;21:3431-3439. Milano MT, Constine LS, Okunieff P. Normal tissue tolerance dose metrics for radiation therapy of major organs. Semin Radiat Oncol. 2007;17:131-140. Martel MK, Sahijdak WM, Ten Haken RK, Kessler ML, Turrisi AT. Fraction size and dose parameters related to the incidence of pericardial effusions. Int J Radiat Oncol Biol Phys. 1998;40:155-161. Wei X, Liu HH, Tucker SL, et al. Risk factors for pericardial effusion in inoperable esophageal cancer patients treated with definitive chemoradiation therapy. Int J Radiat Oncol Biol Phys. 2008;70:707-714. Tripp P, Malhotra HK, Javle M, et al. Cardiac function after chemoradiation for esophageal cancer: comparison of heart dosevolume histogram parameters to multiple gated acquisition scan changes. Dis Esophagus. 2005;18:400-405. Mukherjee S, Aston D, Minett M, Brewster AE, Crosby TDL. The significance of cardiac doses received during chemoradiation of oesophageal and gastro-oesophageal junctional cancers. Clin Oncol (R Coll Radiol). 2003;15:115-120. Bentzen SM, Constine LS, Deasy JO, et al. Quantitative Analyses of Normal Tissue Effects in the Clinic (QUANTEC): an introduction to the scientific issues. Int J Radiat Oncol Biol Phys. 2010;76(Suppl 3): S3-S9. Shapiro CL, Hardenbergh PH, Gelman R, et al. Cardiac effects of adjuvant doxorubicin and radiation therapy in breast cancer patients. J Clin Oncol. 1998;16:3493-3501. Hardenbergh PH, Munley MT, Bentel GC, et al. Cardiac perfusion changes in patients treated for breast cancer with radiation therapy and doxorubicin: preliminary results. Int J Radiat Oncol Biol Phys. 2001;49:1023-1028.
www.practicalradonc.org
Original Report
Women at increased risk for cardiac toxicity following chemoradiation therapy for esophageal carcinoma Lauren M. Tait MD a,⁎, Joshua E. Meyer MD b , Erin McSpadden MPH a , Jonathan D. Cheng MD c , Frank A. Baciewicz MD d , Neal J. Meropol MD e , Steven J. Cohen MD c , Antoinette J. Wozniak MD f , Minsig Choi MD f , Andre A. Konski MD a a
Department of Radiation Oncology, Wayne State University/Barbara A. Karmanos Cancer Center, Detroit, Michigan Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania c Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania d Thoracic Surgical Oncology, Wayne State University/Barbara A. Karmanos Cancer Center, Detroit, Michigan e Division of Hematology and Oncology, University Hospitals/Case Comprehensive Cancer Center, Cleveland, Ohio f Division of Hematology and Oncology, Wayne State University/Barbara A. Karmanos Cancer Center, Detroit, Michigan b
Received 8 June 2012; revised 23 January 2013; accepted 6 February 2013
Abstract Purpose: The purpose of this study was to identify factors associated with cardiac toxicity in patients treated with chemoradiation therapy (CRT) for esophageal carcinoma. Methods and Materials: One hundred twenty-seven patients with adenocarcinoma or squamous cell carcinoma of the esophagus treated from July 2002 to June 2011 at 2 academic institutions with preoperative or definitive CRT were retrospectively reviewed. Association of cardiac toxicity with a number of variables was investigated, including heart disease, cardiac bypass and angioplasty, diabetes, insulin use, smoking, chemotherapy regimen, and tumor location. T test assessed risk of cardiac toxicity secondary to age. Dose volume histograms (DVH) were evaluated for percentage of heart volume receiving N 20, 30, 40, and 50 Gy (V20-V50). The Fisher exact test analyzed for an association between dose volume parameters and cardiac toxicity. Results: Patient population included 100 men and 27 women with a mean age of 64 years. Median follow-up was 12.7 months (range, 0.3-99.6 months). Any cardiac toxicity occurred in 28 patients, the majority of which were pericardial effusion (23/28). Odds ratio for toxicity in women was 4.15 (95% confidence interval [CI], 1.63-10.50; P = .0017) and time to cardiac toxicity by sex was significant (P = .0003). Patients above the median cutoff for V20, V30, and V40 had increased odds of developing cardiac toxicity (P = .03, .008, .002). There was 4.0 increased odds of developing cardiac toxicity with V40 N 57% (95% CI, 1.5-10.3, P = .002). On multivariable logistic regression analysis, sex was the only variable associated with any cardiac toxicity and pericardial effusion (P = .0016, P = .0038). None of the other investigated variables were associated with increased risk of cardiac toxicity. Abstract presented at American Society of Clinical Oncology Gastrointestinal Symposium, January 19-21, 2012, San Francisco, CA. Conflicts of interest: None. ⁎ Corresponding author. Department of Radiation Oncology, 4501 X Street, Suite G-140, Sacramento, CA 95816. E-mail address: [email protected] (L.M. Tait). 1879-8500/$ – see front matter © 2013 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.prro.2013.02.001
e150
L.M. Tait et al
Practical Radiation Oncology: October-December 2013
Conclusions: Female patients and dose greater than the median for V20-V40 were associated with the development of cardiac toxicity, specifically pericardial effusion. These data suggest exercising increased care when designing radiation fields in women undergoing CRT for esophageal carcinoma, as pericardial effusion may be a long-term complication. © 2013 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.
Introduction Radiation therapy in patients treated for breast cancer and Hodgkin lymphoma has been validated as a significant contributing factor for increased risk of cardiac toxicity in numerous previous studies. 1–13 Potential for risk of cardiac injury has been correlated with dose, fraction size, and volume of heart irradiated. 2,7,9,10,12 These findings have contributed to awareness of potential cardiac toxicity following radiation to the thorax for other malignancies, particularly esophageal carcinoma. An estimate of cardiac dose is achievable through the analysis of dose-volume histograms (DVHs). Radiation dose to the heart extrapolated from DVHs has been associated with cardiac toxicity in patients treated for breast cancer and lymphoma. 4,7,9 Milano et al 14 arrived at normal tissue radiation dose constraints for the heart based upon published literature from this collection of patients. An analysis of patients with Hodgkin lymphoma treated with radiation revealed there was an increased risk of developing pericarditis when either the total dose to the heart exceeded 41 Gy or when the radiation dose per fraction was greater than 3 Gy. 9 Martel et al 15 were the first to validate similar findings in a retrospective review of 57 patients with esophageal cancer. Their study evaluated patients treated with nonstandard radiation fractionation schedules and reported 5/57 patients developed nonmalignant pericardial effusions after treatment with chemoradiation therapy (CRT). After correcting for fraction size, a mean and maximum dose to the heart of 27.1 and 47.0 Gy, respectively, was predictive of pericarditis (P = .014). 15 More recently there have been several retrospective analyses of esophageal carcinoma patients treated with CRT. These studies have conflicting outcomes with regard to correlation of heart dose and treatment-associated cardiac toxicity. In a study by Wei et al, 16 patients treated definitively using standard fractionation showed the rate of pericardial effusion to be 73% in a subset of patients who had greater than 46% of their heart volume receiving 30 Gy (V30). On the contrary, a study by Tripp et al 17 did not observe any statistical or clinical significance associated between radiation-dose to the heart, left ventricle, or left anterior descending artery. Tripp and colleagues did observe a short-term decrease in cardiac ejection fraction; however, the magnitude of change was not significant. A small retrospective review by Mukherjee et al 18 supports the relevance of cardiac function preservation in patients who undergo CRT in the preoperative setting. They evaluated posttreatment cardiac ejection fraction and
determined a decreased ejection fraction in 12 of 15 patients assessed (P = .003), but their small sample size was a crucial limiting factor. The discordance in the published data has yet to isolate a critical dose-volume constraint for the heart or patient characteristics associated with an increased risk for cardiac injury with esophageal CRT. The purpose of this study is to identify factors associated with cardiac toxicity in the largest cohort of patients treated with CRT for esophageal carcinoma. Our aims are to characterize the frequency of cardiac toxicity in patients with esophageal cancer treated with standard fractionation CRT and explore the relationships between dose and volume of heart irradiated with cardiac toxicity in this population.
Methods and materials The charts of 144 patients were retrospectively reviewed from 2 academic institutions. Patients evaluated had a pathologically confirmed diagnosis of esophageal carcinoma and received neoadjuvant, definitive, or adjuvant radiation therapy in either institution from July 2002 through June 2011. Of these patients, 127 were eligible for our study; ineligible patients had incomplete DVH data, received a palliative radiation dose, or had previous thoracic radiation. Institutional review board approval was received from both academic centers for this study. Patients underwent a complete history and physical examination prior to treatment. Pretreatment diagnoses of cardiac comorbidities, including hypertension, dyslipidemia, heart failure, myocardial infarction, arrhythmia, cardiomyopathy, or valvular disease were recorded. Incidence of diabetes mellitus, insulin dependence, and smoking history was also recorded. All patients had a pathologic diagnosis of primary squamous cell carcinoma or adenocarcinoma of the esophagus and staging was performed clinically or surgically.
Treatment The majority of patients received combined modality therapy with concurrent CRT. Chemotherapy regimen was administered at the discretion of the treating medical oncologist. Three-dimensional computed tomography (CT) scans were utilized in all radiation treatment plans. Patients were simulated in the supine position, with their arms above their head and were secured with an
Practical Radiation Oncology: October-December 2013
immobilization device. CT scans were performed with normal free breathing in 3-mm slices from the mandible to the upper border of the pelvis. The treating radiation oncologist determined the gross esophageal tumor on the planning CT scan from pretreatment endoscopic ultrasound, diagnostic CT scan, and positron emission tomography scan. A 1-cm circumferential and 3.5-cm superior and inferior expansion were applied, forming the clinical target volume (CTV). Finally, to account for daily setup error, a 1.5-cm expansion from the CTV in all directions formed the planning target volume (PTV). Any involved locoregional lymph nodes were also included in the CTV and a 1- to 2-cm expansion formed the PTV. A commercial treatment planning system was used to develop 3-dimensional conformal or intensity modulated radiation therapy treatment plans, which restricted the maximum spinal cord dose to 45 Gy. Radiation treatment dose was prescribed by the treating physician and ranged from 45 to 59.4 Gy; the median dose was 50.4 Gy in 1.8 Gy fractions.
Dosimetric analysis A single radiation oncologist was responsible for contouring the external heart border on acquired treatment CT scans to control measured cardiac dose. The dose distributions and DVHs of the heart were calculated using the treatment planning system. The volume of the heart receiving at least 20, 30, 40, and 50 Gy (V20, V30, V40, and V50) was recorded for each patient.
Toxicity The primary investigator and co-investigators prospectively recorded the incidence of cardiac toxicity and associated symptoms during patients’ follow-up; if toxicity was reported on an imaging report it was verified on the corresponding CT scan or x-ray. The Radiation Therapy Oncology Group (RTOG) Late Radiation Morbidity Scoring Scheme and the Common Toxicity Criteria Adverse Events (CTCAE), version 3.0, were used to measure cardiac toxicity. Association of cardiac toxicity with a number of clinical variables was investigated, including history of heart disease, cardiac bypass or angioplasty, diabetes mellitus, insulin use, smoking, and tumor location.
Statistical analysis Standard descriptive statistics (with 95% 2-sided confidence intervals [CI]) were used to characterize the study population. We calculated the proportion of study subjects who developed cardiac events (symptomatic or asymptomatic) after the completion of CRT or radiation therapy alone. The χ 2 and Fisher exact tests were used to
Cardiac toxicity and esophageal chemo-TR
e151
test for an association between dose-volume parameters and cardiac toxicity; the median volume for V20 through V50 was used as a dividing point for our dose-volume parameters. Furthermore, the χ 2 and Fisher exact tests evaluated whether individual patient characteristics (gender, heart disease, cardiac bypass or angioplasty, diabetes, insulin use, smoking, chemotherapy regimen, and tumor location) were associated with cardiac toxicity. Logistic regression analyses and a multivariable logistic regression analysis were performed for patient characteristics and DVH parameters that were indicative of cardiac toxicity. These explanatory variables were also separately analyzed with a multivariable logistic regression for pericardial effusion, which was the most frequently observed toxicity. The time to the development of cardiac events and survival was estimated in this population using Kaplan-Meier Table 1
Patient and tumor characteristics
Characteristic Sex Male Female Histology Adenocarcinoma Squamous cell carcinoma Unknown Tumor location Middle Lower Gastroesophageal junction Unknown Preexisting heart disease Yes No Unknown Diabetes Yes No Unknown Insulin dependent Yes No Unknown Current or past smoker Yes No Unknown Previous cardiac bypass Yes No Unknown Age Mean Median Mean male Mean female
Frequency n (%) 100 (79) 27 (21) 97 (76) 23 (18) 7 (6) 23 (18) 37 (29) 61 (48) 6 (5) 71 (56) 54 (43) 2 (1) 24 (19) 101 (80) 2 (1) 5 (4) 120 (95) 2 (1) 87 (69) 38 (30) 2 (1) 15 (12) 110 (87) 2 (1) 64 y 64 y (37-87) 63.5 y (37-85) 64.5 y (42-87)
e152
L.M. Tait et al
Practical Radiation Oncology: October-December 2013
Table 2 Cardiac toxicity frequency and grade according to the Radiation Therapy Oncology Group (RTOG) Late Radiation Morbidity Scoring Scheme and the Common Toxicity Criteria Adverse Events (CTCAE), version 3.0 Cardiac toxicity
N (%)
RTOG (N)
CTCAE (N)
Sympt
Asymp
Men
Women
Angina Cardiomegaly Heart failure Myocardial infarction Pericardial effusion Sick sinus syndrome Total
1 (3.6) 1 (3.6) 1 (3.6) 1 (3.6) 23 (82.1) 1 (3.6) 28
3 1 3 3 3 3 1 3 4
3 1 2 3 1 3 1 2 3 4
1 0 1 1 5 1 9
0 1 0 0 18 0 19
1 0 0 1 13 1 16
0 1 1 0 10 0 12
(1) (1) (1) (1) (20), 4 (3) (1) (1) (24) (3)
(1) (1) (1) (1) (18), 3 (4), 4 (1) (1) (19) (1) (7) (1)
Asymp, asymptomatic; Sympt, symptomatic.
analysis. All statistical analysis was performed with SAS 9.2 software (SAS Institute, Cary, NC).
Results One hundred forty-four patients were included as part of our retrospective review; of this cohort, 127 patients met inclusion criteria. This group included 100 males and 27 females with a mean age of 64 years. The mean follow-up was 24.8 months and the median patient follow-up was 12.7 months (range, 0.3-99.6 months). Patient descriptive factors are provided in Table 1. Cardiac toxicity occurred in 28 patients (crude rate, 22%), with pericardial effusion representing 23 of these. Other toxicities included angina, cardiomegaly, heart failure, myocardial infarction, and sick sinus syndrome (Table 2). As graded by the RTOG scoring system, there were 24 grade 3 and 4 grade 4 cardiac toxicities. When characterized per the CTCAE, version 3.0, system, 7 grade 3 and a single grade 4 toxicity occurred. Median time to
cardiac event was 7.8 months, (range, 0-28.9 months). Cardiac toxicity was documented in 16 men and 12 women. In an analysis of men versus women, median time to cardiac toxicity was 6.4 and 4.1 months, respectively, and was statistically significant on Kaplan-Meier analysis (95% CI, 1.74-7.89; P = .0003 [Fig 1]). Patient demographics were analyzed for correlation to cardiac events. The Fisher exact test, comparing sex and toxicity, demonstrated female patients were 4.15 times more likely to have cardiac toxicity (95% CI, 1.63-10.50; P = .0017). The mean age of patients who had an event was 62.9 years, with a median of 63 years. There was no correlation between age and toxicity (P = .33). Other variables, including histology, tumor location, smoking history, diabetes, insulin dependence, preexisting cardiac disease, and previous cardiac bypass or angioplasty were not correlated to toxicity. Heart and left ventricle dosimetric volumes were evaluated for predictive value of cardiac toxicity. Left ventricle DVH data did not predict for toxicity and therefore was not included in our analysis (data not shown). Analysis of dosimetric factors demonstrated that a significant difference in V20, V30, and V40 existed between patients with cardiac toxicity and those without (Table 3). Patients above the median for V20 (71%) had Table 3 Dose-volume histogram parameters and cardiac toxicity analysis with the Fisher exact test analyzing the significance of the relationship between heart dose-volume histogram data and the development of cardiac events; variables assessed included volume of heart receiving 20-50 Gy (V20, V30, V40, V50)
Figure 1 Kaplan-Meier analysis of time to cardiac toxicity by sex (95% confidence interval, 1.74-7.89; P = .0003). The table beneath the graph depicts the number at risk by sex.
Variable
Median
P value
Odds ratio (OR)
Heart Heart Heart Heart
71 64.5 57 10
.03 (SS) .008 (SS) .002 (SS) .39
2.59 3.39 4.0 1.4
V20 V30 V40 V50
OR 95% confidence interval 1.1-6.3 1.3-8.7 1.5-10.3 0.62-3.4
Practical Radiation Oncology: October-December 2013 Table 4
Cardiac toxicity and esophageal chemo-TR
e153
Multivariable logistic regression analysis for statistically significant variables associated with any cardiac toxicity
Parameter
Standard estimate
Wald error
χ2
P value
Point estimate
95% Wald confidence interval
Intercept V20 V30 V40 Sex
− 2.7544 − 0.5343 0.9386 1.2043 1.6416
0.5338 0.9364 1.1607 0.8015 0.5195
26.6204 0.3255 0.6539 2.2579 9.9842
b .0001 .5683 .4187 .1329 .0016 (SS)
0.586 2.556 3.334 5.164
0.094-3.673 0.263-24.869 0.693-16.040 1.865-14.295
2.59 increased odds of developing cardiac toxicity (95% CI, 1.1-6.3; P = .03). Likewise, median V30 of 64.5% had a significant odds ratio (OR) of 3.39 for cardiac toxicity (95% CI, 1.3-8.7; P = .008). The greatest correlation with toxicity was demonstrated in which 57% of the heart received a dose of at least 40 Gy (OR, 4.0; 95% CI, 1.5-10.3; P = .002). There was no significant difference in the volume of heart irradiated when comparing V20, V30, V40, and V50 between men and women (P = .52, .65, .48, and .47). The mean and median maximum dose to the heart, or Dmax, was 52 and 53 Gy, respectively, ranging from 47 to 62 Gy. When evaluated by sex, the median Dmax in both men and women was 53 Gy as well. A multivariable logistic regression analysis was performed to investigate whether the increased odds ratio for women was a reflection of the relationships between the dose-volume effects and cardiac toxicity. Explanatory variables in the multivariable analysis included sex and DVH values associated with increased cardiac toxicity (V20, V30, and V40). Table 4 summarizes each variable and confirms that sex was independently associated with any cardiac toxicity (P = .0016). Multivariable logistic regression was also analyzed for pericardial effusion. Analyzed variables are summarized in Table 5 and demonstrate an independent likelihood of pericardial effusion associated with sex (P = .0038). Details of chemotherapy regimens are listed in Table 6; most patients received concurrent cisplatin and 5-fluorouracil (5-FU) (n = 81, 64%). Cisplatin/5-FU and paclitaxel-containing chemotherapy regimens did not significantly differ in relation to cardiac toxicity (95% CI, 0.17-2.35; P = .48). Sixty-five patients underwent surgical resection as a component of their multimodality treatment of esophageal carcinoma. Sixty-three patients had surgery following treatment with chemoradiation, 2
Table 5
patients had surgery followed by adjuvant chemoradiation, and 62 patients did not undergo surgery. Mean overall survival for all patients was observed to be 24.8 months with a median of 15 months and a range of 0-100.8 months. Mean survival for patients who experienced cardiac toxicity was 31 months. Comparison of survival by sex did not demonstrate a difference between women versus men despite an increased risk of cardiac toxicity (95% CI, 0.598-2.206; P = .89 [Fig 2]).
Discussion In this study we demonstrated that sex and cardiac dose were 2 factors associated with the development of cardiac toxicity following chemoradiation or radiation alone to the esophagus. Twenty-two percent of patients who were treated with standard fractionation radiation therapy developed some form of cardiac event, which for the majority was pericardial effusion. Additionally, cardiac toxicity was approximately 4 times more likely to occur in women. Women did not demonstrate significant variation in amount of dose or volume of heart receiving radiation and patient sex was confirmed to have an independent relationship with cardiac toxicity, specifically pericardial effusion, on multivariable analysis. Moreover, none of the other clinical variables, cardiac comorbidities, or differences in chemotherapy regimens were associated with cardiac injury. Dosimetric analysis verified patients above the median cutoff for V20, V30, and V40 had increased odds of developing cardiac toxicity. Heart volume greater than 57% receiving 40 Gy had the greatest correlation with development of cardiac toxicity. The only parameter that did not correlate with increased cardiac complications was V50. The median V50 was 10%, which was considerably lower in comparison with the median V20-V40 DVH
Multivariable logistic regression analysis for statistically significant variables associated with pericardial effusion
Parameter
Standard estimate
Wald error
χ2
P value
Point estimate
95% Wald confidence interval
Intercept V20 V30 V40 Sex
− 3.1728 − 0.7604 1.1161 1.3787 1.6724
0.6348 0.9776 1.2244 0.8848 0.5770
24.9798 0.6051 0.8308 2.4282 8.4001
b .0001 .4366 .3620 .1192 .0038 (SS)
0.467 3.053 3.970 5.325
0.069-3.176 0.277-33.648 0.701-22.484 1.718-16.501
e154
L.M. Tait et al
Practical Radiation Oncology: October-December 2013
Table 6 Chemotherapy regimens for patients with esophageal carcinoma treated with concurrent chemoradiation therapy Chemotherapy regimen
n (%)
Cisplatin, 5-FU 5-FU Paclitaxel, carboplatin Paclitaxel, cisplatin, 5-FU Paclitaxel, cisplatin 5-FU, oxaliplatin, Leucovorin Other Induction a No chemotherapy Unknown
81 10 8 7 3 2 6 3 3 7
(64) (8) (6) (5.5) (2) (1.5) (5) (2) (2) (5.5)
FU, fluorouracil. a Three patients had induction chemotherapy in addition to concurrent chemoradiation.
values (V20 71%, V30 64.5%, V40 57%). This may be a reflection of the fact that many patients were treated to 45 or 50.4 Gy and the V50 was relatively low if not zero. Therefore, it is possible that a small subset of patients had a V50 greater than the median explaining why the V50 did not demonstrate a correlation with cardiac toxicity. Previous studies conducted in esophageal CRT correlating dose-volume parameters and risk of cardiotoxicity have significant variation in dose, fractionation, radiation technique, and definition of cardiac volumes for radiation dosimetry. Given the paucity of homogenous data, the Quantitative Analyses of Normal Tissue Effects in the Clinics recommended dose constraint for a mean pericardial dose less than 26 Gy and V30 less than
Figure 2 Survival curves comparing outcomes of women versus men with a diagnosis of esophageal carcinoma treated with radiation therapy. There was no significant difference in survival by Kaplan-Meier analysis (95% confidence interval, .598-2.206; P = .89). The table beneath the graph depicts the number at risk by sex.
46% is based solely on the results from Wei et al. 16,19 Our results parallel several of their findings, including a significant median V30 associated with cardiac risk, and 5-month versus 7.8-month median time to onset of cardiac toxicity in their and our experiences, respectively. In addition, a similar crude rate of cardiac toxicity was observed; 27% versus 22% in our experience. In contrast, they investigated both dose to the pericardium and to the whole heart as individual organs at risk. They concluded that pericardial DVHs more precisely predicted pericardial effusion, and whole heart volume was unreliable. One could argue the reproducibility of contouring pericardial volumes and the practicality of repeatedly exercising this in the clinical setting is extremely unlikely. Additionally, their results were complicated by the fact that many of their dosimetric parameters were highly correlated with each other and numerous dosimetric parameters were indicative of pericardial effusion risk. 16,19 Limitations of our study include a short median follow-up of 12.7 months, which is further complicated by the overall short median survival of patients with esophageal cancer; 15 months in this cohort. Therefore, long-term cardiac complications demonstrated in breast cancer and lymphoma patients, including coronary artery sclerosis, myocardial infarction, cardiomyopathy, valvular disease, and conduction abnormalities, would typically not be observed in this narrow interval. This limitation can partially be attributed to the overall poor prognosis of esophageal cancer and suggests that future studies should more accurately analyze early toxicities such as pericardial effusion, which are more likely to occur in this short follow-up. The definition of “cardiac toxicity” in this study, as mentioned previously, was a heterogeneous group of clinical conditions. The lack of selecting for a specific toxicity obscures the etiology of radiation-induced affects as there are different radiobiologic mechanisms in myocardial infarction versus pericarditis. This may be a specification worth addressing in future studies. In addition, women were only 21% of our population study, which was also a limiting factor. However, esophageal carcinoma is 3 to 4 times more common among men than among women and it is likely that this is an accurate representation of the general patient population. Finally, the vast majority of patients received concurrent chemoradiation therapy (CCT) in our study, which may be a confounding factor attributing to cardiac toxicity. Both induction chemotherapy and CCT were identified risk factors for cardiac injury in breast cancer irradiation trials. 20,21 There have been little data to support this finding in radiation-related cardiac toxicity in patients with esophageal carcinoma receiving CRT. We did not correlate a difference in cardiac risk between cisplatin/ 5-FU and Taxol (Bristol-Meyers Squibb, New York, NY) based regimens. Likewise, other studies did not identify
Practical Radiation Oncology: October-December 2013
significant increased toxicity with induction chemotherapy or CCT. 16,17 Future studies should be directed to determine the mechanism behind the significant amount of women experiencing cardiac toxicity, specifically pericardial effusion, following CRT for esophageal carcinoma. Based on our analysis, this specific finding was independent of the dose-volume relationship with toxicity and women did not have increased comorbidities nor did they demonstrate a significant difference in the volume or dose of heart irradiated compared with men. Potential analyses should investigate the rationale behind the lower cardiac dose tolerance in women versus men. Long-term complications following chemoradiation are still a trivial setback in the bigger picture of the poor overall survival in patients with esophageal carcinoma. In our study, patients who exhibited cardiac toxicity generally had a longer survival suggesting that toxicity likely occurs in patients who live long enough to develop these complications. Nonetheless, treatment-induced toxicities are significant; with the growing utilization of CRT in the preoperative setting and improved survival with this treatment modality, reduction of long-term potential toxicities will become of utter importance. Our data confirm the necessity of exercising increased care when designing radiation fields in the treatment of esophageal carcinoma. Standard dose constraints to the volume of heart irradiated should be considered during treatment planning for patients, especially women. Studies evaluating sophisticated treatment planning with intensity modulated radiation therapy or breath-holding techniques may provide evidence demonstrating less heart dose. Additionally, further investigation is still needed to reproduce the increased morbidity demonstrated in women as well as clarify the biologic mechanism linked with this phenomenon.
References 1. Cuzick J, Stewart H, Rutqvist L, et al. Cause-specific mortality in long-term survivors of breast cancer who participated in trials of radiotherapy. J Clin Oncol. 1994;12:447-453. 2. Gagliardi G, Lax I, Ottolenghi A, Rutqvist LE. Long-term cardiac mortality after radiotherapy of breast cancer—application of the relative seriality model. Br J Radiol. 1996;69:839-846. 3. Paszat LF, Mackillop WJ, Groome PA, Schulze K, Holowaty E. Mortality from myocardial infarction following postlumpectomy radiotherapy for breast cancer: A population-based study in Ontario, Canada. Int J Radiat Oncol Biol Phys. 1999;43:755-762. 4. Clarke M, Collins R, Darby S, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local
Cardiac toxicity and esophageal chemo-TR
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
e155
recurrence and 15-year survival: an over- view of the randomised trials. Lancet. 2005;366:2087-2106. Darby SC, McGale P, Taylor CW, Peto R. Long-term mortality from heart disease and lung cancer after radiotherapy for early breast cancer: prospective cohort study of about 300,000 women in US SEER cancer registries. Lancet Oncol. 2005;6:557-565. Bouillon K, Haddy N, Delaloge S, et al. Long-term cardiovascular mortality after radiotherapy for breast cancer. J Am Coll Cardiol. 2011;57:445-452. Carmel RJ, Kaplan HS. Mantle irradiation in Hodgkin's disease. An analysis of technique, tumour eradication, and complications. Cancer. 1976;37:2813-2825. Gustavsson A, Eskilsson J, Landberg T, et al. Late cardiac effects after mantle radiotherapy in patients with Hodgkin's disease. Ann Oncol. 1990;1:355-363. Cosset JM, Henry-Amar M, Pellae-Cosset B, et al. Pericarditis and myocardial infarctions after Hodgkin's disease therapy. Int J Radiat Oncol Biol Phys. 1991;21:447-449. Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin's disease. J Am Med Assoc. 1993;270:1949-1955. Glanzmann C, Kaufmann P, Jenni R, Hess OM, Huguenin P. Cardiac risk after mediastinal irradiation for Hodgkin's disease. Radiother Oncol. 1998;46:51-62. Eriksson F, Gagliardi G, Liedberg A, et al. Long-term cardiac mortality following radiation therapy for Hodgkin's disease: analysis with the relative seriality model. Radiother Oncol. 2000;55:153-162. Aleman BM, van den Belt-Dusebout AW, Klokman WJ, Van't Veer MB, Bartelink H, van Leeuwen FE. Long-term cause-specific mortality of patients treated for Hodgkin's disease. J Clin Oncol. 2003;21:3431-3439. Milano MT, Constine LS, Okunieff P. Normal tissue tolerance dose metrics for radiation therapy of major organs. Semin Radiat Oncol. 2007;17:131-140. Martel MK, Sahijdak WM, Ten Haken RK, Kessler ML, Turrisi AT. Fraction size and dose parameters related to the incidence of pericardial effusions. Int J Radiat Oncol Biol Phys. 1998;40:155-161. Wei X, Liu HH, Tucker SL, et al. Risk factors for pericardial effusion in inoperable esophageal cancer patients treated with definitive chemoradiation therapy. Int J Radiat Oncol Biol Phys. 2008;70:707-714. Tripp P, Malhotra HK, Javle M, et al. Cardiac function after chemoradiation for esophageal cancer: comparison of heart dosevolume histogram parameters to multiple gated acquisition scan changes. Dis Esophagus. 2005;18:400-405. Mukherjee S, Aston D, Minett M, Brewster AE, Crosby TDL. The significance of cardiac doses received during chemoradiation of oesophageal and gastro-oesophageal junctional cancers. Clin Oncol (R Coll Radiol). 2003;15:115-120. Bentzen SM, Constine LS, Deasy JO, et al. Quantitative Analyses of Normal Tissue Effects in the Clinic (QUANTEC): an introduction to the scientific issues. Int J Radiat Oncol Biol Phys. 2010;76(Suppl 3): S3-S9. Shapiro CL, Hardenbergh PH, Gelman R, et al. Cardiac effects of adjuvant doxorubicin and radiation therapy in breast cancer patients. J Clin Oncol. 1998;16:3493-3501. Hardenbergh PH, Munley MT, Bentel GC, et al. Cardiac perfusion changes in patients treated for breast cancer with radiation therapy and doxorubicin: preliminary results. Int J Radiat Oncol Biol Phys. 2001;49:1023-1028.
