Key PointsQuestion
What radiotherapy dose exposure thresholds to critical cardiac substructures can estimate the probability for major adverse cardiac events and mortality in patients with non–small cell lung cancer?
Findings
In this cohort study of 701 patients with non–small cell lung cancer, left anterior descending coronary artery dose exposure appeared to be an independent factor associated with major adverse cardiac events and all-cause mortality, and baseline coronary heart disease status appeared to interact significantly with dose exposure.
Meaning
These findings suggest that newly identified cardiac substructure dose thresholds should alert clinicians to consider implementing more stringent cardiac radiotherapy planning parameters, but these thresholds require further study for validation, and optimized risk mitigation strategies are warranted.
Importance
Radiotherapy accelerates coronary heart disease (CHD), but the dose to critical cardiac substructures has not been systematically studied in lung cancer.
Objective
To examine independent cardiac substructure radiotherapy factors for major adverse cardiac events (MACE) and all-cause mortality in patients with locally advanced non–small cell lung cancer (NSCLC).
Design, Setting, and Participants
A retrospective cohort analysis of 701 patients with locally advanced NSCLC treated with thoracic radiotherapy at Harvard University–affiliated hospitals between December 1, 2003, and January 27, 2014, was performed. Data analysis was conducted between January 12, 2019, and July 22, 2020. Cardiac substructures were manually delineated. Radiotherapy dose parameters (mean, maximum, and the volume [V, percentage] receiving a specific Gray [Gy] dose in 5-Gy increments) were calculated. Receiver operating curve and cut-point analyses estimating MACE (unstable angina, heart failure hospitalization or urgent visit, myocardial infarction, coronary revascularization, and cardiac death) were performed. Fine and Gray and Cox regressions were adjusted for preexisting CHD and other prognostic factors.
Main Outcomes and Measures
MACE and all-cause mortality.
Results
Of the 701 patients included in the analysis, 356 were men (50.8%). The median age was 65 years (interquartile range, 57-73 years). The optimal cut points for substructure and radiotherapy doses (highest C-index value) were left anterior descending (LAD) coronary artery V15 Gy greater than or equal to 10% (0.64), left circumflex coronary artery V15 Gy greater than or equal to 14% (0.64), left ventricle V15 Gy greater than or equal to 1% (0.64), and mean total coronary artery dose greater than or equal to 7 Gy (0.62). Adjusting for baseline CHD status and other prognostic factors, an LAD coronary artery V15 Gy greater than or equal to 10% was associated with increased risk of MACE (adjusted hazard ratio, 13.90; 95% CI, 1.23-157.21; P = .03) and all-cause mortality (adjusted hazard ratio, 1.58; 95% CI, 1.09-2.29; P = .02). Among patients without CHD, associations with increased 1-year MACE were noted for LAD coronary artery V15 Gy greater than or equal to 10% (4.9% vs 0%), left circumflex coronary artery V15 Gy greater than or equal to 14% (5.2% vs 0.7%), left ventricle V15 Gy greater than or equal to 1% (5.0% vs 0.4%), and mean total coronary artery dose greater than or equal to 7 Gy (4.8% vs 0%) (all P ≤ .001), but only a left ventricle V15 Gy greater than or equal to 1% increased the risk among patients with CHD (8.4% vs 4.1%; P = .046). Among patients without CHD, 2-year all-cause mortality was increased with an LAD coronary artery V15 Gy greater than or equal to 10% (51.2% vs 42.2%; P = .009) and mean total coronary artery dose greater than or equal to 7 Gy (53.2% vs 40.0%; P = .01).
Conclusions and Relevance
The findings of this cohort study suggest that optimal cardiac dose constraints may differ based on preexisting CHD. Although the LAD coronary artery V15 Gy greater than or equal to 10% appeared to be an independent estimator of the probability of MACE and all-cause mortality, particularly in patients without CHD, left ventricle V15 Gy greater than or equal to 1% appeared to confer an increased risk of MACE among patients with CHD. These constraints are worthy of further study because there is a need for improved cardiac risk stratification and aggressive risk mitigation strategies.
Development of cardiac adverse events is a well-recognized risk following radiotherapy in patients with locally advanced non–small cell lung cancer (NSCLC).1-5 Radiation exposure of arterial vessels alters the natural history of coronary artery disease by inducing inflammation that subsequently triggers accelerated progression of the disease.6,7 Cardiac events in patients with NSCLC are common, are associated with cardiac radiation dose and baseline cardiac risk,3,4 and estimate the probability of mortality.8-10 These studies have evaluated radiation exposure to the entire heart with measures such as mean heart dose (MHD). However, MHD has been shown to be a poor surrogate for coronary artery dose exposure,11 thereby precluding direct association of injury with substructures and specific events (eg, coronary dose exposure associating with coronary events). Thus, cardiac substructure dose constraints to determine the probability of major adverse cardiac events (MACE) and mortality in patients with NSCLC are needed.
Locoregional recurrence remains a significant barrier to prolonging survival in patients with locally advanced NSCLC.12-14 In an attempt to overcome this challenge, radiation dose escalation was performed in Radiation Therapy Oncology Group 0617; however, the high-dose arm of the trial was associated with worse survival, and secondary analyses suggested that higher cardiac doses, among other factors, may have contributed to the lower survival rate.8,10 There has since been increasing interest in characterizing the association between cardiac dose and outcomes in lung cancer, although these analyses have been limited by small sample size, nonstandard or inconsistent cardiac end points, absence of cardiac substructure data, and variable assessment of baseline cardiac risk.1,3,4,15,16
A 2019 study showed that MHD was associated with MACE and all-cause mortality in one of the largest cohorts of patients with locally advanced NSCLC, building upon earlier reports with comprehensive inclusion of cardiac risk factors and validated cardiac end points.5 The primary objective of the present study was to expand upon the prior work by determining independent cardiac substructure radiotherapy factors associated with MACE and all-cause mortality in patients with locally advanced NSCLC, adjusting for preexisting heart disease and traditional lung and cardiovascular prognostic factors.
A retrospective analysis of a cohort of 701 consecutive patients with locally advanced NSCLC treated with thoracic radiotherapy between December 1, 2003, and January 27, 2014, was conducted at Brigham and Women’s Hospital and Dana-Farber Cancer Institute, Boston, Massachusetts. Data analysis was conducted between January 12, 2019, and July 22, 2020. Radiotherapy was delivered using 3D-conformal radiotherapy or intensity-modulated radiotherapy techniques (excluding stereotactic body radiotherapy). Radiotherapy was planned (Varian Eclipse, Varian Medical Systems Inc) and delivered in 1.8- to 2.0-Gy fractions.5 Dose constraints were used for the spinal cord (maximum <50 Gy), lungs (mean <17 Gy, volume [V] percentage) receiving a specific gray dose: 20 Gy <30%, V5 Gy <50%), and heart (V30 Gy <50%, V45 Gy <40%, and V60 Gy <20%; adopted in 2008).
This study was approved by the Dana-Farber/Harvard Cancer Center Institutional Review Board under a waiver of consent owing to its retrospective nature with minimal risk. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Cardiac substructures, including the coronary arteries (left main, left anterior descending [LAD], left circumflex, right, and posterior descending), chambers (left and right atria and ventricles), and whole heart, were manually delineated on archived planning computed tomographic scans in MIM Maestro, version 6.9.6 (MIM Software Inc) according to guidelines17 by 4 of us (K.M.A., T.L.C, N.L., and D.S.B.) and independently reviewed by 3 of us (A.N., U.H., and R.H.M.). Radiotherapy dose-volume histograms were recalculated in MIM Maestro; mean (gray), maximum (gray), and volume (percent) receiving a specific (X) gray dose (VX Gy [5-Gy increments]) were calculated.
In-depth manual medical record review was performed.5 Preexisting coronary heart disease (CHD) was defined as coronary artery disease, congestive heart failure, or a CHD risk-equivalent (peripheral vascular disease, stroke, or extensive coronary artery calcifications).5 Ten-year Framingham cardiovascular disease risk was calculated.18 MACE (unstable angina, heart failure hospitalization or urgent visit, myocardial infarction, coronary revascularization, and cardiac death)19 were assessed after initiation of radiotherapy and 30 days or more postoperatively, if applicable, until death or the last follow-up. Patients with preexisting cardiac comorbidities (eg, congestive heart failure) were counted as having a MACE if the event was either greater in severity compared with the 6-month interval preceding radiotherapy or of a different event class (eg, myocardial infarction or cardiac death). Follow-up and determination of cause of death were previously described.5
The distribution of clinical characteristics was assessed using descriptive statistics. Categorical covariables were evaluated using a Fisher exact test and continuous covariables were compared using a Wilcoxon rank-sum test. Area under the receiver operating characteristic curve was calculated. Comparative analyses of area under the curve and cut-point analyses were performed using the Liu method.20 MACE cumulative incidence estimates were adjusted for noncardiac death as a competing risk,21 compared using a 2-sided Gray P value. Kaplan-Meier estimates22 of all-cause mortality were compared using a 2-sided log-rank P value. Fine and Gray23 regressions, adjusted for noncardiac death as a competing risk, and Cox proportional hazards24 regressions were performed. Time 0 was the start date of radiotherapy and concluded by the date of the first MACE, death, or last observation, whichever came first. Multivariable models included covariables with P ≤ .05 determined on univariable analysis and cardiac dose variables. Multicollinearity was assessed by variance inflation factor and tolerance. A P value ≤.05 (2-sided) was considered statistically significant. All analyses were performed with Stata, version 15.1 (StataCorp LLC) statistical software.
Of the 701 patients included in the analysis, 356 were men (50.8%) and 345 were women (49.2%); the median age was 65 years (interquartile range [IQR], 57-73 years). Most patients (405 [57.8%]) were treated with definitive chemoradiotherapy. Among patients without CHD (449 [64.1%]), baseline cardiovascular risk according to the Framingham criteria was high in 217 patients (48.3%), moderate in 111 patients (24.7%), and low in 121 patients (26.9%). There was no significant difference in tumor laterality, radiotherapy technique, or radiotherapy dose between patients with and without CHD (Table 1).
Comparative Area Under the Curve Analysis
Analysis of the area under the curve demonstrated that cardiac dose variables had significantly higher C-index levels using MACE vs all-cause mortality as the end point. For example, for the total population, the median C-index value was 0.54 (IQR, 0.52-0.57) with MACE as the end point vs 0.52 (IQR, 0.51-0.53) for all-cause mortality (P < .001) (eTable 1 in the Supplement). By cardiac substructure, LAD dose variables had significantly higher C-index values for MACE compared with the whole heart, right ventricle, posterior descending artery, left atrium, right atrium, and right coronary artery, but not significantly higher than the left circumflex, total coronary, or left main arteries. Moreover, cardiac dose variables had significantly higher C-indices in patients without vs with CHD for each cardiac substructure. For example, the C-indices for the LAD coronary artery were 0.68 in patients without CHD vs 0.49 in patients with CHD, and for the left circumflex coronary artery were 0.71 in patients without CHD vs 0.49 in patients with CHD (P < .001) (eTable 2, eTable 3, eFigure 1, and eFigure 2 in the Supplement).
The optimal cut-point (C-index) levels for dose variables with the highest C-index level per structure were V15 Gy greater than or equal to 10% (0.64) for the LAD coronary artery, V15 Gy greater than or equal to 14% (0.64) for the left circumflex coronary artery, V15 Gy greater than or equal to 1% (0.64) for the left ventricle, mean greater than or equal to 7 Gy (0.62) for the total coronary arteries, mean greater than or equal to 27 Gy (0.59) for the left main artery, and V25 Gy greater than or equal to 14% (0.59) for the whole heart (included for clinical relevance). Cut points and descriptive statistics for all dose variables are reported (eTable 4 and eTable 5 in the Supplement).
The median follow-up was 20.4 months (IQR, 8.2-44.6 months) overall and 47.8 months (IQR, 31.6-75.4 months) in patients alive. Seventy patients developed greater than or equal to 1 MACE (1-year cumulative incidence, 3.9%; 95% CI, 2.6-5.5), with a median time to first MACE of 20.6 months (IQR, 8.8 months to 43.3 months). After adjusting for age, hypertension, diabetes, arrhythmia, CHD, and radiotherapy technique, a significant increase in the risk of MACE was observed with an LAD coronary artery V15 Gy greater than or equal to 10% (adjusted hazard ratio [aHR], 13.90; 95% CI, 1.23-157.21; P = .03) (Table 2). Although peripheral vascular disease, coronary artery disease, and congestive heart failure were significantly associated with MACE on univariable analysis, given the limited number of MACE, only CHD was included in the multivariable analysis, because the former disease states represent CHD/CHD-risk equivalents. Given modest multicollinearity (eTable 6 in the Supplement), we repeated the multivariable analysis using individual (non-LAD) cardiac dose variables, each of which was significantly associated with MACE. For example, a left circumflex coronary artery V15 Gy greater than or equal to 14% (aHR, 8.13; 95% CI, 2.80-23.58; P < .001) and left ventricle V15 Gy greater than or equal to 1% (aHR, 6.67; 95% CI, 2.30-19.30; P < .001) (eTables 7-12 in the Supplement).
There was a significant interaction between CHD and LAD V15 Gy greater than or equal to 10% (P = .04). Specifically, among patients without CHD (n = 449), LAD V15 Gy greater than or equal to 10% vs <10% was associated with a significantly increased risk of MACE (23 vs 1 event; HR 24.8; 95% CI, 3.49-176.42; P = .001) with 1-year estimates of 4.9% (95% CI, 2.6-8.3) vs 0%, respectively (Figure 1, Table 3). By contrast, among patients with CHD (n = 252), there was not a statistically significant increase in the risk of MACE with LAD coronary artery V15 Gy greater than or equal to 10% vs less than 10% (30 vs 16 events; hazard ratio, 1.42; 95% CI, 0.78-2.60; P = .25) with 1-year estimates of 7.6% (95% CI, 4.0-12.6) for V15 Gy greater than or equal to 10% vs 4.7% (95% CI, 1.8-10.0) for V15 Gy less than 10%. We observed similar findings among other dose variables stratified by cut point, including left circumflex coronary artery V15 Gy greater than or equal to 14% vs less than 14%, with 1-year MACE estimates of 5.2% vs 0.7% in patients without CHD (P < .001), and 9.0% vs 4.3% in patients with CHD (P = .08). Only left ventricle V15 Gy greater than or equal to 1% vs less than 1% was associated with an increased risk of MACE among both patients with and without CHD, with 1-year MACE estimates of 5.0% vs 0.4% (V15 Gy ≥1% vs <1%) in those without CHD (P < .001), and 8.4% vs 4.1% (V15 Gy ≥1% vs <1%) in those with CHD (P = .046) (Table 3).
All-Cause Mortality Analysis
A total of 490 died, 325 (66.3%) of lung cancer, 37 (7.8%) of known noncardiac causes, and 25 (5.1%) of cardiac-specific causes; the remainder (n = 103) were of unknown causes. The median overall survival was 22.2 months (IQR, 9.8-45.1 months). After adjustment for age, sex, performance status, unintentional weight loss, stroke, peripheral vascular disease, myocardial infarction, congestive heart failure, arrhythmia, CHD, treatment paradigm, and treatment year, we observed a significant increase in the risk of all-cause mortality with the LAD coronary artery V15 Gy greater than or equal to 10 Gy (aHR, 1.58; 95% CI, 1.09-2.29; P = .02) (Table 2). Similar to our evaluation of MACE, given modest multicollinearity (eTable 6 in the Supplement), we repeated the analysis using individual (non-LAD coronary artery) cardiac dose variables, of which the mean total coronary artery dose greater than or equal to 7 Gy, MHD greater than or equal to 10 Gy, and whole heart V25 Gy greater than or equal to 14% were significantly associated with all-cause mortality. For example, mean total coronary arteries greater than or equal to 7 Gy (aHR, 1.33; 95% CI, 1.05-1.67; P = .02) and MHD greater than or equal to 10% (aHR, 1.32; 95% CI, 1.04-1.67; P = .02) (eTables 7-12 in the Supplement).
There was a significant interaction between baseline CHD and LAD coronary artery V15 Gy (P = .01). Specifically, for the 449 patients without CHD, the LAD coronary artery V15 Gy greater than or equal to 10% vs less than 10% was associated with a significantly increased risk of all-cause mortality (158 vs 138 deaths; hazard ratio, 1.36; 95% CI, 1.08-1.71; P = .009), with 2-year estimates of 51.2% (95% CI, 44.7%-57.9%) for V15 Gy greater than or equal to 10% vs 42.2% (95% CI, 35.9%-49.0%) V15 Gy less than 10% (Figure 2, Table 3). Among the 252 patients with CHD, there was no statistically significant increase in the risk of all-cause mortality with the LAD coronary artery V15 Gy greater than or equal to 10% vs less than 10% (109 vs 85 deaths; hazard ratio, 0.80; 95% CI, 0.60-1.06; P = .12), with 2-year estimates of 47.9% (95% CI, 40.1%-56.4%) for V15 Gy greater than or equal to 10% vs 61.1% (95% CI, 51.9%-70.4%) for V15 Gy less than 10%. We observed similar findings with mean total coronary artery dose greater than or equal to 10% 7 Gy and whole heart V25 Gy greater than or equal to 14% (Table 3).
Despite the competing risk of lung cancer death, we observed that LAD coronary artery dose exposure (V15 Gy ≥10%) is an independent factor associated with MACE and all-cause mortality and that optimal cardiac dose constraints may differ based on preexisting cardiac status. The LAD coronary artery V15 Gy greater than or equal to 10%, left circumflex coronary artery V15 Gy greater than or equal to 14%, left ventricle V15 Gy greater than or equal to 1%, and mean total coronary artery dose greater than or equal to 7 Gy appeared to confer a 5% absolute increase in 1-year MACE estimates in patients without CHD, but only left ventricle V15 Gy greater than or equal to 1% appeared to confer an increased risk among patients with CHD (Table 3). Similarly, LAD coronary artery V15 Gy greater than or equal to 10% and mean total coronary artery dose greater than or equal to 7 Gy appeared to confer a nearly 10% absolute increase in 2-year all-cause mortality estimates in patients without but not with CHD. These findings support more precise cardiac risk stratification based on coronary and left ventricle dose exposure and baseline status of CHD.
The most robust data for radiotherapy-associated cardiac events are from breast cancer and lymphoma studies.25-27 However, it is challenging to extrapolate these findings to NSCLC given the higher doses of radiotherapy used in NSCLC, with more widely varying dose exposure to cardiac substructures based on primary tumor and nodal locations (eFigure 3 in the Supplement), the more contracted time (1-2 years) during which cardiac events are observed, and the elevated baseline cardiac risk in these patients.3-5,28 Furthermore, it warrants discussion that the most commonly used cardiac dose constraint is MHD. Although Atkins et al5 recently reported that MHD is an independent factor associated with MACE and all-cause mortality in patients with NSCLC, this dose variable is limited because it is unable to account for the variability and steepness of dose distributions across cardiac regions (eFigure 3 in the Supplement) and is insufficient to indicate the probability of left ventricle and coronary artery dose exposure.11 Together, these findings underscore the need for more nuanced cardiac dosimetry to improve cardiac risk estimation.
A recent systematic review failed to identify consistent cardiac radiation dose parameters associated with survival in patients with NSCLC.29 However, this study was performed before Atkins et al5 published findings on what was, to our knowledge, the largest NSCLC cohort describing individual cardiac (MACE and grade ≥3 Common Terminology Criteria for Adverse Events) and survival outcomes with analysis of MHD. The review included 4 small (112-250 patients each) studies with a primary end point of cardiac events.1,3,4,15,16 However, these studies had limited events, used variable and/or nonvalidated cardiac end points, and had inconsistent accounting of baseline cardiac risk. A single study analyzed the coronary (LAD coronary artery) dose, but limited sample size, and thus, limited cardiac events, precluded multivariable analysis of cardiac events and only disease progression was associated with mortality.15 There have been more series analyzing cardiac dose variables associated with survival.8-10,30-41 The largest of these studies was RTOG 0617, showing heart doses of V5, V30, and V40 Gy to be associated with survival,8,10,31,32 and more recently identifying atria dose to 45% greater than 44 Gy, pericardium mean dose to the hottest 55% greater than 51 Gy, and ventricles mean dose to the hottest 5% greater than 56 Gy.39 However, these studies only analyzed pericardium, atria, and ventricles (without coronary arteries), and did not assess cardiac events or account for baseline cardiac risk. Moreover, these NSCLC series largely evaluated whole heart dosimetry,9,30,33-41 and some analyzed the pulmonary artery,35 base of heart,36 and left atrium.41 The only study to evaluate coronary dosimetry used a noncontouring method to match 1161 patients to 5 template anatomies with 14 cardiac substructures delineated (including the LAD, left circumflex coronary, and right coronary arteries), and found that the right atrium, right coronary artery, and ascending aorta were associated with survival, proposing a 23-Gy maximum dose limit, but without assessment of cardiac events or accounting for baseline cardiac risk.37 Together, the limitations of these heterogeneous reports underscore the value of robust clinical data sets with validated and/or standardized cardiac event end points and the need for comprehensive cardiac substructure dosimetry (including coronary arteries).
Given the shared risk profiles between CHD and cancer and the high prevalence of CHD among patients with NSCLC,28,42 careful assessment of preexisting cardiac risk is necessary, because optimal cardiac dose constraints may differ based on preexisting cardiac status. Although we did not observe a significant association between coronary dose and risk of MACE or all-cause mortality in patients with CHD, we posit that this lack of association is the result of exceeding an observable dose response in an ultra-high-risk group. Consistent with this hypothesis, patients with vs without CHD harbor a similar absolute increase in 1-year MACE rates with LAD coronary artery V15 Gy greater than or equal to 10% vs less than 10% (3% vs 5%), left circumflex coronary artery greater than or equal to 14% vs less than 14% (5% vs 5%), left ventricle V15 Gy greater than or equal to 1% vs less than 1% (4% vs 5%), and mean total coronary arteries greater than or equal to 7 Gy vs less than 7 Gy (7% vs 5%) (Table 3). However, only the left ventricle V15 Gy greater than or equal to 1% was statistically significant among patients with CHD. The cause of these differences is unclear, but possible mechanistic reasons include a potential for greater susceptibility to sequalae from radiation-induced microvascular disease or myocardial fibrosis in the presence of baseline epicardial coronary obstructive patterns.6,7,43 Moreover, the absence of a more dominant LAD coronary artery–mediated effect in patients with CHD may, in part, have been confounded by the increased LAD coronary artery dose in patients with vs without CHD (mean, 8.7 vs 7.0 Gy, P = .03) (eTable 5 in the Supplement).
The association between lung dose-volume parameters and symptomatic pneumonitis and mortality has been well characterized.44,45 Together with our current and recent observations demonstrating the association between cardiac and coronary artery radiation dose, cardiac events, and mortality,5 we recommend a balanced approach in radiotherapy planning optimization. Specifically, we do not recommend exceeding lung dose constraints or prioritizing cardiac dose constraints such that tumor coverage is not maintained. Rather, we recommend minimizing cardiac and left coronary artery radiation doses as low as reasonably allowable while achieving safe lung dose-volume exposure and maintaining tumor coverage (eg, with the goal of achieving LAD coronary artery V15 Gy <10% and MHD<10 Gy5). Although modern treatment planning techniques may allow improved optimization of lung and cardiac constraints in a subset of patients, strategies evaluating treatment planning optimization techniques with appropriate cardiac risk stratification are of high clinical interest and are being actively pursued.46 Furthermore, identifying patients at highest risk of cardiac events might help identify who may benefit most from experimental technologies, such as proton radiotherapy.
Strengths and Limitations
Strengths of this study are that, to our knowledge, it represents the largest cohort of individual patient cardiac and coronary dosimetry in thoracic radiotherapy with manual delineation of cardiac substructures.17 Combined with comprehensive detailing of cardiac risk factors and use of a validated MACE end point, the study provides an improved means for identifying patients at high cardiac risk for whom more aggressive cardiac risk mitigation strategies, including more stringent cardiac and coronary artery radiation dose avoidance, may be warranted. The association of precisely calculated radiation dose exposure to critical cardiac substructures (eg, coronary arteries) that have a directly relevant pathophysiologic mechanism of injury (eg, coronary artery stenosis) to clearly defined toxic cardiac events (eg, myocardial infarction, coronary revascularization) is the key novel contribution of this work and an important point of understanding for oncologists and cardiologists.
Limitations of this study include its retrospective design and absence of a validation cohort. Therefore, prospective validation of these newly identified cardiac constraints is warranted. In addition, the known multicollinearity of radiation dose variables precludes thorough direct comparison among these variables. In addition, although the validated MACE end point is ideal for incorporating baseline cardiovascular risk (eg, Framingham or atherosclerotic cardiovascular disease risk) and guideline-based cardiac risk factor optimization, radiotherapy-relevant end points, such as arrhythmias and pericardial events, were not addressed, and the limited number of MACE introduces the potential for data overfitting. Atkins et al5 recently reported the cumulative incidence rates of these individual grade 3 or greater MACE and Common Terminology Criteria for Adverse Events and their differential association with baseline cardiac risk; however, detailed analysis of the relevant cardiac substructure dose variables and their association with these end points was outside of the scope of this report.
Our study suggests that LAD coronary artery V15 Gy greater than or equal to 10% is an independent factor associated with MACE and all-cause mortality in patients with NSCLC and optimal cardiac dose constraints may differ based on preexisting cardiac status. We noted that LAD coronary artery V15 Gy greater than or equal to 10%, left circumflex coronary artery V15 Gy greater than or equal to 14%, left ventricle V15 Gy greater than or equal to 1%, and mean total coronary arteries greater than or equal to 7 Gy appeared to confer a 5% absolute increase in 1-year MACE estimates in patients without CHD, but only left ventricle V15 Gy greater than or equal to 1% conferred an increased risk among patients with CHD. Similarly, an LAD coronary artery V15 Gy greater than or equal to 10% and mean total coronary artery dose greater than or equal to 7 Gy appeared to confer a nearly 10% absolute increase in 2-year all-cause mortality estimates in patients without but not with CHD. These cardiac constraints are worthy of further study for validation because critical cardiac substructure dose constraints for radiotherapy planning are lacking and there is a need for identifying patients for whom more aggressive risk mitigation strategies are warranted.
Accepted for Publication: September 1, 2020.
Published Online: December 17, 2020. doi:10.1001/jamaoncol.2020.6332
Corresponding Author: Raymond H. Mak, MD, Dana-Farber Cancer Institute, Brigham and Women's Hospital, 450 Brookline Ave, Boston, MA 02115 (rmak@partners.org).
Author Contributions: Drs Atkins and Mak had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Atkins, Nohria, Hoffmann, Aerts, Mak.
Acquisition, analysis, or interpretation of data: Atkins, Chaunzwa, Lamba, Bitterman, Rawal, Bredfeldt, Williams, Kozono, Baldini, Aerts, Mak.
Drafting of the manuscript: Atkins, Rawal, Aerts, Mak.
Critical revision of the manuscript for important intellectual content: Atkins, Chaunzwa, Lamba, Bitterman, Bredfeldt, Williams, Kozono, Baldini, Nohria, Hoffmann, Aerts, Mak.
Statistical analysis: Atkins, Rawal.
Administrative, technical, or material support: Atkins, Chaunzwa, Lamba, Williams, Kozono, Hoffmann.
Supervision: Atkins, Bredfeldt, Baldini, Aerts, Mak.
Conflict of Interest Disclosures: Dr Bitterman reported receiving personal fees from Agios Pharmaceuticals outside the submitted work. Dr Williams reported receiving grants from ViewRay Medical Systems outside the submitted work. Dr Nohria reported receiving grants from Amgen Inc and personal fees from Takeda Oncology outside the submitted work. Dr Hoffmann reported receiving consulting fees from Abbott, Duke University (National Institutes of Health [NIH]), Recor Medical and grants on behalf of Kowa Company, MedImmune, HeartFlow, Duke University (Abbott), Oregon Health & Science University (American Hospital Association 13FTF16450001), Columbia University (NIH, 5R01-HL109711), NIH/National Heart, Lung, and Blood Institute [NHLBI] 5K24HL113128, NIH/NHLBI 5T32HL076136, and NIH/NHLBI 5U01HL123339 outside the submitted work. Dr Aerts reported receiving personal fees from Sphera & Genospace and Onc.AI Inc outside the submitted work. Dr Mak reported receiving personal fees from AstraZeneca, grants and personal fees from ViewRay Inc, and personal fees from NewRT outside the submitted work. No other disclosures were reported.
Additional Contributions: Joseph D. Meyers, MSc (Clinical Support Engineer, MIM Software Inc), contributed to data collection by assisting in the development and implementation of the automated workflow for obtaining dose-volume data from radiotherapy plans. No compensation was received.
2.Hardy
D, Liu
CC, Cormier
JN, Xia
R, Du
XL. Cardiac toxicity in association with chemotherapy and radiation therapy in a large cohort of older patients with non–small-cell lung cancer.
Ann Oncol. 2010;21(9):1825-1833. doi:
10.1093/annonc/mdq042
PubMedGoogle ScholarCrossref 3.Wang
K, Eblan
MJ, Deal
AM,
et al. Cardiac toxicity after radiotherapy for stage III non–small-cell lung cancer: pooled analysis of dose-escalation trials delivering 70 to 90 Gy.
J Clin Oncol. 2017;35(13):1387-1394. doi:
10.1200/JCO.2016.70.0229
PubMedGoogle ScholarCrossref 4.Dess
RT, Sun
Y, Matuszak
MM,
et al. Cardiac events after radiation therapy: combined analysis of prospective multicenter trials for locally advanced non–small-cell lung cancer.
J Clin Oncol. 2017;35(13):1395-1402. doi:
10.1200/JCO.2016.71.6142
PubMedGoogle ScholarCrossref 8.Bradley
JD, Paulus
R, Komaki
R,
et al. Standard-dose versus high-dose conformal radiotherapy with concurrent and consolidation carboplatin plus paclitaxel with or without cetuximab for patients with stage IIIA or IIIB non–small-cell lung cancer (RTOG 0617): a randomised, two-by-two factorial phase 3 study.
Lancet Oncol. 2015;16(2):187-199. doi:
10.1016/S1470-2045(14)71207-0
PubMedGoogle ScholarCrossref 10.Bradley
JD, Hu
C, Komaki
RR,
et al. Long-term results of NRG Oncology RTOG 0617: standard- versus high-dose chemoradiotherapy with or without cetuximab for unresectable stage III non–small-cell lung cancer.
J Clin Oncol. 2020;38(7):706-714. doi:
10.1200/JCO.19.01162PubMedGoogle ScholarCrossref 11.Jacob
S, Camilleri
J, Derreumaux
S,
et al. Is mean heart dose a relevant surrogate parameter of left ventricle and coronary arteries exposure during breast cancer radiotherapy: a dosimetric evaluation based on individually-determined radiation dose (BACCARAT study).
Radiat Oncol. 2019;14(1):29. doi:
10.1186/s13014-019-1234-z
PubMedGoogle ScholarCrossref 15.Wang
K, Pearlstein
KA, Patchett
ND,
et al. Heart dosimetric analysis of three types of cardiac toxicity in patients treated on dose-escalation trials for stage III non–small-cell lung cancer.
Radiother Oncol. 2017;125(2):293-300. doi:
10.1016/j.radonc.2017.10.001
PubMedGoogle ScholarCrossref 19.Hicks
KA, Mahaffey
KW, Mehran
R,
et al; Standardized Data Collection for Cardiovascular Trials Initiative (SCTI). 2017 Cardiovascular and stroke endpoint definitions for clinical trials.
J Am Coll Cardiol. 2018;71(9):1021-1034. doi:
10.1016/j.jacc.2017.12.048
PubMedGoogle ScholarCrossref 31.Chun
SG, Hu
C, Choy
H,
et al. Impact of intensity-modulated radiation therapy technique for locally advanced non–small-cell lung cancer: a secondary analysis of the NRG Oncology RTOG 0617 randomized clinical trial.
J Clin Oncol. 2017;35(1):56-62. doi:
10.1200/JCO.2016.69.1378
PubMedGoogle ScholarCrossref 32.Eaton
BR, Pugh
SL, Bradley
JD,
et al. Institutional enrollment and survival among NSCLC patients receiving chemoradiation: NRG Oncology Radiation Therapy Oncology Group (RTOG) 0617.
J Natl Cancer Inst. 2016;108(9):108. doi:
10.1093/jnci/djw034
PubMedGoogle ScholarCrossref 33.Guberina
M, Eberhardt
W, Stuschke
M,
et al. Heart dose exposure as prognostic marker after radiotherapy for resectable stage IIIA/B non–small-cell lung cancer: secondary analysis of a randomized trial.
Ann Oncol. 2017;28(5):1084-1089. doi:
10.1093/annonc/mdx069
PubMedGoogle ScholarCrossref 37.McWilliam
A, Khalifa
J, Vasquez Osorio
E,
et al. Novel methodology to investigate the impact of radiation dose to heart sub-structures on overall survival.
Int J Radiat Oncol Biol Phys. 2020;S0360-3016(20)31318-3.
PubMedGoogle Scholar 46.Atkins
KM, Bitterman
DS, Selesnick
P, Carpenter
C, Cormack
RA, Mak
RH. Dosimetric tradeoffs of mean heart dose reduction predicted by machine learning–guided decision support software in lung cancer.
Int J Radiat Oncol Biol Phys. 2019;105(1):S256. doi:
10.1016/j.ijrobp.2019.06.2537
Google ScholarCrossref