- Email: [email protected]
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
Radiotherapy and Oncology xxx (xxxx) xxx
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Original Article
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy Rebecca Bütof a,b,c,⇑, Steffen Löck a,b,d, Maher Soliman b,e, Robert Haase a, Rosalind Perrin a,f, Christian Richter a,b,d,g, Steffen Appold a,b,c, Mechthild Krause a,b,c,d,g,h, Michael Baumann a,b,d,g,h a OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf; b Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden; c National Center for Tumor Diseases (NCT), Partner Site Dresden, Germany: German Cancer Research Center (DKFZ), Heidelberg, Germany; Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany, and; Helmholtz Association / Helmholtz-Zentrum Dresden - Rossendorf (HZDR), Dresden, Germany; d German Cancer Consortium (DKTK), Partner Site Dresden, Germany; e Oncology Department, Faculty of Medicine, Alexandria University, Egypt; f Strahlenklinik, Universitätsklinikum Erlangen; g Helmholtz-Zentrum Dresden - Rossendorf, Institute of Radiooncology – OncoRay; and h German Cancer Research Center (DKFZ), Heidelberg, Germany
a r t i c l e
i n f o
Article history: Received 24 May 2019 Received in revised form 28 October 2019 Accepted 4 November 2019 Available online xxxx Keywords: Dose–volume parameters Esophagitis Radiotherapy Accelerated Prediction Non-small cell lung cancer
a b s t r a c t Background and purpose: Early radiation-induced esophageal toxicity (RIET) is one of the major side effects in patients with non-small cell lung cancer (NSCLC) and can be a reason for treatment interruptions. As the age of patients with NSCLC and corresponding comorbidities continue to increase, primary radiotherapy alone is a commonly used alternative treatment in these cases. The aim of the present study is to compare dosimetric and clinical parameters from the previously reported CHARTWEL trial for their ability to predict esophagitis and investigate potential differences in the accelerated and conventional fractionation arm. Material and methods: 146 patients of the Dresden cohort of the randomized phase III CHARTWEL trial were included in this post-hoc analysis. Side effects were prospectively scored weekly during the first 8 weeks from start of radiotherapy. To compare both treatment arms, recorded dose–volume parameters were adjusted for the different fractionation schedules. Logistic regression was performed to predict early RIET for the entire study group as well as for the individual treatment arms. Different dosimetric and clinical parameters were tested. Results: Patients receiving the accelerated CHARTWEL schedule experienced earlier and more severe esophagitis (e.g. 20.5% vs. 9.6% grade 2 at week 3, respectively). In contrast, the median time period for recovery of grade 1 esophagitis was significantly longer for patients with conventional fractionation compared to the CHARTWEL group (median [range]: 21 [12–49] days vs. 15 [7–84] days, p = 0.028). In univariable logistic regression none of the dose–volume parameters showed a significant correlation with early RIET grade 2 in the conventional irradiation group. In contrast, for patients receiving CHARTWEL, the physical dose–volumes parameters V40 and V50; and re-scaled values VEQD2,50 and VEQD2,60 were significant predictors of early RIET grade 2. Dose–volume parameters remained different between CHARTWEL and conventional fractionation even after biological rescaling. Conclusion: Our results show a more dominant dose-volume effect in the CHARTWEL arm compared to conventional fractionation, especially for higher esophageal doses. These findings support the notion that dose–volume parameters for radiation esophagitis determined in a specific and time dependent setting of field arrangements can not be easily transferred to another setting. In clinical practice esophageal volumes receiving 40 Gy or more should be strictly limited in hyperfractionated-accelerated fraction schemes. Ó 2019 Elsevier B.V. All rights reserved. Radiotherapy and Oncology xxx (2019) xxx–xxx
⇑ Corresponding author at: Department of Radiotherapy and Radiation Oncology, University Hospital Carl Gustav Carus and Faculty of Medicine of Technische Universität Dresden, Fetscherstraße 74, 01307 Dresden, Germany. E-mail address: [email protected] (R. Bütof).
Non-small cell lung cancer (NSCLC) is one of the most common malignant diseases worldwide [4]. Simultaneous radiochemotherapy in the treatment of locally advanced lung cancer yields better outcome than radiotherapy alone [8]. While this treatment regimen is the standard of care, it is also evoking more side effects
https://doi.org/10.1016/j.radonc.2019.11.002 0167-8140/Ó 2019 Elsevier B.V. All rights reserved.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
2
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
[8,13]. One of the frequent side effects is early radiation-induced esophageal toxicity (RIET), which can be a reason for treatment interruptions. Complete avoidance of esophagus irradiation in lung cancer radiotherapy is impossible due to several factors: the large, irregularly shaped and centrally located lung cancer, frequent involvement of mediastinal lymph nodes (LN), and the central location and length of the esophagus [2,14]. As the age of patients with NSCLC and corresponding comorbidities continues to increase, primary radiotherapy alone remains a commonly used treatment alternative for patients unable to tolerate radiochemotherapy. Hyperfractionated-accelerated irradiation schedules have shown comparable or even improved outcome data of NSCLC patients compared to conventional fractionation, however, also these schedules are associated with increased early esophageal reactions [1,9,12]. Valid doseconstraints for avoidance of esophagitis in hyperfractionatedaccelerated irradiation schedules are missing so far. In the present study, we evaluate esophageal toxicity utilizing data from the prospective CHARTWEL trial [1]. The aim is to compare dosimetric and clinical parameters for their ability to predict early esophagitis and investigate potential differences in the accelerated and conventional fractionation arm. Patients and methods Patients Between 1997 and 2005, 163 patients with NSCLC were treated at the Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Germany, as a part of the randomized phase III CHARTWEL trial, which was carried out by 15 centers in Germany, Poland and Czech Republic. Clinical scoring of esophagitis has been done prospectively in all patients and the results have been previously reported [1]. Treatment planning CT scans and irradiation plans of the patients treated in Dresden could be retrieved and evaluated retrospectively, except for 16 patients because of technical reasons resulting in the exclusion of these patients from this study. In addition, one patient was excluded due to missing esophagitis data. Final results of the CHARTWEL trial as well as a post-hoc analysis of specific questions within the Dresden subcohort have already been published [1,15]. In brief, all patients had inoperable histologically proven NSCLC, they were stratified according UICC staging system (into I: 8%, II: 7%, IIIA: 39%, IIIB: 46%), patients were >18 years and had WHO performance score 0–1. Patients were excluded from the original trial if they had distant metastases, supraclavicular LN metastases, malignant pleural effusion, FEV1 <1 liter under optimized medical therapy, unintended weight loss >15% over the last 6 months, prior radiotherapy, surgical resection if more than biopsy, prior or concurrent other malignant disease except basalioma, T1 squamous cell carcinoma of the skin, T1 glottic carcinoma, and carcinoma in situ of cervix uteri, other grave comorbidities that expected to limit the short term life expectancy of the patient, cardiac pacemaker in the radiation field except if the cardiologist agreed, pregnancy or participation in other clinical trials or if follow up was not possible. Treatment In both treatment arms, the initial planning target volume (PTVinitial) included the mediastinum and the primary tumour with a margin of 1–1.5 cm ipsilaterally and 1 cm contralaterally. The mediastinum was defined as the area from the jugular fossa to 3 cm below the carina. Supraclavicular LN were treated for
tumours of the upper lobes and irradiation of the contralateral hilum was done in case of involvement. The high-dose PTV (PTVboost) included the primary tumour and positive LN with a margin of 1 cm. All patients in both arms were treated with 3D conformal radiotherapy using a linear accelerator with a photon energy 6 MV and shrinking-field technique. The prescribed dose in the conventional arm was 50 Gy/25 fractions/5 weeks to the PTVinitial followed by a boost dose of 16 Gy/8 fractions to the PTVboost, resulting in a total dose of 66 Gy/33 fractions/6.5 weeks. For the CHARTWEL arm, the dose was 39 Gy/26 fractions, 3 fractions daily, excluding weekends, to the PTVinitial and a boost dose to PTVboost of 21 Gy/14 fractions, resulting in a total dose of 60 Gy/40 fractions/2.5 weeks. The interfraction-interval was at least 6 h.
Esophagus contouring and dose volume parameters In all patients, the external surface of the esophagus was retrospectively contoured in the present study in each CT slice from the lower border of the cricoid cartilage to the gastro-esophageal junction for consistency by the same radiation oncologist in the Oncentra MasterPlan planning system (ELEKTA, Sweden) using a mediastinal window setting. Contrast-enhanced planning CTs, after barium swallow, were available in the majority of patients. The mean esophageal dose (Dmean) and absolute esophageal volumes (Vx,abs) receiving at least 20, 30, 40, 50 or 60 Gy were recorded; e.g V40,abs is esophageal volume (in cm3) receiving 40 Gy or higher. In addition, corresponding relative dose–volume parameters Vx were calculated, given by Vx = Vx,abs/V0 with the total esophageal volume V0. To compare the treatment arms, dose–volume parameters were adjusted for the different fractionation schedules. This included a transformation to 2 Gy equivalent doses (EQD2). In addition, dose–volume parameters of the CHARTWEL arm were rescaled by a factor, f ¼ 66 1:2 ¼ 1:15, such that the prescribed tumour dose 60 1:15 of both arms is equal, while for the conventional arm f ¼ 1. This scaling assumes equal clinical efficacy of both treatment arms, as shown previously for tumors [1]. It accounts for differing repopulation and incomplete repair of the mucosa in both arms. The resulting dose–volume parameters (VEQD2,x) were calculated on a voxel-wise basis from the physical doses using
EQD2v ¼ f n dv
v 1 þ ad=b
1 þ 2Gy a=b
where dv is the physical dose per fraction in voxel v , a=b ¼ 10 and n is the number of fractions (33 for the conventional arm and 40 for CHARTWEL).
Dysphagia Normal tissue effects were scored weekly during the first 8 weeks from start of radiotherapy, then every 3 months up to 2 years, after that every 6 months up to 5 years. Dysphagia was scored using the MRC–CHART scoring system [12]: grade 0 = no dysphagia, grade 1 = some discomfort on swallowing – no disturbance of diet, grade 2 = soft diet required, grade 3 = fluids only, 4 = severe difficulties even with fluids. Dysphagia that occurred within 3 months from start of radiotherapy was considered as early radiation-induced esophageal toxicity (RIET) and thereafter as late RIET. The maximum dysphagia score found in the respective time interval was used.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
3
R. Bütof et al. / Radiotherapy and Oncology xxx (xxxx) xxx
Statistical analysis
Results
Logistic regression was performed to predict the endpoint early RIET using dose–volume parameters and other clinical variables. Early RIET was dichotomized into the two categories, grade < 2 vs. grade 2. Since there were only 3 patients who experienced late RIET grade 2, this endpoint was not considered. The analysis was done for the entire study group as well as for the individual treatment arms. Each parameter was tested in univariable analysis. Since clinical variables did not show significant correlations with early RIET grade 2, multivariable regression was not performed. The dosimetric parameters were tested independently as they were highly correlated with each other. The latent period of early RIET (time interval from the start of radiotherapy to the onset of esophagitis) was estimated by the Kaplan–Meier method and compared between both treatment arms using the log-rank test. Parameters were compared between the treatment arms using the chi-squared test (for categorical variables) or the Mann–Whitney–U test (for continuous variables). Due to the exploratory nature of this study, multiple testing corrections were not applied. SPSS version 25 was used, p-values below 0.05 were considered as statistically significant and two-sided tests were performed.
146 patients were included in this post-hoc analysis of prospectively collected data. Characteristics of the whole patient cohort showed no significant differences between both treatment schedules (Table 1). The median tumor volume was 66.1 cm3 (range: 5.2–422.0). The median age of patients was 69.6 years (range: 49.6–89.4) and the median weight loss before start of therapy was 2.6%. Only 8 patients were female. Half of the patients received CHARTWEL (73/146). Two thirds of patients (67.8%) had squamous cell carcinoma (SCC) and only 12.3% had node-negative tumors. More than 80% of the cohort has been classified as stage III. Around 20% of patients had received chemotherapy prior to radiotherapy. Only 13 patients (8.9%) did not develop early radiation-induced esophageal toxicity (RIET), 84 patients (57.5%) had maximum grade 1 esophagitis, 41 patients (28.1%) had maximum grade 2, 6 patients (4.1%) developed maximum grade 3 and 2 patients (1.4%) suffered from grade 4 early esophagitis. Thus, a total of 49 patients (33.6%) developed early RIET grade 2 or higher. Only 3 patients (2.1%) experienced late RIET grade 2.
Table 1 Patient characteristics for both treatment arms and the combined cohort. Variable
Conventional arm
CHARTWEL arm
of 73
%
of 73
%
p-Value
All patients of 146
%
Gender
Male Female
68 5
93.2 6.8
70 3
95.9 4.1
0.47
138 8
94.5 5.5
T stage
1 2 3 4
7 16 15 35
9.6 21.9 20.5 47.9
5 24 20 24
6.8 32.9 27.4 32.9
0.20
12 40 35 59
8.2 27.4 24.0 40.4
N stage
0 1 2 3
7 13 45 8
9.6 17.8 61.6 11.0
11 14 34 14
15.1 19.2 46.6 19.1
0.25
18 27 79 22
12.3 18.5 54.1 15.1
WHO
0 1
11 62
15.1 84.9
16 57
21.9 78.1
0.29
27 119
18.5 81.5
Stage
I II IIIA IIIB
5 5 26 37
6.8 6.8 35.6 50.7
7 5 31 30
9.6 6.8 42.5 41.1
0.68
12 10 57 67
8.2 6.8 39.0 46.0
Grading
1 2 3 Missing
1 38 24 10
1.4 52.1 32.9 13.7
1 44 18 10
1.4 60.3 24.7 13.7
0.52
2 82 42 20
143 56.2 28.8 13.7
Histology
SCC Adeno Large cell Other Missing
51 13 4 2 3
69.9 17.8 5.5 2.7 4.1
48 17 2 0 6
65.8 23.3 2.7 0.0 8.2
0.36
99 30 6 2 9
67.8 20.5 4.1 1.4 6.2
Previous chemotherapy
No Yes
57 16
78.1 21.9
59 14
80.8 19.2
0.68
116 30
79.5 20.5
RIET
0 1 2 3 4
10 42 18 3 0
13.7 57.5 24.7 4.1 0.0
3 42 23 3 2
4.1 57.5 31.5 4.1 2.7
0.17
13 84 41 6 2
8.9 57.5 28.1 4.1 1.4
Late RIET
0 1 2 3 4 Missing
45 6 1 0 1 20
61.6 8.2 1.4 0.0 1.4 27.4
48 8 1 0 0 16
65.8 11.0 1.4 0.0 0.0 21.9
0.74
93 14 2 0 1 36
63.7 9.6 1.4 0.0 0.7 24.7
Variable
Median (range)
Median (range)
p-Value
Median (range)
Age (years) Volume Tumour (cm3) Weight loss (%)
68.9 (49.6–89.4) 70.5 (5.4–422.0) 2.8 (-17.0–12.0)
70.1 (54.4–86.0) 66.1 (5.2–326.6) 1.5 (-14.0–9.0)
0.15 0.70 0.14
69.6 (49.6–89.4) 66.1 (5.2–422.0) 2.6 (17.0–12.0)
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
4
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
Table 2 Dose–volume parameters (physical dose and EQD2) of the esophagus for both treatment arms and the combined cohort. Comparisons between cohorts were performed by MannWhitney-U tests. Dose–volume parameter
Conventional arm Median (Range)
CHARTWEL arm Median (Range)
p-Value
All patients Median (Range)
Dmean (Gy) EQD2mean (Gy) Total esophageal volume (cm3) V20 (%) V30 (%) V40 (%) V50 (%) V60 (%) VEQD2,20 (%) VEQD2,30 (%) VEQD2,40 (%) VEQD2,50 (%) VEQD2,60 (%)
30.0 29.1 48.3 51.8 49.7 47.3 39.6 24.9 51.8 48.8 44.2 36.8 23.9
27.0 (15.9–41.8) 28.9 (16.5–45.0) 48.1 (34.5–92.3) 52.6 (32.5–85.5) 47.5 (26.7–79.8) 40.0 (1.8–65.8) 28.1 (0.0–50.6) 1.1 (0.0–35.1) 52.6 (32.5–85.5) 48.5 (27.8–81.7) 43.5 (19.4–72.8) 31.7 (0.0–60.0) 21.1 (0.0–45.3)
<0.001 0.43 0.35 0.40 0.38 <0.001 <0.001 <0.001 0.27 0.98 0.24 0.002 0.094
29.3 29.0 48.2 52.4 49.5 43.2 34.7 13.0 52.3 48.7 43.9 33.4 22.4
(19.0–55.1) (17.9–54.7) (26.3–102.7) (32.7–98.5) (29.3–88.4) (26.4–85.8) (14.9–83.5) (0.0–70.7) (32.7–97.6) (28.3–85.8) (26.4–85.8) (6.7–83.5) (0.0–68.1)
(15.9–55.1) (16.5–54.7) (26.3–102.7) (32.5–98.5) (16.7–88.4) (1.8–85.8) (0.0–83.5) (0.0–70.7) (32.5–97.6) (27.8–85.8) (19.4–85.8) (0.0–83.5) (0.0–68.1)
grade 2 (p = 0.075 and p = 0.078, respectively) for the combined treatment arms. Since no clinical parameter was significantly correlated with early RIET, multivariable analyses were not performed. The latency period of early RIET was calculated and compared between both treatment arms (Fig. 3). Patients receiving CHARTWEL developed early RIET grade 1 significantly earlier than those receiving conventional radiotherapy (p = 0.011). In contrast, the median time period for recovery of grade 1 esophagitis was significantly longer for patients with conventional fractionation compared to the CHARTWEL schedule (median [range]: 21 [12–49] days vs. 15 [7–84] days, p = 0.028). For grade 2 toxicity, median times were 28 [20–56] days and 21 [7–84] days (p = 0.74), respectively, confirming this tendency.
Discussion Fig. 1. Average dose volume histograms of the esophagus using the physical dose distribution (solid lines) or the rescaled dose distributions (dashed lines) for conventional treatment (black) and CHARTWEL (grey).
In line with the results of the CHART trial and our previous report on the complete multicenter study cohort [1,12], patients receiving the accelerated schedule experienced earlier and more severe esophagitis (20.5% vs. 9.6% grade 2 in the accelerated vs. conventional fractionation arm at week 3 of treatment, respectively). Cumulative esophagitis rates after 90 days have been 95.9% (accelerated group) vs. 86.3% (conventional group) for grade 1 and 6.8% vs. 4.1% for grade 3 (Fig. 3). The mean volume of the esophagus was 48.2 cm3 (26.3–102.7). Table 2 shows the relative esophageal volume receiving different dose levels amongst other dosimetric parameters. While dose–volume parameters based on the physical dose distribution showed highly significant differences between the treatment arms, in particular for high doses, parameters based on the rescaled dose distribution possessed smaller differences between the arms (Fig. 1). In univariable logistic regression none of the dose–volume parameters showed a significant association with early RIET grade 2 in patients treated with a shrinking-field conventional irradiation within this trial. In contrast, for patients receiving CHARTWEL, the physical dose–volumes parameters V40 and V50; and re-scaled values VEQD2,50 and VEQD2,60 were significant predictors of early RIET grade 2 (Table 3). Exemplarily, the logistic models using either V50 or VEQD2,50 are presented in Fig. 2 for both treatment arms separately and for the combined data. Among the clinical parameters, gender and prior chemotherapy revealed a statistical trend for association with early RIET
Early RIET is one of the major radiation-induced toxicities in patients with NSCLC and the primary cause of treatment interruptions as, unlike pneumonitis, it occurs early during radiotherapy. So far, several studies have addressed a wide range of dose–volume parameters tested for correlation with early RIET, but there has been no clear dose or dose–volume parameter defined as significant and practically relevant predictor. Furthermore, most of the reported data focused on conventional fractionation and on concurrent radiochemotherapy in lung cancer patients [11,20], while dose-constraints for minimizing the risk of esophagitis in hyperfractionated-accelerated irradiation schedules are missing so far. As the age of patients with NSCLC and corresponding comorbidities continues to increase, primary radiotherapy without simultaneous chemotherapy remains a commonly used treatment alternative in clinical practice. Based on the results of the CHART or CHARTWEL trial [1,12] as well as individual patient data metaanalysis [9] hyperfractionated-accelerated irradiation schedules may be superior options for patients not fit for simultaneous radiochemotherapy compared to conventional fractionation. Therefore, predictors of early esophageal toxicity in those treatment regimes are highly desirable. We derived our results from a post-hoc analysis of a prospective randomized trial and found, as expected from the higher dose intensity and our previous report on the complete multicenter study cohort [1], that patients receiving CHARTWEL developed acute RIET more often and significantly earlier than those receiving conventional radiotherapy (Fig. 3). Interestingly, higher rates of early RIET occurred in the CHARTWEL cohort although esophageal high-dose volumes were smaller compared to the conventional fractionation cohort (Fig. 1). The cumulative incidence of esophagi-
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
5
R. Bütof et al. / Radiotherapy and Oncology xxx (xxxx) xxx Table 3 Exemplary univariable logistic regression models for early RIET grade 2 for both treatment arms and the combined cohorts. Parameter
Conventional arm
CHARTWEL arm
All patients
OR (95% CI)
p-Value
OR (95% CI)
p-Value
OR (95% CI)
p-Value
V40 Constant
9.94 (0.09–1130) 0.14 (0.01–1.39)
0.34 0.093
91.7 (1.25–6734) 0.10 (0.02–0.63)
0.039 0.014
13.6 (0.72–256) 0.16 (0.04–0.62)
0.082 0.008
V50 Constant
9.78 (0.08–1186) 0.16 (0.02–1.24)
0.35 0.079
26.9 (1.03–705) 0.27 (0.10–0.73)
0.048 0.010
4.23 (0.43–41.3) 0.31 (0.14–0.73)
0.22 0.007
VEQD2,50 Constant
2.77 (0.04–181.8) 0.28 (0.05–1.45)
0.63 0.13
30.6 (1.23–764) 0.23 (0.08–0.69)
0.037 0.008
6.63 (0.61–72.3) 0.27 (0.11–0.65)
0.12 0.004
VEQD2,60 Constant
1.05 (0.03–33.6) 0.40 (0.15–1.05)
0.98 0.062
40.9 (1.11–1501) 0.30 (0.12–0.73)
0.043 0.008
4.63 (0.42–51.0) 0.36 (0.19–0.68)
0.21 0.002
Fig. 2. Logistic regression of early RIET grade 2 for both treatment arms (conventional: long dashed lines, red; CHARTWEL: short dashed lines, blue) and the combined cohorts (solid lines, black) depending on the relative volume of the oesophagus receiving 50 Gy physical dose [V50, left] and 50 Gy rescaled dose (including transformation to 2 Gy equivalent doses adjusted for repopulation and incomplete repair) [VEQD2,50; right]. Lines show the regression results while dots show the proportion of early RIET grade 2 in pooled patient groups. Bars represent the standard error of the mean.
Fig. 3. Cumulative incidence of early RIET grade 1 (top), grade 2 (center) and grade 3 (bottom) after the start of radiotherapy for both treatment arms (conventional: black dashed line, CHARTWEL: solid gray line).
tis grade 2 of 38 % reported here after CHARTWEL is in line with the incidence of 33% observed in the recently published DART trial on non-small cell lung cancer, which used two daily fractions of 1.8 Gy, and dose constraints of a mean esophageal dose of 34 Gy with a maximum of 80 Gy in the center of the esophagus [23]. Although the incidence of esophagitis grade 3 was higher in the CHART trial reported by Saunders et al. [12] who used 2D treatment planning, the kinetics of RIET over time in that study compares well to the observations after CHARTWEL [1]. Notably, the
median period of time for regression of grade 1 esophagitis was significantly shorter for patients treated with CHART or CHARTWEL compared to conventional fractionation [1,12]. This suggests a greater stimulation of repopulation of mucosal cells by hyperfractionated-accelerated radiation which, although probably negletable during the short treatment courses, promotes healing of mucositis by repopulation within a few weeks after end of treatment [16]. In addition to the mucositis scores by radiation oncologists [1,12], also patient reported data from the CHARTWEL trial indicate that dysphagia at 8 weeks after start of treatment or later was not increased and the quality of life not impaired after accelerated-hyperfractionated fractionation relative to conventional fractionation [6]. In our study, univariate analysis did not reveal any dose–volume parameter that was significantly associated with early RIET in those patients treated with shrinking-field conventional fractionation. In contrast, for patients receiving CHARTWEL, several dose–volume parameters such as V40 and V50, were significant predictors of early RIET. As a clinical consequence, all efforts should therefore be made to restrict esophageal volumes receiving 40 Gy or higher when hyperfractionated-accelerated treatment schedules are applied. Similar dose–volume parameters have also been reported to be associated with esophagitis in patients receiving hypofractionated radiochemotherapy [7,17,18] or conventional simultaneous radiochemotherapy [10] and are in line with QUANTEC based recommendations for esophageal dose–volume limits [21]. This suggests that dose–volume parameters for prediction of esophagitis may be more readily detectable in more intense treatment regimes. In contrast, in less dose-intensive schedules other factors, e.g. repopulation commencing already during the long treatment course, might mask these possible correlations with dose–volume parameters. In this context it is interesting to
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
6
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
note, that also the volume parameters predicting local tumor control were more dominant after hyperfractionated-accelerated fractionation compared to standard fractionation in the CHARTWEL trial [15]. To account for different repopulation rates, effects of dose per fraction and incomplete repair between the treatment arms we rescaled the dose–volume parameters to biological doses assuming equal clinical efficacy of both treatment arms. As shown by our results, physical doses alone correlated with early RIET in the CHARTWEL cohort but not after conventional fractionation. The used scaling factor of 1.15 increased the dose values in the CHARTWEL arm in comparison to the conventional arm, e.g. compare V60 with VEQD2,60 in table 2. Even after rescaling, however, dose parameters were somewhat larger for the conventional arm in contrast to the higher toxicity in the CHARTWEL arm, suggesting that the scaling factor might also be higher. To test the presented NTCP models, we thus performed additional logistic regression analyses of RIET grade 2 on the combined patient cohort including one physical dose–volume parameter and its interaction with the treatment arm, resembling the scaling factor. Also in these analyses, V40 and V50 were still significantly related to RIET grade 2. In general, one would expect similar dose–volume parameters for CHARTWEL and conventional fractionation after a perfect biological rescaling of doses. As already mentioned this is not the case in our analysis (Fig. 1), which at least in parts is caused by the fact that perfect rescaling is counteracted by the different time points and dose levels of downsizing the PTVinitial to the PTVboost. This finding supports the notion that dose–volume parameters for radiation esophagitis determined in a specific and time dependent setting of field arrangements cannot be easily transferred to another setting. Although target volume concept and planning technique used in the evaluated trial is not longer comparable with clinical standard today, the CHARTWEL fractionation schedule itself is still clinically applied. Therefore, our dose–volume parameter findings for early RIET are still applicable for this kind of hyperfractionated-accelerated irradiation. Among the clinical parameters, gender and prior chemotherapy revealed only a statistical trend for association with early RIET when combining both treatment arms, but not separately in one group. Former studies showed a significant impact of chemotherapy application and number of cycles for esophageal toxicity, especially in combined sequential or simultaneous radichemotherapy schedules [18,22]. In line with these results, recently published data suggest neutropenia as a possible risk factor and biomarker for higher grades of RIET in patients treated with concurrent radiochemotherapy [3]. Further perspectives for more individualized prediction of radiation toxicity include combined models of dose–volume parameters and potential biomarkers, e.g. cytokine blood levels. So far, conflicting results of predicitive value of IL-8 or other cytokine blood levels for radiation esophagitis have been reported [19,5]. Therefore, the final value of such biomarkers is currently unclear; they could possibly contribute to improvement of individualized NTCP models in the future. For clinical implementation of individualized dose-constraints, further prospective trials, especially on combination with possible biomarkers, are needed. In conclusion, for prediction of early RIET more dominant dose– volume effects were found after hyperfractionated-accelerated compared to conventional fractionation, especially for higher esophageal doses. This suggests that dose–volume parameters for correlation with esophagitis may be more readily detectable in more intense treatment regimes. In contrast, in less dose-intensive schedules other factors, e.g. repopulation commencing during long treatment times, may mask these possible correlations with dose– volume parameters. Our findings support the notion that dose–volume parameters for radiation esophagitis determined in a specific and time dependent setting of field arrangements can not be easily
transferred to another setting. In clinical practice esophageal volumes receiving 40 Gy or more should be strictly limited in hyperfractionated-accelerated fractionation schemes. Conflict of interest None. References [1] Baumann M, Herrmann T, Koch R, et al. Final results of the randomized phase III CHARTWEL-trial (ARO 97-1) comparing hyperfractionated-accelerated versus conventionally fractionated radiotherapy in non-small cell lung cancer (NSCLC). Radiother Oncol 2011;100:76–85. [2] Bradley J, Graham MV, Winter K, et al. Toxicity and outcome results of RTOG 9311: a phase I-II dose-escalation study using three-dimensional conformal radiotherapy in patients with inoperable non-small-cell lung carcinoma. Int J Radiat Oncol Biol Phys 2005;61:318–28. [3] Everitt S, Duffy M, Bressel M, et al. Association of oesophageal radiation dose volume metrics, neutropenia and acute radiation oesophagitis in patients receiving chemoradiotherapy for non-small cell lung cancer. Radiat Oncol 2016;11:20. [4] Goeckenjan G, Sitter H, Thomas M, et al. Prevention, diagnosis, therapy, and follow-up of lung cancer: interdisciplinary guideline of the German Respiratory Society and the German Cancer Society. Pneumologie 2011;65:39–59. [5] Hawkins PG, Boonstra PS, Hobson ST, et al. Prediction of radiation esophagitis in non-small cell lung cancer using clinical factors, dosimetric parameters, and pretreatment cytokine levels. Transl Oncol 2018;11:102–8. [6] Hechtner M, Krause M, Konig J, et al. Long-term quality of life in inoperable non-small cell lung cancer patients treated with conventionally fractionated compared to hyperfractionated accelerated radiotherapy - Results of the randomized CHARTWEL trial. Radiother Oncol 2018;126:283–90. [7] Kwint M, Uyterlinde W, Nijkamp J, et al. Acute esophagus toxicity in lung cancer patients after intensity modulated radiation therapy and concurrent chemotherapy. Int J Radiat Oncol Biol Phys 2012;84:e223–8. [8] Le Chevalier T, Arriagada R, Quoix E, et al. Radiotherapy alone versus combined chemotherapy and radiotherapy in nonresectable non-small-cell lung cancer: first analysis of a randomized trial in 353 patients. J Natl Cancer Inst 1991;83:417–23. [9] Mauguen A, Le Pechoux C, Saunders MI, et al. Hyperfractionated or accelerated radiotherapy in lung cancer: an individual patient data meta-analysis. J Clin Oncol 2012;30:2788–97. [10] Rodriguez N, Algara M, Foro P, et al. Predictors of acute esophagitis in lung cancer patients treated with concurrent three-dimensional conformal radiotherapy and chemotherapy. Int J Radiat Oncol Biol Phys 2009;73:810–7. [11] Rose J, Rodrigues G, Yaremko B, Lock M, D’Souza D. Systematic review of dosevolume parameters in the prediction of esophagitis in thoracic radiotherapy. Radiother Oncol 2009;91:282–7. [12] Saunders M, Dische S, Barrett A, Harvey A, Gibson D, Parmar M. Continuous hyperfractionated accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small-cell lung cancer: a randomised multicentre trial. CHART Steering Committee. Lancet 1997;350:161–5. [13] Sause W, Kolesar P, Taylor SI, et al. Final results of phase III trial in regionally advanced unresectable non-small cell lung cancer: Radiation Therapy Oncology Group, Eastern Cooperative Oncology Group, and Southwest Oncology Group. Chest 2000;117:358–64. [14] Socinski MA, Morris DE, Halle JS, et al. Induction and concurrent chemotherapy with high-dose thoracic conformal radiation therapy in unresectable stage IIIA and IIIB non-small-cell lung cancer: a dose-escalation phase I trial. J Clin Oncol 2004;22:4341–50. [15] Soliman M, Yaromina A, Appold S, et al. GTV differentially impacts locoregional control of non-small cell lung cancer (NSCLC) after different fractionation schedules: subgroup analysis of the prospective randomized CHARTWEL trial. Radiother Oncol 2013. [16] Stewart FA, Dorr W. Milestones in normal tissue radiation biology over the past 50 years: from clonogenic cell survival to cytokine networks and back to stem cell recovery. Int J Radiat Biol 2009;85:574–86. [17] Uyterlinde W, Belderbos J, Baas C, et al. Prediction of acute toxicity grade >/=3 in patients with locally advanced non-small-cell lung cancer receiving intensity modulated radiotherapy and concurrent low-dose Cisplatin. Clin Lung Cancer 2013;14:541–8. [18] Uyterlinde W, Chen C, Kwint M, et al. Prognostic parameters for acute esophagus toxicity in intensity modulated radiotherapy and concurrent chemotherapy for locally advanced non-small cell lung cancer. Radiother Oncol 2013;107:392–7. [19] Wang S, Campbell J, Stenmark MH, et al. A model combining age, equivalent uniform dose and IL-8 may predict radiation esophagitis in patients with nonsmall cell lung cancer. Radiother Oncol 2018;126:506–10. [20] Werner-Wasik M, Paulus R, Curran Jr WJ, Byhardt R. Acute esophagitis and late lung toxicity in concurrent chemoradiotherapy trials in patients with locally advanced non-small-cell lung cancer: analysis of the radiation therapy oncology group (RTOG) database. Clin Lung Cancer 2011;12:245–51.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
R. Bütof et al. / Radiotherapy and Oncology xxx (xxxx) xxx [21] Werner-Wasik M, Yorke E, Deasy J, Nam J, Marks LB. Radiation dose-volume effects in the esophagus. Int J Radiat Oncol Biol Phys 2010;76:S86–93. [22] Wijsman R, Dankers F, Troost EG, et al. Multivariable normal-tissue complication modeling of acute esophageal toxicity in advanced stage nonsmall cell lung cancer patients treated with intensity-modulated (chemo-) radiotherapy. Radiother Oncol 2015;117:49–54.
7
[23] Wurstbauer K, Zehentmayr F, Deutschmann H, et al. DART-bid for locoregionally advanced NSCLC: summary of acute and late toxicity with longterm follow-up; experiences with pulmonary dose constraints. Strahlenther Onkol 2017;193:315–23.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Original Article
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy Rebecca Bütof a,b,c,⇑, Steffen Löck a,b,d, Maher Soliman b,e, Robert Haase a, Rosalind Perrin a,f, Christian Richter a,b,d,g, Steffen Appold a,b,c, Mechthild Krause a,b,c,d,g,h, Michael Baumann a,b,d,g,h a OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf; b Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden; c National Center for Tumor Diseases (NCT), Partner Site Dresden, Germany: German Cancer Research Center (DKFZ), Heidelberg, Germany; Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany, and; Helmholtz Association / Helmholtz-Zentrum Dresden - Rossendorf (HZDR), Dresden, Germany; d German Cancer Consortium (DKTK), Partner Site Dresden, Germany; e Oncology Department, Faculty of Medicine, Alexandria University, Egypt; f Strahlenklinik, Universitätsklinikum Erlangen; g Helmholtz-Zentrum Dresden - Rossendorf, Institute of Radiooncology – OncoRay; and h German Cancer Research Center (DKFZ), Heidelberg, Germany
a r t i c l e
i n f o
Article history: Received 24 May 2019 Received in revised form 28 October 2019 Accepted 4 November 2019 Available online xxxx Keywords: Dose–volume parameters Esophagitis Radiotherapy Accelerated Prediction Non-small cell lung cancer
a b s t r a c t Background and purpose: Early radiation-induced esophageal toxicity (RIET) is one of the major side effects in patients with non-small cell lung cancer (NSCLC) and can be a reason for treatment interruptions. As the age of patients with NSCLC and corresponding comorbidities continue to increase, primary radiotherapy alone is a commonly used alternative treatment in these cases. The aim of the present study is to compare dosimetric and clinical parameters from the previously reported CHARTWEL trial for their ability to predict esophagitis and investigate potential differences in the accelerated and conventional fractionation arm. Material and methods: 146 patients of the Dresden cohort of the randomized phase III CHARTWEL trial were included in this post-hoc analysis. Side effects were prospectively scored weekly during the first 8 weeks from start of radiotherapy. To compare both treatment arms, recorded dose–volume parameters were adjusted for the different fractionation schedules. Logistic regression was performed to predict early RIET for the entire study group as well as for the individual treatment arms. Different dosimetric and clinical parameters were tested. Results: Patients receiving the accelerated CHARTWEL schedule experienced earlier and more severe esophagitis (e.g. 20.5% vs. 9.6% grade 2 at week 3, respectively). In contrast, the median time period for recovery of grade 1 esophagitis was significantly longer for patients with conventional fractionation compared to the CHARTWEL group (median [range]: 21 [12–49] days vs. 15 [7–84] days, p = 0.028). In univariable logistic regression none of the dose–volume parameters showed a significant correlation with early RIET grade 2 in the conventional irradiation group. In contrast, for patients receiving CHARTWEL, the physical dose–volumes parameters V40 and V50; and re-scaled values VEQD2,50 and VEQD2,60 were significant predictors of early RIET grade 2. Dose–volume parameters remained different between CHARTWEL and conventional fractionation even after biological rescaling. Conclusion: Our results show a more dominant dose-volume effect in the CHARTWEL arm compared to conventional fractionation, especially for higher esophageal doses. These findings support the notion that dose–volume parameters for radiation esophagitis determined in a specific and time dependent setting of field arrangements can not be easily transferred to another setting. In clinical practice esophageal volumes receiving 40 Gy or more should be strictly limited in hyperfractionated-accelerated fraction schemes. Ó 2019 Elsevier B.V. All rights reserved. Radiotherapy and Oncology xxx (2019) xxx–xxx
⇑ Corresponding author at: Department of Radiotherapy and Radiation Oncology, University Hospital Carl Gustav Carus and Faculty of Medicine of Technische Universität Dresden, Fetscherstraße 74, 01307 Dresden, Germany. E-mail address: [email protected] (R. Bütof).
Non-small cell lung cancer (NSCLC) is one of the most common malignant diseases worldwide [4]. Simultaneous radiochemotherapy in the treatment of locally advanced lung cancer yields better outcome than radiotherapy alone [8]. While this treatment regimen is the standard of care, it is also evoking more side effects
https://doi.org/10.1016/j.radonc.2019.11.002 0167-8140/Ó 2019 Elsevier B.V. All rights reserved.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
2
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
[8,13]. One of the frequent side effects is early radiation-induced esophageal toxicity (RIET), which can be a reason for treatment interruptions. Complete avoidance of esophagus irradiation in lung cancer radiotherapy is impossible due to several factors: the large, irregularly shaped and centrally located lung cancer, frequent involvement of mediastinal lymph nodes (LN), and the central location and length of the esophagus [2,14]. As the age of patients with NSCLC and corresponding comorbidities continues to increase, primary radiotherapy alone remains a commonly used treatment alternative for patients unable to tolerate radiochemotherapy. Hyperfractionated-accelerated irradiation schedules have shown comparable or even improved outcome data of NSCLC patients compared to conventional fractionation, however, also these schedules are associated with increased early esophageal reactions [1,9,12]. Valid doseconstraints for avoidance of esophagitis in hyperfractionatedaccelerated irradiation schedules are missing so far. In the present study, we evaluate esophageal toxicity utilizing data from the prospective CHARTWEL trial [1]. The aim is to compare dosimetric and clinical parameters for their ability to predict early esophagitis and investigate potential differences in the accelerated and conventional fractionation arm. Patients and methods Patients Between 1997 and 2005, 163 patients with NSCLC were treated at the Department of Radiotherapy and Radiation Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Germany, as a part of the randomized phase III CHARTWEL trial, which was carried out by 15 centers in Germany, Poland and Czech Republic. Clinical scoring of esophagitis has been done prospectively in all patients and the results have been previously reported [1]. Treatment planning CT scans and irradiation plans of the patients treated in Dresden could be retrieved and evaluated retrospectively, except for 16 patients because of technical reasons resulting in the exclusion of these patients from this study. In addition, one patient was excluded due to missing esophagitis data. Final results of the CHARTWEL trial as well as a post-hoc analysis of specific questions within the Dresden subcohort have already been published [1,15]. In brief, all patients had inoperable histologically proven NSCLC, they were stratified according UICC staging system (into I: 8%, II: 7%, IIIA: 39%, IIIB: 46%), patients were >18 years and had WHO performance score 0–1. Patients were excluded from the original trial if they had distant metastases, supraclavicular LN metastases, malignant pleural effusion, FEV1 <1 liter under optimized medical therapy, unintended weight loss >15% over the last 6 months, prior radiotherapy, surgical resection if more than biopsy, prior or concurrent other malignant disease except basalioma, T1 squamous cell carcinoma of the skin, T1 glottic carcinoma, and carcinoma in situ of cervix uteri, other grave comorbidities that expected to limit the short term life expectancy of the patient, cardiac pacemaker in the radiation field except if the cardiologist agreed, pregnancy or participation in other clinical trials or if follow up was not possible. Treatment In both treatment arms, the initial planning target volume (PTVinitial) included the mediastinum and the primary tumour with a margin of 1–1.5 cm ipsilaterally and 1 cm contralaterally. The mediastinum was defined as the area from the jugular fossa to 3 cm below the carina. Supraclavicular LN were treated for
tumours of the upper lobes and irradiation of the contralateral hilum was done in case of involvement. The high-dose PTV (PTVboost) included the primary tumour and positive LN with a margin of 1 cm. All patients in both arms were treated with 3D conformal radiotherapy using a linear accelerator with a photon energy 6 MV and shrinking-field technique. The prescribed dose in the conventional arm was 50 Gy/25 fractions/5 weeks to the PTVinitial followed by a boost dose of 16 Gy/8 fractions to the PTVboost, resulting in a total dose of 66 Gy/33 fractions/6.5 weeks. For the CHARTWEL arm, the dose was 39 Gy/26 fractions, 3 fractions daily, excluding weekends, to the PTVinitial and a boost dose to PTVboost of 21 Gy/14 fractions, resulting in a total dose of 60 Gy/40 fractions/2.5 weeks. The interfraction-interval was at least 6 h.
Esophagus contouring and dose volume parameters In all patients, the external surface of the esophagus was retrospectively contoured in the present study in each CT slice from the lower border of the cricoid cartilage to the gastro-esophageal junction for consistency by the same radiation oncologist in the Oncentra MasterPlan planning system (ELEKTA, Sweden) using a mediastinal window setting. Contrast-enhanced planning CTs, after barium swallow, were available in the majority of patients. The mean esophageal dose (Dmean) and absolute esophageal volumes (Vx,abs) receiving at least 20, 30, 40, 50 or 60 Gy were recorded; e.g V40,abs is esophageal volume (in cm3) receiving 40 Gy or higher. In addition, corresponding relative dose–volume parameters Vx were calculated, given by Vx = Vx,abs/V0 with the total esophageal volume V0. To compare the treatment arms, dose–volume parameters were adjusted for the different fractionation schedules. This included a transformation to 2 Gy equivalent doses (EQD2). In addition, dose–volume parameters of the CHARTWEL arm were rescaled by a factor, f ¼ 66 1:2 ¼ 1:15, such that the prescribed tumour dose 60 1:15 of both arms is equal, while for the conventional arm f ¼ 1. This scaling assumes equal clinical efficacy of both treatment arms, as shown previously for tumors [1]. It accounts for differing repopulation and incomplete repair of the mucosa in both arms. The resulting dose–volume parameters (VEQD2,x) were calculated on a voxel-wise basis from the physical doses using
EQD2v ¼ f n dv
v 1 þ ad=b
1 þ 2Gy a=b
where dv is the physical dose per fraction in voxel v , a=b ¼ 10 and n is the number of fractions (33 for the conventional arm and 40 for CHARTWEL).
Dysphagia Normal tissue effects were scored weekly during the first 8 weeks from start of radiotherapy, then every 3 months up to 2 years, after that every 6 months up to 5 years. Dysphagia was scored using the MRC–CHART scoring system [12]: grade 0 = no dysphagia, grade 1 = some discomfort on swallowing – no disturbance of diet, grade 2 = soft diet required, grade 3 = fluids only, 4 = severe difficulties even with fluids. Dysphagia that occurred within 3 months from start of radiotherapy was considered as early radiation-induced esophageal toxicity (RIET) and thereafter as late RIET. The maximum dysphagia score found in the respective time interval was used.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
3
R. Bütof et al. / Radiotherapy and Oncology xxx (xxxx) xxx
Statistical analysis
Results
Logistic regression was performed to predict the endpoint early RIET using dose–volume parameters and other clinical variables. Early RIET was dichotomized into the two categories, grade < 2 vs. grade 2. Since there were only 3 patients who experienced late RIET grade 2, this endpoint was not considered. The analysis was done for the entire study group as well as for the individual treatment arms. Each parameter was tested in univariable analysis. Since clinical variables did not show significant correlations with early RIET grade 2, multivariable regression was not performed. The dosimetric parameters were tested independently as they were highly correlated with each other. The latent period of early RIET (time interval from the start of radiotherapy to the onset of esophagitis) was estimated by the Kaplan–Meier method and compared between both treatment arms using the log-rank test. Parameters were compared between the treatment arms using the chi-squared test (for categorical variables) or the Mann–Whitney–U test (for continuous variables). Due to the exploratory nature of this study, multiple testing corrections were not applied. SPSS version 25 was used, p-values below 0.05 were considered as statistically significant and two-sided tests were performed.
146 patients were included in this post-hoc analysis of prospectively collected data. Characteristics of the whole patient cohort showed no significant differences between both treatment schedules (Table 1). The median tumor volume was 66.1 cm3 (range: 5.2–422.0). The median age of patients was 69.6 years (range: 49.6–89.4) and the median weight loss before start of therapy was 2.6%. Only 8 patients were female. Half of the patients received CHARTWEL (73/146). Two thirds of patients (67.8%) had squamous cell carcinoma (SCC) and only 12.3% had node-negative tumors. More than 80% of the cohort has been classified as stage III. Around 20% of patients had received chemotherapy prior to radiotherapy. Only 13 patients (8.9%) did not develop early radiation-induced esophageal toxicity (RIET), 84 patients (57.5%) had maximum grade 1 esophagitis, 41 patients (28.1%) had maximum grade 2, 6 patients (4.1%) developed maximum grade 3 and 2 patients (1.4%) suffered from grade 4 early esophagitis. Thus, a total of 49 patients (33.6%) developed early RIET grade 2 or higher. Only 3 patients (2.1%) experienced late RIET grade 2.
Table 1 Patient characteristics for both treatment arms and the combined cohort. Variable
Conventional arm
CHARTWEL arm
of 73
%
of 73
%
p-Value
All patients of 146
%
Gender
Male Female
68 5
93.2 6.8
70 3
95.9 4.1
0.47
138 8
94.5 5.5
T stage
1 2 3 4
7 16 15 35
9.6 21.9 20.5 47.9
5 24 20 24
6.8 32.9 27.4 32.9
0.20
12 40 35 59
8.2 27.4 24.0 40.4
N stage
0 1 2 3
7 13 45 8
9.6 17.8 61.6 11.0
11 14 34 14
15.1 19.2 46.6 19.1
0.25
18 27 79 22
12.3 18.5 54.1 15.1
WHO
0 1
11 62
15.1 84.9
16 57
21.9 78.1
0.29
27 119
18.5 81.5
Stage
I II IIIA IIIB
5 5 26 37
6.8 6.8 35.6 50.7
7 5 31 30
9.6 6.8 42.5 41.1
0.68
12 10 57 67
8.2 6.8 39.0 46.0
Grading
1 2 3 Missing
1 38 24 10
1.4 52.1 32.9 13.7
1 44 18 10
1.4 60.3 24.7 13.7
0.52
2 82 42 20
143 56.2 28.8 13.7
Histology
SCC Adeno Large cell Other Missing
51 13 4 2 3
69.9 17.8 5.5 2.7 4.1
48 17 2 0 6
65.8 23.3 2.7 0.0 8.2
0.36
99 30 6 2 9
67.8 20.5 4.1 1.4 6.2
Previous chemotherapy
No Yes
57 16
78.1 21.9
59 14
80.8 19.2
0.68
116 30
79.5 20.5
RIET
0 1 2 3 4
10 42 18 3 0
13.7 57.5 24.7 4.1 0.0
3 42 23 3 2
4.1 57.5 31.5 4.1 2.7
0.17
13 84 41 6 2
8.9 57.5 28.1 4.1 1.4
Late RIET
0 1 2 3 4 Missing
45 6 1 0 1 20
61.6 8.2 1.4 0.0 1.4 27.4
48 8 1 0 0 16
65.8 11.0 1.4 0.0 0.0 21.9
0.74
93 14 2 0 1 36
63.7 9.6 1.4 0.0 0.7 24.7
Variable
Median (range)
Median (range)
p-Value
Median (range)
Age (years) Volume Tumour (cm3) Weight loss (%)
68.9 (49.6–89.4) 70.5 (5.4–422.0) 2.8 (-17.0–12.0)
70.1 (54.4–86.0) 66.1 (5.2–326.6) 1.5 (-14.0–9.0)
0.15 0.70 0.14
69.6 (49.6–89.4) 66.1 (5.2–422.0) 2.6 (17.0–12.0)
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
4
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
Table 2 Dose–volume parameters (physical dose and EQD2) of the esophagus for both treatment arms and the combined cohort. Comparisons between cohorts were performed by MannWhitney-U tests. Dose–volume parameter
Conventional arm Median (Range)
CHARTWEL arm Median (Range)
p-Value
All patients Median (Range)
Dmean (Gy) EQD2mean (Gy) Total esophageal volume (cm3) V20 (%) V30 (%) V40 (%) V50 (%) V60 (%) VEQD2,20 (%) VEQD2,30 (%) VEQD2,40 (%) VEQD2,50 (%) VEQD2,60 (%)
30.0 29.1 48.3 51.8 49.7 47.3 39.6 24.9 51.8 48.8 44.2 36.8 23.9
27.0 (15.9–41.8) 28.9 (16.5–45.0) 48.1 (34.5–92.3) 52.6 (32.5–85.5) 47.5 (26.7–79.8) 40.0 (1.8–65.8) 28.1 (0.0–50.6) 1.1 (0.0–35.1) 52.6 (32.5–85.5) 48.5 (27.8–81.7) 43.5 (19.4–72.8) 31.7 (0.0–60.0) 21.1 (0.0–45.3)
<0.001 0.43 0.35 0.40 0.38 <0.001 <0.001 <0.001 0.27 0.98 0.24 0.002 0.094
29.3 29.0 48.2 52.4 49.5 43.2 34.7 13.0 52.3 48.7 43.9 33.4 22.4
(19.0–55.1) (17.9–54.7) (26.3–102.7) (32.7–98.5) (29.3–88.4) (26.4–85.8) (14.9–83.5) (0.0–70.7) (32.7–97.6) (28.3–85.8) (26.4–85.8) (6.7–83.5) (0.0–68.1)
(15.9–55.1) (16.5–54.7) (26.3–102.7) (32.5–98.5) (16.7–88.4) (1.8–85.8) (0.0–83.5) (0.0–70.7) (32.5–97.6) (27.8–85.8) (19.4–85.8) (0.0–83.5) (0.0–68.1)
grade 2 (p = 0.075 and p = 0.078, respectively) for the combined treatment arms. Since no clinical parameter was significantly correlated with early RIET, multivariable analyses were not performed. The latency period of early RIET was calculated and compared between both treatment arms (Fig. 3). Patients receiving CHARTWEL developed early RIET grade 1 significantly earlier than those receiving conventional radiotherapy (p = 0.011). In contrast, the median time period for recovery of grade 1 esophagitis was significantly longer for patients with conventional fractionation compared to the CHARTWEL schedule (median [range]: 21 [12–49] days vs. 15 [7–84] days, p = 0.028). For grade 2 toxicity, median times were 28 [20–56] days and 21 [7–84] days (p = 0.74), respectively, confirming this tendency.
Discussion Fig. 1. Average dose volume histograms of the esophagus using the physical dose distribution (solid lines) or the rescaled dose distributions (dashed lines) for conventional treatment (black) and CHARTWEL (grey).
In line with the results of the CHART trial and our previous report on the complete multicenter study cohort [1,12], patients receiving the accelerated schedule experienced earlier and more severe esophagitis (20.5% vs. 9.6% grade 2 in the accelerated vs. conventional fractionation arm at week 3 of treatment, respectively). Cumulative esophagitis rates after 90 days have been 95.9% (accelerated group) vs. 86.3% (conventional group) for grade 1 and 6.8% vs. 4.1% for grade 3 (Fig. 3). The mean volume of the esophagus was 48.2 cm3 (26.3–102.7). Table 2 shows the relative esophageal volume receiving different dose levels amongst other dosimetric parameters. While dose–volume parameters based on the physical dose distribution showed highly significant differences between the treatment arms, in particular for high doses, parameters based on the rescaled dose distribution possessed smaller differences between the arms (Fig. 1). In univariable logistic regression none of the dose–volume parameters showed a significant association with early RIET grade 2 in patients treated with a shrinking-field conventional irradiation within this trial. In contrast, for patients receiving CHARTWEL, the physical dose–volumes parameters V40 and V50; and re-scaled values VEQD2,50 and VEQD2,60 were significant predictors of early RIET grade 2 (Table 3). Exemplarily, the logistic models using either V50 or VEQD2,50 are presented in Fig. 2 for both treatment arms separately and for the combined data. Among the clinical parameters, gender and prior chemotherapy revealed a statistical trend for association with early RIET
Early RIET is one of the major radiation-induced toxicities in patients with NSCLC and the primary cause of treatment interruptions as, unlike pneumonitis, it occurs early during radiotherapy. So far, several studies have addressed a wide range of dose–volume parameters tested for correlation with early RIET, but there has been no clear dose or dose–volume parameter defined as significant and practically relevant predictor. Furthermore, most of the reported data focused on conventional fractionation and on concurrent radiochemotherapy in lung cancer patients [11,20], while dose-constraints for minimizing the risk of esophagitis in hyperfractionated-accelerated irradiation schedules are missing so far. As the age of patients with NSCLC and corresponding comorbidities continues to increase, primary radiotherapy without simultaneous chemotherapy remains a commonly used treatment alternative in clinical practice. Based on the results of the CHART or CHARTWEL trial [1,12] as well as individual patient data metaanalysis [9] hyperfractionated-accelerated irradiation schedules may be superior options for patients not fit for simultaneous radiochemotherapy compared to conventional fractionation. Therefore, predictors of early esophageal toxicity in those treatment regimes are highly desirable. We derived our results from a post-hoc analysis of a prospective randomized trial and found, as expected from the higher dose intensity and our previous report on the complete multicenter study cohort [1], that patients receiving CHARTWEL developed acute RIET more often and significantly earlier than those receiving conventional radiotherapy (Fig. 3). Interestingly, higher rates of early RIET occurred in the CHARTWEL cohort although esophageal high-dose volumes were smaller compared to the conventional fractionation cohort (Fig. 1). The cumulative incidence of esophagi-
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
5
R. Bütof et al. / Radiotherapy and Oncology xxx (xxxx) xxx Table 3 Exemplary univariable logistic regression models for early RIET grade 2 for both treatment arms and the combined cohorts. Parameter
Conventional arm
CHARTWEL arm
All patients
OR (95% CI)
p-Value
OR (95% CI)
p-Value
OR (95% CI)
p-Value
V40 Constant
9.94 (0.09–1130) 0.14 (0.01–1.39)
0.34 0.093
91.7 (1.25–6734) 0.10 (0.02–0.63)
0.039 0.014
13.6 (0.72–256) 0.16 (0.04–0.62)
0.082 0.008
V50 Constant
9.78 (0.08–1186) 0.16 (0.02–1.24)
0.35 0.079
26.9 (1.03–705) 0.27 (0.10–0.73)
0.048 0.010
4.23 (0.43–41.3) 0.31 (0.14–0.73)
0.22 0.007
VEQD2,50 Constant
2.77 (0.04–181.8) 0.28 (0.05–1.45)
0.63 0.13
30.6 (1.23–764) 0.23 (0.08–0.69)
0.037 0.008
6.63 (0.61–72.3) 0.27 (0.11–0.65)
0.12 0.004
VEQD2,60 Constant
1.05 (0.03–33.6) 0.40 (0.15–1.05)
0.98 0.062
40.9 (1.11–1501) 0.30 (0.12–0.73)
0.043 0.008
4.63 (0.42–51.0) 0.36 (0.19–0.68)
0.21 0.002
Fig. 2. Logistic regression of early RIET grade 2 for both treatment arms (conventional: long dashed lines, red; CHARTWEL: short dashed lines, blue) and the combined cohorts (solid lines, black) depending on the relative volume of the oesophagus receiving 50 Gy physical dose [V50, left] and 50 Gy rescaled dose (including transformation to 2 Gy equivalent doses adjusted for repopulation and incomplete repair) [VEQD2,50; right]. Lines show the regression results while dots show the proportion of early RIET grade 2 in pooled patient groups. Bars represent the standard error of the mean.
Fig. 3. Cumulative incidence of early RIET grade 1 (top), grade 2 (center) and grade 3 (bottom) after the start of radiotherapy for both treatment arms (conventional: black dashed line, CHARTWEL: solid gray line).
tis grade 2 of 38 % reported here after CHARTWEL is in line with the incidence of 33% observed in the recently published DART trial on non-small cell lung cancer, which used two daily fractions of 1.8 Gy, and dose constraints of a mean esophageal dose of 34 Gy with a maximum of 80 Gy in the center of the esophagus [23]. Although the incidence of esophagitis grade 3 was higher in the CHART trial reported by Saunders et al. [12] who used 2D treatment planning, the kinetics of RIET over time in that study compares well to the observations after CHARTWEL [1]. Notably, the
median period of time for regression of grade 1 esophagitis was significantly shorter for patients treated with CHART or CHARTWEL compared to conventional fractionation [1,12]. This suggests a greater stimulation of repopulation of mucosal cells by hyperfractionated-accelerated radiation which, although probably negletable during the short treatment courses, promotes healing of mucositis by repopulation within a few weeks after end of treatment [16]. In addition to the mucositis scores by radiation oncologists [1,12], also patient reported data from the CHARTWEL trial indicate that dysphagia at 8 weeks after start of treatment or later was not increased and the quality of life not impaired after accelerated-hyperfractionated fractionation relative to conventional fractionation [6]. In our study, univariate analysis did not reveal any dose–volume parameter that was significantly associated with early RIET in those patients treated with shrinking-field conventional fractionation. In contrast, for patients receiving CHARTWEL, several dose–volume parameters such as V40 and V50, were significant predictors of early RIET. As a clinical consequence, all efforts should therefore be made to restrict esophageal volumes receiving 40 Gy or higher when hyperfractionated-accelerated treatment schedules are applied. Similar dose–volume parameters have also been reported to be associated with esophagitis in patients receiving hypofractionated radiochemotherapy [7,17,18] or conventional simultaneous radiochemotherapy [10] and are in line with QUANTEC based recommendations for esophageal dose–volume limits [21]. This suggests that dose–volume parameters for prediction of esophagitis may be more readily detectable in more intense treatment regimes. In contrast, in less dose-intensive schedules other factors, e.g. repopulation commencing already during the long treatment course, might mask these possible correlations with dose–volume parameters. In this context it is interesting to
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
6
Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy
note, that also the volume parameters predicting local tumor control were more dominant after hyperfractionated-accelerated fractionation compared to standard fractionation in the CHARTWEL trial [15]. To account for different repopulation rates, effects of dose per fraction and incomplete repair between the treatment arms we rescaled the dose–volume parameters to biological doses assuming equal clinical efficacy of both treatment arms. As shown by our results, physical doses alone correlated with early RIET in the CHARTWEL cohort but not after conventional fractionation. The used scaling factor of 1.15 increased the dose values in the CHARTWEL arm in comparison to the conventional arm, e.g. compare V60 with VEQD2,60 in table 2. Even after rescaling, however, dose parameters were somewhat larger for the conventional arm in contrast to the higher toxicity in the CHARTWEL arm, suggesting that the scaling factor might also be higher. To test the presented NTCP models, we thus performed additional logistic regression analyses of RIET grade 2 on the combined patient cohort including one physical dose–volume parameter and its interaction with the treatment arm, resembling the scaling factor. Also in these analyses, V40 and V50 were still significantly related to RIET grade 2. In general, one would expect similar dose–volume parameters for CHARTWEL and conventional fractionation after a perfect biological rescaling of doses. As already mentioned this is not the case in our analysis (Fig. 1), which at least in parts is caused by the fact that perfect rescaling is counteracted by the different time points and dose levels of downsizing the PTVinitial to the PTVboost. This finding supports the notion that dose–volume parameters for radiation esophagitis determined in a specific and time dependent setting of field arrangements cannot be easily transferred to another setting. Although target volume concept and planning technique used in the evaluated trial is not longer comparable with clinical standard today, the CHARTWEL fractionation schedule itself is still clinically applied. Therefore, our dose–volume parameter findings for early RIET are still applicable for this kind of hyperfractionated-accelerated irradiation. Among the clinical parameters, gender and prior chemotherapy revealed only a statistical trend for association with early RIET when combining both treatment arms, but not separately in one group. Former studies showed a significant impact of chemotherapy application and number of cycles for esophageal toxicity, especially in combined sequential or simultaneous radichemotherapy schedules [18,22]. In line with these results, recently published data suggest neutropenia as a possible risk factor and biomarker for higher grades of RIET in patients treated with concurrent radiochemotherapy [3]. Further perspectives for more individualized prediction of radiation toxicity include combined models of dose–volume parameters and potential biomarkers, e.g. cytokine blood levels. So far, conflicting results of predicitive value of IL-8 or other cytokine blood levels for radiation esophagitis have been reported [19,5]. Therefore, the final value of such biomarkers is currently unclear; they could possibly contribute to improvement of individualized NTCP models in the future. For clinical implementation of individualized dose-constraints, further prospective trials, especially on combination with possible biomarkers, are needed. In conclusion, for prediction of early RIET more dominant dose– volume effects were found after hyperfractionated-accelerated compared to conventional fractionation, especially for higher esophageal doses. This suggests that dose–volume parameters for correlation with esophagitis may be more readily detectable in more intense treatment regimes. In contrast, in less dose-intensive schedules other factors, e.g. repopulation commencing during long treatment times, may mask these possible correlations with dose– volume parameters. Our findings support the notion that dose–volume parameters for radiation esophagitis determined in a specific and time dependent setting of field arrangements can not be easily
transferred to another setting. In clinical practice esophageal volumes receiving 40 Gy or more should be strictly limited in hyperfractionated-accelerated fractionation schemes. Conflict of interest None. References [1] Baumann M, Herrmann T, Koch R, et al. Final results of the randomized phase III CHARTWEL-trial (ARO 97-1) comparing hyperfractionated-accelerated versus conventionally fractionated radiotherapy in non-small cell lung cancer (NSCLC). Radiother Oncol 2011;100:76–85. [2] Bradley J, Graham MV, Winter K, et al. Toxicity and outcome results of RTOG 9311: a phase I-II dose-escalation study using three-dimensional conformal radiotherapy in patients with inoperable non-small-cell lung carcinoma. Int J Radiat Oncol Biol Phys 2005;61:318–28. [3] Everitt S, Duffy M, Bressel M, et al. Association of oesophageal radiation dose volume metrics, neutropenia and acute radiation oesophagitis in patients receiving chemoradiotherapy for non-small cell lung cancer. Radiat Oncol 2016;11:20. [4] Goeckenjan G, Sitter H, Thomas M, et al. Prevention, diagnosis, therapy, and follow-up of lung cancer: interdisciplinary guideline of the German Respiratory Society and the German Cancer Society. Pneumologie 2011;65:39–59. [5] Hawkins PG, Boonstra PS, Hobson ST, et al. Prediction of radiation esophagitis in non-small cell lung cancer using clinical factors, dosimetric parameters, and pretreatment cytokine levels. Transl Oncol 2018;11:102–8. [6] Hechtner M, Krause M, Konig J, et al. Long-term quality of life in inoperable non-small cell lung cancer patients treated with conventionally fractionated compared to hyperfractionated accelerated radiotherapy - Results of the randomized CHARTWEL trial. Radiother Oncol 2018;126:283–90. [7] Kwint M, Uyterlinde W, Nijkamp J, et al. Acute esophagus toxicity in lung cancer patients after intensity modulated radiation therapy and concurrent chemotherapy. Int J Radiat Oncol Biol Phys 2012;84:e223–8. [8] Le Chevalier T, Arriagada R, Quoix E, et al. Radiotherapy alone versus combined chemotherapy and radiotherapy in nonresectable non-small-cell lung cancer: first analysis of a randomized trial in 353 patients. J Natl Cancer Inst 1991;83:417–23. [9] Mauguen A, Le Pechoux C, Saunders MI, et al. Hyperfractionated or accelerated radiotherapy in lung cancer: an individual patient data meta-analysis. J Clin Oncol 2012;30:2788–97. [10] Rodriguez N, Algara M, Foro P, et al. Predictors of acute esophagitis in lung cancer patients treated with concurrent three-dimensional conformal radiotherapy and chemotherapy. Int J Radiat Oncol Biol Phys 2009;73:810–7. [11] Rose J, Rodrigues G, Yaremko B, Lock M, D’Souza D. Systematic review of dosevolume parameters in the prediction of esophagitis in thoracic radiotherapy. Radiother Oncol 2009;91:282–7. [12] Saunders M, Dische S, Barrett A, Harvey A, Gibson D, Parmar M. Continuous hyperfractionated accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small-cell lung cancer: a randomised multicentre trial. CHART Steering Committee. Lancet 1997;350:161–5. [13] Sause W, Kolesar P, Taylor SI, et al. Final results of phase III trial in regionally advanced unresectable non-small cell lung cancer: Radiation Therapy Oncology Group, Eastern Cooperative Oncology Group, and Southwest Oncology Group. Chest 2000;117:358–64. [14] Socinski MA, Morris DE, Halle JS, et al. Induction and concurrent chemotherapy with high-dose thoracic conformal radiation therapy in unresectable stage IIIA and IIIB non-small-cell lung cancer: a dose-escalation phase I trial. J Clin Oncol 2004;22:4341–50. [15] Soliman M, Yaromina A, Appold S, et al. GTV differentially impacts locoregional control of non-small cell lung cancer (NSCLC) after different fractionation schedules: subgroup analysis of the prospective randomized CHARTWEL trial. Radiother Oncol 2013. [16] Stewart FA, Dorr W. Milestones in normal tissue radiation biology over the past 50 years: from clonogenic cell survival to cytokine networks and back to stem cell recovery. Int J Radiat Biol 2009;85:574–86. [17] Uyterlinde W, Belderbos J, Baas C, et al. Prediction of acute toxicity grade >/=3 in patients with locally advanced non-small-cell lung cancer receiving intensity modulated radiotherapy and concurrent low-dose Cisplatin. Clin Lung Cancer 2013;14:541–8. [18] Uyterlinde W, Chen C, Kwint M, et al. Prognostic parameters for acute esophagus toxicity in intensity modulated radiotherapy and concurrent chemotherapy for locally advanced non-small cell lung cancer. Radiother Oncol 2013;107:392–7. [19] Wang S, Campbell J, Stenmark MH, et al. A model combining age, equivalent uniform dose and IL-8 may predict radiation esophagitis in patients with nonsmall cell lung cancer. Radiother Oncol 2018;126:506–10. [20] Werner-Wasik M, Paulus R, Curran Jr WJ, Byhardt R. Acute esophagitis and late lung toxicity in concurrent chemoradiotherapy trials in patients with locally advanced non-small-cell lung cancer: analysis of the radiation therapy oncology group (RTOG) database. Clin Lung Cancer 2011;12:245–51.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002
R. Bütof et al. / Radiotherapy and Oncology xxx (xxxx) xxx [21] Werner-Wasik M, Yorke E, Deasy J, Nam J, Marks LB. Radiation dose-volume effects in the esophagus. Int J Radiat Oncol Biol Phys 2010;76:S86–93. [22] Wijsman R, Dankers F, Troost EG, et al. Multivariable normal-tissue complication modeling of acute esophageal toxicity in advanced stage nonsmall cell lung cancer patients treated with intensity-modulated (chemo-) radiotherapy. Radiother Oncol 2015;117:49–54.
7
[23] Wurstbauer K, Zehentmayr F, Deutschmann H, et al. DART-bid for locoregionally advanced NSCLC: summary of acute and late toxicity with longterm follow-up; experiences with pulmonary dose constraints. Strahlenther Onkol 2017;193:315–23.
Please cite this article as: R. Bütof, S. Löck, M. Soliman et al., Dose–volume predictors of early esophageal toxicity in non-small cell lung cancer patients treated with accelerated-hyperfractionated radiotherapy, Radiotherapy and Oncology, https://doi.org/10.1016/j.radonc.2019.11.002