Comparison and Prognostic Analysis of Elective Nodal Irradiation Using Definitive Radiotherapy versus Chemoradiotherapy for Treatment of Esophageal Cancer

Abstract

Objective: To investigate the prognostic factors for esophageal cancer (EC) patients treated by elective nodal irradiation (ENI) using chemo radiotherapy (CRT) and radiotherapy (RT)-alone. Methods: Data from 340 patients with EC were randomized to receive RTalone or CRT between January 2008 and December 2012. All patients received ENI either with late course RT or simultaneous integrated boost (SIB) - Intensity Modulated Radiotherapy (IMRT). The impact of clinic pathological factors and treatment modality on the overall survival (OS), and progression-free survival (PFS) were analyzed using Logrank test, Cox proportional regression model, and propensity score matching (PSM). P<0.05 was considered statistically significant. Results: A total of 340 patients were included, 174 patients (51.2%) underwent RT- alone and 166 patients (48.8%) received CRT. After the PSM, the median OS and median PFS times were 37.3 and 13.0 months for the RT–group, while those of the CRTgroup were 39.0 and 16.2 months, respectively. The 5-year OS rates was 32.9% for the RT-group, while those of the CRT-group was 31.3%, respectively (χ2=0.002, p=0.961). The 5-year PFS rate was 7.8% for the RT - group whereas, those of the CRT- group was 22.9%, respectively (χ2=3.911, p=0.048). Subgroup analysis showed that, late-course RT was significantly associated with improve PFS in CRT – group for patients within ≤ 60 years, female gender with cT3-4, N0- status, cTNM- stage III-IV, T- length> 5 cm, SCC subtype, GTV volume >30 cm3, (p<0.05 for all analysis). Conclusion: Compared with RT- alone, ENI using CRT and late- course RT provides a PFS benefit to EC patients, especially in those within ≤ 60 years old, cT3-4, N0- status, cTNM- stage III-IV, SCC subtype, T- length >5 cm, and GTV- volume >30 cm3 but it did not improve OS. Therefore, this finding could be of a particularly important pathway to the stratification parameters for a personalized treatment.

Keywords

Esophageal cancer, Chemoradiotherapy, Radiotherapy, Elective node irradiation, Late- course RT, SIB- IMRT.

Introduction

Esophageal cancer (EC) is a major cause of cancer morbidity and mortality throughout the world. In China, there were an estimated 477,900 new cases and, despite notable technical advances in diagnosis and treatment, 375,000 deaths occurred due to EC in 2015 [1]. Although the overall prognosis of patients has been improved among digestive cancers, the 5-year overall survival (OS) remains dismal about 10-20% [2,3]. Evidence indicates that surgery significantly improves survival in patients with early-stage EC. While in advanced forms, its role becomes limited to palliative indications for patients comfort rather than the survival improvement [4]. As in surgery, chemoradiotherapy (CRT) has been well established as a standard approach to treat EC [3,4]. However, the two treatments methods are often limited because of patients’ poor general conditions, advanced age or comorbidities. In this clinical situation, definitive radiotherapy (RT) - alone remains the only curative treatment option. In contrast, the 5-year OS rates after RT-alone has been usually associated with disappointing (OS range: 0% to 20%) [5]. However, the main concern today is three-fold. First, the definition of the target volume for the primary lesion. Second, the target volume delineation for the lymph nodes. Three, the optimal radiation therapy dose prescription. Recently, some researchers favorable to an elective node irradiation (ENI) claim their position, in addition to the high risk of micro metastatic invasion of the regional nodes at diagnosis, on the high frequency of relapses in the non-irradiated lymph node zones [6-9]. Studies from the radiotherapy-oncology group (RTOG 85-01/94-05) suggested that a combination of CRT and ENI might reduce the locoregional relapse rate in conventional fractionation compared with RT-alone [5,6]. In contrast, substantial late toxicities with the concomitant combination of chemotherapy may alleviate its benefits of quality of life and on the patient survival [5,6-11]. Therefore, the decision for ENI using CRT should depend on the primary tumor volume, clinical TNM- stage and tumor localization. Nowadays, there is no clear consensus regarding whether ENI should be required as a standard in the curative treatment for EC. For instance, the target volume delineation at the author’s institution is defined according to the habits and experience of the physician. In this setting, we sought to retrospectively analyze the clinicopathological characteristics and identify prognostic factors for OS, and progression-free survival (PFS) in EC patients treated by ENI using CRT and RT- alone.

Materials and Methods

Patient’s characteristics

Data from patients who were treated at the Fourth Affiliated Hospital of Hebei Medical University from January 2008 to December 2012 were screened and 924 patients with EC were analyzed retrospectively. Of these, 340 patients who received ENI were included according to the following selection criteria: histological diagnosis of EC; Performance status of 0-2; Clinical stage T1-4, N0-1, M0 according to the TNM staging system proposed by the American Joint Committee on Cancer (7th edition); Patients with no history of malignancy or prior surgical treatment related to EC. All patients diagnosed with tracheoesophageal or esophagomediastinal fistula or with a documented distant metastasis (Lung, Liver and Bone) were excluded from the study

Treatment protocol

RT-planning and target volume definition

The images of enhanced computed tomography and endoscopic representation were used to define the range of GTV volume. The irradiated fields were extended from the supraclavicular fossa to the esophagogastric junction, including the total mediastinum and the drainage region. The GTV was contoured on the planning CT by referring to the CT-scanning images and barium esophagography. The clinical target volume (CTV) margin was routinely created by expanding the GTV by 3.0-5.0 cm in the craniocaudal direction and 0.5-1.0 cm in the other four directions for the primary lesion. When metastatic lymph node was distant from the primary oesophageal lesion, the lymph node was contoured alone and noted as GTV-nd. The criterion for lymph node involvement was at least one of the following: diameter of the short axis ≥ 10 mm, the diameter of the long axis ≥ 15 mm on CT images. The presence of extra-nodal tumor extension and peripheral enhanced lymph node helped us to determine metastatic status. The planning target volume (PTV/PTV-nd) was created by expanding the CTV and the GTV-nd by a minimum of 0.5-1.0 cm radial margin. The nodal region that received ENI was noted as CTV1. Elective treatment of nodal regions depended upon the primary lesion location: 1-Primary lesion localized to the cervical and upper thoracic esophagus, 2-Primary lesion localized to the middle thoracic esophagus, 3 Primary lesion located in the lower thoracic esophagus [12]. The PTV1 margin was created by expanding the CTV1 by a minimum of 0.5-1.0 cm radial margin. All patients were treated with 3DCRT/IMRT, and the total dose of RT was delivered using a linear accelerator with beams 6 MeV photon. Treatment plans were generated with a three-dimensional planning system (ADAC-Pinnacle 3, version 5.0). Patients were treated 5 days a week, for those who had SIB- IMRT, the prescribed dose was 58.05-65.1Gy/28-31 fractions of 1.95-2.15 Gy, with the requirement that 95% of the PTV receives the prescribed dose, and 95% PTV1 receive 48.6-57.6Gy/28-31 fractions with a single dose of 1.7-1.8Gy. For those who received late- course RT: 95% PTV1 received 46-54 Gy/23-28 fractions, with a single dose of 1.8-2Gy during the first course of RT. For the late course, a booster dose was further administered to the primary esophageal lesion and the metastatic lymph nodes up to a total dose of 58-66 Gy/29-33 fractions at a single dose of 1.8-2Gy.

Systemic therapy regimen

With respect to chemotherapy in the entire group, a total of 166 patients (48.8%) had chemotherapy, including concurrent CRT 88 patients (53.0%) and sequential chemoradiotherapy 78 patients (47.0%). For patients who received concurrent CRT, the first course was started from the first day of irradiation. For those receiving sequential CRT, the first course began 1 week after the RT was completed. Typically, the main protocol consisted of platinum-based chemotherapy with combined 5-fluorouracil and a taxane (docetaxel or paclitaxel). According to the protocol in our cancer institute, three weeks for a cycle of chemotherapy and each patient received at least 2 cycles. Before and after every cycle of chemotherapy, a complete blood count was obtained. If the white blood cell (WBC) down to 2000/mm³ to 2500/mm³ after the CT, patients were treated with the Recombinant Human Granulocyte Colony-stimulating Factor Injection. If there was a grade 4 hematological reaction or grade 3/4 gastrointestinal reactions or radiation-induced pneumonitis/esophagitis occurred, symptomatic treatment was delivered and the CRT dose was adjusted in the subsequent course of therapy.

Statistical analysis

Patients treated with RT-alone were compared with those treated with CRT using the χ² test, Fisher’s exact or Student test according to the type of data. The treatment response evaluation was performed 4 to 12 weeks after the end of RT or CRT. The events considered to define event-free survival were: locoregional relapse, distant metastases, and all deaths. The OS was defined as the time from diagnosis to death date or the latest news of patients lost to follow-up. The PFS was defined as the time interval between the starting date of first-line treatment and the date of disease progression, last follow-up or death of any cause. Propensity score matching (PSM) was performed for further comparison. The OS and PFS rates were calculated with the Kaplan Meier method, and both groups were compared using the log-rank test. All survival curves were generated by the Kaplan-Meier method and compared with the Log-rank test. The potential prognostic factors for OS and PFS at 1, 3 and 5 years were investigated in univariate analysis. Prognostic factors with p < 0.05 were further evaluated in a multivariate Cox regression analysis. All these analyses were possible with SPSS software version 22.0 P <0.05 was considered statistically significant.

Results

Description of the whole cohort

From January 2008 to December 2012, altogether 924 patients with pathologically- proven EC were investigated at the Fourth Affiliated Hospital of Hebei Medical University. Of these, 584 patients (63.2%) received involved - field radiotherapy. These patients were not included in this study. Case histories of the remaining 340 patients (36.8%) were available in the hospital archives and all of them were also fitted to our inclusion criteria Figure 1.

Figure 1: Diagram showing flow of participants through the study (IFRT: Involved-field radiotherapy; ENI: Elective nodal irradiation; RT: radiotherapy; CRT: radiochemotherapy).

Clinicopathological characteristics (Table 1).

Table 1: Characteristics of All Patients, According to Treatment Groups (n=340).

Survival outcomes

The mean age in the entire group was 64 ± 0.4 (range: 41-85 years), with a relatively higher mean age in RT- group (65.6 ± 0.6 vs. 63.1 ± 0.7 years, if CRT, χ²=18.808, p=0.000). Over 65.3% (n=222 patients) of patients were male, and the M/F sex ratio was high in CRT- group (2.5 vs. 1.4, if RT- alone, χ²=5.850, p=0.016). The percentages of patients with cTNM- stage III-IV was 50.6% (n=84 patients) in CRT-group vs. 39.7% (n=69 patients) in RT- group (χ²=4.414, p=0.043). Two hundred and three patients experienced late- course RT including, 123 patients (70.7%) in RT-group vs. 80 patients (48.2%) in CRT-group (χ²=17.872, p=0.000). The most common digestive toxicity was esophagitis, including 88 patients (50.6%) of grade 1-2 esophagitis in RTgroup vs. 110 patients (66.3%) of grade 1-2 esophagitis in CRT-group (χ²=8.950, p=0.011). The incidence of hematological toxicities was obviously higher in patients who had CRT rather than those who had RT-alone, especially for grade 3-4 anemia (18.1% vs. 3.4%; χ²=24.217, p=0.000).

During the study period, the median OS time for the whole group was 38.2 ± 3.4 months (range: 31.6-44.8 months). The median and 5-year OS rate for those patients treated with RT- alone was 35.2 ± 4.9 months (95% CI: 31.6-44.8 months) and 33.4%, compared to 43.0 ± 3.3 months (95% CI: 36.0-49.9 months) and 31.2% for those treated with CRT. There was no statistical difference between the two groups (χ²=0.337, p=0.562; Figure 2).

Figure 2: Kaplan Meier curves comparing overall survival for oesophageal cancer patients treated by ENI using RT- alone and CRT (n=340). The 5-year Kaplan-Meier showed that no significant difference was observed between RT- group and CRT- group, OS rates (33.4% and 31.2%, χ²=0.337, P=0.562).

By contrast, the median and 5-year PFS in patients who had CRT was relatively longer than those who had RT- alone, with the median PFS of 16.7 ± 1.4 months (95% CI, 13.9-19.5 months) and the 5- year PFS rate of 22.7% for those who had CRT vs. 13.0 ± 0.9 months (95% CI, 11.3–14.7 months) and 10.8%, if RT- alone. This finding was statistically significant (χ²=6.099, p=0.014; Figure 3).

Figure 3: Kaplan Meier curves comparing progression-free survival for oesophageal cancer patients treated by ENI using RT- alone and CRT (n=340). The 5-year Kaplan-Meier showed that a significant difference was observed between RT- group and CRT- group, PFS rates (10.8% and 22.7%, χ²=6.099, P=0.014).

Propensity-score matching

In order to minimize the error caused by the influencing factors and to make the results more reliable, the clinicopathological characteristics and the potential prognostic factors for both groups were reassessed for further comparison. The PSM models were then developed for each comparative level. In this context, PSM confirmed the absence of a statistically significant difference in OS between the two groups (p>0.05; Figure 4).

Figure 4: Propensity-matched Kaplan Meier curves comparing overall survival for oesophageal cancer patients treated by ENI using RT- alone and CRT (n=240). The 5-year Kaplan-Meier after matching score showed that no significant difference was observed between RT- group and CRT- group, OS rates (32.9% and 31.3%, χ²=0.002, P=0.961).

Additional analysis was also performed for PFS. Overall, the CRT- group was remained associated with improving PFS, with a 5-year PFS rate of 7.8% in RT- group and 22.9% in CRT- group, respectively (χ²=3.911, p=0.048, Figure 5).

Figure 5: Propensity-matched Kaplan Meier curves comparing progression-free survival for oesophageal cancer patients treated by ENI using RT- alone and CRT (n=240). The 5-year Kaplan-Meier after matching score showed a significant difference was observed between RT- group and CRT- group, PFS rates ( 7.8% and 22.8%, χ²=3.961, P=0.048).

Prognostic factors for the whole cohort after PSM

The group analysis conducted in order to reexamine the effect of the treatment on potential prognostic factors. After propensity score matching, 240 well-balanced pairs of patients were available for outcome comparison. In univariate analysis, prognostic factors identified were (χ²=4.445, p=0.035), cT- stage (χ²=11.477, p=0.001), cTNM - stage (χ²=19.180, p=0.000), Tlength (χ²=6.170, p=0.013) Table 2.

Table 2: Characteristics of Patients After Matching, According to Treatment Group (n= 240).

Of this statistical significance persisted only for cTNM - stage (p=0.000) upon multivariate analysis. However, although there was no statistically significant difference in OS for cT- stage, N- status and the GTV- volume, there was a trend toward a higher ratio of 5- year OS in favor of cT1-2 (39.6% vs. 14.5%, if cT3-4, χ²=11.477, p=0.001), N0- status (35.3% vs. 29.9%, if N1+ 2 - status, χ²=3.383, p=0.066) and the GTV- volume ≤ 30 cm³ (39.8% vs. 25.4%, if GTV- volume >30 cm³; χ²=3.667, p=0.055, Table 3).

Table 3: Univariate analysis of the effect of potential prognostic factors on OS oesophageal cancer patients treated by ENI using RT- alone and CRT after PSM.

Additional PSM was also performed for PFS in the whole group and the outcomes revealed that male – gender was an unfavorable prognostic factor for PFS (p=0.034, Tables 4 and 5).

Table 4: Univariate analysis of the effect of potential prognostic factors on PFS in patients with esophageal cancer treated by ENI using RT- alone and CRT after PSM (n=240).

Table 5: Multivariate analysis of the effect of potentials prognostic factors on OS and LRRFS in patients with esophageal cancer treated by ENI using RT- alone and CRT after PSM (n=240).

Prognostic factors for the subgroup after PSM

In order to ascertain the specific factors affecting PFS and OS, we conducted a further study of the clinical stage and radiation dose. For the subgroup, various prognostic factors were evaluated by log- rang test and the data indicated that age ≤ 60 years old (χ²=7.074, p=0.008), cT3-4 (χ²=3.901, p=0.048), N0 status (χ²=7.449, p=0.006), cTNM stage III-IV (χ²=3.833, p=0.050), SCC (χ²=4.207, p=0.040), T- length >5 cm (χ²=3.003, p=0.083), GTV- volume >30 cm³ (χ²=7.910, p=0.007), and late- course RT (χ²=6.863, p=0.009) were marginally or significatively associated with a better PFS prognosis in CRT- group (Table 6).

Table 6: Subgroup Comparison of Patients with Different Characteristics (n=240).

Furthermore, in addition to age ≤ 60 years old (χ²=3.423, p=0.064), patients with N0-status (χ²=3.025, p=0.082), cTNM stage III-IV (χ²=2.776, p=0.096), and T- length ≤ 5 cm (χ²=3.579, p=0.059) were also more likely to demonstrate an OS benefit from CRT without statistical significance Table 7. As well, it confirmed that non- squamous cell carcinoma was associated with improve OS after RT-alone (χ²=5.687, p=0.017, Table 7).

Table 7: Subgroup Comparison of Patients with Different Characteristics (n=240).

Discussion

In order to ascertain the specific factors affecting PFS and OS, we conducted a further study of the clinical characteristics and radiation dose. Our subgroup analysis also showed that, compared to RT- alone, CRT improved PFS in the following subgroups: age ≤ 60 years (χ²=7.074, p=0.008), cT3-4 (χ²=3.901, p=0.048), N0 -status (χ²=7.449, p=0.006), cTNM stage III-IV (χ²=3.833, p=0.050), SCC (χ²=4.207, p=0.040), T-length >5 cm (χ²=3.003, p=0.083), GTV-volume >30 cm³ (χ²=7.910, p=0.007), and latecourse RT (χ²=6.863, p=0.009) Table 7. Moreover, in addition to age ≤ 60 years (χ²=3.423, p=0.064), patients with N 0-status (χ²=3.025, p= 0.082), cTNM stage III-IV (χ²=2.776, p=0.096), and T-length ≤ 5 cm (χ²=3.579, p=0.059) were more likely to demonstrate an OS benefit from CRT rather than RT- alone, although the difference was not statistically significant Table 6. Indeed, we observe survival superiority of RT- alone in non- squamous esophageal carcinoma. One possible reason for this result might be that the number of non-squamous esophageal carcinoma included in the study was small (33 patients) compare to the SCC subtype (307 patients).

The results of this population-based study reveal that the use of ENI using CRT is associated with significantly improved PFS for EC. After the propensity score matching, the median PSF times were 13.0 (95% CI:11.3-14.7) and 16.2 (95% CI:13.0- 19.4) months in RT and CRT-group, respectively, and the 1, 3 and 5-year survival rates were 57.1%, 20.5, and 7.8% in RT-group and 65.5%, 28.5%, and 22.9% in CRT- group, respectively (χ²=3.911, p=0.048). Patients within poor prognostic factors, including, age ≤ 60 years, with cT3-4, N1+2 - status, cTNM stage III-IV, SCC, T- length >5 cm, and GTV- volume >30 cm³, were more likely to benefit from CRT and have trended toward better survival than those receiving ENI using RT- alone. However, our results highlighted the lack of a statistically significant difference in OS between the two groups. The 1, 3 and 5 year OS rate in RT – group were 87.1%, 50.6% and 33.4%, respectively vs. 90.1%, 58.9% and 31.2%, if CRT- group (χ²=0.337, p=0.562, Figure 5). Pouliquen et al. and Walsh et al. [13,14], also reported the absence of survival benefit between the two treatment methods with ORR of 72% (95% CI, 0.53-0.97; p=0.05). Inversely, Herskovic et al. in a phase III study reported a superiority of CRT over RT- alone [15]. The results at one-year from the study describe an increase in locoregional control rate (38 versus 56%) as well as a significant increase in survival at 2 years (10 versus 38%). The number of metastases at 2 years decreased from 26 to 12%. By contrast, for some investigators, even though CRT may prevent elective nodal failure or improve local control, it is not clear whether the association of CRT and ENI improves OS [16]. These might mean that CRT using ENI contribute to increasing PFS benefit by controlling minor metastases, but does not contribute to increasing significantly OS.

Generally, it has been found that acute complications in patients undergoing CRT are more severe than those of patients receiving RT- alone. These complications include anemia, pneumonitis, and esophagitis along with other common symptoms like fatigue, nausea, vomiting, ulceration, epidermitis. Our results showed 31 patients (18.7%) of acute grade 3- 4 anemia in CRT- group vs. 10 patients (5.7%) in RT- group (χ²=24.217, p=0.000) and 110 patients (66.3%) of grade 1-2 esophagitis in CRT –group vs 88 patients (50.6%), if RT-alone ( χ²=8.950, p=0.011). There were no patients with an esophageal stricture. Sang Jun Byun, et al. reported 11.6% of esophagitis, 2.3% of ulceration and 27.9% of esophageal stricture after CRT [17].

Regarding the prognostic role of clinical factors on OS, and PFS, several clinical characteristics related variables were identified in a univariate analysis by log-rank test. For the whole group, although cT- stage, N- status, cTNM stage, T-length, and GTV-volume were of significant prognostic relevance for OS, the propensity-matched analysis, concluded that cTNM- stage (p=0.000) was the only independent prognostic factor for OS (Table 5). Several studies have reported that the cTNM stage is one of the most important prognostic factors in estimating survival rates including depth of tumor invasion, nodal involvement, GTV-volume and distant metastases. However, although difficulties and variation in the TNM staging undoubtedly exist, the present finding was translated into a survival advantage for those patients with cTNM stage I-II. The 1, 3 and 5 years OS rates were 92.3%, 65.7% and 43.7%, respectively in patients with cTNM stage I-II vs. 85.6% 41.2% and 17.3% for those with cTNM stage III-IV, respectively ( χ²=19.180, p=0.000, Table 3). Besides, in the study by Boggs DH et al. [18] published in 2015, the GTV- volume was a significant predictor for improved OS (p=0.001). They concluded that GTV-volume was a more powerful predictor of patient outcome than the traditional TNM staging. This finding consistent with our result; however, the target volume in Boggs study was contoured as separate regions rather than together [18] Our finding suggests that the conventional T- classification only is not invariably the best indicator of the real nature of the disease; in addition to age, a classification based on the actual cTNM-stage and the primary GTV- volume could have a better prognostic value.

Furthermore, as many variables could potentially be responsible for the difference observed in PFS between the two groups. In accordance with previous findings, factor independently found in the present study to be predictive of improved PFS after PSM was female gender. The PFS rate in female patients was significantly better than in male patients (χ²=4.594, p=0.032, Table 4). This result was also translated into a survival advantage for female patients. The OS rate at 3 and 5 years for the female patients were 65.6% and 41.2% vs. 49.3% and 27.1%, if male patients ( χ²=4.445, p=0.035) (Table 3). For Halperin EC et al. esophageal carcinoma in men tend to have a more aggressive nature with poorer outcomes [19].

To our knowledge, potential prognostic factors including cTNM stage, cT- stage, primary tumor location, histologic type, gender, patient age, tumor oxygenation and hemoglobin level before treatment have been already reported to possess prognostic value for EC [20,21]. In our subgroup analysis after PSM, in addition to female gender, age ≤ 60 years (χ²=7.074, p=0.008), cT3- 4 (χ²=3.901, p=0.048), N 0-status (χ²=7.449, p=0.006), cTNM- stage III-IV (χ²=3.833, p=0.050), T- length >5 cm (χ²=3.003, p=0.083), SCC (χ²=4.207, p=0.040), GTV volume >30 cm³(χ²=7.190, p=0.007), and late- course RT technique (χ²=6.863, p=0.009) were associated with a better PFS in CRT- group. Compared to others previously published studies, some authors have admitted that advanced age at diagnosis, the depth of tumor invasion, tumor length >5 cm, GTV–volume and middle and lower esophagus localization were correlated with a poor PFS prognosis in CRT- group [18,22-24]. However, the most important message in this study is that compared to RT- alone, ENI using CRT and late-course RT improved PFS prognosis in the following subgroups female gender, age ≤ 60 years, cT3- 4, N0- status, cTNM- stage III-IV, T- length >5 cm, SCC, GTV volume >30 cm³. This could be a particularly important pathway to the stratification parameters for a personalized treatment.

References

1. W. Chen, et al. Cancer statistics in China, 2015, CA: A Cancer Journal for Clinicians. 2016;66:115-132.
2. Eslick GD. Epidemiology of esophageal cancer. Gastroenterol Clin North Am. 2009;38:17-25.
3. Pera M. Epidemiology of esophageal cancer, especially adenocarcinoma of the esophagus and esophagogastric junction. Recent results in Cancer Research: Esophageal carcinoma. State of the art. LANGE J & SIEWERT JR. Eds, Springer-Verlag, Berlin 2000;155:1-14.
4. Thorban S, et al. Immunocytochemical detection of disseminated tumor cells in the bone marrow of patients with esophageal carcinoma. J Natl Cancer Inst 1996;88:1222-1227.
5. Cooper JS, et al. Chemoradiotherapy of locally advanced esophageal cancer: long-term follow-up of a prospective randomized trial (RTOG 85-01). Radiation Therapy Oncology Group. JAMA. 1999;281:1623-1627.
6. Minsky BD, et al. INT 0123 (Radiation Therapy Oncology Group 94-05) phase III trial of combined-modality therapy for esophageal cancer: high-dose versus standard-dose radiation therapy. J Clin Oncol. 2002;20:1167-1174.
7. Higuchi K, et al. Current management of esophageal squamous-cell carcinoma in Japan and other countries. Gastrointest Cancer Res. 2009;3:153-161.
8. Meng X, et al. Cetuximab in combination with chemoradiotherapy in Chinese patients with non- resectable, locally advanced esophageal squamous cell carcinoma: a prospective, multicenter phase II trial. Radiother Oncol. 2013;109:275-280.
9. Ishikura S, et al. Long-term toxicity after definitive chemoradiotherapy for squamous cell carcinoma of the thoracic esophagus. J Clin Oncol 2003;21:2697-2702.
10. Zhao KL, et al. Three-dimensional conformal radiation therapy for esophageal squamous cell carcinoma: is elective nodal irradiation necessary? Int J Radiat Oncol Biol Phys. 2010;76:446-451.
11. Yamashita H, et al. Involved-field irradiation concurrently combined with nedaplatin/5-fluorouracil for inoperable esophageal cancer on basis of (18) FDG-PET scans: a phase II study. Radiotherapy Oncol. 2014;113:182-187.
12. Amin MB, et al. eds. AJCC Cancer Staging Manual. 8th ed. New York: Springer; 2017.
13. Pouliquen X, et al. 5-Fluorouracil and cisplatin therapy after palliative surgical resection of squamous cell carcinoma of the esophagus: a multicenter randomized trial. Ann Surg 1996;223:127-133.
14. Walsh TN, et al. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. The New England journal of medicine. 1996;335:462-467.
15. Herskovic A, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med. 1992;326:1593-1598.
16. Yamashita H, et al. Details of recurrence sites after elective nodal irradiation (ENI) using 3D-conformal radiotherapy (3D-CRT) combined with chemotherapy for thoracic esophageal squamous cell carcinoma a retrospective analysis. Radiother Oncol 2011;98:255-260.
17. Sang Jun Byun, et al. Concurrent Chemoradiotherapy in Locally Advanced Esophageal Cancer. Jkstro.2011;29.1.20.
18. Boggs DH, et al. Primary gross tumor volume is an important prognostic factor in locally advanced esophageal cancer patients treated with trimodality therapy Gastrointest Cancer. 2015;46:131-7.
19. Halperin EC, et al. Perez and Brady's principles and practice for radiation oncology. 5th ed. New York: Lippincott Williams & Wilkins. 2008;1131-1153.
20. Grigiene R, et al. Prognostic value of anemia for patients with cervical cancer treated with irradiation. Medicina (Kaunas) 2005;41:916-924.
21. Neuhof D, et al. Retrospective evaluation of combined modality treatment and prognostic factors in patients with esophageal cancer. Acta Oncol 2005;44:168-173.
22. Polee MB, et al. Prognostic factors for survival in patients with advanced esophageal cancer treated with cisplatin-based combination chemotherapy. Br J Cancer. 2003;89:2045-2050.
23. Lightdale CJ. Esophageal cancer. Am J Gastroenterol. 1999;94:20-29.
24. Japanese Committee for Registration of Esophageal Carcinoma. A proposal for a new TNM classification of esophageal carcinoma. Jpn J Clin Oncol. 1985;15:625-636.