doi: /j.ijrobp

Transcription

1 doi: /j.ijrobp Int. J. Radiation Oncology Biol. Phys., Vol. 82, No. 1, pp , 2012 Copyright Ó 2012 Elsevier Inc. Printed in the USA. All rights reserved /$ - see front matter CLINICAL INVESTIGATION Thoracic Cancer NUMBER AND LOCATION OF POSITIVE NODES, POSTOPERATIVE RADIOTHERAPY, AND SURVIVAL AFTER ESOPHAGECTOMY WITH THREE-FIELD LYMPH NODE DISSECTION FOR THORACIC ESOPHAGEAL SQUAMOUS CELL CARCINOMA JUNQIANG CHEN, M.D.,* JIANJI PAN, M.D.,* XIONGWEI ZHENG, M.D., y KUNSHOU ZHU, M.D., z JIANCHENG LI, M.D.,* MINGQIANG CHEN, M.D.,* JIEZHONG WANG, M.D.,* AND ZHONGXING LIAO, M.D. { Departments of *Radiation Oncology, y Pathology, and z Surgery, the Teaching Hospital of Fujian Medical University, Fujian Provincial Cancer Hospital, Fuzhou, China, and { Department of Radiation Oncology, the University of Texas M. D. Anderson Cancer Center, Houston, Texas Purpose: To analyze influences of the number and location of positive lymph nodes and postoperative radiotherapy on survival for patients with thoracic esophageal squamous cell carcinoma (TE-SCC) treated with radical esophagectomy with three-field lymphadenectomy. Methods and Materials: A total of 945 patients underwent radical esophagectomy plus three-field lymph node dissection for node-positive TE-SCC at Fujian Provincial Tumor Hospital between January 1993 and March Five hundred ninety patients received surgery only (S group), and 355 patients received surgery, followed 3 to 4 weeks later by postoperative radiotherapy (S+R group) to a median total dose of 50 Gy in 25 fractions. We assessed potential associations among patient-, tumor-, and treatment-related factors and overall survival. Results: Five-year overall survival rates were 32.8% for the entire group, 29.6% for the S group, and 38.0% for the S+R group (p = for S vs. S+R). Treatment with postoperative radiotherapy was particularly beneficial for patients with $3 positive nodes and for those with metastasis in the upper (supraclavicular and upper mediastinal) region or both the upper and lower (mediastinal and abdominal) regions (p < 0.05). Postoperative radiotherapy was also associated with lower recurrence rates in the supraclavicular and upper and middle mediastinal regions (p < 0.05). Sex, primary tumor length, number of positive nodes, pathological T category, and postoperative radiotherapy were all independent predictors of survival. Conclusions: Postoperative radiotherapy was associated with better survival for patients with node-positive TE- SCC, particularly those with three or more positive nodes and positive nodes in the supraclavicular and superior mediastinal regions. Ó 2012 Elsevier Inc. Esophageal squamous cell carcinoma, Three-field lymphadenectomy, Lymph node metastasis, Nodal basins, Adjuvant radiotherapy. INTRODUCTION Esophageal carcinoma is an aggressive tumor and a leading cause of cancer-related death worldwide. In Europe and the United States, most esophageal tumors are adenocarcinomas and most commonly arise in the distal end of the esophagus and at the gastroesophageal junction (1). In Asia, especially in China, most esophageal carcinomas (95%) are the squamous cell histology type, and most tumors arise in the thoracic esophagus (2). The presence of lymph node metastasis is an important factor influencing prognosis after surgery, and the number of involved nodes is a strong prognostic factor for survival as well (3 6). Surgery is a mainstay of treatment for esophageal carcinoma. However, many patients who undergo only esophagectomy develop local recurrence and hematogenous metastasis at distant locations (7, 8). Despite some controversy about whether postoperative radiotherapy improves survival in all cases (5, 9 11), postoperative radiotherapy does seem to improve survival rates in cases involving lymph node metastasis. However, little information is available for the relationship between numbers and locations of nodal metastases and potential benefits from postoperative radiotherapy. In this study, we analyzed the effects of Reprint requests to: Jianji Pan, M.D., Department of Radiation Oncology, the Teaching Hospital of Fujian Medical University, Fujian Provincial Cancer Hospital, 91 Maluding, Fuma Road, Fuzhou , China. Tel: (+86) ; Fax: (+86) ; panjianji@126.com Conflict of interest: none. Acknowledgment We thank Prof. Guoliang Jiang, M.D., Department of Radiation Oncology, Fudan University Cancer Hospital, 475 and Prof. Xianglin Yuan, Department of Cancer Medicine, Tongji Hospital, Tongji Medical School, Huazhong University of Technology, for their guidance and suggestions for study design, data analysis, and manuscript preparation. Received April 19, 2010, and in revised form Aug 10, Accepted for publication Aug 27, 2010.

2 476 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 METHODS AND MATERIALS Patients Subjects in this retrospective analysis consisted of 945 patients with pathologically confirmed node-positive TE-SCC treated at Fujian Provincial Hospital from January 1993 through March All patients underwent radical esophagectomy with three-field lymphadenectomy (inferior cervical, right thorax, and superior abdominal regions). No patient received neoadjuvant or adjuvant chemotherapy or neoadjuvant radiotherapy. No patient appeared to have distant hematogenous metastasis at diagnosis. Fig. 1. Definition of lymph node regions. various patient-, tumor-, and treatment-related factors on overall survival after radical esophagectomy with threefield lymph node dissection for node-positive thoracic esophageal squamous cell carcinoma (TE-SCC) at Fujian Provincial Tumor Hospital. We were particularly interested in determining whether the number and location of nodal metastases and the extent of radiation fields affected prognosis. Surgical procedure and definition of nodal regions All patients underwent extended esophagectomy and three-field nodal dissection as described by Chen et al. (12). The operation consisted of a right thoracotomy, laparotomy, bilateral cervical U-shaped incision, and en bloc removal of the esophagus and gastric cardia (partial gastrectomy). A gastric tube was advanced from the greater curvature of the stomach to the neck for an esophagogastric anastomosis. The three-field nodal dissection removed all the lymph nodes from the regions of the cervical and superior mediastinum, including those along the right and left laryngeal recurrent nerves throughout their mediastinal course, the nodes at the upper thoracic paraesophagus, both right and left tracheobronchial recesses, pretrachea, the deep nodes in the lower cervical area located posteriorly and laterally to the carotid sheath, and supraclavicular foci; nodes in the middle and lower mediastinum, including the Table 1. Patient and tumor characteristics Variable Total no. of cases (% of total) No. of S cases (% of total) No. of S+R cases (% of total) c 2 value p value Sex Male 709 (75.0) 431 (76.1) 278 (78.3) Female 236 (25.0) 159 (26.9) 77 (21.7) Age <60 years 576 (61.0) 335 (56.8) 241 (67.9) $60 years 369 (39.0) 255 (43.2) 114 (32.1) Tumor locations < Upper thorax 146 (15.4) 63 (10.7) 83 (23.4) Mid thorax 712 (75.3) 459 (77.8) 253 (71.3) Lower thorax 87 (9.2) 68 (11.5) 19 (5.4) Length of tumor on X-ray #5 cm 488 (51.6) 306 (51.9) 182 (51.3) >5 cm 457 (48.4) 284 (48.1) 173 (48.7) Tumor differentiation Low (G3) 133 (22.5) 77 (21.7) Moderate (G2) 372 (63.1) 225 (63.4) High (G1) 85 (14.4) 53 (14.9) pt category T 1/2 162 (17.1) 92 (15.6) 70 (19.7) T (72.2) 445 (75.4) 237 (66.8) T (10.7) 53 (9.0) 48 (13.5) No. of positive lymph nodes (51.5) 306 (51.9) 181 (51.0) (29.8) 173 (29.3) 109 (30.7) >5 176 (18.6) 111 (18.8) 65 (18.3) Location of positive nodes < Upper 315 (33.3) 156 (26.4) 159 (44.8) Lower 189 (20.0) 156 (26.4) 33 (9.3) Both 441 (46.7) 278 (47.1) 163 (45.9) Abbreviations: S = surgery only; S+R = surgery plus postoperative radiotherapy; G = grade. Node locations (upper, lower, or both) are shown in Fig. 1.

3 Postoperative RT and survival in esophageal cancer d J. CHEN et al. 477 periesophageal, parahiatal, subcarinal, and aortopulmonary window nodes; and nodes in the upper abdominal and retroperitoneal areas, including celiac, splenic, common hepatic, left gastric, lesser curvature, and parahiatal nodes (Fig. 1). For analytical purposes, nodes were grouped as upper or lower as follows. Nodes superior to the inferior surface of the tracheal bifurcation (i.e., cervical/ supraclavicular and upper mediastinal nodes) were the upper nodes ; all other nodes from below that line to the gastroduodenal junction, including the those of the middle and lower mediastinal regions and the upper abdominal region, were considered lower nodes. Radiotherapy Radiotherapy was started 3 to 4 weeks after surgery. Of the 355 patients who received postoperative radiotherapy, 50 patients who had had surgery before January 1996 received irradiation in large T-shaped fields that included bilateral supraclavicular upper foci, mediastinum, drainage region of the left gastric artery lymph nodes adjacent to the gastric cardia orifice, and the original esophageal tumor bed. Use of these large fields produced unacceptably high complication rates, so we later modified the radiation fields by omitting the left gastric artery drainage region, and 305 patients underwent smaller T-field irradiation. Radiation to 36 Gy to the isocenter was given in 18 fractions using a linear accelerator with 6- to 8-MV X- rays via parallel opposed anteroposterior T-shaped fields. Then, to spare the spinal cord, an additional 14 Gy of radiation was given to the bilateral supraclavicular and upper mediastinal regions with 12- MeVelectron beams and to the middle and lower mediastinum with 18-MV X-rays in parallel opposed lateral fields. The median radiation dose to the tumor bed was 50 Gy (range, Gy) in 25 fractions at 2 Gy per fraction, 5 days per week. 77 years old (median, 54 years). The two groups were unbalanced in terms of age, tumor category, and location of nodal metastases: patients in the S+R group were younger (<60 years) and had more tumors in the upper thorax, more stage pt4 tumors, and more upper-region nodal metastases at presentation. Survival Survival rates for the entire population were 81.6% at 1 year, 46.2% at 3 years, and 32.8% at 5 years. Median survival time was 31.5 months. Survival rates according to type of treatment (S vs. S+R) are shown in Fig. 2. Survival rates for the S group were 77.8% at 1 year, 41.8% at 3 years, and 29.6% at 5 years. Median survival time was 25.6 months. Corresponding survival rates for the S+R group were 87.8% at 1 year, 53.1% at 3 years, and 38% at 5 years. Median survival time was 38.7 months. Survival rates in the S+R group were significantly better than those of the S group (p = 0.001). A total of 25,344 lymph nodes was removed from the 945 patients (mean, 26.8 nodes per patient; range, nodes). Positive nodes were found in the upper region in 315 patients, in the lower region in 189 patients, and in both upper and lower regions in 441 patients. Corresponding 5-year survival rates were 40.7%, 46.8%, and 21.5% for patients with nodal metastases in the upper region, lower region, and both regions, respectively. Patients with metastases in both upper and lower regions had worse survival rates than those with metastases in the upper or lower regions (p < ), Follow-up A combination of clinic service records, phone calls, letters, and survival records from the Census Register Center of the Fujian Province Public Security Department was used to determine the vital status of each patient as of May 1, Fifty-two patients lost to follow-up were considered dead at the date of last known contact. The numbers of cases with follow-up information available at 1, 3, and 5 years were 704, 355, and 189 patients, respectively. Statistical analysis Patient- and disease-related factors were considered according to type of treatment received (surgery only [S] or surgery plus postoperative radiotherapy [S+R]) and analyzed with SPSS version 15.0 software (SPSS Inc, Chicago, IL). Survival times were calculated by month from the date of surgery to the date of death or final follow-up. The c 2 test was used to compare effects of clinical factors on survival between the two groups. The Kaplan-Meier method was used to compare survival rates between treatment groups. The log-rank test was used as the significance test. Cox regression was used to evaluate the hazard ratios (HR) as well as the 95% confidence intervals. Statistical significance was defined as a p value of < RESULTS Clinical characteristics of the study population Patient characteristics are shown in Table 1. The 590 patients in the S group were 32 to 82 years old (median, 57 years), and the 355 patients in the S+R group were 27 to Fig. 2. Overall survival rates for patients who had surgery only (S) and for those who had surgery followed by radiation (S+R) for thoracic esophageal squamous cell carcinoma.

4 478 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 indicating that extensive nodal metastases was an indicator of poor survival (Fig. 3). Of the 458 patients found to have three or more positive nodes, 284 patients were in the S group, and 174 patients were in the S+R group. Corresponding 5-year survival rates were 20.4% for all patients, 17.8% for the S group, and 25.2% for the S+R group, with corresponding median survival times of 21.7, 18.7, and 29.8 months, respectively. Postoperative radiation improved overall survival rates among patients with three or more positive nodes (p = 0.001) (Fig 4) but not for those with one or two positive nodes (p > 0.05, data not shown). To explore the influence of anatomic location of nodal involvement on survival, we compared the effects of postoperative radiation on patients presenting with positive nodes in the upper (supraclavicular and upper mediastinal) regions, the lower (middle and inferior mediastinal and upper abdominal regions) regions, and in both the upper and the lower regions. Receipt of postoperative radiotherapy did not affect 5- year survival rates for patients with only lower-region metastases (52.9% for the S+R group vs. 45.5% for the S group; p = 0.724), but postoperative radiotherapy had a positive effect on 5-year survival for those with upper-region metastases (45.5% for the S+R group vs. 34.9% for the S group; p = 0.021) and for those with metastases in both the upper and the lower regions (27.6% for the S+R group vs. 17.9% for Fig. 4. Overall survival rates for patients presenting with $3 positive lymph nodes treated with surgery (S) or surgery plus postoperative radiation (S+R). the S group; p = 0.002) (Table 2; Fig. 5). The radiation field size (large T vs. small T) did not affect survival (p > 0.05) (Table 2). Univariate and multivariate analyses of prognostic factors Results of univariate analyses are shown in Table 3. Being male and having a tumor 5 cm or longer on barium swallow X-ray, three or more positive nodes, positive nodes in both Table 2. Five-year survival rates for S and S+R patients Variable S S+R HR (95% CI) p value No. of positive lymph nodes ( ) ( ) > ( ) Location of positive nodes Upper region ( ) Lower region ( ) Both ( ) Radiation field size Large T field (ref) Small T field ( ) Fig. 3. Overall survival for patients presenting with positive nodes in the lower region (i.e., middle and lower mediastinal and upper abdominal beds), upper region (i.e., cervical and upper mediastinal beds), and in both upper and lower regions. Abbreviations: HR = hazard ratio; CI = confidence interval; ref = reference value. Five-year survival rates for patients given surgery (S) or surgery plus postoperative radiation (S+R) according to number and location of positive lymph nodes and radiation field size.

5 Postoperative RT and survival in esophageal cancer d J. CHEN et al. 479 the upper and lower regions, stage pt3 or pt4 tumor, and not receiving postoperative radiotherapy were associated with worse 5-year survival rates (p = to < ). Receipt of postoperative radiotherapy reduced the HR of dying at 5 years by 34% (HR = 0.763; 95% confidence limit, ; p = 0.001). Age, primary tumor location, and tumor differentiation were not predictive of prognosis according to univariate analysis. Findings from multivariate analyses are shown in Table 4. Sex, length of primary tumor, number of positive nodes, pt category, and receipt of postoperative radiotherapy were independent prognostic factors for survival, but location of nodal metastases was not. Causes of treatment failure Of the total 945 patients, 604 patients had died by the last follow-up, including 374 patients in the S group and 230 patients from the S+R group. Cause of death was unknown for 127 patients in the S group and for 81 patients in the S+R group (9 patients who had undergone large-t-field irradiation and 72 patients who had undergone small-t-field irradiation). The most common cause of treatment failure in both groups was hematogenous distant metastases (Table 5). Receipt of postoperative radiotherapy reduced the likelihood of cervical and mediastinal recurrence by more than 50% (p < ). For those cases in which nodal failure sites were known, recurrence in the mediastinal nodes was located mainly in the superior mediastinum, in the upper mediastinum at the paratracheal station above the aortic arch, and below the carina. Inferior mediastinal lymph node metastasis was never discovered in the entire group. Only 1 patient in the S group developed recurrence at the anastomotic stoma. Receipt of postoperative radiotherapy was associated with lower rates of supraclavicular and mediastinal recurrence (p < 0.05). Toxicity of postoperative radiotherapy Toxicities related to postoperative radiation, scored according to Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer (1992) criteria (13), are summarized in Table 6. Receiving radiation to the larger T field, as opposed to the smaller T field, was associated with higher risk of acute toxicity in the upper aerodigestive tract, stomach, and cardiopulmonary system (p < 0.05). Patients who received large-t-field irradiation were at higher risk of late complications at the grade-5 level, including cardiac, pulmonary, and grade 5 gastric bleeding (p <0.05). Fig. 5. Overall survival for patients presenting with positive nodes in the upper (cervical/upper mediastinal) region (A) or in the lower region (B) or in both the upper and the lower regions (C) treated with surgery (S) or surgery plus postoperative radiation (S+R). DISCUSSION Our results indicated that the addition of radiotherapy after radical esophagectomy with three-field lymphadenectomy for node-positive TE-SCC was associated with improved survival rates, particularly for patients with three or more positive nodes and involvement of only the supraclavicular and superior mediastinal regions. Postoperative

6 480 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 Table 3. Univariate analysis of prognostic factors and survival Variable 5-year survival rate (%) Median survival time (months) HR (95% CI) p value Age (years) < (ref) $ ( ) Sex Male (ref) Female ( ) Primary tumor location Upper thorax (ref) Mid thorax ( ) Lower thorax ( ) Length of tumor on barium-swallow X-ray #5 cm (ref) >5 cm ( ) Tumor differentiation Low (G3) (ref) Moderate (G2) ( ) High (G1) ( ) No. of lymph node metastases (ref) ( ) < > ( ) < Location of nodal metastases Upper region (ref) Lower region ( ) Both regions ( ) < pt category T 1 or (ref) T ( ) < T ( ) < Postoperative radiotherapy No (ref) Yes ( ) Abbreviations: HR = hazard ratio; CI = confidence interval; G = grade; ref = reference value. radiotherapy also reduced the risk of recurrence in the supraclavicular and middle and superior mediastinal regions. Local recurrence and distant metastasis are the main causes of failure after surgical treatment of esophageal carcinoma, with total recurrence rates after radical esophagectomy and extensive three-field lymph node dissection ranging from 27% to 43% and local recurrence rates from 41.5% to 49% (7, 14 17). Distant metastases rates among Table 4. Multivariate analysis of prognostic factors for survival Variable HR (95% CI) p value Sex (female vs. male) ( ) Length of primary tumor ( ) (> vs. #5 cm) No. of nodal metastases ( ) < (1-2 vs. $ 3) Location of nodal metastases ( ) (upper vs. both upper and lower) pt category (pt1-2 vs. pt$3) ( ) Postoperative radiotherapy (no vs. yes) ( ) Abbreviations: HR = hazard ratio; CI = confidence interval. patients who underwent such surgery for node-positive disease are still higher at 41.5% to 61.3% (7, 15, 16). The number of positive nodes correlates with increased total recurrence and local-regional recurrence (14, 16, 18) but not with distant metastasis (p > 0.05) (16, 18). Kimura et al. (18) reported total recurrence rates after surgery of 17.6% for patients with 0 positive nodes, 48.5% for those with 1 to 3 positive nodes, and 78.6% for those with $4 positive nodes. Bhansali et al. (16) reported similar total recurrence rates after surgery, namely, 23% for those with 0 positive nodes, 33% for 1 positive node, 48% for 2 to 4 nodes, and 73% for $5 nodes. Beta et al. (14) also reported similar rates of 29% for those with 0 positive nodes, 42% for 1 to 5 nodes, and 71% for $5 nodes. Xiao et al. (5, 9) found in a randomized study that postoperative radiotherapy improved survival rates among patients with esophageal cancer and nodal metastasis and reduced the rates of nodal recurrence in the thoracic cavity and clavicular region. We also found that postoperative radiotherapy after esophagectomy for node-positive TE- SCC was associated with better 5-year survival rates than surgery alone (38.0% vs. 29.6%, p = 0.001). Furthermore, we found that postoperative radiotherapy was of particular benefit for patients with 3 or more positive lymph nodes;

7 Postoperative RT and survival in esophageal cancer d J. CHEN et al. 481 Table 5. Causes of treatment failure No. of patients experiencing failure (%) Disease recurrence Surgery only (n = 590) Surgery + radiation (n = 355) c 2 value p value Large T field (n = 50) Small T field (n = 305) c 2 value p value Total 216 (36.6) 122 (34.7) (46.0) 99 (32.5) Cervical region 76 (35.2) 15 (12.3) < (8.7) 13 (13.1) Mediastinal region 97 (44.9) 18 (14.8) < (13.0) 15 (15.2) < Celiac region 34 (15.7) 19 (15.6) (13.0) 16 (16.2) Tumor bed 17 (7.9) 4 (3.3) (4.3) 3 (3.0) NA Hematogenous metastasis 134 (62.0) 84 (68.9) (65.2) 69 (69.7) Deaths from radiation-related causes 15 (4.2) 13 (26.0) 2 (0.7) < Abbreviation: NA = not applicable (from Fisher s exact test). the apparent improvement for those with 1 to 2 positive nodes (50.7% vs. 41.2% after surgery alone) was not statistically significant (p = 0.070). In terms of nodal recurrence patterns, Nakagawa et al. (7) reported a study of 171 patients who underwent radical esophagectomy and three-field dissection for TE-SCC; of the 30 patients who experienced locoregional recurrence, 28.6% of recurrences appeared in the cervical region, 62.9% in the mediastinum, 17.1% in the abdominal region, and 2.9% at the anastomotic stoma. By comparison, Chen et al. (8), reporting recurrence among 191 patients after radical esophagectomy for TE-SCC, found that cervical/supraclavicular and mediastinal nodes accounted for 76.8% of Table 6. Toxicity of large-t-field vs. small-t-field irradiation Variable No. patients experiencing toxicity (%) Large T field (n = 50) Small T field (n = 305) c 2 value p value Acute toxicities Hematologic 18 (36.0) 106 (34.8) Grade (34.0) 100 (32.8) Grade $3 1 (2.0) 6 (2.0) NA Esophagus 14 (28.0) 78 (25.6) Upper digestive 6 (12.0) 7 (2.3) tract Pulmonary 14 (28.0) 67 (22.0) Late toxicities Cardiac All grades 14 (28.0) 2 (0.7) < Grade (14.0) 1 (0.3) < Grade 5 7 (14.0) 1 (0.3) < Pulmonary All grades 7 (14.0) 4 (1.3) < Grade (8.0) 4 (1.3) Grade 5 3 (6.0) 0 (0.0) N/A Gastric bleeding All grades 3 (6.0) 2 (0.7) NA Grade (0.0) 1 (0.3) NA Grade 5 3 (6.0) 1 (0.3) NA Abbreviation: NA = not applicable (from Fisher s exact test). recurrences. In the current study, 35.2% of locoregional recurrences after surgery alone appeared in the inferior cervical supraclavicular nodes, 44.9% in middle and upper mediastinal nodes, 15.7% in the epigastric (celiac) nodes, and 7.9% in the tumor bed. Receipt of postoperative radiotherapy reduced those rates to 12.3%, 14.8%, 15.3%, and 3.3%, respectively. Our finding of statistically significant reductions in recurrence in the supraclavicular and middle and upper mediastinal regions (p < ) agrees with the findings of Xiao et al. (5, 9). Finally, with regard to survival according to disease location at presentation, we found that postoperative radiotherapy was associated with improved survival rates for patients with nodal disease in the upper region and in both the upper and lower regions but not for those with only lower-region node disease. A possible reason for this result is the technical difficulty of lymph node dissection in the lower neck and upper mediastinal regions owing to the complex anatomy in those regions, with the abundance of nerves and lymphatic vessels and adjoining large blood vessels and critical organs; incomplete dissection in these regions might have left behind subclinical foci of disease. The middle and lower mediastinum and upper abdominal areas, on the other hand, can be well exposed, and lymph node dissection is comparatively more thorough. This explanation is supported by our finding that no patient experienced failure in the abdomen regardless of whether or not they received postoperative radiation, and regardless of our minimizing coverage of the abdominal regions in the smaller-t-field radiation delivered to most of the patients in this study. CONCLUSIONS Currently, no definitive conclusions have been reached regarding the ideal field size for postoperative prophylactic irradiation. Some investigators (9, 10) have used large fields that encompass the bilateral supraclavicular region, the mediastinal region, and the left gastric artery nodal drainage area adjacent to the gastroesophageal junction. However, other investigators have argued that reducing the size of radiation fields has not reduced survival rates (19, 20). In our study, the 5-year survival rate for those given

8 482 I. J. Radiation Oncology d Biology d Physics Volume 82, Number 1, 2012 large-field irradiation was 39.6% versus 37.7% for those given smaller-field radiation (p = 0.823). In a study reported by Fok et al. (21), late effects of postoperative radiotherapy for esophageal carcinoma have included radiation fibrosis of the lung, noncancerous pericardial and pleural effusion, and alimentary tract hemorrhage. In our study, 13 of 50 patients in the large-field group (26%) died of radiotherapy late complications compared with only 2 of 305 patients (0.7%) given smaller-field radiation (p < ). Beginning in 1996, we modified our treatment fields to exclude the abdominal-node region, and since that time, we have seen significant reductions in acute and late radiotherapy complications without compromising local control. The fact that no patient in the current study developed recurrence in the abdomen suggests that irradiating that area was not necessary. Indeed, between 62% and 70% of patients in our study developed distant metastases, suggesting that esophageal carcinoma is a local presentation of a systemic disease and that the addition of systemic therapy for this disease should be investigated to further improve patient outcome. REFERENCES 1. Daly JM, Fry WA, Little AG, et al. Esophageal cancer: Results of an American College of Surgeons Patient Care Evaluation Study. J Am Coll Surg 2000;190: , discussion Mei G, Yi-dian Z, Hai-jun Y, et al. Analysis of clinicopathological characteristics for 5406 cases of esophageal neoplasm. Chin J Cancer Prev Treat 2008;15: Tachibana M, Kinugasa S, Yoshimura H, et al. Clinical outcomes of extended esophagectomy with three-field lymph node dissection for esophageal squamous cell carcinoma. Am J Surg 2005;189: Hofstetter W, Correa AM, Bekele N, et al. Proposed modification of nodal status in AJCC esophageal cancer staging system. Ann Thorac Surg 2007;84: Xiao ZF, Yang ZY, Miao YJ, et al. Influence of number of metastatic lymph nodes on survival of curative resected thoracic esophageal cancer patients and value of radiotherapy: report of 549 cases. Int J Radiat Oncol Biol Phys 2005;62: Shimada H, Okazumi S, Matsubara H, et al. Impact of the number and extent of positive lymph nodes in 200 patients with thoracic esophageal squamous cell carcinoma after three-field lymph node dissection. World J Surg 2006;30: Nakagawa S, Kanda T, Kosugi S, et al. Recurrence pattern of squamous cell carcinoma of the thoracic esophagus after extended radical esophagectomy with three-field lymphadenectomy. J Am Coll Surg 2004;198: Chen G, Wang Z, Liu XY, et al. Recurrence pattern of squamous cell carcinoma in the middle thoracic esophagus after modified Ivor-Lewis esophagectomy. World J Surg 2007;31: Xiao ZF, Yang ZY, Liang J, et al. Value of radiotherapy after radical surgery for esophageal carcinoma: a report of 495 patients. Ann Thorac Surg 2003;75: Ténière P, Hay J-M, Fingerhut A, et al. Postoperation radiation therapy dose not increase survival after curative resection for squamous as shown by a multicenter controlled trial. Surg Gynecol Obstet 1991;173: Lu J C QPD, ZhaWW, Zhang Y Q. The meta-analysis of randomized controlled trial of prophylactic radiotherapy for esophageal carcinoma after curative resection. Xunzheng Yixue 2005;5: Chen J, Liu S, Pan J, et al. The pattern and prevalence of lymphatic spread in thoracic oesophageal squamous cell carcinoma. Eur J Cardiothorac Surg 2009;36: Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31(5): Baba M, Aikou T, Yoshinaka H, et al. Long-term results of subtotal esophagectomy with three-field lymphadenectomy for carcinoma of the thoracic esophagus. Ann Surg 1994;219: Kyriazanos ID, Tachibana M, Shibakita M, et al. Pattern of recurrence after extended esophagectomy for squamous cell carcinoma of the esophagus. Hepatogastroenterology 2003;50: Bhansali MS, Fujita H, Kakegawa T, et al. Pattern of recurrence after extended radical esophagectomy with three-field lymph node dissection for squamous cell carcinoma in the thoracic esophagus. World J Surg 1997;21: Matsubara T, Ueda M, Takahashi T, et al. Localization of recurrent disease after extended lymph node dissection for carcinoma of the thoracic esophagus. J Am Coll Surg 1996;182: Kimura H, Konishi K, Arakawa H, et al. Number of lymph node metastases influences survival in patients with thoracic esophageal carcinoma: Therapeutic value of radiation treatment for recurrence. Dis Esophagus 1999;12: Lu JC, Tao H, Zhang YQ, et al. Extent of prophylactic postoperative radiotherapy after radical surgery of thoracic esophageal squamous cell carcinoma. Dis Esophagus 2008;21: Qiao XY, Wang W, Zhou ZG, et al. Comparison of efficacy of regional and extensive clinical target volumes in postoperative radiotherapy for esophageal squamous cell carcinoma. Int J Radiat Oncol Biol Phys 2008;70: Fok M, Sham JST, Choy D, et al. Postoperative radiotherapy for carcinoma of the esophagus: a prospective, randomized controlled study. Surgery 1993;113: