Surgery Combined with Radiotherapy Improved Survival in Metastatic Esophageal Cancer in a Surveillance Epidemiology and End Results Population-based Study

Abstract

This retrospective study used a population-based national registry to determine the impact of local treatment modalities on survival in patients with metastatic esophageal cancer (EC). The Surveillance Epidemiology and End Results (SEER) database was used to identify patients with metastatic EC from 1988 to 2012. A total of 9,125 patients were identified. There were 426 patients underwent primary surgery, 4,786 patients were administered radiotherapy (RT) alone, 847 patients underwent surgery plus RT, and 3,066 patients without any local treatment. Multivariate analysis results indicated that year of diagnosis, age, race, histologic subtype, grade, and local treatment modalities were independent prognostic factors for overall survival (OS). The 5-year OS were 8.4%, 4.5%, 17.5%, and 3.4% in primary surgery, RT only, surgery plus RT, and no local treatment, respectively (P < 0.001). Subgroup analyses showed that the impact of RT was mainly reflected by preoperative radiotherapy, as patients received preoperative radiotherapy had significantly better OS than patients who underwent primary surgery alone and postoperative RT, the 5-year OS rates were 24.7%, 6.5%, and 7.8%, respectively, respectively (P < 0.001). Surgery plus RT, especially preoperative RT, may improve long-term survival of patients with metastatic EC.

Surgery Combined with

Radiotherapy Improved Survival in Metastatic Esophageal Cancer in a Surveillance Epidemiology and End Results Population-based Study

OPEN

R A

San-GangWu,*, Wei-Hao Xie,*, Zhao-Qiang Zhang,*, Jia-Yuan Sun, Feng-Yan Li, Huan-Xin Lin, Yong Bao & Zhen-Yu He

This retrospective study used a population-based national registry to determine the impact of local treatment modalities on survival in patients with metastatic esophageal cancer (EC). The Surveillance Epidemiology and End Results (SEER) database was used to identify patients with metastatic EC from that year of diagnosis, age, race, histologic subtype, grade, and local treatment modalities were P<

respectively, respectively (P<

term survival of patients with metastatic EC.

Esophageal cancer (EC) is a highly lethal malignancy, and the incidence is increasing1. In 2015, it has been estimated that there would be 16,980 new EC cases and 15,590 deaths in the United States2. Approximately 50% of patients had metastases to distant lymph nodes or organs at the initial diagnosis3,4. The prognosis of metastatic EC is poor, and the 5-year survival rate is lower than 5%5. The palliative treatment in metastatic EC depends mainly on patients clinical situation with the goal of reducing cancer-related symptoms and extending survival without compromising quality of life. Systemic treatment consists of chemotherapy, targeted therapy, and best supportive care. Local treatment mainly includes feeding tubes, beam radiation, brachytherapy, and endoscopic management techniques such as dilation and stenting6,7.

Preoperative chemoradiotherapy may significantly increase the radical resection rate and improve survival for advanced esophageal carcinoma8, but the value of surgery plus radiotherapy (RT) for metastatic EC has not yet been claried. RT is not a rst-line treatment for metastatic EC, but RT may improve the patients symptoms of obstruction9. Studies with small sample sizes have shown that local treatments including surgery could prolong survival in metastatic EC1013. Studies have shown that surgery and/or radiotherapy can improve survival in patients with stage IV malignant tumors14,15. In this study, we analyzed the metastatic EC using a

Republic of China. Republic of China. *These authors contributed equally to this work. Correspondence and requests for materials should be addressed to Y.B. (email:

SCIENTIFIC REPORTS

1

www.nature.com/scientificreports/

Figure 1. Patients included in analysis.

population-based national registry (Surveillance Epidemiology and End Results, SEER) to determine the impact of local treatment strategies on survival in metastatic EC.

Results

Patient characteristics and treatment. The SEER database included a total of 63,759 patients with EC in 19882012, and 31.6% (20,168 patients) had a distant stage; 9,125 patients met the inclusion criteria of this study (Fig.1). The patient characteristics are shown in Table1. The median age of initial diagnosis was 64 years (range, 2196); 83.5% (7,621/9,125) were white; 82.0% (7,486/9,125) were male; 59.2% (5,406/9,125) had adenocarcinoma; and 76.7% (6,995/9,125) had a lower thoracic esophageal cancer. Local treatment modalities were as follows: 426 (4.7%) patients underwent primary cancer-directed surgery (CDS); 4,786 (52.4%) were primary RT alone; 847 (9.3%) underwent CDS plus RT; and 3,066 (33.6%) were not administered any local treatment. Among patients who underwent CDS plus RT, 57.3% (485/847) were administered preoperative radiotherapy,

SCIENTIFIC REPORTS

2

www.nature.com/scientificreports/

Characteristic n CDS (%) RT (%) CDS+RT (%) None (%) P-value

Year of diagnosis

19881992 514 67 (15.7) 288 (6.0) 55 (6.5) 104 (3.4) <0.001 19931997 668 68 (16.0) 406 (8.5) 55 (6.5) 139 (4.5)19982002 1755 111 (26.1) 898 (18.8) 204 (24.1) 542 (17.7) 20032007 2854 102 (23.9) 1489 (31.1) 258 (30.5) 1005 (32.8) 20082012 3334 78 (18.3) 1705 (35.6) 275 (32.5) 1276 (41.6)

Race

Black 1016 49 (11.5) 645 (13.5) 59 (7.0) 263 (8.6) <0.001

White 7621 356 (83.6) 3832 (80.1) 750 (88.5) 2683 (87.5)

Other 488 21 (4.9) 309 (6.5) 38 (4.5) 120 (3.9)

Age

60 3515 159 (37.3) 1773 (37.0) 419 (49.5) 1164 (38.0) <0.001 >60 5610 267 (62.7) 3013 (63.0) 428 (50.5) 1902 (62.0)

Sex

Male 7486 349 (81.9) 3867 (80.8) 736 (86.9) 2534 (82.6) <0.001

Female 1639 77 (18.1) 919 (19.2) 111 (13.1) 532 (17.4)

Tumor histology

Squamous 2757 111 (26.1) 1769 (37.0) 207 (24.4) 670 (21.9) <0.001

Adenocarcinoma 5406 270 (63.4) 2574 (53.8) 543 (64.1) 2019 (65.9)

Other 962 45 (10.6) 443 (9.3) 97 (11.5) 377 (12.3)

Tumor location

Upper thoracic 483 8 (1.9) 340 (7.1) 27 (3.2) 108 (3.5) <0.001

Middle thoracic 1647 62 (14.6) 1029 (21.5) 110 (13.0) 446 (14.5)

Lower thoracic 6995 356 (83.6) 3417 (71.4) 710 (83.8) 2512 (81.9)

Grade (n=7653)

G1 270 14 (3.6) 143 (3.6) 24 (3.2) 89 (3.5) 0.271

G2 2826 138 (35.8) 1522 (38.0) 287 (38.4) 879 (34.9)

G3-4 4557 234 (60.6) 2339 (58.4) 436 (58.4) 1548 (61.5)

Table 1. Patient characteristics. CDS, cancer-directed surgery; RT, radiotherapy; G1, well dierentiated; G2, moderately dierentiated; G3, poorly dierentiated; G4, undierentiated.

while 38.3% (324/847) were received postoperative radiotherapy and 4.5% (38/847) were underwent both preoperative and postoperative RT.

Survival. The median follow-up time for all patients was 9 months (range, 4261 months) with a median survival time was 10 months. The 1 year, 2 years, 3 years, 5 years, and 10 years OS rates were 40.5%, 14.6%, 8.4%, 5.4%, and 3.5%, respectively (Fig.2).

Prognostic factors analysis. Univariate analysis showed that year of diagnosis, age, race, tumor histology, grade, and local treatment modalities were risk factors for OS (Table2).

Multivariate analysis indicated that year of diagnosis, age at diagnosis, race, tumor histology, grade, and local treatment modalities were independent prognostic factors for OS. Patients who underwent primary CDS was signicantly better OS than that of patients who were primary RT alone (HR, 1.440; 95% CI, 1.2871.611; P<0.001) and who were not received any local treatment (HR, 1.602; 95% CI, 1.4271.799; P< 0.001). Surgery combined with RT could further improve survival (HR, 0.793; 95% CI, 0.6930.908; P=0.001) (Table3).

Survival after local treatment. The 5-year OS rates were 8.4%, 4.5%, 17.5%, and 3.4% for primary CDS, RT alone, CDS plus RT, and no local treatment, respectively, with a median survival time of 11.0, 9.0, 15.0, and 9.0 months, respectively (P< 0.001) (Fig.3). Patients who were received preoperative RT was signicantly better OS than that of patients who underwent primary CDS and CDS plus postoperative RT, with 5-year OS rates of 24.7%, 6.5%, and 7.8%, respectively, and a median survival time of 20.0, 11.0, and 12.0 months, respectively (P<0.001) (Fig.4).

We determined the eect of local treatment modalities on OS by year of diagnosis. It was signicantly associated with OS from 1988 to 1999 (log-rank test P< 0.001) (Figure S1A) and from 2000 to 2012 (P< 0.001) (Figure S1B). However, the survival benet was signicantly better from 2000 to 2012 for those treated with CDS plus RT.

The prognostic eect of local treatment modalities was examined based on dierent ethnicities. In black patients, CDS with or without RT improved OS than that of patients who underwent primary RT alone or did not have any local treatment (P < 0.001) (Figure S2A). In white patients, OS was better for patients who underwent CDS plus RT, as compared with other local treatment modalities (P< 0.001) (Figure S2B).

SCIENTIFIC REPORTS

3

www.nature.com/scientificreports/

Figure 2. Overall survival of patients with metastatic esophageal cancer.

Characteristic HR 95%CI P-value

Year of diagnosis 0.975 0.9710.979 <0.001

Age 1.008 1.0061.010 <0.001Race

Black 1

White 0.922 0.8600.988 0.021Other 0.89 0.7930.998 0.047Sex

Male 1

Female 0.974 0.9201.031 0.359Tumor histology

Squamous 1

Adenocarcinoma 0.977 0.9311.026 0.347Other 1.092 1.0101.180 0.027Tumor location

Upper thoracic 1

Middle thoracic 1.061 0.9531.181 0.280Lower thoracic 0.982 0.8901.083 0.716Grade

G1 1

G2 1.028 0.9001.174 0.684G3-4 1.179 1.0351.344 0.013Local treatment modalities

CDS 1

RT 1.291 1.1621.435 <0.001CDS+RT 0.690 0.6080.784 <0.001None 1.384 1.2421.542 <0.001

Table 2. Univariate analysis of overall survival. CDS, cancer-directed surgery; RT, radiotherapy; G1, well dierentiated; G2, moderately dierentiated; G3, poorly dierentiated; G4, undierentiated; HR, hazard ratio; CI, condence interval.

The eect of local treatment modalities on OS was examined based on sex. In patients who underwent CDS plus RT, OS was signicantly better than that of patients who were received other local treatment modalities

SCIENTIFIC REPORTS

4

www.nature.com/scientificreports/

Characteristic HR 95% CI P-value

Year of diagnosis 0.971 0.9670.975 <0.001

Age 1.007 1.0051.009 <0.001Race

Black 1

White 0.906 0.8340.985 0.021Other 0.884 0.7791.003 0.055Tumor histology

Squamous 1

Adenocarcinoma 1.099 1.0351.167 0.002Other 1.174 1.0721.286 0.001Grade

G1 1

G2 1.067 0.9340.985 0.337G3-4 1.202 1.0551.371 0.006Local treatment modalities

CDS 1

RT 1.440 1.2871.611 <0.001CDS+RT 0.793 0.6930.908 0.001None 1.602 1.4271.799 <0.001

Table 3. Multivariate analyses of overall survival. CDS, cancer-directed surgery; RT, radiotherapy; G1, well dierentiated; G2, moderately dierentiated; G3, poorly dierentiated; G4, undierentiated; HR, hazard ratio; CI, condence interval.

Figure 3. Overall survival of patients with metastatic esophageal cancer undergoing dierent local treatment modalities (CDS, cancer-directed surgery; RT, radiotherapy).

(P < 0.001 for male patients; P < 0.001 for female patients) (Figure S3A,S3B). The results were also signicant dierence in patients who were aged 60 years (P< 0.001) and aged >60 years (P< 0.001) (Figure S4A,S4B).

In addition, CDS plus RT provided the OS benet in patients with squamous cell carcinoma (P < 0.001) (Figure S5A), adenocarcinoma (P < 0.001) (Figure S5B), grade I-II (P< 0.001) (Figure S6A), and grade III-IV P< 0.001) metastatic EC patients. (Figure S6B).

The OS rates were compared based on tumor location for patients who were underwent dierent local treatment modalities (Figure S7AS7C). The prognostic eect of local treatment modalities was also found in patients with tumors located in the middle thoracic esophagus (P < 0.001) and lower thoracic esophagus (P < 0.001). However, in patients with tumors located in the upper thoracic esophagus, local treatment modalities were not associated with OS (P=0.272).

SCIENTIFIC REPORTS

5

www.nature.com/scientificreports/

Figure 4. Overall survival of patients with metastatic esophageal cancer undergoing surgery combined with radiotherapy (CDS, cancer-directed surgery; RT, radiotherapy; pre, preoperative; post, postoperative).

Discussion

Given the limited of studies with small sample sizes investigating the eect of local treatment in metastatic EC1013.

In this study, we explored the prognostic value of local treatment modalities including CDS and RT in metastatic EC based on 9,125 metastatic EC patients in the SEER database and our results found that surgery plus RT could signicantly improve survival in metastatic EC.

Systematic therapy is still the rst-line treatment for metastatic EC. The main purpose of local treatment lies in eective control of dysphagia, pain, bleeding, and other symptoms. The potential value of surgery and RT in metastatic EC remains controversial. In a study by Schauer et al., 19 patients with stage IV Barretts adenocarcinoma received multimodality therapy including resection of the primary tumor. No signicant dierence was found in postoperative morbidity and mortality between metastatic EC and locally advanced EC, but the median survival was only 9 months16. Tanaka et al. also found that surgery did not improve survival in stage IVB EC with distant organ metastasis (P=0.1291)5. Wang et al. included 96 patients with stage IV EC who were received palliative chemotherapy and concurrent chemoradiotherapy (CRT), of which 14 patients underwent surgery aer neoadjuvant therapy and surgery had signicantly better survival than those who did not11. Two related studies also showed that long-term survival could be achieved aer resection of the primary tumor and metastases of stage IV EC12,13. In our study, 1,273 patients received surgery with or without RT, and surgery combined with RT could signicantly improve survival. Thus, multimodality therapy including surgery and RT has the potential to prolong survival in metastatic EC.

Multimodality therapy is the dominant research direction in metastatic EC. Our subgroup analysis showed that in 20002012, patients who underwent surgery plus RT obtained a signicantly better survival than patients in 19881999. Although the SEER data could not reect specic conditions in patients regarding chemotherapy and targeted therapy, we speculated that it was closely correlated with of the eect of systemic treatment in metastatic EC1720. Systemic therapy is the primary treatment of metastatic EC, but local treatment including surgery or RT aer eective systemic therapy could further reduce the tumor burden. Therefore, we recommend for future prospective studies to investigate the eect of local treatment in metastatic EC.

Our study showed that patients with upper thoracic esophageal cancer did not benet from local treatment, which might be related to greater difficulties in surgical treatment in upper thoracic esophageal cancer than middle and lower thoracic esophageal cancer. We could not clarify the eect of surgical treatment in upper thoracic metastatic EC, as only 30 patients underwent surgery with or without RT in this study.

In this study, the 5-year OS for preoperative RT plus CDS could reach 24.7%, while no signicant dierence in survival was seen for primary CDS and CDS plus postoperative RT (5-year OS, 6.5% and 7.8%, respectively), indicating that preoperative neoadjuvant therapy has a greater value in metastatic EC. Our study found that the OS improvement for surgery plus RT was mainly reected by preoperative RT which could provide the best chance for the complete resection of primary tumors.

There are several limitations in our study. First, inherent biases exist in any retrospective study. Second, due to the limitations of SEER data, we could not obtain related information including chemotherapy, indications for surgery and RT, and range of non-regional lymph node metastases and distant metastases. In addition, patients with distant SEER stage were intended to approximate stage IV in the TNM staging system, and our results also promoted that OS of distant stage in SEER was substantially similar to that of stage IV esophageal carcinoma.

SCIENTIFIC REPORTS

6

www.nature.com/scientificreports/

Several dierent extent of disease schemes have been used in the SEER database. Therefore, a potential dierence in the two staging systems should be considered. However, the primary strength of this study was the ability to assess the epidemiology, prognostic factors, and local treatment modalities in metastatic EC using a SEER registry. Although retrospective reviews are generally considered inferior to prospective studies, no prospective study design has been performed to assess the clinical value of local treatment in metastatic EC.

In conclusion, surgery plus RT, especially preoperative RT, may improve long-term survival of patients with metastatic EC. A prospective study on metastatic EC should be conducted to investigate the eect of local treatment in metastatic EC. Our ndings may play an important role in local treatment considerations in metastatic EC if further conrmed in studies with larger sample sizes.

Methods and Materials

Patients. Data were obtained from the current SEER database to identify patients with EC diagnosed in 19882012. We obtained permission to access research data les with the reference number 11252-Nov 201421.

Patients included in this study had the following criteria: 1) metastatic thoracic esophageal cancer with a known tumor location; and 2) local treatment modalities including cancer-directed surgery (CDS), beam radiotherapy, CDS plus RT, or no local treatment. Metastatic disease was dened as having a distant stage at diagnosis according to the SEER historic stage. Distant stage was dened as a neoplasm that had spread to parts of the body remote from the primary tumor through direct extension, discontinuous metastasis (e.g., implantation or seeding) to distant organs and tissues, or the lymphatic system to distant lymph nodes21. Patients were excluded from the analysis if they had an estimated survival of 3 months aer diagnosis. SEER data did not require informed consent, and this study was approved by the ethics committee of the Sun Yat-sen University Cancer Center.

Clinicopathological and treatment factors. The following clinicopathological and treatment factors were collected from the SEER database: year of diagnosis, age at diagnosis, race, tumor histology, tumor location, grade, and local treatment modalities. Vital status including cause of death and duration of follow-up was recorded.

Statistical analysis. The 2 and Fishers exact probability tests were used to analyze dierences between qualitative data. Univariate and multivariate Cox regression analyses were generated to analyze risk factors for overall survival (OS). Multivariable analyses were performed for factors that were signicantly associated with OS in univariate analyses. Survival rates were calculated and plotted using the KaplanMeier method and compared using the log-rank test. All data were analyzed using the SPSS statistical soware package, version 21.0 (IBM Corporation, Armonk, NY, USA). A P value of <0.05 was considered signicant.

References

1. Rustgi, A. K. & El-Serag, H. B. Esophageal carcinoma. N Engl J Med 371, 2499509 (2014).2. Siegel, R. L., Miller, K. D. & Jemal, A. Cancer statistics, 2015. CA Cancer J Clin 65, 529 (2015).3. Enzinger, P. C. & Mayer, R. J. Esophageal cancer. N Engl J Med 349:22412252 (2003).4. Horner, M. J. et al. SEER Cancer Statistics Review, 1975-2006, National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/ csr/1975_2006/, based on November 2008 SEER data submission, posted to the SEER web site, 2009. (2009) (Date of access: 02/11/2015).

5. Tanaka, T. et al. Outcomes of multimodality therapy for stage IVB esophageal cancer with distant organ metastasis (M1-Org). Dis Esophagus 23, 646651 (2010).

6. Berry, M. F. Esophageal cancer: staging system and guidelines for staging and treatment. J Thorac Dis 6, S289297 (2014).7. Stah, M. et al. Oesophageal cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 24, vi5156 (2013).

8. Shapiro, J. et al. Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer(CROSS): long-term results of a randomised controlled trial. Lancet Oncol 16, 10901098 (2015).

9. Nieman, D. R. & Peters, J. H. Treatment strategies for esophageal cancer. Gastroenterol Clin North Am 42, 187197 (2013).10. Yamashita, K. et al. Multimodality Treatment and Prognosis in Esophageal Squamous Cell Carcinoma Requiring Esophagectomy. Hepatogastroenterology 61, 10421048 (2014).

11. Wang, J. et al. Factors Predictive of Improved Outcomes With Multimodality Local Therapy Aer Palliative Chemotherapy forStage IV Esophageal Cancer. Am J Clin Oncol. 39, 228235 (2016).

12. Chao, Y. K. et al. Distant nodal metastases from intrathoracic esophageal squamous cell carcinoma: characteristics of long-term survivors aer chemoradiotherapy. J Surg Oncol 102, 158162 (2010).

13. Blank, S. et al. A reliable risk score for stage IV esophagogastric cancer. Eur J Surg Oncol 39, 823830 (2013).14. Ouyang, W. W. et al. Radiation dose and survival of patients with stage IV non-small cell lung cancer undergoing concurrentchemotherapy and thoracic three-dimensional radiotherapy: reanalysis of the ndings of a single-centerprospective study. BMC Cancer 14, 491 (2014).

15. Khan, H., Khan, N., Ahmad, A., Olszewski, A. J. & Somasundar, P. Surgical management of metastatic colon cancer: A population-based analysis. J Geriatr Oncol 6, 446453 (2015).

16. Schauer, M. et al. Results of a multimodal therapy in patients with stage IV Barretts adenocarcinoma. World J Surg 32, 26552660 (2008).17. Anderson, S. E. et al. Combined modality therapy in esophageal cancer: the Memorial experience. Semin Surg Oncol 21, 228232 (2003).18. Dutton, S. J. et al. Getinib for oesophageal cancer progressing aer chemotherapy (COG): a phase 3, multicentre, double-blind, placebo-controlled randomised trial. Lancet Oncol 15, 894904 (2014).

19. Ford, H. E. et al. Docetaxel versus active symptom control for refractory oesophagogastric adenocarcinoma (COUGAR-02): an open-label, phase 3 randomised controlled trial. Lancet Oncol 15, 7886 (2014).

20. Bang, Y. J. et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positiveadvanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlledtrial. Lancet 376, 687697 (2010).

21. Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: Incidence-SEER 18 Regs Research Data+Hurricane Katrina Impacted Louisiana Cases, Nov 2014 Sub (1973-2012 varying)-Linked To County Attributes-Total U.S., 1969-2013 Counties, National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2015, based on the November 2014 submission. (2015) (Date of access: 30/05/2015).

SCIENTIFIC REPORTS

7

www.nature.com/scientificreports/

This work was supported by grants from the National Natural Science Foundation of China (No. 81402527), the Sci-Tech Office of Guangdong Province (No. 2013B021800157, 2013B021800458) and the Youth Foundation of Fujian Provincial Health and Family Planning Commission (No. 2014-2-63).

S.-G.W., W.-H.X., Z.-Q.Z., Y.-B. and Z.-Y.H. was involved in study concept and design, draing of the manuscript, obtained funding and study supervision. S.-G.W. and W.-H.X. was involved in acquisition of data. Z.-Q.Z., J.-Y.S. and H.-X.L. were involved in statistical analysis. S.-G.W., W.-H.X., F.-Y.L., Z.-Q.Z. and Z.-Y.H. were involved in interpretation of data revision of manuscript. All authors were involved in the interpretation of the data; revision of the manuscript; and the decision to submit the manuscript for publication.

Supplementary information accompanies this paper at http://www.nature.com/srep

Competing nancial interests: The authors declare no competing nancial interests.

How to cite this article: Wu, S.-G. et al. Surgery Combined with Radiotherapy Improved Survival in Metastatic Esophageal Cancer in a Surveillance Epidemiology and End Results Population-based Study. Sci. Rep. 6, 28280; doi: 10.1038/srep28280 (2016).

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

SCIENTIFIC REPORTS

8