Skip to main content

Some NLM-NCBI services and products are experiencing heavy traffic, which may affect performance and availability. We apologize for the inconvenience and appreciate your patience. For assistance, please contact our Help Desk at info@ncbi.nlm.nih.gov.

NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Dec 1.
Published in final edited form as: Ann Thorac Surg. 2019 Jul 26;108(6):1633–1639. doi: 10.1016/j.athoracsur.2019.05.091

Surgery is Associated with Survival Benefit in T4a Esophageal Adenocarcinoma: A National Analysis

Vignesh Raman 1, Oliver K Jawitz 1, Soraya L Voigt 1, Norma E Farrow 1, Chi-Fu J Yang 2, Thomas A D’Amico 1, David H Harpole Jr 1
PMCID: PMC7257596  NIHMSID: NIHMS1589670  PMID: 31356800

Abstract

Background

The National Comprehensive Cancer Network (NCCN) guidelines recommend consideration of surgery for clinical T4a esophageal adenocarcinoma. However, there are limited data on the outcomes of patients with T4a adenocarcinoma treated with surgery vs. definitive chemoradiation.

Methods

The National Cancer Database (NCDB) was used to identify patients from 2010–2015 with clinical T4aN0–3M0 esophageal adenocarcinoma, and grouped by receipt of surgery (with or without perioperative therapy) or definitive, concurrent chemoradiation. Patients receiving incomplete definitive therapy or with missing survival information were excluded. Overall survival (OS) was evaluated with Kaplan-Meier and Cox proportional hazard analyses.

Results

Of 182 patients in the study, 85 (47%) underwent esophagectomy and 97 (53%) underwent chemoradiation. In the surgery cohort, 79 patients (93%) received perioperative chemotherapy. Unadjusted and multivariable analyses demonstrated a significant survival benefit associated with surgery compared to definitive chemoradiotherapy (adjusted hazard ratio [HR] 0.32; 95%CI 0.21, 0.50). A 1:1 propensity score-matched analysis of 63 patient pairs also revealed a significant OS benefit with surgery compared to chemoradiotherapy alone (HR 0.26; 95%CI 0.16, 0.43).

Conclusion

In this national analysis, surgery for cT4a esophageal adenocarcinoma was associated with improved outcomes when compared to definitive chemoradiation. Surgery should be considered for medically fit patients with cT4aN0–3M0 esophageal adenocarcinoma.

Introduction

Esophageal cancer has a projected incidence of 17,000 in 2018, with 16,000 deaths1. It carries a poor prognosis, with an all-stage five-year survival of 19%. Induction chemoradiotherapy followed by surgery is the accepted treatment for locally advanced esophageal cancer 2,3. Stage T4 esophageal cancer portends poor survival, estimated between five and 23 months, with a five-year survival of five to 60% 49. The National Comprehensive Cancer Network guidelines recommend consideration of surgery with induction or adjuvant therapy for patients with clinical stage T4a cancer, which includes spread to the pleura, pericardium, and diaphragm but not the great vessels or trachea. However, there is limited and conflicting literature about the best management strategy for clinical stage T4 esophageal cancer. Three prospective trials in patients with locally advanced squamous cell cancer have not demonstrated a survival benefit for surgery compared to definitive chemotherapy and radiation 1012. However, cohort studies also in squamous cell cancer have demonstrated a survival benefit with surgery following chemoradiation 7,13. There are no prospective or retrospective studies examining patients with T4 adenocarcinoma. We performed an analysis of the National Cancer Database (NCDB) to compare the outcomes of patients with T4 esophageal adenocarcinoma treated with surgery vs. definitive chemotherapy and radiation.

Methods

National Cancer Database

The NCDB is a collaborative effort of the American College of Surgeons and American Cancer Society, representing about 80% of newly diagnosed cancers across 1500 institutions in the United States and Puerto Rico 14,15.

Study Design

The study was reviewed and deemed exempt by our Institutional Review Board. All patients in the NCDB diagnosed with clinical stage T4a esophageal adenocarcinoma were identified using the International Classification of Diseases for Oncology, Third Edition (ICD-0–3) histology codes. The American Joint Commission on Cancer introduced the T4a stage in 2009, so only patients diagnosed between 2010 and 2015 were included. The cohort was limited to patients with clinical T4aN0–3M0 disease treated with either surgical resection with or without perioperative therapy or definitive chemotherapy and radiation (Fig. 1). Patients who did not receive complete definitive therapy by NCCN recommendations, namely those who received definitive chemotherapy alone or radiation alone, who did not receive concurrent chemoradiation, and who received <= 40 Gy of radiation, along with patients with missing survival data were excluded. The primary outcome of interest was overall survival (OS).

Figure 1.

Figure 1.

Patient selection scheme

Statistical Analysis

Patients in the study were grouped by type of treatment received (surgery or definitive chemoradiation). The Wilcoxon rank sum test and Pearson’s chi-squared test were used for comparison of continuous variables and categorical variables, respectively. OS was computed with the Kaplan-Meier product-limit and long-rank tests. Multivariable regression was performed using a Cox proportional hazards model. A nearest-neighbor 1:1 propensity score-matched analysis with a caliper of 0.216 based on patient- and tumor-related variables (age, sex, race, insurance status, urban or rural status, treatment at academic center, Charleson-Deyo comorbidity index, and clinical N stage) was used to compare survival of two groups of patients with similar characteristics. In an effort to mitigate immortal time bias in patients undergoing surgery compared to chemoradiation, conditional landmark survival analyses17,18 were performed on the propensity score-matched groups using landmark times of median time from diagnosis to surgery (4.6 months) and the 90th percentile of time to surgery (6.1 months). All statistical analyses were performed with R version 3.5 for Mac (Vienna, Austria). A two-sided p value less than or equal to 0.05 was considered statistically significant.

Results

A total of 182 patients met study criteria (Fig. 1), of whom 85 (47%) underwent esophagectomy and 97 (53%) underwent definitive chemoradiation. The demographic characteristics of patients in either group are summarized in Table 1. Patients undergoing esophagectomy were more likely to be younger and treated at an academic institution. Amongst patients undergoing surgery, 79 (93%) received perioperative chemotherapy: 72 received induction chemoradiation (85%), 1(1%) received adjuvant chemoradiation and six patients (7%) received induction chemotherapy alone. Six patients (7%) received adjuvant radiation alone. Patients undergoing surgery had a median length of stay of 10 days (IQR 7–12 days), with a 30-day readmission rate of 5%. The complete resection rate was 91%. Ninety-day postoperative mortality was 12%. Of patients undergoing surgery, 30% had pathologic T0 (pT0) disease, 4% pT1, 20% pT2, 31% pT3, and 13% pT4a, respectively.

Table 1.

Demographic characteristics of study patients

Chemoradiation (n=97) (%) Surgery (n=85) (%) p value
Age (years, median) 69 62 <0.001
Sex (female) 18 (18.6) 19 (22.4) 0.65
Race 0.40
White 94 (97.9) 79 (94.0)
Black 1 (1.0) 3 (3.6)
Other 1 (1.0) 2 (2.4)
Year of diagnosis, median (inter-quartile range) 2012 (2011–2014) 2012 (2011–2013) 0.68
CDCC Score 0.90
0 74 (76.3) 63 (74.1)
1 18 (18.6) 18 (21.2)
2+ 5 (5.2) 4 (4.7)
Insurance status 0.20
Private 37 (38.1) 43 (51.2)
Government 59 (60.8) 40 (47.6)
None 1 (1.0) 1 (1.2)
Facility location 0.46
Metro 79 (82.3) 65 (78.3)
Urban 16 (16.7) 18 (21.7)
Rural 1 (1.0) 0 (0)
Facility type 0.01
Academic/research program 34 (35.1) 46 (55.4)
Clinical N stage 0.65
N0 34(35.1) 35(41.2)
N1 44(45.4) 31(36.5)
N2 15(15.5) 16(18.8)
N3 4(4.1) 3(3.5)
Perioperative chemotherapy 79(92.9)
Number of radiation doses (IQR) 28(28–30) 28(25–28) 0.19
Median radiation dose (Gy) (IQR) 50.0(45.0–50.4) 50.4(45.0–50.4) 0.99

Patients undergoing esophagectomy had a median survival of 36 months (95% confidence interval [CI] 25, N/A) while those treated with chemoradiation had a median survival of 13 months (95% CI: 11, 17). Univariable analysis demonstrated a significant advantage in OS for patients treated with surgery compared to chemoradiation (Fig. 2). Cox multivariable regression also demonstrated a significant survival benefit for surgery patients (hazard ratio [HR] 0.32; 95%CI 0.21, 0.50) compared to chemoradiation (Table 2).

Figure 2.

Figure 2.

Kaplan-Meier survival curves of patients with cT4a esophageal adenocarcinoma grouped by treatment

Table 2.

Cox multivariable regression of independent predictors of survival in patients with T4a esophageal adenocarcinoma treated with esophagectomy vs. definitive chemoradiation

95% Confidence Interval

Predictor Hazard Ratio Lower Upper p-value
Age (per year) 1.00 0.97 1.02 0.87
Sex (female) 1.01 0.61 1.67 0.97
Race (reference: White)
Black 3.84 0.88 16.8 0.07
Other 0.73 0.10 5.50 0.76
Year of diagnosis (per year) 1.15 0.99 1.34 0.07
CDCC score (reference: 0)
1 1.25 0.77 2.03 0.36
2+ 1.52 0.55 4.23 0.42
Insurance status (reference: private)
Government 1.56 0.93 2.63 0.09
None 0.60 0.08 4.51 0.62
Facility location (reference: metro)
Urban 1.71 1.11 2.64 0.01
Rural 0.53 0.07 4.04 0.54
Facility type (reference: non-academic)
Academic/Research Program 1.15 0.77 1.73 0.48
Clinical N stage (reference: N0)
N1 1.52 0.98 2.36 0.06
N2 1.39 0.77 2.51 0.28
N3 3.20 1.35 7.60 0.009
Treatment (reference: definitive chemoradiation)
Esophagectomy 0.32 0.21 0.50 <0.001

A propensity-score matched analysis revealed 63 pairs of matched patients (Table 3). Surgery was associated with a significantly higher survival compared to definitive therapy on unadjusted matched analysis (Fig. 3). Multivariable analysis demonstrated that surgery independently predicted survival (HR 0.26; 95%CI 0.16, 0.43; Table 4). In a landmark analysis, using a landmark analysis time point of 4.6 months, which was the median time from diagnosis to surgery in the surgery group, surgery was associated with improved survival in unadjusted (Fig. 4) and multivariable analysis (HR 0.32; 95% CI: 0.19, 0.54) when compared to definitive chemoradiation. In a second landmark analysis at the 90th percentile of time to surgery (6.1 months), surgery was again associated with improved survival compared to chemoradiation (HR 0.31; 95%CI 0.18, 0.54).

Table 3.

Demographic characteristics of propensity score-matched patients

Chemoradiation (n=63) (%) Surgery (n=63)(%) p value
Age (years, median) 65 63 0.77
Sex (female) 12(19) 14(22) 0.83
Race 0.60
White 62(98) 61(97)
Black 0(0) 1(1.5)
Other 1(2) 1(1.5)
Year of diagnosis, median (inter-quartile range) 2012(2011–2014) 2012(2011–2013) 0.49
CDCC Score 0.52
0 49(78) 45(71)
1 10(16) 15(24)
2+ 4(6) 3(5)
Insurance status 0.77
Private 26(41) 30(48)
Government 36(57) 32(51)
None 1(2) 1(1.5)
Facility location 0.60
Metro 50(79) 50(79)
Urban 12(19) 13(21)
Rural 1(1.5) 0(0)
Facility type 0.72
Academic/research program 27(43) 30(48)
Clinical N stage 0.78
N0 23(37) 23(37
N1 27(43) 25(40)
N2 12(19) 12(19)
N3 1(1.5) 3(5)
Perioperative chemotherapy N/A 58(92) N/A

Figure 3.

Figure 3.

Kaplan-Meier survival curves of propensity score-matched patients with cT4a esophageal adenocarcinoma grouped by treatment

Table 4.

Cox multivariable regression of independent predictors of survival in propensity score-matched patients with T4a esophageal adenocarcinoma treated with surgery vs. definitive chemoradiation

95% Confidence Interval
Predictor Hazard Ratio Lower Upper p-value
Age (per year) 0.99 0.96 1.03 0.59
Sex (female) 1.19 0.68 2.06 0.54
CDCC score (reference: 0)
1 1.39 0.77 2.49 0.28
2+ 5.80 1.97 17.1 0.001
Insurance status (reference: private)
Government 1.30 0.71 2.37 0.40
None 0.42 0.05 3.26 0.41
Facility location (reference: metro)
Urban 1.57 0.92 2.68 0.10
Rural 0.48 0.06 3.75 0.49
Facility type (reference: non-academic)
Academic/Research Program 0.91 0.57 1.47 0.70
Clinical N stage (reference: N0)
N1 1.75 1.00 3.06 0.05
N2 1.74 0.91 3.33 0.09
N3 2.95 0.94 9.25 0.06
Treatment (reference: chemoradiation)
Esophagectomy 0.26 0.16 0.43 <0.001

Figure 4.

Figure 4.

Kaplan-Meier survival curves of propensity score-matched patients with cT4a esophageal adenocarcinoma at landmark analysis timepoint of median time to surgery (4.6 months)

Discussion

We present the largest analysis, to our knowledge, of outcomes for patients with clinical stage T4a esophageal adenocarcinoma. In our study, patients who underwent esophagectomy had improved OS compared to patients who received definitive chemoradiation, even after adjustment for comorbidities and nodal staging, and with a propensity score-matched and conditional landmark analysis.

While the NCCN guidelines recommend consideration of surgery for clinical stage T4a esophageal cancer, the limited literature on locally advanced esophageal cancer is equivocal and described in predominantly squamous cell cancer populations. Three prospective studies failed to identify an OS benefit to adding surgery following chemoradiation in locally advanced squamous cell cancer. Fujita and colleagues described a prospective trial of 53 patients with T4N0–1M0 squamous cell cancer who underwent either definitive chemoradiation or surgery with both pre- and postoperative chemoradiation 12. There was no significant difference in five-year OS between the two groups, although a subgroup analysis revealed a trend towards a survival benefit for patients receiving surgery who were non-responders to chemoradiotherapy. Stahl and colleagues reported a prospective, randomized controlled trial (RCT) of 172 patients with T3–4N0M0 squamous cell cancer treated with either chemoradiation followed by surgery or extended, definitive chemoradiation 8. They found that OS did not differ between the two groups, but there was improved local control with surgery (two-year progression-free survival of 64% vs. 41%). Bedenne et al. performed an RCT of 259 patients with T3N0–1M0 squamous cell cancer, who were treated with either chemoradiation alone or chemoradiation followed by surgery 11. Importantly, only patients who were deemed responders to chemoradiation were enrolled in this trial. They reported no difference in two-year OS between the groups. However, three small Japanese cohort studies in patients with locally advanced squamous cell cancer report a significant survival benefit with surgery. Ikeda and colleagues describe a median survival of about 30 months for 13 patients receiving surgery following chemoradiation for T4 esophageal cancer, compared to six months for those receiving definitive chemoradiation 19. Noguchi et al. found a significant survival benefit (median 14 months vs. 3 months) for 24 patients treated with surgery following chemoradiation in T4 esophageal cancer 6, while Yano and colleagues found in a cohort of 45 patients that chemoradiation with surgery confers a survival of 30 months compared to six months without surgery 7.

The existing data have numerous confounding factors that limit interpretation. All of the major literature describe outcomes in largely squamous cell populations, and there is no study to our knowledge that examines an adenocarcinoma-predominant population. Additionally, the published studies are at least one to two decades old. As a result, they also pre-date the T4a-T4b distinction identified by the AJCC in 2009, and therefore include patients who would now be considered T4b and unresectable. Indeed, the aorta and main airway are the most common sites of invasion in T4 esophageal cancer, and are most often associated with a margin-positive resection and poor survival 3,20. Each study also has a different chemotherapy and radiation regimen. They describe small patient populations. By contrast, our study examines a relatively large contemporary cohort of patients with clinical stage T4a esophageal adenocarcinoma.

Nevertheless, our study has its own limitations. It is a retrospective cohort study, with inherent confounding variables. Because the NCDB does not provide granular information on clinical decision-making, there is likely significant selection bias. One of the outstanding questions from this study, for instance, is the reasons patients with cT4a disease were assigned to one of two treatment modalities, which the NCDB does not capture. The NCCN recommends consideration of surgery for patients with cT4a adenocarcinoma without providing specific guidance on patient selection given the limited data. At our institution, we consider patients with cT4a disease for surgery if they have adequate pulmonary function, with an FEV1 and DLCO exceeding 60%, if they have low cardiac risk (either no risk factors or a negative cardiac stress test), and if the have an ECOG performance status of zero. The NCDB, unfortunately, does not catalogue these variables, which is an important limitation of our analysis. In addition, the NCDB does not provide re-staging information following induction chemoradiotherapy, and this is likely an important factor in determining a patient’s suitability for surgery, given the prospective literature suggesting that patients who are poor responders to chemoradiotherapy may better benefit from surgery than those who do respond. For instance, in our institution, we sometimes offer surgery to high risk surgical candidates if they have residual local disease without evidence of distant metastasis on PET following induction therapy. This information is not available in the NCDB, along with the type of staging performed. The NCDB also does not identify patients with metastatic disease following induction therapy who never received surgery, and does not include data about the number of cycles and type of chemotherapy, which is an important confounder in our analysis. The survival analysis in this study might also suffer from immortal time bias as survival is measured from time of diagnosis, and patients who receive surgery necessarily survived until surgery while patients receiving chemoradiation might have died before receiving surgery, therefore inflating the outcomes of the surgery population. To minimize this bias, we performed a conditional landmark survival analysis examining only patients in either group who remained at risk at the median time from diagnosis to surgery, and this analysis recapitulated the finding that overall survival is improved with surgery.

Using a large national database, we report that patients with clinical stage T4a esophageal adenocarcinoma treated with surgery have improved survival compared to patients treated with definitive chemoradiotherapy. Guideline-concordant induction therapy and surgery should be considered for patients with cT4a esophageal cancer who can tolerate surgery. Larger prospective randomized data are needed, especially in adenocarcinoma, but patient accrual may prove challenging given the incidence.

Acknowledgements and Funding

The American College of Surgeons is in a Business Associate Agreement that includes a data use agreement with each of its Commission on Cancer accredited hospitals. The data used in the study are derived from a de-identified National Cancer Data Base file. The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology used or the conclusions drawn from these data by the investigators.

Drs. Jawitz was supported by a National Institutes of Health T32 grant in clinical research (T32HL069749). Drs. Raman, Farrow and Voigt were supported by a National Institutes of Health T32 grant in surgical oncology (T32 CA093245).

References

  • 1.Esophageal Cancer - Cancer Stat Facts. https://seer.cancer.gov/statfacts/html/esoph.html. Accessed September 3, 2018.
  • 2.Akutsu Y, Matsubara H. Chemoradiotherapy and surgery for T4 esophageal cancer in Japan. Surg Today. 2015;45(11):1360–1365. doi: 10.1007/s00595-015-1116-4 [DOI] [PubMed] [Google Scholar]
  • 3.Siersema PD, Van RH. Treatment of locally advanced esophageal cancer with surgery and chemoradiation. Curr Opin Gastroenterol. 2008;24(4):535–540. doi: 10.1097/MOG.0b013e3283025ca2 [DOI] [PubMed] [Google Scholar]
  • 4.Chak A, Canto M, Gerdes H, et al. Prognosis of esophageal cancers preoperatively staged to be locally invasive (T4) by endoscopic ultrasound (EUS): a multicenter retrospective cohort study. Gastrointest Endosc. 1995;42(6):501–506. doi: 10.1016/S0016-5107(95)70001-3 [DOI] [PubMed] [Google Scholar]
  • 5.Galandiuk S, Hermann RE, Cosgrove DM, Gassman JJ. Cancer of the esophagus. The Cleveland Clinic experience. Ann Surg. 1986;203(1):101–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Noguchi T, Moriyama H, Wada S, et al. Resection surgery with neoadjuvant chemoradiotherapy improves outcomes of patients with T4 esophageal carcinoma. Dis Esophagus. 2003;16(2):94–98. doi: 10.1046/j.1442-2050.2003.00304.x [DOI] [PubMed] [Google Scholar]
  • 7.Yano M, Tsujinaka T, Shiozaki H, et al. Concurrent chemotherapy (5-fluorouracil and cisplatin) and radiation therapy followed by surgery for T4 squamous cell carcinoma of the esophagus. J Surg Oncol. 1999;70(1):25–32. doi: [DOI] [PubMed] [Google Scholar]
  • 8.Stahl M, Stuschke M, Lehmann N, et al. Chemoradiation with and without surgery in patients with locally advanced squamous cell carcinoma of the esophagus. J Clin Oncol Off J Am Soc Clin Oncol. 2005;23(10):2310–2317. doi: 10.1200/JCO.2005.00.034 [DOI] [PubMed] [Google Scholar]
  • 9.Karimata H, Shimoji H, Nishimaki T. Clinicopathological factors predicting R0 resection and long-term survival after esophagectomy in patients with T4 esophageal cancer undergoing induction chemotherapy or chemoradiotherapy. Surg Today. 2015;45(4):479–486. doi: 10.1007/s00595-014-0980-7 [DOI] [PubMed] [Google Scholar]
  • 10.Shimoji H, Karimata H, Nagahama M, Nishimaki T. Induction chemotherapy or chemoradiotherapy followed by radical esophagectomy for T4 esophageal cancer: results of a prospective cohort study. World J Surg. 2013;37(9):2180–2188. doi: 10.1007/s00268-013-2074-x [DOI] [PubMed] [Google Scholar]
  • 11.Bedenne L, Michel P, Bouché O, et al. Chemoradiation followed by surgery compared with chemoradiation alone in squamous cancer of the esophagus: FFCD 9102. J Clin Oncol Off J Am Soc Clin Oncol. 2007;25(10):1160–1168. doi: 10.1200/JCO.2005.04.7118 [DOI] [PubMed] [Google Scholar]
  • 12.Fujita H, Sueyoshi S, Tanaka T, et al. Esophagectomy: Is It Necessary after Chemoradiotherapy for a Locally Advanced T4 Esophageal Cancer? Prospective Nonrandomized Trial Comparing Chemoradiotherapy with Surgery versus without Surgery. World J Surg. 2005;29(1):25–31. doi: 10.1007/s00268-004-7590-2 [DOI] [PubMed] [Google Scholar]
  • 13.Chemoradiotherapy followed by surgery for thoracic esophageal cancer potentially or actually involving adjacent organs | Diseases of the Esophagus | Oxford Academic. https://academic.oup.com/dote/article-abstract/14/3-4/197/2419959. Accessed September 6, 2018. [DOI] [PubMed]
  • 14.National Cancer Database. American College of Surgeons. https://www.facs.org/quality-programs/cancer/ncdb. Accessed August 24, 2018.
  • 15.Bilimoria KY, Stewart AK, Winchester DP, Ko CY. The National Cancer Data Base: a powerful initiative to improve cancer care in the United States. Ann Surg Oncol. 2008;15(3):683–690. doi: 10.1245/s10434-007-9747-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.MatchIt: Nonparametric Preprocessing for Parametric Causal Inference. https://gking.harvard.edu/matchit. Accessed April 18, 2019.
  • 17.Anderson JR. Analysis of Survival by Tumor Response. 1983:10. [DOI] [PubMed] [Google Scholar]
  • 18.Urania Dafni. Landmark Analysis at the 25-Year Landmark Point. Circ Cardiovasc Qual Outcomes. 2011;4(3):363–371. doi: 10.1161/CIRCOUTCOMES.110.957951 [DOI] [PubMed] [Google Scholar]
  • 19.Ikeda K, Ishida K, Sato N, et al. Chemoradiotherapy followed by surgery for thoracic esophageal cancer potentially or actually involving adjacent organs. Dis Esophagus Off J Int Soc Dis Esophagus. 2001;14(3–4):197–201. [DOI] [PubMed] [Google Scholar]
  • 20.Shimada H, Shiratori T, Okazumi S, et al. Have surgical outcomes of pathologic T4 esophageal squamous cell carcinoma really improved? Analysis of 268 cases during 45 years of experience. J Am Coll Surg. 2008;206(1):48–56. doi: 10.1016/j.jamcollsurg.2007.06.013 [DOI] [PubMed] [Google Scholar]

RESOURCES